Massive access or communication, emerging as one of six usage scenarios in 6G, has attracted considerable recent attention due to its potential to empower next-generation industrial cyber-physical systems such as smart grids, factory automation, industrial internet-of-things (IIoT), etc. However, to guarantee its QoS, the associated power control becomes computationally intractable with a huge number of users. In this paper, we present a tractable algorithm for power control in massive access, based on mean-field approximations. In particular, our aim is to maximize the overall throughput in each scheduling period, at the beginning of which each user has a finite number of backlogged bits. To achieve this goal and overcome the curse of dimensionality, a mean-field game (MFG) is formulated. Unfortunately, the formulated MFG is still a non-convex optimization problem. Enlightened by MAPEL, an efficient solver for non-convex power control problem, we leverage multiplicative linear fractional programming (MLFP) to tackle the non-convexity in our formulated MFG. Furthermore, the mean-field approximation assisted power control strategy requires low signaling overhead consumed for estimation and feedback of channel state information (CSI). Simulation results demonstrate that the proposed tractable power control attains substantial performance gains in both the overall throughput and computational complexity.

Massive connectivity supports the sporadic access of a vast number of devices without requiring prior permission from the base station (BS). In such scenarios, the BS must perform joint activity detection and channel estimation (JADCE) prior to data reception. Message passing algorithms have emerged as a prominent solution for JADCE under a Bayesian inference framework. The existing message passing algorithms, however, typically rely on some hand-crafted and overly simplistic priors of the wireless channel, leading to significant channel estimation errors and reduced activity detection accuracy. In this paper, we focus on the problem of JADCE in a multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) grant-free random access network. We propose to incorporate a more accurate channel prior learned by score-based generative models (a.k.a. diffusion models) into message passing, so as to push towards the performance limit of JADCE. Specifically, we develop a novel turbo message passing (TMP) framework that models the entire channel matrix as a super node, rather than factorizing it element-wise. This design enables the seamless integration of score-based generative models as a minimum mean-squared error (MMSE) denoiser. The variance of the denoiser, which is essential in message passing, can also be learned through score-based generative models. Our approach, termed score-based TMP for JADCE (STMP-JADCE), takes full advantages of the powerful generative prior and, meanwhile, benefits from the fast convergence speed of message passing. Numerical simulations show that STMP-JADCE drastically enhances the activity detection and channel estimation performance compared to the state-of-the-art baseline algorithms.

Reinforcement learning (RL) has been used for uncrewed aerial vehicle (UAV)-based jamming detection by selecting hypothesis-test thresholds on received signal strength (RSS) to safeguard large language model (LLM) inference over wireless links. However, existing methods struggle against smart jammers with dynamic jamming probability. In this paper, we propose a UAV-based jamming detection scheme for LLM-enabled wireless networks to detect smart jamming. The proposed scheme augments traditional wireless features with an LLM-inferred communication environment indicator derived from multimodal sensing (e.g., images/videos from mobile devices), improving accuracy beyond RSS alone. Our approach jointly optimizes the test threshold and the UAV’s sniffer channel assignment, even with a limited number of antennas. We achieve this via a RL policy that leverages UAV imagery, e.g., suspicious radio devices and measuring relative distances, to accelerate detection speed. We implement the proposed scheme on a UAV equipped with a universal software radio peripheral to alert on jamming attacks in a wireless network integrated with a 7-billion-parameter vision-language multimodal LLM. Experiments with a UAV at 3 m detecting a smart jammer during image–text LLM inference across three mobile devices and an edge server demonstrate improvements in both detection accuracy and speed.

Extremely large-scale arrays (XL-arrays) have emerged as a key enabler in achieving the unprecedented performance requirements of future wireless networks, leading to a significant increase in the range of the near-field region. This transition necessitates the spherical wavefront model for characterizing the wireless propagation rather than the far-field planar counterpart, thereby introducing extra degrees-of-freedom (DoFs) to wireless system design. In this paper, we explore the beam focusing-based physical layer security (PLS) in the near field, where multiple legitimate users and one eavesdropper are situated in the near-field region of the XL-array base station (BS). First, we consider a special case with one legitimate user and one eavesdropper to shed useful insights into near-field PLS. In particular, it is shown that 1) Artificial noise (AN) is crucial to near-fieldsecurity provisioning, transforming an insecure system to a secure one; 2) AN can yield numeroussecurity gains, which considerably enhances PLS in the near field as compared to the case without AN taken into account. Next, for the general case with multiple legitimate users, we propose an efficient low-complexity approach to design the beamforming with AN to guarantee near-field secure transmission. Specifically, the low-complexity approach is conceived starting by introducing the concept ofinterference domainto capture the inter-user interference level, followed by athree-step identification frameworkfor designing the beamforming. Finally, numerical results reveal that 1) the PLS enhancement in the near field is pronounced thanks to the additional spatial DoFs; 2) the proposed approach can achieve close performance to that of the computationally-extensive conventional method yet with a significantly lower computational complexity.

Edge intelligence (EI) allows resource-constrained edge devices (EDs) to offload computation-intensive AI tasks (e.g., visual object detection) to edge servers (ESs) for fast execution. However, transmitting high-volume raw task data (e.g., 4K video) over bandwidth-limited wireless networks incurs significant latency. While EDs can reduce transmission latency by degrading data before transmission (e.g., reducing resolution from 4K to 720p or 480p), it often deteriorates inference accuracy, creating a critical accuracy-latency tradeoff. The difficulty in balancing this tradeoff stems from the absence of closed-form models capturing content-dependent accuracy-latency relationships. Besides, under bandwidth sharing constraints, the discrete degradation decisions among the EDs demonstrate inherent combinatorial complexity. Mathematically, it requires solving a challengingblack-boxmixed-integer nonlinear programming (MINLP). To address this problem, we propose LAB, a novel learning framework that seamlessly integrates deep reinforcement learning (DRL) and Bayesian optimization (BO). Specifically, LAB employs: (a) a DNN-based actor that maps input system state to degradation actions, directly addressing the combinatorial complexity of the MINLP; and (b) a BO-based critic with an explicit model built from fitting a Gaussian process surrogate with historical observations, enabling model-based evaluation of degradation actions. For each selected action, optimal bandwidth allocation is then efficiently derived via convex optimization. Numerical evaluations on real-world self-driving datasets demonstrate that LAB achieves near-optimal accuracy-latency tradeoff, exhibiting only 1.22% accuracy degradation and 0.07s added latency compared to exhaustive search. Notably, it outperforms conventional DRL with 3.29% higher accuracy and 42.60% lower latency, demonstrating its advantageous performance in handling black-box optimization problems. The complete source code for LAB will be published on GitHub upon acceptance.

Spatial multiplexing in far-field (FF) multiple-input multiple-output (MIMO) systems typically relies on the multipath provided by scatterers. Recent studies have revealed that near-field (NF) MIMO systems can exploit spatial multiplexing even in line-of-sight (LoS) propagation. Through singular value decomposition (SVD), an orthogonal basis consisting of the channel’s right singular vectors can be obtained to formulate a beamforming matrix. This matrix maps multiple data streams into orthogonal sub-channels, enabling spatial multiplexing. However, the SVD-based beamforming matrix, composed of entries with varying magnitudes and phases, generally requires a costly full digital array architecture. To achieve spatial multiplexing with low-complexity beamforming, this paper explores the channel’s structural characteristics of NF-LoS MIMO systems. Specifically, we consider a general point-to-point MIMO setup for a pair of uniform linear arrays (ULAs). We first investigate the ability of the transmitter (Tx) array to resolve paths to different Rx antennas under the NF spherical wave model. Based on the derived resolution, we explicitly formulate a quasi-orthogonal (QO) basis consisting of NF beamfocusing vectors, ensuring that the normalized magnitude of any two QO vectors remains sufficiently small to meet practical application requirements. By appropriately arranging these QO vectors into a beamforming matrix, we transform the spatial channel into a beamspace channel, where the beams serve as QO sub-channels. Furthermore, we utilize these QO beams or sub-channels for beamspace modulation (BM), where the Tx dynamically selects beams to deliver information via beam index modulation and via spatial multiplexing. Importantly, the beamformer designed for selecting QO beams, consisting of constant-magnitude entries, can be implemented using only analog phase shifters. Compared with the existing SVD based BM, the proposed scheme simplifies implementation in hybrid array architectures and achieves high spectrum efficiency with reduced computation complexity. We also show the feasibility of the proposed scheme in practical scenarios, including imperfect channel state information (CSI), finite-resolution phase shifters, mixed LoS and none-line-of-sight (NLoS) channels, and uniform planar arrays (UPAs).

Extremely large-scale antenna arrays (ELAAs) and millimeter wave (mmWave) communication are prominent enablers in advanced 6G networks. To accommodate the unique characteristics introduced by these trends in communication system design, near-field spherical-wave propagation modeling is imperative, departing from the classical far-field planar-wave model. In the millimeter wave band, where wavelengths are small compared to the array aperture, many propagation phenomena resemble those in optics. Inspired by Fresnel diffraction analysis in classical optics, we explore the utilization of the Fresnel Transform for channel modeling, especially in the near-field region. In this paper, we propose a Discrete Fresnel Transform (DFnT)-based dictionary towards effectively characterizing and sparsely representing the channel vector in both near and far-field scenarios. Specifically, we bridge the gap between the continuous Fresnel Transform and the near-field DFnT dictionary by discretizing and parameterizing the transform, adapting it to ELAA codewords. We assess the near-field representation capability of the dictionary by employing it for channel estimation using Orthogonal Matching Pursuit, a compressive channel estimation algorithm whose performance relies heavily on the sparsity of the representation. Simulation results underscore the superiority of the DFnT dictionary over the existing far-field and the Polar dictionaries.

Finite-blocklength coding (FBC) is a promising technology to achieve ultra-reliable and low-latency communications (URLLC) in emerging applications, e.g., autonomous driving and extended reality. The challenge lies in scheduling random arriving traffic in URLLC due to the blocklength constraint imposed by low latency. This paper designs a cross-layer mechanism, called the joint scheduling and FBC policy, to meet URLLC requirements for bursty traffic under AWGN and block fading channels. First, we model a single-user transmission system as a tandem queue model. In this model, a packet queue buffers randomly arriving packets. After these packets are encoded using FBC, the resulting encoded symbols are buffered in a symbol queue. To analyze the latency and reliability performance, we represent the system as a Markov chain and conduct steady-state and transition analyses. After that, we construct a non-convex problem to minimize delay subject to reliability and power constraints. Using a variable combination method, we convert this problem into a linear-fractional programming (LFP) problem. Notice that extremely high reliable requirement significantly increases the computational complexity of standard LFP method.We approximate the objective function and packet drop ratio constraint, and obtain a lower bound of the minimum average delay effectively with a marginal performance loss.

With explosive increase of data at the mobile edge, federated learning (FL) emerges as a promising technique to reduce data transmission costs and privacy leakage risks. Nevertheless, the huge communication overhead for an increasing volume of edge devices still restricts the FL performance. Over-the-air computation (AirComp) is viable for alleviating the communication burden in FL systems. However, there consequently appears a straggler issue restraining the performance of the over-the-air FL (OA-FL) framework, which is even worse especially when devices training a machine learning model are distributed over a relatively large service area. In this paper, we propose an unmanned aerial vehicle (UAV) enabled OA-FL scheme, where the UAV acts as a parameter server (PS) to aggregate the local gradients hierarchically for global model updating. The global aggregation frequency is tunable in the hierarchical aggregation approach, enabling it to balance the resource consumption between communication and learning. Building on this approach, we carry out a gradient-correlation-aware FL performance analysis and jointly optimize the trajectory of UAV-PS, the device selection state, and the aggregation coefficients. An algorithm based on alternating optimization (AO) is developed to solve the formulated problem, where successive convex approximation (SCA) and fractional programming (FP) are utilized for the convexification of the non-convex problem. Numerical simulation results demonstrate the effectiveness of our UAV enabled hierarchical aggregation scheme compared with several existing baselines.

The distortion-perception (DP) tradeoff reveals a fundamental conflict between distortion metrics (e.g., MSE and PSNR) and perceptual quality. Recent research has increasingly concentrated on evaluating denoising algorithms within the DP framework. However, existing algorithms either prioritize perceptual quality by sacrificing acceptable distortion, or focus on minimizing MSE for faithful restoration. When the goal shifts or noisy measurements vary, adapting to different points on the DP plane needs retraining or even re-designing the model. Inspired by recent advances in solving inverse problems using score-based generative models, we explore the potential of flexibly and optimally traversing DP tradeoffs using a single pre-trained score-based model. Specifically, we introduce a variance-scaled reverse diffusion process and theoretically characterize the marginal distribution. We then prove that the proposed sample process is an optimal solution to the DP tradeoff for conditional Gaussian distribution. Experimental results on two-dimensional and image datasets illustrate that a single score network can effectively and flexibly traverse the DP tradeoff for general denoising problems.

In future 6G communication systems, large-scale antenna arrays promise enhanced signal strength and spatial resolution, but they also increase the complexity of beam training. Moreover, as antenna counts grow and carrier wavelengths shrink, the channel model transits from far-field (FF) planar waves to near-field (NF) spherical waves, further complicating the beam training process. This paper focuses on typical millimeter-wave (mmWave) systems equipped with large-scale uniform planar arrays (UPAs), which produce 3D beam patterns and introduce additional challenges for NF beam training. Existing methods primarily rely on either FF steering or NF focusing codewords, both of which are highly sensitive to mismatches in user equipment (UE) position, leading to degraded performance or excessive training overhead. In contrast, we introduce a novel beam training approach leveraging the beam-diverging effect, which enables adjustable wide-beam coverage using only a single radio frequency (RF) chain. Specifically, we first analyze the spatial characteristics of this effect in UPA systems and leverage them to construct hierarchical codebooks for coarse UE localization. Then, we develop a 3D sampling mechanism to build an NF refinement codebook for precise beam training. Numerical results demonstrate that the proposed algorithm achieves superior beam training performance while maintaining low training overhead.

Millimeter wave channels exhibit sparsity under certain basis in the far-field, such as Discrete Fourier Transform (DFT) basis. This property enables the use of compressed sensing (CS) techniques to reconstruct channel vectors with reduced pilot overhead. However, with increasing antenna array sizes and higher frequencies, users more frequently operate in the radiating near-field, where spherical wavefronts dominate. In this regime, the DFT basis no longer yields exact sparsity. Instead, the channel exhibits a block-sparse structure, where non-zero elements tend to cluster in the angular domain. To exploit this structure, we propose a block sparsity-aware CS approach for near-field channel estimation. Specifically, we derive the block sparse signal recovery condition based on block coherence measurements. Based on the condition, we adopt a mixed l2/l1norm-based optimization that extends the conventional l1-based methods by explicitly modeling block sparsity of the near-field channel. Additionally, we design an optimized sensing matrix that outperforms conventional random sensing matrices. This method effectively exploits the block structure of non-zero elements, leading to more accurate channel estimation in both near-field and far-field regions.

In this paper, we consider reusing a rotary-wing UAV as both an airborne base station (BS) and radar to provide integrated sensing and communication (ISAC) wideband service to a ground mobile user. Specifically, the UAV transmits orthogonal frequency-division multiplexing (OFDM) signals where a part of the sub-carriers are assigned for communication purposes. We formulate an online optimization problem that jointly optimizes the UAV trajectory and power allocation of the OFDM sub-carriers to provide a balanced communication and localization service to the ground user. The problem is very challenging because of the non-convex localization accuracy metric with respect to the trajectory and transmit power. For this, we decouple the original problem into a sub-carrier power allocation sub-problem and a trajectory design sub-problem, and propose efficient algorithms to solve them respectively. Simulation results show that the proposed algorithm reduces the localization error by more than 66% at the cost of affordable decrease of communication rate compared to the representative benchmark method considered.

As a key enabling technique for extremely largescale antenna array (ELAA), near-field (NF) beam training aims to estimate the channel conditions across all antenna-user equipment (UE) links, which is inherently complex and therefore challenging. Existing NF beam training methods primarily rely on beam focusing techniques, where numerous codewords that concentrate the signal beam on different points or directions are sequentially deployed as pilots to match the channel conditions, resulting in a trade-off between pilot overhead and beam training accuracy. In this paper, we introduce the concept of the beam diverging effect and leverage it to enable more efficient beam training. First, we propose the mirror codeword and demonstrate that it can trigger beam diverging effect using only one radio frequency (RF) chain. Then, we introduce the mirror polardomain codebook (MPC) along with a hierarchical method that enables rough localization of the UE using only$2 \log _{2} N$pilots. Finally, we present two supplementary approaches to enhance beam training performance: a pilot set expansion method to enable precise beam training and an MPC angular range reduction method to ensure the effectiveness of the proposed algorithm. Numerical results affirm that our proposed algorithm achieves high beam training accuracy while using relatively low overhead.

Asynchronous federated learning (AFL) has emerged as a promising solution to address system heterogeneity and improve the training efficiency of federated learning. However, existing AFL methods face two critical limitations: 1) they rely on strong assumptions about bounded data heterogeneity across clients, and 2) they require meticulous tuning of learning rates based on unknown system parameters. In this paper, we tackle these challenges by leveraging momentum-based optimization and adaptive learning strategies. We first propose MasFL, a novel momentum-driven AFL framework that successfully eliminates the need for data heterogeneity bounds by effectively utilizing historical descent directions across clients and iterations. By mitigating the staleness accumulation caused by asynchronous updates, we prove that MasFL achieves state-of- the-art convergence rates with linear speedup in both the number of participating clients and local updates. Building on this foundation, we further introduce AdaMasFL, an adaptive variant that incorporates gradient normalization into local updates. Remarkably, this integration removes all dependencies on problem-specific parameters, yielding a fully tuning-free AFL approach while retaining theoretical guarantees. Extensive experiments demonstrate that AdaMasFL consistently outperforms state-of-the-art AFL methods in run- time efficiency and exhibits exceptional robustness across diverse learning rate configurations and system conditions.

Decentralized bilevel optimization has garnered significant attention due to its critical role in solving large-scale machine learning problems. However, existing methods often rely on prior knowledge of problem parameters—such as smoothness, convexity, or communication network topologies—to determine appropriate stepsizes. In practice, these problem parameters are typically unavailable, leading to substantial manual effort for hyperparameter tuning. In this paper, we propose \textbf{AdaSDBO}, a fully problem-parameter-free algorithm for decentralized bilevel optimization with a single-loop structure. AdaSDBO leverages adaptive stepsizes based on cumulative gradient norms to update all variables simultaneously, dynamically adjusting its progress and eliminating the need for problem-specific hyperparameter tuning. Through rigorous theoretical analysis, we establish that AdaSDBO achieves a convergence rate of $\widetilde{\mathcal{O}}\left(\frac{1}{T}\right)$, matching the performance of well-tuned state-of-the-art methods up to polylogarithmic factors. Extensive numerical experiments demonstrate that AdaSDBO delivers competitive performance compared to existing decentralized bilevel optimization methods while exhibiting remarkable robustness across diverse stepsize configurations.

This paper addresses the problem of end-to-end (E2E) design of learning and communication in a task-oriented semantic communication system. In particular, we consider a multi-device cooperative edge inference system over a wireless multiple-input multiple-output (MIMO) multiple access channel, where multiple devices transmit extracted features to a server to perform a classification task. We formulate the E2E design of feature encoding, MIMO precoding, and classification as a conditional mutual information maximization problem. However, it is notoriously difficult to design and train an E2E network that can be adaptive to both the task dataset and different channel realizations. Regarding network training, we propose a decoupled pretraining framework that separately trains the feature encoder and the MIMO precoder, with a maximuma posteriori(MAP) classifier employed at the server to generate the inference result. The feature encoder is pretrained exclusively using the task dataset, while the MIMO precoder is pretrained solely based on the channel and noise distributions. Nevertheless, we manage to align the pretraining objectives of each individual component with the E2E learning objective, so as to approach the performance bound of E2E learning. By leveraging the decoupled pretraining results for initialization, the E2E learning can be conducted with minimal training overhead. Regarding network architecture design, we develop two deep unfolded precoding networks that effectively incorporate the domain knowledge of the solution to the decoupled precoding problem. Simulation results on both the CIFAR-10 and ModelNet10 datasets verify that the proposed method achieves significantly higher classification accuracy compared to various baselines.

By extracting task-relevant information while maximally compressing the input, the information bottleneck (IB) principle has provided a guideline for learning effective and robust representations of the target inference. However, extending the idea to the multi-task learning scenario with joint consideration of generative tasks and traditional reconstruction tasks remains unexplored. This paper addresses this gap by reconsidering the lossy compression problem with diverse constraints on data reconstruction, perceptual quality, and classification accuracy. Firstly, we study two ternary relationships, namely, therate-distortion-classification (RDC)andrate-perception-classification (RPC). For both RDC and RPC functions, we derive the closed-form expressions of the optimal rate for binary and Gaussian sources. These new results complement the IB principle and provide insights into effectively extracting task-oriented information to fulfill diverse objectives. Secondly, unlike prior research demonstrating a tradeoff between classification and perception in signal restoration problems, we prove that such a tradeoff does not exist in the RPC function and reveal that the source noise plays a decisive role in the classification-perception tradeoff. Finally, we implement a deep-learning-based image compression framework, incorporating multiple tasks related to distortion, perception, and classification. The experimental results coincide with the theoretical analysis and verify the effectiveness of our generalized IB in balancing various task objectives.

This paper addresses the challenge of achieving high average User-Perceived Throughput (UPT) in multi-user downlink transmissions, which is essential for immersive experience in emerging 6G applications like extended reality and digital twins. Achieving high UPT involves effectively managing burst data traffic over fluctuating wireless channels, a topic not extensively explored in current literature. We approach this by modeling the system as a Markov Decision Process (MDP), focusing on maximizing average UPT by minimizing the average time users have non-empty queues while maintaining system stability. We introduce a multi-user scheduling scheme for a system with a single Resource Block (RB). Specifically, the scheduling metric is a fluid approximation of the minimum average time with non-empty queues, which is obtained by analytically solving a convex optimization problem. We next extend our scheduling scheme to handle multiple spectrum RBs through successive allocation of available RBs, achieving polynomial complexity. Using a quadratic Lyapunov drift method, we ensure scheduler stability. The scheduler is also modified to operate online, adapting to time-varying channel conditions. Simulation results demonstrate that our scheduler not only achieves near-optimal performance but also significantly outperforms existing benchmarks.

Reconfigurable intelligent surfaces (RISs) have shown the potential to improve signal-to-interference-plus-noise ratio (SINR) related coverage, especially at high-frequency communications. However, assessing electromagnetic field exposure (EMFE) and establishing EMFE regulations in RIS-assisted large-scale networks remain open issues. This paper proposes a stochastic geometry (SG) based framework to characterize SINR and EMFE in such networks for downlink and uplink scenarios. Particularly, we carefully consider the association rule with the presence of RISs, accurate antenna pattern at base stations (BSs), fading model, and power control mechanism at mobile devices in the system model. Under the proposed framework, we derive the marginal and joint distributions of SINR and EMFE in downlink and uplink, respectively. The first moment of EMFE is also provided. Additionally, we design the compliance distance (CD) between a BS/RIS and a user to comply with the EMFE regulations. To facilitate efficient identification, we further provide approximate closed-form expressions for CDs. From numerical results of the marginal distributions, we find that in the downlink scenario, deploying RISs may not always be beneficial, as the improved SINR comes at the cost of increased EMFE. However, in the uplink scenario, RIS deployment is promising to enhance coverage while still maintaining EMFE compliance. By simultaneously evaluating coverage and compliance metrics through joint distributions, we demonstrate the feasibility of RISs in improving uplink and downlink performance. Insights from this framework can contribute to establishing EMFE guidelines and achieving a balance between coverage and compliance when deploying RISs.

In millimeter wave communication systems, achieving high-quality data transmission demands efficient and rapid beam alignment. Conventional deep learning-based methods, although promising, often rely on the assumption that training and testing channels share identical distribution. This assumption may not hold in practical settings, potentially leading to significant performance degradation when the deployment environment changes. To address this issue, we introduce SAMBA, a novel meta-learning-based approach for adaptive beam alignment without requiring Channel State Information (CSI). SAMBA enables swift adaptation to unknown scenarios using a minimal set of newly labeled data. Specifically, we adopt a probing beam-search strategy to obviate the need for CSI. Furthermore, we employ Model-Agnostic Meta-Learning (MAML) for parameter pre-training and fine-tuning to enhance our model’s adaptability. Confronted with the challenge of numerous beam candidates in the narrow beam selection problem, which complicates the straightforward replication of MAML, we develop a novel training task generation strategy. In our experimental assessments, we subjected SAMBA to a wide range of challenging scenarios using ray-tracing simulations. These scenarios encompassed various frequency bands, distinct base station layouts, and outdoor-to-indoor transitions. Our results demonstrate that SAMBA consistently outperforms learning-based baseline models, showcasing its superior domain adaptation capabilities in dynamic and diverse channel settings.

In this paper, we consider a task-oriented communication system for multi-device edge inference over a multiple-input multiple-output (MIMO) multiple-access channel, where the learning (feature encoding and classification) and communication (precoding) modules are designed to achieve the same goal of inference accuracy maximization. End-to-end learning of these modules involves both the task dataset and the time-varying wireless channel, which may result in unaffordable training overhead and complexity. Instead, we advocate a modular design of learning and communication to achieve the consistent goal. Specifically, we leverage the maximal coding rate reduction (MCR2) objective as a surrogate to represent the inference accuracy, which allows us to explicitly formulate the precoding optimization problem separated from the learning design. We develop a block coordinate ascent (BCA) algorithm for efficient problem-solving. Moreover, the MCR2objective also serves the loss function of the feature encoding network, thus unifying the objectives of learning and communication. Simulation results demonstrate the superior performance of the proposed method compared to various baselines. As such, our work paves the way for further exploration into the synergistic alignment of learning and communication objectives in task-oriented communication systems. Code is available at https://github.com/chang-cai/TaskCommMCR2.

Balancing diverse task objectives under limited rate is crucial for developing robust multitask deep learning (DL) models and improving performance across various domains. In this paper, we consider the lossy compression problem with human-centric and task-oriented metrics, such as perceptual quality and classification accuracy. We investigate two ternary relationships, namely, the rate-distortion-classification (RDC) and rate-perception-classification (RPC). For both RDC and RPC functions, we derive the closed-form expressions of the optimal rate for both binary and Gaussian sources. Notably, both RDC and RPC relationships exhibit distinct characteristics compared to the previous RDP tradeoff proposed by Blau et al. Then, we conduct experiments by implementing a DL-based image compression framework, incorporating rate, distortion, perception, and classification constraints. The experimental results verify the theoretical characteristics of RDC and RPC tradeoffs, providing information-theoretical insights into the design of loss functions to balance diverse task objectives in deep learning.

The rapid proliferation of artificial intelligence (AI) services gives rise to the development of federated learning (FL), enabling the cooperative learning among wireless devices (WDs) with only local model parameters communicated. Nevertheless, the current emergence of large AI models renders the existing FL approaches inefficient, due to the huge communication overhead. In this paper, we propose a novel over-the-air federated distillation (FD) framework by synergizing the strength of FL and knowledge distillation to avoid the heavy local model transmission. Instead of sharing model parameters, only WDs' model outputs, referred to as knowledge, are shared and aggregated over-the-air by exploiting the superposition property of the multiple-access channel. Accordingly, we study the communication-learning co-design in over-the-air FD, aiming to maximize the learning convergence rate while meeting the power constraints of the transceivers. The main challenge lies in the intractability of the learning performance analysis, as well as the non-convex nature and the optimization spanning the whole FD training period. To tackle this problem, we propose an efficient algorithm to jointly optimize the transmit power of the WDs, estimator for over-the-air aggregation, and receiver beamforming per training round. Numerical results demonstrate that the proposed over-the-air FD achieves significant communication overhead reduction, with only a slight compensation of testing accuracy compared to conventional FL benchmarks.

Deploying reconfigurable intelligent surfaces (RISs) offers the potential to improve coverage performance in mil-limeter wave (mmWave) communications. However, electric and magnetic field (EMF) exposure related to base station (BS) trans-mission and RIS reflection is still unclear. This paper provides an analytical framework to evaluate the EMF exposure in RIS-assisted mmWave cellular networks. The proposed framework provides insights to establish technical guidelines for ensuring EMF exposure within a safe limit. For example, by considering the compliance distance (CD) set for BSs, we explore the necessity of designing a similar CD for RISs.

This article considers deploying a dual-functional unmanned aerial vehicle (UAV) as both an aerial data collector and aerial anchor node (AN) to assist the ground base stations in providing integrated localization and communication (ILAC) service in urban areas. A major challenge to the urban ILAC service quality lies in the severe blockage of ground-to-air links by densely located buildings. To improve the service quality, we leverage the recent advance in urban physical environment map (PEM), also known as the three-dimensional (3-D) city map, to aid in optimizing the 3-D deployment of UAV. This allows a UAV to avoid blockages and establish strong acrlong LoS links to all target ground users. We propose a PEM-aided ILAC service model and formulate a UAV 3-D deployment optimization problem. The aim is to maximize the sum communication rate of ground users while satisfying individual localization accuracy and communication rate constraints. The problem is very challenging to solve mainly because the localization accuracy and blockage-avoiding constraints are both nonconvex with respect to UAV position. To tackle the problem, we first adopt a new localization accuracy metric and subsequently derive a convex expression of the localization constraint. Then, we convert the blockage-avoiding constraints into an equivalent and analytically tractable form and propose an efficient iterative algorithm to solve the UAV deployment optimization problem. Simulation results show that the proposed method achieves close-to-optimal performance under dense urban blockage setups, while significantly reducing the computational complexity.

Due to its agility, reusability, and programmability, the unmanned aerial vehicle (UAV) can be utilized as a flying base station in mobile edge computing (MEC) systems, providing cost-effective computation services to distributed ground devices in the absence of terrestrial infrastructure. A defect of traditional UAVs is that they rely heavily on the onboard limited battery for the power supply, severely restricting UAVs’ operating endurance and flying range. To tackle this problem, we consider using a solar-powered UAV as the edge server for sensing data collection and processing. However, owing to the atmospheric absorption, the amount of harvested solar energy increases with the flying altitude, resulting in a non-trivial tradeoff between energy harvesting and communication performance. In addition, the dynamics of the moving clouds also make energy harvesting exhibit stochastic variations in the solar panel’s output, rendering the instability of the energy conversion. In this paper, given the randomness of energy and data arrivals, we propose a Lyapunov-based method to maximize the long-term system throughput, subject to the time average constraints on the solar power supply, the data queue stability, and the energy consumption of the devices. Specifically, without knowing the future system knowledge, we formulate the problem as a multi-stage online stochastic optimization and decompose the original problem into per-slot deterministic optimization problems. In each slot, we iteratively optimize the data sensing rate, the computation offloading, the communication resource allocation, and the 3D trajectory of the UAV. The proposed algorithm can adaptively adjust the UAV’s altitude according to its residual energy, thus striking a balance between energy harvesting and system throughput. Furthermore, it has low complexity which makes it suitable for online implementation. Extensive simulations have demonstrated the effectiveness of the algorithm, in that, it significantly outperforms the benchmark schemes in the system throughput, while satisfying the prescribed time average constraints at the same time.

Artificial intelligence (AI) technologies have emerged as pivotal enablers across a multitude of industries, including consumer electronics, healthcare, and manufacturing, largely due to their significant resurgence over the past decade. The transformative power of AI is primarily derived from the utilization of deep neural networks (DNNs), which require extensive data for training and substantial computational resources for processing. Consequently, DNN models are typically trained and deployed on resource-rich cloud servers. However, due to potential latency issues associated with cloud communications, deep learning (DL) workflows (e.g., DNN training and inference) are increasingly being transitioned to wireless edge networks in proximity to end-user devices (EUDs). This shift is designed to support latency-sensitive applications and has given rise to a new paradigm of edge AI, which will play a critical role in upcoming sixth-generation (6G) networks to support ubiquitous AI applications. Despite its considerable potential, edge AI faces substantial challenges, mostly due to the dichotomy between the resource limitations of wireless edge networks and the resource-intensive nature of DL. Specifically, the acquisition of large-scale data, as well as the training and inference processes of DNNs, can rapidly deplete the battery energy of EUDs. This necessitates an energy-conscious approach to edge AI to ensure both optimal and sustainable performance. In this article, we present a contemporary survey on green edge AI. We commence by analyzing the principal energy consumption components of edge AI systems to identify the fundamental design principles of green edge AI. Guided by these principles, we then explore energy-efficient design methodologies for the three critical tasks in edge AI systems, including training data acquisition, edge training, and edge inference. Finally, we underscore potential future research directions to further enhance the energy efficiency (EE) of edge AI.

Millimeter-wave (mmWave) is highly susceptible to obstacles and requires significant directivity of beams. To address these challenges, a promising solution is to deploy antenna arrays at base stations (BSs) and reconfigurable intelligent surfaces (RISs). Antenna arrays enable 3D beamforming and provide highly directional beams, while RISs create favorable transmission environments. In this paper, we propose an analytical framework to quantify the coverage performance of the RIS-assisted mmWave cellular network with 3D beamforming. Modeling obstacles and RISs with a line Boolean model and BSs with a Poisson point process (PPP), we provide the distance distribution between a UE and its associated RIS by evaluating the impact of the RIS orientation. Moreover, we adopt a 3D sectorized flat-top antenna model to characterize the interference introduced by dynamic directional beams (i.e., mainlobe/sidelobe). Furthermore, we derive the signal-to-interference ratio (SIR) coverage probability, where an equivalent PPP is proposed to enhance the computation efficiency. Numerical results validate the accuracy of our analysis and show that the proposed network achieves significant coverage improvement. We also investigate the impacts of the key parameters on the coverage probability, providing useful insights for deploying RISs and antenna arrays in mmWave cellular networks.

In this paper, we investigate the uplink channel estimation problem in intelligent reflecting surface (IRS)-assisted broadband millimeter wave multi-input multi-output communication systems. Considering the channel sparsity in both the angle and delay domains, we decouple the channel estimation problem into several sub-problems, namely, sequential estimations of the angles at the user, the base station, and the IRS, alongside the propagation path delays. By exploiting the Vandermonde structure of both the array manifold and the phase difference among multiple subcarriers caused by delays, we propose a multi-stage atomic norm minimization-based (MS-ANM) channel estimation algorithm. In particular, we take full advantage of the sharing of angle information among all subcarriers in the broadband channel to enhance the estimation accuracy. To reduce the computational complexity, we further propose a multi-stage orthogonal matching pursuit-based (MS-OMP) algorithm. Simulation results show that the MS-ANM algorithm significantly outperforms the benchmark algorithms, and the MS-OMP algorithm strikes a good balance between performance and computational complexity. By fully exploring the channel’s sparsity in the angle and delay domains along with the Vandermonde structure, both proposed algorithms remarkably reduce the training overhead and thus greatly improve the spectral efficiency over the benchmark algorithms.

Integrated Sensing and Communications (ISAC) is an emerging 6G technology to address the increasing demands for ubiquitous sensing and communication. Achieving Ultra-Reliable and Low-Latency Communication (URLLC) with ISAC has been relatively under-investigated in the existing literature. In this paper, we investigate joint beamforming and scheduling for an ISAC-enabled system to fulfill URLLC requirements. We focus on a typical URLLC scenario where periodic and aperiodic traffic flows coexist. Specifically, the aperiodic traffic is triggered by sensing the stochastic environment. We propose a joint beamforming and scheduling scheme to sense the environment and transmit traffic simultaneously, where efficient scheduling is obtained by exploiting sensing results. To analyze the latency and reliability performance of the proposed scheme, we first describe the system as a Partially Observable Markov Decision Process (POMDP). Then, we analytically express the latency and reliability performance as the probability of successful transmission within the latency constraint. Based on this, we formulate an optimal performance tradeoff for the periodic and aperiodic traffic and obtain an optimal tradeoff by a Dynamic Programming (DP)-based algorithm. Furthermore, we reveal some insightful properties of the optimal tradeoff and the optimal policy. Simulation results show that the optimal tradeoff outperforms other tradeoffs of benchmark algorithms.

Extended reality (XR) is an emerging 6G application with unique traffic characteristics and requirements, calling for innovative Ultra-Reliable and Low-Latency Communication (URLLC) technologies. This paper investigates multi-user scheduling to meet XR services’ hard-latency constraints. Specifically, we focus on a periodical traffic model, where the latency constraint for transmitting each XR frame is less than the inter-arrival time. We describe the system as a periodic Markov Decision Process (MDP) with the performance metric being the probability of successful transmission within the latency constraint. We then obtain the maximum success probability and the optimal scheduling based on the optimal value function. In the case of homogeneous arrivals, we construct a lower bound of the optimal value function. Based on this, we propose an online multi-user scheduling policy that determines scheduling decisions by solving a series of nonlinear Knapsack Problems (KPs) in polynomial time. Our analysis demonstrates that the scheduling scheme is asymptotically optimal with increasing users. Furthermore, we extend the online scheduling scheme to heterogeneous arrivals and present extensions for practical scenarios with multiple resource blocks, quasi-periodical arrivals, random frame sizes, and time-correlated channel fading. Finally, simulation results show that the proposed scheduler achieves near-optimal performance and outperforms other benchmark schedulers.

In task-oriented communications, most existing work designed the physical-layer communication modules and learning based codecs with distinct objectives: learning is targeted at accurate execution of specific tasks, while communication aims at optimizing conventional communication metrics, such as throughput maximization, delay minimization, or bit error rate minimization. The inconsistency between the design objectives may hinder the exploitation of the full benefits of task-oriented communications. In this paper, we consider a task-oriented multi-device edge inference system over a multiple-input multiple-output (MIMO) multiple-access channel, where the learning (i.e., feature encoding and classification) and communication (i.e., precoding) modules are designed with the same goal of inference accuracy maximization. Instead of end-to-end learning which involves both the task dataset and wireless channel during training, we advocate a separate design of learning and communication to achieve the consistent goal. Specifically, we leverage the maximal coding rate reduction (MCR2) objective as a surrogate to represent the inference accuracy, which allows us to explicitly formulate the precoding optimization problem. We cast valuable insights into this formulation and develop a block coordinate ascent (BCA) algorithm for efficient problem-solving. Moreover, the MCR2 objective serves the loss function for feature encoding and guides the classification design. Simulation results on the synthetic features explain the mechanism of MCR2 precoding at different SNRs. We also validate on the CIFAR-10 and ModelNet10 datasets that the proposed design achieves a better latency-accuracy tradeoff compared to various baselines with inconsistent learning-communication objectives. As such, our work paves the way for further exploration into the synergistic alignment of learning and communication objectives in task-oriented communication systems.

Recent years have witnessed the development of federated learning (FL) that allows wireless devices (WDs) to collaboratively learn a global model under the coordination of a parameter server without sharing their local datasets. To meet the communication efficiency and privacy requirements, over-the-air computation and differential privacy (DP) have been incorporated in FL by leveraging the signal-superposition property of multiple-access channels and using artificial noises to perturb local model updates, thereby preserving DP. In this paper, we propose an exploration into device sampling with replacement as a potential mechanism for augmenting the DP levels of WDs in over-the-air FL. In particular, we delve into the joint optimization of device sampling strategy, the number of training rounds, and over-the-air transceiver design. Our goal is to maximize the learning performance while ensuring each WD meets the DP requirement. The problem is challenging due to the intractable FL convergence rate and privacy losses under random sampling, coupled with the strong interconnection among mixed continuous and integer decision variables. To tackle this problem, we first analyze the learning convergence rate and privacy losses of WDs. The analysis allows us to derive the optimal transceiver design per round in closed forms. Then, we propose an efficient alternating optimization algorithm by deriving the optimal device sampling strategy and the number of training rounds in semi-closed forms. Our numerical results, based on real-world learning tasks, showcase the effectiveness of our proposed approach compared with representative baselines.

Federated learning (FL) enables edge devices to collaboratively train machine learning models, with model communication replacing direct data uploading. While over-the-air model aggregation improves communication efficiency, uploading models to an edge server over wireless networks can pose privacy risks. Differential privacy (DP) is a widely used quantitative technique to measure statistical data privacy in FL. Previous research has focused on over-the-air FL with a single-antenna server, leveraging communication noise to enhance user-level DP. This approach achieves the so-called "free DP" by controlling transmit power rather than introducing additional DP-preserving mechanisms at devices, such as adding artificial noise. In this paper, we study differentially private over-the-air FL over a multiple-input multiple-output (MIMO) fading channel. We show that FL model communication with a multiple-antenna server amplifies privacy leakage when the multiple-antenna server employs separate receive combining for model aggregation and information inference. Consequently, relying solely on communication noise, as done in the multiple-input single-output system, cannot meet high privacy requirements, and a device-side privacy-preserving mechanism is necessary for optimal DP design. We analyze the learning convergence and privacy loss of the studied FL system and propose a transceiver design algorithm based on alternating optimization. Numerical results demonstrate that the proposed method achieves a better privacy-learning trade-off compared to prior work.

In future B5G/6G broadband communication systems, non-linear signal distortion caused by the impairment of transmit power amplifier (PA) can severely degrade the communication performance, especially when uplink users share the wireless medium using non-orthogonal multiple access (NOMA) schemes. This is because the successive interference cancellation (SIC) decoding technique, used in NOMA, is incapable of eliminating the interference caused by PA distortion. Consequently, each user’s decoding process suffers from the cumulative distortion noise of all uplink users. In this paper, we establish a new and tractable DPD-PA distortion signal model based on real-world measurements, where the distortion noise power is a polynomial function of PA transmit power diverging from the oversimplified linear function commonly employed in existing studies. Applying the proposed signal model, we characterize the capacity rate region of multi-user uplink NOMA by optimizing the user transmit power. Our findings reveal a significant contraction in the capacity region of NOMA, attributable to polynomial distortion noise power. For practical engineering applications, we formulate a general weighted sum rate maximization (WSRMax) problem under individual user rate constraints. We further propose an efficient power control algorithm to attain the optimal performance. Numerical results show that the optimal power control policy under the proposed non-linear PA model achieves on average 13% higher throughput compared to the policies assuming an ideal linear PA model. Overall, our findings demonstrate the importance of accurate PA distortion modeling to the performance of NOMA and provide efficient optimal power control method accordingly.

Recent years have witnessed the development of federated learning (FL) that allows wireless devices (WDs) to collaboratively learn a global model under the coordination of a parameter server without sharing local datasets. To meet the communication efficiency and privacy requirements, over-the-air computation and differential privacy (DP) are further incorporated in FL by leveraging the signal-superposition property of multiple-access channels, as well as artificial noises to perturb local model updates for DP preservation. In this paper, we consider the device sampling with replacement, as an amplifier for the DP levels of WDs, in differentially private over-the-air FL. Accordingly, we study the joint optimization of device sampling strategy and over-the-air transceiver design that maximizes the learning performance while satisfying the DP requirement of each WD. The problem is challenging due to the intractable FL convergence rate and privacy losses under the sampling randomness, and the strong coupling among mixed decision variables. To tackle this problem, we first derive the analytical learning convergence rate and privacy losses of WDs, based on which the optimal transceiver design and device sampling strategy are obtained in closed forms. Numerical results demonstrate the effectiveness of our proposed approach compared with representative baselines.

Federated learning (FL) allows edge devices to collaboratively train machine learning models without directly sharing data. While over-the-air model aggregation improves communication efficiency, model uploading can lead to privacy risks. Previous research has focused on over-the-air FL with a single-antenna server, leveraging communication noise to enhance user-level privacy. This method achieves the so-called “free” privacy by decreasing transmit power instead of introducing additional privacy-preserving mechanisms at the devices. In this paper, we analyze the privacy leakage of over-the-air FL over a multiple-input multiple-output (MIMO) fading channel. We show that FL model aggregation with a multiple-antenna server amplifies privacy leakage. Consequently, relying solely on communication noise is inefficient to meet high privacy requirements, particularly when the receive antenna array is large. This calls for a joint optimization algorithm for the device-side privacy-preserving mechanism and the receiving protocol to achieve a better privacy-learning tradeoff. Numerical results validate our analysis and highlight the impact of the transmit power and the receive antenna array size on the privacy leakage.

Achieving high reliability and low latency is a critical challenge for a wide range of applications that demand strict performance guarantees, such as real-time systems, industrial automation, and autonomous vehicles. Our primary focus is on ultra-reliable low-latency communication (URLLC), which aims to ensure a real-time requirement. We propose a solution for hard delay-constrained communication, which meets strict latency requirements by incorporating variable-length coding in short-packet transmission systems. Our approach utilizes a cross-layer design using truncated channel inversion transmission across parallel channels. This system can be characterized as a two-dimensional Markov chain, which consists of both the packet queue buffering bits to be encoded and symbol queue buffering coded symbols to be transmitted. By leveraging embedded Markov chains, we formulate an optimization problem to minimize the average power consumption while converting the problem into a one-dimensional Markov chain. We present a heuristic algorithm to obtain hard delay-constrained policies and utilize gradient descent policy to refine the policies and explore the trade-offs between hard delay constraints and power consumption.

In the forthcoming 6G era, Extended reality (XR) is an emerging application with unique traffic characteristics requirements, calling for innovative Ultra-Reliable and Low-Latency Communication (URLLC) technologies. In this paper, we investigate multi-user scheduling to meet hard-latency constraints for XR services. Specifically, we focus on a periodical XR traffic model, where the latency constraint for transmitting each XR frame is less than the inter-arrival time. To find an optimal multi-user scheduling scheme, we first describe the system as a periodic Markov Decision Process (MDP), where the scheduling performance is expressed as the probability of successful transmission within the latency constraint. Then, we obtain the maximum success probability and the optimal scheduling based on the optimal value function. Inspired by the properties of the optimal value function, we construct a lower bound of it and propose an online multi-user scheduling scheme. In particular, scheduling decisions under the proposed scheme are determined by solving a series of nonlinear Knapsack Problem (KP) in polynomial time. Finally, simulation results show that the proposed scheduler achieves nearly optimal performance and outperforms other benchmark schedulers.

In future B5G/6G wideband communication systems, non-linear signal distortion caused by the impairment of transmit power amplifier (PA) can severely degrade the communication performance. The performance impact is especially significant when uplink users share the wireless medium using Non-orthogonal Multiple Access (NOMA) scheme. This is because the successive interference cancellation (SIC) information decoding technique of NOMA cannot eliminate the interference caused by the PA non-linear distortion, such that the decoding of each user will suffer from the aggregate distortion noise of all the uplink users. In this paper, we study the impact of PA non-linear distortion on the performance of uplink NOMA. In particular, we first establish a new PA distortion signal model based on real-world measurements, where the distortion noise power is a polynomial function of PA transmit power, instead of a simplified linear function in most existing studies. Under the proposed signal model, we then accurately characterize the capacity region of a two-user uplink NOMA by optimizing the user transmit power. We show that the polynomial distortion noise power significantly shrinks the achievable capacity region of NOMA. This indicates that existing studies may have overestimated the communication performance of NOMA in practical wideband systems. Besides, the non-linear noise power also leads to a rather different optimal power allocation strategy to attain maximum throughput. Simulation results show that, for a PA following the polynomial distortion noise power model, the proposed optimal power allocation method achieves on average 12.2% higher sum throughput than that obtained from ideal PA model. Overall, our results demonstrate the importance of accurate PA distortion modeling to the performance of NOMA and provide an efficient power allocation method to attain the optimal performance.

Differentially private federated learning enables clients with privacy concerns to collaboratively train a model while preserving their privacy. Clients' locally available data and maximum tolerable privacy budgets will affect their contributions to the training performance. To date, existing studies focus on homogeneous privacy budgets and thus there lack systematic studies regarding the impact of clients' diverse privacy budgets (privacy heterogeneity). This paper represents the first step toward filling this gap. Through rigorous convergence analysis, we illustrate that the influence of privacy protection on training loss is affected by client selection probabilities. In addition, client selection and privacy protection together induce a non-vanishing training error in federated learning. Our analysis then shows that the non-vanishing training error is a convex function of client selection probabilities and thus allows us to formulate privacy-aware client selection as a convex optimization problem. Numerical results demonstrate that the privacy-aware client selection strategy can significantly improve learning performance. For example, compared with unbiased selection, the privacy-aware client selection strategy decreases the test loss by up to 63% on MNIST convolutional neural network classifier.

Due to the wide and license-free bandwidth, visible light communication (VLC) functions as a potential technology to meet the exponentially expanding traffic demands in wireless communications. However, the sensitivity to obstacles and high-path loss are the key issues that practical VLC systems must carefully deal with. In this article, the utilization of optical intelligent reflecting surface (OIRS) array in VLC is considered to create additional light propagation paths, thereby achieving a remarkable performance gain. In the OIRS-assisted VLC system, though the power of the signal at receiver can be increased, the resource allocation is relatively complex. Besides, the OIRS also causes time delays among signals received via various propagation paths, which is usually overlooked in existing works. To overcome these issues, the OIRS-assisted VLC system is interpreted as an autoencoder (AE), named OIRS-AE, whose architecture is enhanced according to both the model-driven and data-driven perspectives. By this way, the processing modules at the transmitter, OIRS, and receiver, including the corresponding encoding, resource management, and decoding schemes, can be simultaneously optimized, which is expected to achieve more reliable communication. Moreover, the impact of the OIRS-induced time delay spread on system performance is explored under various situations. The simulation results show that the proposed OIRS-AE can outperform the traditional OIRS-assisted VLC systems in terms of bit error rate performance.

In next-generation wireless networks, reconfigurable intelligent surface (RIS)-assisted multiple-input multiple-output (MIMO) systems are foreseeable to support a large number of antennas at the transceiver as well as a large number of reflecting elements at the RIS. To fully unleash the potential of RIS, the phase shifts of RIS elements should be carefully designed, resulting in a high-dimensional non-convex optimization problem that is hard to solve with affordable computational complexity. In this paper, we address this scalability issue by partitioning RIS into sub-surfaces, so as to optimize the phase shifts in sub-surface levels to reduce complexity. Specifically, each sub-surface employs a linear phase variation structure to anomalously reflect the incident signal to a desired direction, and the sizes of sub-surfaces can be adaptively adjusted according to channel conditions. We formulate the achievable rate maximization problem by jointly optimizing the transmit covariance matrix and the RIS phase shifts. Under the RIS partitioning framework, the RIS phase shifts optimization reduces to the manipulation of the sub-surface sizes, the phase gradients of sub-surfaces, as well as the common phase shifts of sub-surfaces. Then, we characterize the asymptotic behavior of the system with an infinitely large number of transceiver antennas and RIS elements. The asymptotic analysis provides useful insights on the understanding of the fundamental performance-complexity tradeoff in RIS partitioning design. We show that in the asymptotic domain, the achievable rate maximization problem has a rather simple form with an explicit physical meaning of optimization variables. We develop an efficient algorithm to find an approximately optimal solution to the asymptotic problem via a one-dimensional (1D) grid search. Moreover, we discuss the insights and impacts of the asymptotic result on finite-size system design. By applying the asymptotic result to a finite-size system with necessary modifications, we show by numerical results that the proposed design achieves a favorable tradeoff between system performance and computational complexity.

Future wireless networks are expected to support diverse mobile services, including artificial intelligence (AI) services and ubiquitous data transmissions. Federated learning (FL), as a revolutionary learning approach, enables collaborative AI model training across distributed mobile edge devices. By exploiting the superposition property of multiple-access channels, over-the-air computation allows concurrent model uploading from massive devices over the same radio resources, and thus significantly reduces the communication cost of FL. In this paper, we study the coexistence of over-the-air FL and traditional information transfer (IT) in a mobile edge network, where an access point (AP) coordinates a set of devices for over-the-air FL and serves multiple devices for information transfer in the meantime. We propose a coexisting federated learning and information transfer (CFLIT) communication framework, where the FL and IT devices share the wireless spectrum in an orthogonal frequency division multiplexing (OFDM) system. Under this framework, we aim to maximize the IT data rate and guarantee a given FL convergence performance by optimizing the long-term radio resource allocation. A key challenge that limits the spectrum efficiency of the coexisting system lies in the large overhead incurred by frequent communication between the server and edge devices for FL model aggregation. To address the challenge, we rigorously analyze the impact of the computation-to-communication ratio on the convergence of over-the-air FL in wireless fading channels. The analysis reveals the existence of an optimal computation-to-communication ratio that minimizes the amount of radio resources needed for over-the-air FL to converge to a given error tolerance. Based on the analysis, we propose a low-complexity online algorithm to jointly optimize the radio resource allocation for both the FL devices and IT devices. We further derive an analytical expression of the achievable data rate of IT users. Extensive numerical simulations verify the superior performance of the proposed design for the coexistence of FL and IT devices in wireless cellular systems.

With growing popularity of enormous public safety and transportation infrastructure cameras, there are increasing demands for automatic mobile video analytics. The emerging multi-access edge computing (MEC) technology has been recently applied to improve the accuracy-latency tradeoff of mobile video analytics. In this paper, we study an MEC-enabled multi-device video analytics system and formulate the problem as a Markov decision process (MDP) to meet two practical challenges: i) the absence of ground truth in real-time and ii) the content-varying degradation-accuracy relation. In particular, we aim to design an online joint frame degradation and bandwidth allocation algorithm with the time-varying function and limited feedback from each device. Thanks to the MDP formulation and$n$-step return technique, the long-term goal offers adaptive information gathering and thus improves the average accuracy and latency. For sample efficiency, we decompose the MDP problem into discrete degradation adaptation subproblems and continuous bandwidth allocation subproblems. Based on the decomposition, we propose a deep reinforcement learning (DRL) based framework, referred to as DBAG, to solve the decomposed subproblems. DBAG integrates model-based optimization and model-free DRL to solve the MDP problem with a discrete-continuous hybrid action space. Under various network setups and public datasets, DBAG greatly improves the accuracy-latency tradeoff.

In emergency scenarios, unmanned aerial vehicles (UAVs) can be deployed to assist localization and communication services for ground terminals. In this paper, we propose a new integrated air-ground networking paradigm that uses dual-functional UAVs to assist the ground networks for improving both communication and localization performance. We investigate the optimization problem of deploying the minimal number of UAVs to satisfy the communication and localization requirements of ground users. The problem has several technical difficulties including the cardinality minimization, the non-convexity of localization performance metric regarding UAV location, and the association between user and communication terminal. To tackle the difficulties, we adopt$D$-optimality as the localization performance metric, and derive the geometric characteristics of the feasible UAV hovering regions in 2D and 3D based on accurate approximation values. We solve the simplified 2D projection deployment problem by transforming the problem into a minimum hitting set problem, and propose a low-complexity algorithm to solve it. Through numerical simulations, we compare our proposed algorithm with benchmark methods. The number of UAVs required by the proposed algorithm is close to the optimal solution, while other benchmark methods require much more UAVs to accomplish the same task.

In reconfigurable intelligent surface (RIS) aided communications, how to optimize a large number of RIS reflecting elements in a scalable and efficient manner remains an open challenge. This paper addresses the scalability issue by partitioning RIS into sub-surfaces, so as to optimize the phase shifts in sub-surface levels to reduce complexity. In the proposed design, each sub-surface employs a linear phase variation structure to anomalously reflect the incident signal to a desired direction. We formulate the achievable rate maximization problem by jointly optimizing the transmit covariance matrix and the structured RIS phase shifts. Then, we characterize the asymptotic behavior of the system with an infinitely large number of transceiver antennas and reflecting elements. We develop an efficient algorithm to find the optimal solution to the asymptotic problem via one-dimensional (1D) search. By applying the asymptotic result to a finite-size system with necessary modifications, we show by numerical results that the proposed design achieves a favorable tradeoff between system performance and computational complexity.

Finite blocklength coding has attracted considerable recent attention because it holds the promise of ultra-reliable and low-latency communications (URLLC) in smart grids, autonomous driving, tele-surgery, and industrial internet of things (IIoT). However, as the instantaneous blocklength is constrained by the number of backlogged bits, short packet transmission with random packet arrival becomes a challenging issue. In this paper, we present a cross-layer mechanism referred to as the buffer-aware variable-blocklength coding to minimize the average delay of bursty traffics. To optimize and analyze the buffer-aware short packet transmission, we formulate a tandem queue model consisting of both the packet queue buffering bits to be encoded and the symbol queue buffering coded symbols to be sent. By deriving the transition probability matrix and the steady state probability of the two-dimensional Markov chain characterizing the tandem queue, we obtain the average latency as a function of the arrival rate and transmission power. Simulation results verify our theoretical analysis and demonstrate the potential of the buffer-aware variable-blocklength coding scheme.

Integrated Sensing and Communications (ISAC) has been regarded as a technological trend in 6G networks to obtain significant performance gain with deep integration of sensing and communication functionalities. However, achieving Ultra-Reliable and Low-Latency Communication (URLLC) with ISAC designs has been under-investigated. In this paper, we investigate a joint beamforming and scheduling design for an ISAC-enabled system fulfilling URLLC requirements. In particular, we employ an ISAC design to sense stochastic events, through which an aperiodic traffic is generated. The event-triggered aperiodic traffic is next scheduled with a periodic traffic. Meanwhile, the system provides reliability for both periodic and aperiodic traffics within a hard latency constraint. To understand the latency and reliability performance under the joint design, we first describe the sensing and communication processes under a Partially Observable Markov Decision Process (POMDP) framework. Then, we analytically express the two traffics' latency and reliability performance under the joint beamforming and scheduling. Based on the analysis, we formulate an optimal tradeoff between the performance of the periodic and aperiodic traffics, and obtain it by a Dynamic Programming (DP)-based algorithm. Simulation results finally show that the optimal tradeoff outperforms other tradeoffs of benchmark algorithms.

The emerging mobile edge computing (MEC) technology has been recently applied to improve the Quality of Experience (QoE) of network services, such as live video streaming. In this paper, we study an energy-aware adaptive live streaming scheme in wireless edge networks. In particular, we aim to design a joint uplink transmission and edge transcoding algorithm maximizing the video followers’ QoE, while minimizing the energy consumption of the video streamer. We formulate the problem as a Markov decision process (MDP), and propose a deep reinforcement learning (DRL) based framework, named SACCT, to determine the streamer’s encoding bitrate, the uploading power as well as the edge transcoding bitrates and frequency. We decompose the MDP problem into inter-frame and intra-frame problems to address the key design challenges that arise from continuous-discrete hybrid action space, time-varying state and action spaces, and unknown network variation. By doing so, SACCT integrates model-based optimization and model-free DRL to determine the intra-frame continuous resource allocation decisions and the inter-frame discrete bitrate adaptation decisions, respectively. To integrate both the numerical features (e.g., channel gain) and the categorical features (e.g., bitrate), we propose a communication Transformer (CT) as a backbone of SACCT by representing network states as communication tokens and running Transformers to model multi-scale dependencies. Extensive simulations manifest that compared with state-of-the-art approaches, SACCT can provide 128.23% (on average) extra reward. As such, by leveraging joint uplink adaption and edge transcoding, the proposed scheme enables an intelligent wireless network edge with QoE-assured and energy-aware live streaming services.

Federated learning (FL) has recently emerged as a promising technology to enable artificial intelligence (AI) at the network edge, where distributed mobile devices collaboratively train a shared AI model under the coordination of an edge server. To significantly improve the communication efficiency of FL, over-the-air computation allows a large number of mobile devices to concurrently upload their local models by exploiting the superposition property of wireless multi-access channels. Due to wireless channel fading, the model aggregation error at the edge server is dominated by the weakest channel among all devices, causing severe straggler issues. In this paper, we propose a relay-assisted cooperative FL scheme to effectively address the straggler issue. In particular, we deploy multiple half-duplex relays to cooperatively assist the devices in uploading the local model updates to the edge server. The nature of the over-the-air computation poses system objectives and constraints that are distinct from those in traditional relay communication systems. Moreover, the strong coupling between the design variables renders the optimization of such a system challenging. To tackle the issue, we propose an alternating-optimization-based algorithm to optimize the transceiver and relay operation with low complexity. Then, we analyze the model aggregation error in a single-relay case and show that our relay-assisted scheme achieves a smaller error than the one without relays provided that the relay transmit power and the relay channel gains are sufficiently large. The analysis provides critical insights on relay deployment in the implementation of cooperative FL. Extensive numerical results show that our design achieves faster convergence compared with state-of-the-art schemes.

The emerging mobile edge computing (MEC) technology has been recently applied to improve adaptive bitrate (ABR) streaming service quality under time-varying wireless channels. In this paper, we consider a heterogeneous multi-user MEC-enabled video streaming network with time-varying wireless channels in sequential time frames. In particular, we aim to design an online joint transcoding and transmission resource allocation algorithm to maximize the ABR streaming user’s quality of experience (QoE) subject to the bandwidth and CPU constraints. The algorithm is “online” in the sense that the bitrate and resource allocation decisions made at each frame depend only on the observation of past events. We formulate the problem as a mixed integer non-linear programming (MINLP) that jointly determines bitrate adaptation, bandwidth allocation, and CPU cycle assignment. To cope with the challenge arising from the coupling decisions of adjacent frames, we propose a low-complexity online algorithm, named OCCA. Specifically, by introducing queueing model constraints, we transform the offline non-conex MINLP problem into a multi-frame problem. Then, we analytically decouple the multi-stage(frame) problem to multiple per-frame convex subproblems that can be solved with high robustness and low computational complexity. We perform simulations with realistic scenarios to evaluate the performance of the proposed algorithm. Results manifest that compared with state-of-the-art approaches, our proposed algorithm can provide 97.84% (on average) extra QoE.

Device-edge co-inference opens up new possibilities for resource-constrained wireless devices (WDs) to execute deep neural network (DNN)-based applications with heavy computation workloads. In particular, the WD executes the first few layers of the DNN and sends the intermediate features to the edge server that processes the remaining layers of the DNN. By adapting the model splitting decision, there exists a tradeoff between local computation cost and communication overhead. In practice, the DNN model is re-trained and updated periodically at the edge server. Once the DNN parameters are regenerated, part of the updated model must be placed at the WD to facilitate on-device inference. In this paper, we study the joint optimization of the model placement and online model splitting decisions to minimize the energy-and-time cost of device-edge co-inference in presence of wireless channel fading. The problem is challenging because the model placement and model splitting decisions are strongly coupled, while involving two different time scales. We first tackle online model splitting by formulating an optimal stopping problem, where the finite horizon of the problem is determined by the model placement decision. In addition to deriving the optimal model splitting rule based on backward induction, we further investigate a simple one-stage look-ahead rule, for which we are able to obtain analytical expressions of the model splitting decision. The analysis is useful for us to efficiently optimize the model placement decision in a larger time scale. In particular, we obtain a closed-form model placement solution for the fully-connected multilayer perceptron with equal neurons. Simulation results validate the superior performance of the joint optimal model placement and splitting with various DNN structures.

The recent development of unmanned aerial vehicle (UAV) and mobile edge computing (MEC) technologies provides flexible and resilient computation services to mobile users out of the terrestrial computing service coverage. In this paper, we consider a UAV-enabled MEC platform that serves multiple mobile ground users with random movements and task arrivals. We aim to minimize the average weighted energy consumption of all users subject to the average UAV energy consumption and data queue stability constraints. We formulate the problem as a multi-stage stochastic optimization, and adopt Lyapunov optimization to convert it into per-slot deterministic problems with fewer optimizing variables. We design two reduced-complexity methods that solve the resource allocation and the UAV movement either in two sequential steps or jointly in one step. Both methods can guarantee to satisfy the average UAV energy and queue stability constraints, meanwhile achieving a tradeoff between the user energy consumption and the length of queue backlog. Simulation results show that the two methods significantly outperform the other benchmark methods including a learning-based method in reducing the energy consumption of ground users. In between, the proposed joint optimization method achieves better performance than the two-stage method at the cost of higher computational complexity.

Unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) has recently emerged as a cost-effective solution to provide computation service to distributed devices in the absence of terrestrial infrastructure. In this paper, we consider a UAV-enabled MEC system serving multiple energy harvesting (EH) devices, where the energy and task data arrive at the users stochastically. Without any future knowledge of task data and energy arrivals, our objective is to design an online algorithm to jointly optimize the UAV energy and task processing rate, meanwhile satisfying the long-term data queue stability. We formulate the problem as a multi-stage stochastic programming and propose an online algorithm, named PLOT, based on perturbed Lyapunov optimization technique. In particular, PLOT resolves the coupling effect of sequential control actions, and converts the stochastic problem into per-slot deterministic optimization problem. For each per-slot problem, we design a low-complexity algorithm to solve it. We show that the PLOT algorithm can derive a feasible solution to the original problem and achieve an$[O(1/V),O(V)]$trade-off between the system cost and the data queue length. Simulation results justify our analysis and demonstrate that the PLOT algorithm achieves better performance in terms of system utility and maintains queue stability that is not achieved by other benchmark methods.

In this paper, we study an unmanned aerial vehicle (UAV)-enabled mobile edge computing (MEC) system with multiple energy harvesting (EH) devices. Considering the stochastic energy and data arrivals in sequential time slots, we formulate the UAV propulsion energy minimization problem with long-term data queue stability and battery causality constraints as a multi-stage stochastic optimization programming. To facilitate online control without any prior knowledge of future information, we adopt the perturbed Lyapunov optimization method that decouples the control decisions made in sequential time slots and determines the real-time control decisions by solving a deterministic problem in each time slot. For the per-slot deterministic problem, we decouple it into three sub-problems: the optimal energy harvesting, the computation resource allocation and the UAV trajectory control, and propose a reduced-complexity method to solve them sepa-rately. Simulation results demonstrate that the proposed algorithm guarantees the data queue stability that is not achievable by the benchmark method when the two methods consume identical UAV propulsion energy.

In this paper, we consider a multi-user mobile-edge computing (MEC) network with time-varying wireless channels and stochastic user task data arrivals in sequential time frames. In particular, we aim to design an online computation offloading algorithm to maximize the network data processing capability subject to the long-term data queue stability and average power constraints. The online algorithm is practical in the sense that the decisions for each time frame are made without the assumption of knowing future channel conditions and data arrivals. We formulate the problem as a multi-stage stochastic mixed integer non-linear programming (MINLP) problem that jointly determines the binary offloading (each user computes the task either locally or at the edge server) and system resource allocation decisions in sequential time frames. To address the coupling in the decisions of different time frames, we propose a novel framework, named LyDROO, that combines the advantages of Lyapunov optimization and deep reinforcement learning (DRL). Specifically, LyDROO first applies Lyapunov optimization to decouple the multi-stage stochastic MINLP into deterministic per-frame MINLP subproblems of much smaller size. Then, it integrates model-based optimization and model-free DRL to solve the per-frame MINLP problems with very low computational complexity. Simulation results show that the proposed LyDROO achieves optimal computation performance while satisfying all the long-term constraints. Besides, it induces very low execution latency that is particularly suitable for real-time implementation in fast fading environments.

This paper studies semi-blind channel estimation for a reconfigurable intelligent surface (RIS) aided uplink massive multiple-input multiple-output (MIMO) system, in which the base station simultaneously estimates the channel coefficients and detects the partially unknown transmit symbols. We formulate the semi-blind channel estimation task as a trilinear inference problem. Based on the approximate message passing (AMP) principle, we develop a computationally efficient approach, called Trilinear AMP, to calculate the marginal posterior mean estimators of the trilinear inference problem. Simulation results demonstrate the effectiveness of the proposed Trilinear AMP approach.

Intelligent reflecting surface (IRS) is an emerging technology to enhance the energy efficiency and spectrum efficiency of wireless-powered communication networks (WPCNs). In this article, we investigate an IRS-assisted multiuser multiple-input single-output (MISO) WPCN, where the single-antenna wireless devices (WDs) harvest wireless energy in the downlink (DL) and transmit their information simultaneously in the uplink (UL) to a common hybrid access point (HAP) equipped with multiple antennas. Our goal is to maximize the weighted sum rate (WSR) of all the energy-harvesting users. To make full use of the beamforming gain provided by both the HAP and the IRS, we jointly optimize the active beamforming of the HAP and the reflecting coefficients (passive beamforming) of the IRS in both DL and UL transmissions, as well as the transmit power of the WDs to mitigate the interuser interference at the HAP. To tackle the challenging optimization problem, we first consider fixing the passive beamforming, and converting the remaining joint active beamforming and user transmit power control problem into an equivalent weighted minimum mean-square error problem, where we solve it using an efficient block-coordinate descent method. Then, we fix the active beamforming and user transmit power, and optimize the passive beamforming coefficients of the IRS in both the DL and UL using a semidefinite relaxation method. Accordingly, we apply a block-structured optimization method to update the two sets of variables alternately. The numerical results show that the proposed joint optimization achieves significant performance gain over other representative benchmark methods and effectively improves the throughput performance in multiuser MISO WPCNs.

In this paper, we investigate over-the-air model aggregation in a federated edge learning (FEEL) system. We introduce a Markovian probability model to characterize the intrinsic temporal structure of the model aggregation series. With this temporal probability model, we formulate the model aggregation problem as to infer the desired aggregated update given all the past observations from a Bayesian perspective. We develop a message passing based algorithm, termed temporal-structure-assisted gradient aggregation (TSA-GA), to fulfil this estimation task with low complexity and near-optimal performance. We further establish the state evolution (SE) analysis to characterize the behaviour of the proposed TSA-GA algorithm, and derive an explicit bound of the expected loss reduction of the FEEL system under certain standard regularity conditions. In addition, we develop an expectation maximization (EM) strategy to learn the unknown parameters in the Markovian model. We show that the proposed TSA-GA significantly outperforms the state-of-the-art analog compression scheme, and is able to achieve comparable learning performance as the error-free benchmark in terms of final test accuracy.

This article proposes a reinforcement-learning (RL) approach for optimizing charging scheduling and pricing strategies that maximize the system objective of a public electric vehicle (EV) charging station. The proposed algorithm is “online” in the sense that the charging and pricing decisions made at each time depend only on the observation of past events, and is “model-free” in the sense that the algorithm does not rely on any assumed stochastic models of uncertain events. To cope with the challenge arising from the time-varying continuous state and action spaces in the RL problem, we first show that it suffices to optimize the total charging rates to fulfill the charging requests before departure times. Then, we propose a feature-based linear function approximator for the state-value function to further enhance the efficiency and generalization ability of the proposed algorithm. Through numerical simulations with real-world data, we show that the proposed RL algorithm achieves on average 138.5% higher charging-station profit than representative benchmark algorithms.

In this article, we propose an online datadriven approach that leverages the isolation mechanism for fast event detection with low-quality data measurement. The proposed adaptive and online isolation forest (iForest)based detection (AOIFD) method adopts a hierarchical subspace feature selection scheme to design two levels of detectors. As such, it is capable of differentiating events from low-quality data measurements, preventing false alarms in the presence of low-quality data measurements. We further propose a data augmentation method to address the training data imbalance, which is caused by the rare occurrence of events. Moreover, we propose an adaptive training process to update the AOIFD method so that it can adapt to the time-varying operating conditions of power systems. The proposed AOIFD algorithm is practical in the sense that it is a fast-response method that requires no system modeling information and no global communications. Case studies with both synthetic and realistic PMU data are conducted to validate the effectiveness of the proposed method.

Opportunistic computation offloading is an effective method to improve the computation performance of mobile-edge computing (MEC) networks under dynamic edge environment. In this paper, we consider a multi-user MEC network with time-varying wireless channels and stochastic user task data arrivals in sequential time frames. In particular, we aim to design an online computation offloading algorithm to maximize the network data processing capability subject to the long-term data queue stability and average power constraints. The online algorithm is practical in the sense that the decisions for each time frame are made without the assumption of knowing the future realizations of random channel conditions and data arrivals. We formulate the problem as a multi-stage stochastic mixed integer non-linear programming (MINLP) problem that jointly determines the binary offloading (each user computes the task either locally or at the edge server) and system resource allocation decisions in sequential time frames. To address the coupling in the decisions of different time frames, we propose a novel framework, named LyDROO, that combines the advantages of Lyapunov optimization and deep reinforcement learning (DRL). Specifically, LyDROO first applies Lyapunov optimization to decouple the multi-stage stochastic MINLP into deterministic per-frame MINLP subproblems. By doing so, it guarantees to satisfy all the long-term constraints by solving the per-frame subproblems that are much smaller in size. Then, LyDROO integrates model-based optimization and model-free DRL to solve the per-frame MINLP problems with very low computational complexity. Simulation results show that under various network setups, the proposed LyDROO achieves optimal computation performance while stabilizing all queues in the system. Besides, it induces very low computation time that is particularly suitable for real-time implementation in fast fading environments.

Leveraging recent advances on mobile edge computing (MEC), edge intelligence has emerged as a promising paradigm to support mobile artificial intelligence (AI) applications at the network edge. In this paper, we consider the AI service placement problem in a multi-user MEC system, where the access point (AP) places the most up-to-date AI program at user devices to enable local computing/task execution at the user side. To fully utilize the stringent wireless spectrum and edge computing resources, the AP sends the AI service program to a user only when enabling local computing at the user yields a better system performance. We formulate a mixed-integer non-linear programming (MINLP) problem to minimize the total computation time and energy consumption of all users by jointly optimizing the service placement (i.e., which users to receive the program) and resource allocation (on local CPU frequencies, uplink bandwidth, and edge CPU frequency). To tackle the MINLP problem, we derive analytical expressions to calculate the optimal resource allocation decisions with low complexity. This allows us to efficiently obtain the optimal service placement solution by search-based algorithms such as meta-heuristic or greedy search algorithms. To enhance the algorithm scalability in large-sized networks, we further propose an ADMM (alternating direction method of multipliers) based method to decompose the optimization problem into parallel tractable MINLP subproblems. The ADMM method eliminates the need of searching in a high-dimensional space for service placement decisions and thus has a low computational complexity that grows linearly with the number of users. Simulation results show that the proposed algorithms perform extremely close to the optimum and significantly outperform the other representative benchmark algorithms.

To exploit massive amounts of data generated at mobile edge networks,federated learning(FL) has been proposed as an attractive substitute for centralized machine learning (ML). By collaboratively training a shared learning model at edge devices, FL avoids direct data transmission and thus overcomes high communication latency and privacy issues as compared to centralized ML. To improve the communication efficiency in FL model aggregation,over-the-air computationhas been introduced to support a large number of simultaneous local model uploading by exploiting the inherent superposition property of wireless channels. However, due to the heterogeneity of communication capacities among edge devices, over-the-air FL suffers from the straggler issue in which the device with the weakest channel acts as a bottleneck of the model aggregation performance. This issue can be alleviated by device selection to some extent, but the latter still suffers from a tradeoff between data exploitation and model communication. In this paper, we leverage thereconfigurable intelligent surface(RIS) technology to relieve the straggler issue in over-the-air FL. Specifically, we develop a learning analysis framework to quantitatively characterize the impact of device selection and model aggregation error on the convergence of over-the-air FL. Then, we formulate a unified communication-learning optimization problem to jointly optimize device selection, over-the-air transceiver design, and RIS configuration. Numerical experiments show that the proposed design achieves substantial learning accuracy improvement compared with the state-of-the-art approaches, especially when channel conditions vary dramatically across edge devices.

Provided with mobile edge computing (MEC) services, wireless devices (WDs) no longer have to experience long latency in running their desired programs locally, but can pay to offload computation tasks to the edge server. Given its limited storage space, it is important for the edge server at the base station (BS) to determine which service programs to cache by meeting and guiding WDs' offloading decisions. In this article, we propose an MEC service pricing scheme to coordinate with the service caching decisions and control WDs' task offloading behavior in a cellular network. We propose a two-stage dynamic game of incomplete information to model and analyze the two-stage interaction between the BS and multiple associated WDs. Specifically, in Stage I, the BS determines the MEC service caching and announces the service program prices to the WDs, with the objective to maximize its expected profit under both storage and computation resource constraints. In Stage II, given the prices of different service programs, each WD selfishly decides its offloading decision to minimize individual service delay and cost, without knowing the other WDs' desired program types or local execution delays. Despite the lack of WD's information and the coupling of all the WDs' offloading decisions, we derive the optimal threshold-based offloading policy that can be easily adopted by the WDs in Stage II at the Bayesian equilibrium. In particular, a WD is more likely to offload when there are fewer WDs competing for the edge server's computation resource, or when it perceives a good channel condition or low MEC service price. Then, by predicting the WDs' offloading equilibrium, we jointly optimize the BS' pricing and service caching in Stage I via a low-complexity algorithm. In particular, we first study the differentiated pricing scheme and prove that the same price should be charged to the cached programs of the same workload. Motivated by this analysis, we further propose a low-complexity uniform pricing heuristics.

In this paper, we consider the service placement problem in a multi-user MEC system, where the access point (AP) places the most up-to-date artificial intelligent (AI) program at user devices via a broadcast channel. In particular, a user that successfully receives the program can execute its tasks both locally and remotely at the AP via partial task offloading. Otherwise, all its computations must be offloaded to and executed at the AP. We formulate a mixed-integer non-linear programming (MINLP) problem to minimize the total computation time and energy consumption of all users. The problem is particularly challenging because the service placement solution (i.e., which users to receive the program) is combinatorial in nature and strongly coupled with the computation offloading decision of each user (how much task to be executed at the AP) and resource allocation (on local CPU frequencies and uplink bandwidth). We tackle the problem with an ADMM (alternating direction method of multipliers) based method that effectively decomposes the problem into parallel smaller and tractable MINLP subproblems. Simulation results show that the proposed method achieves a performance extremely close to the optimum and has a low computational complexity that grows linearly with the number of users.

In this paper, we consider a mobile-edge computing (MEC) system, where an access point (AP) assists a mobile device (MD) to execute an application consisting of multiple tasks following a general task call graph. The objective is to jointly determine the offloading decision of each task and the resource allocation (e.g., CPU computing power) under time-varying wireless fading channels and stochastic edge computing capability, so that the energy-time cost (ETC) of the MD is minimized. Solving the problem is particularly hard due to the combinatorial offloading decisions and the strong coupling among task executions under the general dependency model. To address the issue, we propose a deep reinforcement learning (DRL) framework based on the actor-critic learning structure. In particular, the actor network utilizes a deep neural network (DNN) to learn the optimal mapping from the input states (i.e., wireless channel gains and edge CPU frequency) to the binary offloading decision of each task. Meanwhile, for the critic network, we show that given the offloading decision, the remaining resource allocation problem becomes convex, where we can quickly evaluate the ETC performance of the offloading decisions output by the actor network. Accordingly, we select the best offloading action and store the state-action pair in an experience replay memory as the training dataset to continuously improve the action generation DNN. Numerical results show that for various types of task graphs, the proposed algorithm achieves up to 99.5% of the optimal performance while significantly reducing the computational complexity compared to the existing optimization methods.

In this paper, we study the channel acquisition problem in a reconfigurable intelligent surface (RIS) assisted multiuser multiple-input multiple-output (MIMO) system, where an RIS with fully passive phase-shift elements is deployed to assist the MIMO communication. The state-of-the-art channel acquisition approach in such a system estimates the cascaded transmitter-to-RIS and RIS-to-receiver channels by adopting excessively long training sequences. To estimate the cascaded channels with an affordable training overhead, we formulate the channel estimation problem as a matrix-calibration based matrix factorization task. By exploiting the information on the slow-varying channel components and the hidden channel sparsity, we propose a novel message-passing based algorithm to factorize the cascaded channels.

Intelligent reflecting surface (IRS) can effectively enhance the energy and spectral efficiency of wireless communication system through the use of a large number of low-cost passive reflecting elements. In this paper, we investigate throughput optimization of IRS-assisted user cooperation in a wireless powered communication network (WPCN), where the two WDs harvest wireless energy and transmit information to a common hybrid access point (HAP). In particular, the two WDs first exchange their independent information with each other and then form a virtual antenna array to transmit jointly to the HAP. We aim to maximize the common (minimum) throughput performance by jointly optimizing the transmit time and power allocations of the two WDs on wireless energy and information transmissions and the passive array coefficients on reflecting the wireless energy and information signals. By comparing with some existing benchmark schemes, our results show that the proposed IRS-assisted user cooperation method can effectively improve the throughput performance of cooperative transmission in WPCNs.

State estimation is critical to the monitoring and control of smart grids. Recently, the false data injection attack (FDIA) is emerging as a severe threat to state estimation. Conventional FDIA detection approaches are limited by their strong statistical knowledge assumptions, complexity, and hardware cost. Moreover, most of the current FDIA detection approaches focus on detecting the presence of FDIA, while the important information of the exact injection locations is not attainable. Inspired by the recent advances in deep learning, we propose a deep-learning-based locational detection architecture (DLLD) to detect the exact locations of FDIA in real time. The DLLD architecture concatenates a convolutional neural network (CNN) with a standard bad data detector (BDD). The BDD is used to remove the low-quality data. The followed CNN, as a multilabel classifier, is employed to capture the inconsistency and co-occurrence dependency in the power flow measurements due to the potential attacks. The proposed DLLD is “model-free” in the sense that it does not leverage any prior statistical assumptions. It is also “cost-friendly” in the sense that it does not alter the current BDD system and the runtime of the detection process is only hundreds of microseconds on a household computer. Through extensive experiments in the IEEE bus systems, we show that DLLD can perform locational detection precisely under various noise and attack conditions. In addition, we also demonstrate that the employed multilabel classification approach effectively enhances the presence-detection accuracy.

Reconfigurable intelligent surface (RIS) is envisioned to be an essential component of the paradigm for beyond 5G networks as it can potentially provide similar or higher array gains with much lower hardware cost and energy consumption compared with the massive multiple-input multiple-output (MIMO) technology. In this paper, we focus on one of the fundamental challenges, namely the channel acquisition, in a RIS-assisted multiuser MIMO system. The state-of-the-art channel acquisition approach in such a system with fully passive RIS elements estimates the cascaded transmitter-to-RIS and RIS-to-receiver channels by adopting excessively long training sequences. To estimate the cascaded channels with an affordable training overhead, we formulate the channel estimation problem in the RIS-assisted multiuser MIMO system as a matrix-calibration based matrix factorization task. By exploiting the information on the slow-varying channel components and the hidden channel sparsity, we propose a novel message-passing based algorithm to factorize the cascaded channels. Furthermore, we present an analytical framework to characterize the theoretical performance bound of the proposed estimator in the large-system limit. Finally, we conduct simulations to verify the high accuracy and efficiency of the proposed algorithm.

The Smart Grid represents an unprecedented opportunity to move the electric grid into a new era of reliability, availability, and efficiency. It uses two-way communications, digital technologies, advanced sensing and computing infrastructure, and software abilities to provide improved monitoring, protection and optimization of all the grids’ components including generation, transmission, distribution and consumers.

In this paper, we study joint antenna activity detection, channel estimation, and multiuser detection for massive multiple-input multiple-output (MIMO) systems with general spatial modulation (GSM). We first establish a double-sparsity massive MIMO model by considering the channel sparsity of the massive MIMO channel and the signal sparsity of GSM. Based on the double-sparsity model, we formulate a blind detection problem. To solve the blind detection problem, we develop message-passing based blind channel-and-signal estimation (BCSE) algorithm. The BCSE algorithm basically follows the affine sparse matrix factorization technique, but with critical modifications to handle the double-sparsity property of the model. We show that the BCSE algorithm significantly outperforms the existing blind and training-based algorithms, and is able to closely approach the genie bounds (with either known channel or known signal). In the BCSE algorithm, short pilots are employed to remove the phase and permutation ambiguities after sparse matrix factorization. To utilize the short pilots more efficiently, we further develop the semi-blind channel-and-signal estimation (SBCSE) algorithm to incorporate the estimation of the phase and permutation ambiguities into the iterative message-passing process. We show that the SBCSE algorithm substantially outperforms the counterpart algorithms including the BCSE algorithm in the short-pilot regime.

Wireless powered mobile-edge computing (MEC) has recently emerged as a promising paradigm to enhance the data processing capability of low-power networks, such as wireless sensor networks and internet of things (IoT). In this paper, we consider a wireless powered MEC network that adopts a binary offloading policy, so that each computation task of wireless devices (WDs) is either executed locally or fully offloaded to an MEC server. Our goal is to acquire an online algorithm that optimally adapts task offloading decisions and wireless resource allocations to the time-varying wireless channel conditions. This requires quickly solving hard combinatorial optimization problems within the channel coherence time, which is hardly achievable with conventional numerical optimization methods. To tackle this problem, we propose a Deep Reinforcement learning-based Online Offloading (DROO) framework that implements a deep neural network as a scalable solution that learns the binary offloading decisions from the experience. It eliminates the need of solving combinatorial optimization problems, and thus greatly reduces the computational complexity especially in large-size networks. To further reduce the complexity, we propose an adaptive procedure that automatically adjusts the parameters of the DROO algorithm on the fly. Numerical results show that the proposed algorithm can achieve near-optimal performance while significantly decreasing the computation time by more than an order of magnitude compared with existing optimization methods. For example, the CPU execution latency of DROO is less than 0.1 second in a 30-user network, making real-time and optimal offloading truly viable even in a fast fading environment.

Heterogeneous Network (HetNet), where Small cell Base Stations (SBSs) are densely deployed to offload traffic from macro Base Stations (BSs), is identified as a key solution to meet the unprecedented mobile traffic demand. The high density of SBSs are designed for peak traffic hours and consume an unnecessarily large amount of energy during off-peak time. In this paper, we propose a Deep Reinforcement-Learning (DRL) based SBS activation strategy that activates the optimal subset of SBSs to significantly lower the energy consumption without compromising the quality of service. In particular, we formulate the SBS on/off switching problem into a Markov Decision Process that can be solved by Actor Critic (AC) reinforcement learning methods. To avoid prohibitively high computational and storage costs of conventional tabular-based approaches, we propose to use deep neural networks to approximate the policy and value functions in the AC approach. Moreover, to expedite the training process, we adopt a Deep Deterministic Policy Gradient (DDPG) approach together with a novel action refinement scheme. Through extensive numerical simulations, we show that the proposed scheme greatly outperforms the existing methods in terms of both energy efficiency and computational efficiency. We also show that the proposed scheme can scale to large system with polynomial complexities in both storage and computation.

In mobile edge computing (MEC) systems, edge service caching refers to pre-storing the necessary programs for executing computation tasks at MEC servers. Service caching effectively reduces the real-time delay/bandwidth cost on acquiring and initializing service applications when computation tasks are offloaded to the MEC servers. The limited caching space at resource-constrained edge servers calls for careful design of caching placement to determine which programs to cache over time. This is in general a complicated problem that highly correlates to the computation offloading decisions of computation tasks, i.e., whether or not to offload a task for edge execution. In this paper, we consider a single edge server that assists a mobile user (MU) in executing a sequence of computation tasks. In particular, the MU can upload and run its customized programs at the edge server, while the server can selectively cache the previously generated programs for future reuse. To minimize the computation delay and energy consumption of the MU, we formulate a mixed integer non-linear programming (MINLP) that jointly optimizes the service caching placement, computation offloading decisions, and system resource allocation (e.g., CPU processing frequency and transmit power of MU). To tackle the problem, we first derive the closed-form expressions of the optimal resource allocation solutions, and subsequently transform the MINLP into an equivalent pure 0-1 integer linear programming (ILP) that is much simpler to solve. To further reduce the complexity in solving the ILP, we exploit the underlying structures of caching causality and task dependency models, and accordingly devise a reduced-complexity alternating minimization technique to update the caching placement and offloading decision alternately. Extensive simulations show that the proposed joint optimization techniques achieve substantial resource savings of the MU compared to other representative benchmark methods considered.

In this paper, we study the beamforming design problem in frequency-division duplexing (FDD) downlink massive MIMO systems, where instantaneous channel state information (CSI) is assumed to be unavailable at the base station (BS). We propose to extract the information of the angle-of-departures (AoDs) and the corresponding large-scale fading coefficients (a.k.a. spatial information) of the downlink channel from the uplink channel estimation procedure, based on which a novel downlink beamforming design is presented. By separating the subpaths for different users based on the spatial information and the hidden sparsity of the physical channel, we construct near-orthogonal virtual channels in the beamforming design. Furthermore, we derive a sum-rate expression and its approximations for the proposed system. Based on these closed-form rate expressions, we develop two low-complexity beam selection schemes and carry out asymptotic analysis to provide valuable insights on the system design. Numerical results demonstrate a significant performance improvement of our proposed algorithm over the state-of-the-art beamforming approach.

In this paper, we consider a mobile-edge computing (MEC) system, where an access point (AP) assists a mobile device (MD) to execute an application consisting of multiple tasks following a general task call graph. The objective is to jointly determine the offloading decision of each task and the resource allocation (e.g., CPU computing power) under time-varying wireless fading channels and stochastic edge computing capability, so that the energy-time cost (ETC) of the MD is minimized. Solving the problem is particularly hard due to the combinatorial offloading decisions and the strong coupling among task executions under the general dependency model. Conventional numerical optimization methods are inefficient to solve such a problem, especially when the problem size is large. To address the issue, we propose a deep reinforcement learning (DRL) framework based on the actor-critic learning structure. In particular, the actor network utilizes a DNN to learn the optimal mapping from the input states (i.e., wireless channel gains and edge CPU frequency) to the binary offloading decision of each task. Meanwhile, by analyzing the structure of the optimal solution, we derive a low-complexity algorithm for the critic network to quickly evaluate the ETC performance of the offloading decisions output by the actor network. With the low-complexity critic network, we can quickly select the best offloading action and subsequently store the state-action pair in an experience replay memory as the training dataset to continuously improve the action generation DNN. To further reduce the complexity, we show that the optimal offloading decision exhibits an one-climb structure, which can be utilized to significantly reduce the search space of action generation. Numerical results show that for various types of task graphs, the proposed algorithm achieves up to 99.1% of the optimal performance while significantly reducing the computational complexity compared to the existing optimization methods.

Mobile-edge computing (MEC) has recently emerged as a cost-effective paradigm to enhance the computing capability of hardware-constrained wireless devices (WDs). In this paper, we first consider a two-user MEC network, where each WD has a sequence of tasks to execute. In particular, we consider task dependency between the two WDs, where the input of a task at one WD requires the final task output at the other WD. Under the considered task-dependency model, we study the optimal task offloading policy and resource allocation (e.g., on offloading transmit power and local CPU frequencies) that minimize the weighted sum of the WDs' energy consumption and task execution time. The problem is challenging due to the combinatorial nature of the offloading decisions among all tasks and the strong coupling with resource allocation. To tackle this problem, we first assume that the offloading decisions are given and derive the closed-form expressions of the optimal offloading transmit power and local CPU frequencies. Then, an efficient bi-section search method is proposed to obtain the optimal solutions. Furthermore, we prove that the optimal offloading decisions follow an one-climb policy, based on which a reduced-complexity Gibbs Sampling algorithm is proposed to obtain the optimal offloading decisions. We then extend the investigation to a general multi-user scenario, where the input of a task at one WD requires the final task outputs from multiple other WDs. Numerical results show that the proposed method can significantly outperform the other representative benchmarks and efficiently achieve low complexity with respect to the call graph size.

In this paper, we study the beamforming design problem in frequency-division duplexing (FDD) massive MIMO downlink systems, where instantaneous channel state information (CSI) is unavailable at the base station (BS). We propose to extract the spatial information (i.e., the angle parameters and the large-scale fading coefficients) of the downlink channel from the uplink channel estimation procedure, based on which a novel downlink beamforming design is provided. Furthermore, we derive a sum-rate expression and its approximations for the proposed system. By maximizing the resultant sum-rate, we develop a low-complexity beam selection scheme. Numerical results demonstrate that our proposed algorithm has a significant improvement compared to the existing statistical beamforming (SBF) approaches.

In this paper, we study blind signal detection by exploiting the hidden sparsity of angular-domain propagation channels in massive MIMO systems. The state-of-the-art approach utilizes the channel sparsity by representing the angular-domain channel with a uniform angle-sampling grid. However, this approach is only applicable to uniform linear arrays and may cause a substantial performance loss due to the energy leakage problem. In contrast to this approach, we deploy a sparse channel representation with a fixed general sampling grid. Based on that, we formulate the blind signal detection problem as an affine matrix factorization task and develop a novel message passing algorithm to estimate the channel and the user signals simultaneously. Unlike the existing approach, the proposed algorithm is applicable to general antenna arrays. Numerical results show that our proposed method significantly reduces the estimation error compared to the state-of-the-art approach by avoiding the leakage of energy.

The Paradigm of Internet of Things (IoT) is increasingly integrated with real-world applications. There are many critical IoT applications in the energy sector. Worldwide energy systems and infrastructure are experiencing tremendous transformation. There has been a drastic surge in the global energy consumption, which has tripled in the past 50 years. As a result, new measures have been introduced to improve the responsiveness and robustness of the energy systems, along with the global trends of deregulation and decarbonization.

In this paper, we consider an Internet of Electric Vehicles (IoEV) powered by heterogeneous charging facilities in the transportation network. In particular, we take into account the state-of-the-art vehicle-to-grid (V2G) charging and renewable power generation technologies implemented in the charging stations, such that the charging stations differ from each other in their energy capacities, electricity prices, and service types (i.e., with or without V2G capability). In this case, each electric vehicle (EV) user needs to decide which path to take (i.e., the routing problem) and where and how much to charge/discharge its battery at the charging stations in the chosen path (i.e., the charging scheduling problem) such that its journey can be accomplished with the minimum monetary cost and time delay. From the system operator's perspective, we formulate a joint routing and charging scheduling optimization problem for an IoEV network, and show that the problem is NP-hard in general. To tackle the NP-hardness, we propose an approximate algorithm that can achieve affordable computational complexity in large-size IoEV networks. The proposed algorithm allows the routing and charging solution to be calculated in a distributed manner by the system operator and EV users, which can effectively reduce the computational complexity at the system operator and protect the EV users' privacy and autonomy. Besides, a proximal method is introduced to improve the convergence rate of the proposed algorithm. Extensive simulations using real world data show that the proposed distributed algorithm can achieve near-optimal performance with relatively low computational complexity in different system set-ups.

Small-cell architecture is widely adopted by cellular network operators to increase spectral spatial efficiency. However, this approach suffers from low spectrum temporal efficiency. When a cell becomes smaller and covers fewer users, its total traffic fluctuates significantly due to insufficient traffic aggregation and exhibits a large “peak-to-mean” ratio. As operators customarily provision spectrum for peak traffic, large traffic temporal fluctuation inevitably leads to low spectrum temporal efficiency. To address this issue, in this paper, we advocate device-to-device (D2D) load-balancing as a useful mechanism. The idea is to shift traffic from a congested cell to its adjacent under-utilized cells by leveraging inter-cell D2D communication, so that the traffic can be served without using extra spectrum, effectively improving the spectrum temporal efficiency. We provide theoretical modeling and analysis to characterize the benefit of D2D load balancing, in terms of total spectrum requirements and the corresponding cost, in terms of incurred D2D traffic overhead. We carry out empirical evaluations based on real-world 4G data traces and show that D2D load balancing can reduce the spectrum requirement by 25% as compared to the standard scenario without D2D load balancing, at the expense of negligible 0.7% D2D traffic overhead.

Banded linear systems arise in many communication scenarios, e.g., those involving inter-carrier interference and inter-symbol interference. Motivated by recent advances in deep learning, we propose to design a high-accuracy low-complexity signal detector for banded linear systems based on convolutional neural networks (CNNs). We develop a novel CNN-based detector by utilizing the banded structure of the channel matrix. Specifically, the proposed CNN-based detector consists of three modules: the input preprocessing module; the CNN module; and the output postprocessing module. With such an architecture, the proposed CNN-based detector is adaptive to different system sizes, and can overcome the curse of dimensionality, which is a ubiquitous challenge in deep learning. Through extensive numerical experiments, we demonstrate that the proposed CNN-based detector outperforms conventional deep neural networks and existing model-based detectors in both accuracy and computational time. Moreover, we show that the CNN is flexible for systems with large sizes or wide bands. We also show that the proposed CNN-based detector can be easily extended to near-banded systems such as doubly selective orthogonal frequency division multiplexing (OFDM) systems and 2-D magnetic recording (TDMR) systems, in which the channel matrices do not have a strictly banded structure.

Motivated by the recent advances in deep learning, we propose to design high-accuracy low-complexity signal detectors for banded linear systems based on deep neural networks (DNNs). We first design a fully connected DNN for signal detection. Then, to deal with the curse of dimensionality, we propose a novel convolutional neural network (CNN) based on the banded structure of the channel matrix. From simulations, we observe that the proposed CNN outperforms the fully connected DNN in both accuracy and computational time. Moreover, CNN is more robust for the extension to channel matrices with a large size or a wide band. We also run extensive numerical experiments to show that both fully connected DNN and CNN perform much better than existing detectors with comparable complexity.

In this paper, we consider a two-user mobile-edge computing (MEC) network, where each wireless device (WD) has a sequence of tasks to execute. In particular, we consider task dependency between the two WDs, where the input of a task at one WD requires the final task output at the other WD. Under the considered task-dependency model, we study the optimal task offloading policy and resource allocation (on offloading transmit power and local CPU frequencies) that minimize the weighted sum of the WDs' energy consumption and execution time. The problem is challenging due to the combinatorial nature of the offloading decision among all tasks and the strong coupling with resource allocation among subsequent tasks. When the offloading decision is given, we obtain the closed-form expressions of the offloading transmit power and local CPU frequencies and propose an efficient method to obtain the optimal solutions. Furthermore, we prove that the optimal offloading decision follows an one-climb policy, based on which a reduced-complexity algorithm is proposed to obtain the optimal offloading decision in polynomial time. Numerical results validate the effectiveness of our proposed methods.

In this paper, we consider a wireless powered mobile edge computing (MEC) network, where the distributed energy-harvesting wireless devices (WDs) are powered by means of radio frequency (RF) wireless power transfer (WPT). In particular, the WDs follow a binary computation offloading policy, i.e., data set of a computing task has to be executed as a whole either locally or remotely at the MEC server via task offloading. We are interested in maximizing the (weighted) sum computation rate of all the WDs in the network by jointly optimizing the individual computing mode selection (i.e., local computing or offloading) and the system transmission time allocation (on WPT and task offloading). The major difficulty lies in the combinatorial nature of multi-user computing mode selection and its strong coupling with transmission time allocation. To tackle this problem, we propose a joint optimization method based on the ADMM (alternating direction method of multipliers) decomposition technique. Simulation results show that the proposed method can efficiently achieve near- optimal performance under various network setups, and significantly outperform the other representative benchmark methods considered. Besides, using both theoretical analysis and numerical study, we show that the proposed method enjoys low computational complexity against the increase of networks size.

In this paper, we consider the profit-maximizing demand response of an energy customer in the real-time electricity market. In a real-time electricity market, the market clearing price is determined by the random deviation of actual power supply and demand from the predicted values in the day-ahead market. An energy customer, which requires a total amount of energy over a certain period of time, has the flexibility of shifting its energy usage in time, and therefore is in perfect position to exploit the volatile real-time market price through demand response. We show that the profit-maximizing demand response strategy can be obtained by solving a finite-horizon continuous-state Markov decision process (MDP) problem. Through rigorous analysis, we show that the optimal actual demand policy exhibits a threshold structure, which can solve the MDP without the need of discretizing the state and action spaces. We demonstrate through extensive simulations that the proposed demand response strategy not only maximizes the profit of the energy customer, but also alleviates the supply-demand imbalance in the power grid, and even reduces the bills of other market participants. On average, the proposed demand response strategy increases the energy customer's profit by 53.8% and saves the bills of other utilities by 80.4% comparing with the benchmark algorithms.

Wireless network virtualization enables mobile virtual network operators (MVNOs) to develop new services on a low-cost platform by leasing virtual resources from mobile network owners. In this paper, we investigate a two-stage spectrum leasing framework, where an MVNO acquires spectrum resources through both advance reservation and on-demand request. To maximize its surplus, the MVNO needs to jointly optimize the amount of spectrum resources to lease in the two stages by taking into account traffic intensity, random user locations, wireless channel statistics, quality-of-service requirements, and the price differences. Meanwhile, to maximize the utilization of the acquired resources, the MVNO dynamically allocates the spectrum resources to its mobile subscribers (users) according to fast wireless channel fading. We formulate the MVNO's surplus maximization problem as a tri-level nested optimization problem consisting of dynamic resource allocation (DRA), on-demand request, and advance reservation subproblems. To solve the problem efficiently, we first analyze the DRA problem, and then use the optimal solution to find the optimal leasing decisions in the two stages. In particular, we derive a closed-form expression of the optimal on-demand request, and develop a stochastic gradient descent algorithm to find the optimal advance reservation. For a special case when the proportional fairness utility function is adopted, we show that the optimal two-stage leasing scheme is related to the number of users and is irrelevant to user locations. Simulation results show that the two-stage spectrum leasing scheme can adapt to different levels of traffic and on-demand price variations, and achieve higher surplus than conventional one-stage leasing schemes.

In practical massive MIMO systems, a substantial portion of system resources are consumed to acquire channel state information (CSI), leading to a drastically lower system capacity compared with the ideal case, where perfect CSI is available. In this paper, we show that the overhead for CSI acquisition can be largely compensated by the potential gain due to the sparsity of the massive MIMO channel in a certain transformed domain. To this end, we propose a novel blind detection scheme that simultaneously estimates the channel and data by factorizing the received signal matrix. We show that by exploiting the channel sparsity, our proposed scheme can achieve a degree of freedom (DoF) very close to the ideal case, provided that the channel is sufficiently sparse. Specifically, the achievable DoF has a fractional gap of only 1/T from the ideal DoF, where T is the channel coherence time. This is a remarkable advance for understanding the performance limit of the massive MIMO system. We further show that the performance advantage of our proposed scheme in the asymptotic SNR regime carries over to the practical SNR regime. Numerical results demonstrate that our proposed scheme significantly outperforms its counterpart schemes in the practical SNR regime under various system configurations.

Finite battery lifetime and low computing capability of size-constrained wireless devices (WDs) have been longstanding performance limitations of many low-power wireless networks, e.g., wireless sensor networks and Internet of Things. The recent development of radio frequency-based wireless power transfer (WPT) and mobile edge computing (MEC) technologies provide a promising solution to fully remove these limitations so as to achieve sustainable device operation and enhanced computational capability. In this paper, we consider a multi-user MEC network powered by the WPT, where each energy-harvesting WD follows a binary computation offloading policy, i.e., the data set of a task has to be executed as a whole either locally or remotely at the MEC server via task offloading. In particular, we are interested in maximizing the (weighted) sum computation rate of all the WDs in the network by jointly optimizing the individual computing mode selection (i.e., local computing or offloading) and the system transmission time allocation (on WPT and task offloading). The major difficulty lies in the combinatorial nature of the multi-user computing mode selection and its strong coupling with the transmission time allocation. To tackle this problem, we first consider a decoupled optimization, where we assume that the mode selection is given and propose a simple bi-section search algorithm to obtain the conditional optimal time allocation. On top of that, a coordinate descent method is devised to optimize the mode selection. The method is simple in implementation but may suffer from high computational complexity in a large-size network. To address this problem, we further propose a joint optimization method based on the alternating direction method of multipliers (ADMM) decomposition technique, which enjoys a much slower increase of computational complexity as the networks size increases. Extensive simulations show that both the proposed methods can efficiently achieve a near-optimal performance under various network setups, and significantly outperform the other representative benchmark methods considered.

In this paper, we endeavor to seek a fundamental understanding of the throughput limit for a training-based uplink multiuser multiple-input multiple-output (MIMO) system. In a multiuser MIMO system, users are geographically separated and experience significantly different large-scale fading. As large-scale fading varies order of magnitude more slowly than small-scale fading, the system resource consumed for acquiring large-scale fading coefficients is usually marginal as compared with that for small-scale fading coefficients. Such a difference between large-scale and small-scale fadings, however, has not been appropriately handled in the existing approaches for training-based multiuser MIMO systems. In this paper, we ignore the overhead for large-scale fading acquisition by assuming that the large-scale fading coefficients are known a priori, whereas the small-scale fading coefficients are estimated via training. Our target is to maximize the mutual information lower bound (MILB) of a training-based uplink multiuser MIMO system over various system parameters, including the training length, the training signals, the number of users, and the power allocation between training and data transmission. As the exact solution is difficult, we establish upper and lower bounds of the MILB. Specifically, we use the majorization theory to establish an upper bound of MILB. We then propose a lower bound of the MILB by deactivating a portion of users with the smallest large-scale fading gains and assigning orthogonal training sequences to the rest of the users. We show that this simple lower bound is asymptotically optimal for MILB maximization in the high signal-to-noise ratio (SNR) regime. We also show that the upper and lower bounds are reasonably tight in the finite SNR regime under various system configurations. Furthermore, we derive the optimal training design for MILB maximization in uplink multiuser MIMO with uniform large-scale fading. Numerical results are presented to verify our analysis.

Load balancing is an effective approach to address the spatial-temporal fluctuation problem of mobile data traffic for cellular networks. The existing schemes that focus on channel borrowing from neighboring cells cannot be directly applied to the future 5G wireless networks, because the neighboring cells will reuse the same spectrum band in 5G systems. In this paper, we consider an orthogonal frequency division multiple access ultra-dense small cell network, where device-to-device (D2D) communication is advocated to facilitate load balancing without extra spectrum. Specifically, the data traffic can be effectively offloaded from a congested small cell to other underutilized small cells by D2D communications. The problem is naturally formulated as a joint resource allocation and D2D routing problem that maximizes the system sum-rate. To efficiently solve the problem, we decouple the problem into a resource allocation subproblem and a D2D routing subproblem. The two subproblems are solved iteratively as a monotonic optimization problem and a complementary geometric programming problem, respectively. Simulation results show that the data sum-rate in the neighboring small cells increases 20% on average by offloading the data traffic in the congested small cell to the neighboring small cell base stations.

In this paper, we investigate the impact of channel sparsity on the fundamental limits of massive multiple-input multiple-output (MIMO) systems with K single-antenna transmit terminals and an N-antenna receiver, where "massive" means N >> K >> 1, and "sparsity" means that a large portion of the channel coefficients are zeros. We propose a novel blind detection scheme that simultaneously estimates the channel and data through factorizing the received signal matrix. We show that the proposed scheme can achieve a degrees of freedom (DoF) arbitrarily close to K(1-1/T) with T being the channel coherence time, provided that N is sufficiently large and the channel is sufficiently sparse. This achievable DoF has a fractional gap of only 1/T from the ideal DoF of K, which is a remarkable advance for understanding the performance limit of the massive MIMO system. Furthermore, we consider the algorithm design for the proposed blind detection scheme. Numerical results demonstrate that the proposed blind detection scheme significantly outperforms its counterpart schemes in the practical SNR regime.

The massive integration of electric vehicles (EVs) will pose great challenges to the stability and efficiency of both the conventional power networks and transportation systems. The recently emerging smart grid technology, which integrates advanced communication, control, and charging infrastructures, provides promising solutions to tackle these challenges. In this paper, we consider a smart-grid enabled EV network with heterogeneous charging facilities of different charging costs and capabilities, e.g., allowing EV to sell energy back to the grid. In this case, an EV user needs to decide which path to take, and where and how much to charge/discharge its battery at charging stations in the chosen path such that its journey can be accomplished with the minimum monetary cost. From the system operator's perspective, we study a joint optimization of the routing selection and charging schedules to maximize the overall consumer surplus of a set of EVs. To reduce the computational complexity of the system operator and signaling exchange, we propose a distributed scheme such that each user can maximize its own profit, and the system operator can also achieve the maximum consumer surplus through limited signaling exchange with the EV users. Our simulation shows that the proposed algorithm could efficiently save energy cost of the users and improve the usage of renewable energy in the power network.

Load balancing is an effective approach to address the spatial-temporal fluctuation of mobile traffic in cellular networks. However, existing schemes that rely on channel borrowing from neighboring cells cannot be directly applied to current LTE-A systems where all cells are deployed with the same spectrum band. In this paper, we consider a multi-cell OFDMA network, where Device-to-Device (D2D) communication is opted for load balancing without the need of extra spectrum. Specifically, the traffic can be steered from a congested cell to another underutilized cells by D2D communications. The problem is naturally formulated as a joint resource allocation and D2D routing problem that maximizes the system sum-rate. In particular, we formulate the resource allocation as a Monotonic optimization problem and the D2D routing as a complementary Geometric Programming problem. The two subproblems are solved iteratively to obtain the joint resource allocation and D2D routing solution. Simulation results demonstrate that the D2D load balancing can improve the system sum-rate efficiently.

We consider efficient communications over the multiple-input multiple-output (MIMO) multiway distributed relay channel (MDRC) with full data exchange, where each user, equipped with multiple antennas, broadcasts its message to all the other users via the help of a number of distributive relays. We propose a physical-layer network coding (PNC) based scheme involving linear precoding for channel alignment, nested lattice coding for PNC, and lattice-based precoding for interference mitigation. We show that distributed relaying achieves the same sum-rate as cooperative relaying in the high SNR regime in most scenarios, which implies that the proposed scheme with distributed relays is more suitable for practical systems than the schemes with cooperative relays.

Wireless powered communication network (WPCN) is a novel networking paradigm that uses radio frequency (RF) wireless energy transfer (WET) technology to power the information transmissions of wireless devices (WDs). When energy and information are transferred in the same frequency band, a major design issue is transmission scheduling to avoid interference and achieve high communication performance. Commonly used centralized scheduling methods in WPCN may result in high control signaling overhead and thus are not suitable for wireless networks constituting a large number of WDs with random locations and dynamic operations. To tackle this issue, we propose in this paper a distributed scheduling protocol for energy and information transmissions in WPCN. Specifically, we allow a WD that is about to deplete its battery to broadcast an energy request buzz (ERB), which triggers WET from its associated hybrid access point (HAP) to recharge the battery. If no ERB is sent, the WDs contend to transmit data to the HAP using the conventional p-persistent CSMA (carrier sensing multiple access). In particular, we propose an energy queueing model based on an energy decoupling property to derive the throughput performance. Our analysis is verified through simulations under practical network parameters, which demonstrate good throughput performance of the distributed scheduling protocol and reveal some interesting design insights that are different from conventional contention-based communication network assuming the WDs are powered with unlimited energy supplies.

Cloud radio access network (C-RAN) is a promising architecture for unprecedented capacity enhancement in next-generation wireless networks thanks to the centralization and virtualization of base station processing. However, centralized signal processing in C-RANs involves high computational complexity that quickly becomes unaffordable when the network grows to a huge size. First, this paper endeavors to design a scalable uplink signal detection algorithm, in the sense that both the complexity per unit network area and the total computation time remain constant when the network size grows. To this end, we formulate the signal detection in C-RAN as an inference problem over a bipartite random geometric graph. By passing messages among neighboring nodes, message passing (a.k.a. belief propagation) provides an efficient way to solve the inference problem over a sparse graph. However, the traditional message-passing algorithm is not guaranteed to converge, because the corresponding bipartite random geometric graph is locally dense and contains many short loops. As a major contribution of this paper, we propose a randomized Gaussian message passing (RGMP) algorithm to improve the convergence. Instead of exchanging messages simultaneously or in a fixed order, we propose to exchange messages asynchronously in a random order. The proposed RGMP algorithm demonstrates significantly better convergence performance than conventional message passing. The randomness of the message updating schedule also simplifies the analysis, and allows the derivation of the convergence conditions for the RGMP algorithm. In addition, we generalize the RGMP algorithm to a blockwise RGMP (B-RGMP) algorithm, which allows parallel implementation. The average computation time of B-RGMP remains constant when the network size increases.

We consider training-based channel estimation for a cloud radio access network (CRAN), in which a large amount of remote radio heads and users are randomly scattered over the service area. In this model, assigning orthogonal training sequences to all users will incur a substantial overhead to the overall network, and is even impossible when the number of users is large. Therefore, in this paper, we introduce the notion of local orthogonality, under which the training sequence of a user is orthogonal to those of the other users in its neighborhood. We model the design of locally orthogonal training sequences as a graph coloring problem. Then, based on the theory of random geometric graph, we show that the minimum training length scales in the order of ln K, where K is the number of users covered by a CRAN. This implies that the proposed training design yields a scalable solution to sustain the need of largescale cooperation in CRANs.

We consider training-based channel estimation for cloud radio access networks (CRANs), in which a large amount of remote radio heads (RRHs) and users are randomly scattered over a certain service area. In this model, assigning orthogonal training sequences to all users, if possible, will cause a substantial overhead to the overall network. Instead, we introduce the notion of local orthogonality, in which the training sequence of a user is required to be orthogonal to the training sequences of other users in its neighborhood. We model the design of locally orthogonal training sequences as a graph coloring problem. Then, based on the theory of random geometric graph, we show that the minimum training length scales in the order of ln K, where K is the number of users covered by the CRAN. Therefore, the proposed training design yields a scalable solution to sustain the need of large-scale cooperation in CRANs.

In Cloud Radio Access Networks (C-RANs), the high computational complexity of signal processing becomes unaffordable due to the large number of remote radio heads (RRHs) and users. This paper proposes a randomized Gaussian message-passing (RGMP) algorithm to reduce the complexity of uplink signal processing in C-RANs. Specifically, we first propose to use Gaussian message passing to reduce the computational complexity. In C-RANs, RRHs only need to detect signals from nearby users as the signals from distant users are very weak and can be ignored. Thus, in message-passing algorithms, messages only need to be exchanged among nearby RRHs and users. This leads to a linear computational complexity with the number of RRHs and users. Then, to improve the convergence of message passing, we propose to exchange messages in a random order instead of exchanging them simultaneously or in a fixed order. Numerical results show that the proposed RGMP algorithm has better convergence performance than conventional message passing. The randomness of the message update schedule also simplifies the analysis, which allows us to derive some convergence conditions for the RGMP algorithm. Besides analysis, we also compare the convergence rate of RGMP with existing low-complexity algorithms through extensive simulations.

Featured by centralized processing and cloud-based infrastructure, cloud radio access network (C-RAN) has emerged as a promising solution to handle the data proliferation in future wireless networks. However, the attractive capacity enhancement brought by large-scale centralized processing comes along with increased computing resource requirement in the baseband unit (BBU) pool. Thus, computing resource as another dimension of manageable resource needs to be considered in resource allocation and C-RAN system design. In this paper, we first characterize the relationship between PHY transmission characteristics and the required computing resource in the BBU pool. Based on this, we propose a feasible algorithm to maximize the network sum-rate under limited computing resource constraint, which is a binary-integer non-linear programming (BINLP) problem with non-convex constraints in nature. Numerical results show the significant impact of computing resource on both user-RRH association strategy and the achievable sum-rate performance.

In this paper, we endeavour to maximize the throughput of training-based multiuser multiple-input multiple-output (MIMO) systems. In a multiuser MIMO system, users are geographically separated. So, the near-far effect plays an indispensable role in channel fading. The existing optimal training design for conventional MIMO does not take the near-far effect into account, and thus is not applicable to a multiuser MIMO system. In this work, we use the majorization theory as a basic tool to study the tradeoff between the channel estimation quality and the information throughput. We establish tight upper and lower bounds of the throughput. Due to the near-far effect, the optimal training design for throughput maximization is to deactivate a portion of users with the weakest channels in transmission. This observation shed light on the practical design of training-based multiuser MIMO systems.

Featured by centralized processing and cloud based infrastructure, Cloud Radio Access Network (C-RAN) is a promising solution to achieving an unprecedented system capacity in future wireless cellular networks. The huge capacity gain mainly comes from the centralized and coordinated signal processing at the cloud server. However, full-scale coordination in a large-scale C-RAN requires the processing of very large channel matrices, leading to high computational complexity and channel estimation overhead. To tackle this challenge, we establish a unified theoretical framework for dynamic clustering by exploiting the near-sparsity of large C-RAN channel matrices. Based on this framework, we propose a dynamic nested clustering (DNC) algorithm that greatly improves the system scalability in terms of baseband-processing and channel-estimation complexity. With the proposed DNC algorithm, we show that the computational complexity (i.e., the computation time with serial processing) for the optimal linear detector is significantly reduced from O(N3) to O(N2), where N is the number of remote radio heads (RRHs) in the C-RAN. Moreover, the proposed DNC algorithm is also amenable to parallel processing, which further reduces the computation time to O(N42/23).

The multi-hop multi-flow transmission has been proposed as a promising solution to cope with the spectrum scarcity in densely populated user environments. Due to the mutual interference between different flows and different hops of the same flow, the resource allocation for multi-hop multi-flow wireless networks is in general non-convex, and thus cannot be solved by conventional convex optimization techniques. In this paper, we propose an algorithm to effectively solve the resource allocation problem by jointly optimizing the rate control and scheduling. Specifically, we show that the problem can be decomposed into a set of problems that maximizes the weight-sum-flow rate at each slot. Furthermore, to solve the non-convex weighted sum flow rate maximization problem, we exploit its hidden monotonicity and develop a global optimal rate control and scheduling (G-RCS) algorithm based on the theory of monotonic optimization. Our analysis shows that the proposed G-RCS algorithm is guaranteed to converge to an optimal solution in a finite number of iterations. To reduce the complexity, we propose an accelerated algorithm, referred to as the A-G-RCS, based on the inherent symmetry of the optimal solution. Numerical results validate that the proposed algorithms can serve as a performance benchmark for the existing heuristic algorithms.

Cloud radio access network (C-RAN) emerges as a promising solution to sustain the mobile data explosion with low cost and high energy efficiency. The centralized base band processing of C-RAN facilitates coordinated signal processing at the cloud server, which can potentially lead to huge capacity gain. However, full-scale coordination in a large-scale system inevitably results in high computational complexity that limits the scalability of the system. To address this issue, this paper proposes a scalable uplink signal processing algorithm based on message passing. By exploiting near-sparsity of large C- RAN channel matrices, we derive a sparse message- passing algorithm that reduces the computational complexity of signal detection to be linear with the number of RRHs and users. This implies that the average computational complexity per user does not grow with the network size, and hence the system is scalable. In addition, we discuss the convergence of the sparse message-passing algorithm and propose a block-wise message-passing algorithm that significantly improves the probability of convergence.

Device-to-Device (D2D) is an emerging technology that is typically employed as an underlay of the uplink (UL) cellular networks. The downlink cellular spectrum, however, is rarely reused by D2D links due to the strong interference from the base station to the D2D receivers. To this end, we propose to enable D2D transmissions in the downlink by multiplexing D2D and cellular users using beamforming techniques. In particular, a cross-layer design approach is adopted to jointly optimize beamforming, spectrum allocation, and power control, so that the total system transmission power is minimized. To deal with the non-convex and combinatorial nature of the problem, we transform the formulation into a convex optimization problem by identical deformation and relaxations, and propose a semidefinite relaxation (SDR) based algorithm to approximate the optimal solution. Moreover, we focus on the feasibility of the convex problem, and propose an improved algorithm which reduces the probability of users out of service. The simulation results show that the performance of our proposed algorithm is close to the optimal solution, and the improved algorithm gives a higher efficiency on user admission control.

With the emergence of new applications, e.g., video gaming, file sharing and proximity aware social networks, high data-rate traffics will be likely to occur between devices in the vicinity of each other. Enabling Device to Device (D2D) communication between such devices can efficiently offload these traffics from the base station. Meanwhile, by reusing the cellular spectrum, D2D has the potential to significantly improve the spectral efficiency. However, if not controlled carefully, the interference caused by D2D communication may severely degrade the performance of regular cellular services. To resolve this issue, this paper proposes a simple guard-zone based mode selection scheme as an efficient and scalable mechanism to manage the interference in D2D underlaid cellular systems. With the help of stochastic geometry, we rigorously analyze the interference and throughput of such system. The results indicate the existence of an optimal guard zone radius that maximizes the total system throughput. Thanks to the tractability of our analysis, the optimal guard zone radius can be efficiently obtained. Through extensive numerical analysis, we show that the success probability and aggregate throughput obtained at the optimal guard zone radius are significantly greater than those without guard zone.

Small-cell architecture is widely adopted by cellular network operators to increase network capacity. By reducing the size of cells, operators can pack more (low-power) base stations in an area to better serve the growing demands, without causing extra interference. However, this approach suffers from low spectrum temporal efficiency. When a cell becomes smaller and covers fewer users, its total traffic fluctuates significantly due to insufficient traffic aggregation and exhibiting a large "peak to-mean" ratio. As operators customarily provision spectrum for peak traffic, large traffic temporal fluctuation inevitably leads to low spectrum temporal efficiency. In this work, we first carryout a case-study based on real-world 3G data traffic traces and confirm that 90% of the cells in a metropolitan district are less than 40% utilized. Our study also reveals that peak traffic of adjacent cells are highly asynchronous. Motivated by these observations, we advocate device-to-device (D2D) load-balancing as a useful mechanism to address the fundamental drawback of small-cell architecture. The idea is to shift traffic from a congested cell to its adjacent under-utilized cells by leveraging inter-cell D2D communication, so that the traffic can be served without using extra spectrum, effectively improving the spectrum temporal efficiency. We provide theoretical modeling and analysis to characterize the benefit of D2D load balancing, in terms of sum peak traffic reduction of individual cells. We also derive the corresponding cost, in terms of incurred D2D traffic overhead. We carry out empirical evaluations based on real-world 3G data traces to gauge the benefit and cost of D2D load balancing under practical settings. The results show that D2D load balancing can reduce the sum peak traffic of individual cells by 35% as compared to the standard scenario without D2D load balancing, at the expense of 45% D2D traffic overhead.

In spectrum underlay cognitive radio networks, secondary users (SUs) are allowed to reuse the spectrum allocated to a primary system. The interference between SUs actually carries information and can potentially be exploited to improve the network performance through information-theoretic interference processing. In this paper, we design an optimal joint power and rate control algorithm that maximizes the secondary system utility subject to the interference temperature constraints of primary users based on the capacity-approaching interference processing scheme called as the Han-Kobayashi scheme. The optimal solution is difficult to achieve because the optimization problem is in general non-convex. To make the optimization problem tractable, this paper first transforms the problem into a monotonic optimization problem through exploiting its hidden monotonicity. We then devise an effective algorithm to obtain the global optimal solution to the joint power and rate control problem in the Han-Kobayashi scheme. The key idea behind the proposed algorithm is to construct a sequence of shrinking polyblocks that approximate the upper boundary of the feasible region with increasing precision. Numerical results further show that the achieved utility of our scheme significantly outperforms the utility of conventional schemes which treat the interference between SUs as the noise.

Featured by centralized processing and cloud based infrastructure, Cloud Radio Access Network (C-RAN) is a promising solution to achieve an unprecedented system capacity in future wireless cellular networks. The huge capacity gain mainly comes from the centralized and coordinated signal processing at the cloud server. However, full-scale coordination in a large-scale C-RAN requires the processing of very large channel matrices, leading to high computational complexity and channel estimation overhead. To resolve this challenge, we show in this paper that the channel matrices can be greatly sparsified without substantially compromising the system capacity. Through rigorous analysis, we derive a simple threshold-based channel matrix sparsification approach. Based on this approach, for reasonably large networks, the non-zero entries in the channel matrix can be reduced to a very low percentage (say 0.13% ∼ 2%) by compromising only 5% of SINR. This means each RRH only needs to obtain the CSI of a small number of closest users, resulting in a significant reduction in the channel estimation overhead. On the other hand, the high sparsity of the channel matrix allows us to design detection algorithms that are scalable in the sense that the average computational complexity per user does not grow with the network size.

We consider large-scale virtual multiple-input multiple-output (MIMO) systems, in which a large number of user terminals communicate with a large number of cooperative bases stations (BSs). We focus on a training-based scheme and investigate the throughput maximization over various system parameters including pilot symbols, the time allocation coefficient a, the power allocation coefficient γ, and the user number K. Our main contribution is to derive simple throughput expressions by utilizing the random matrix theory, based on which closed-form optimal solutions of (K, γ, α) are obtained. We show that, for a large but finite coherent time T, the optimal K for throughput optimization satisfies K > T/2 and converges to T/2 as the signal-to-noise ratio goes to infinity.

Accurate state estimation is of paramount importance to maintain the power system operating in a secure and efficient state. The recently identified coordinated data injection attacks to meter measurements can bypass the current security system and introduce errors to the state estimates. The conventional wisdom to mitigate such attacks is by securing meter measurements to evade malicious injections. In this paper, we provide a novel alternative to defend against false data injection attacks using covert power network topological information. By keeping the exact reactance of a set of transmission lines from attackers, no false data injection attack can be launched to compromise any set of state variables. We first investigate from the attackers' perspective the necessary condition to perform an injection attack. Based on the arguments, we characterize the optimal protection problem, which protects the state variables with minimum cost, as a well-studied Steiner tree problem in a graph. In addition, we also propose a mixed defending strategy that jointly considers the use of covert topological information and secure meter measurements when either method alone is costly or unable to achieve the protection objective. A mixed-integer linear programming formulation is introduced to obtain the optimal mixed defending strategy. To tackle the NP-hardness of the problem, a tree-pruning-based heuristic is further presented to produce an approximate solution in polynomial time. The advantageous performance of the proposed defending mechanisms is verified in IEEE standard power system test cases.

A well-functioning power system relies on its accurate state estimation. However, recent research shows that well-structured false-data injection attacks can bypass the current security system and introduce arbitrary errors to state estimates. In this paper, we propose a novel defending mechanism against false-data injection attacks using covert power network topological information. By keeping the exact reactance of a set of transmission lines from attackers, no false data injection attack can be launched to compromise any set of critical state estimates. We first investigate from the attackers' perspective the necessary condition to perform injection attack. Based on the arguments, we characterize the optimal protection problem, which protects the state estimates with minimum number of covert transmission lines, into a minimum Steiner tree problem in a graph, where some off-the-shelf algorithms can be applied. Besides, we also propose a mixed defending strategy that jointly considers the use of covert topological information (CTI) and secure meter measurements when CTI protection alone fails to achieve the protection objective due to technical constraints. The advantageous performance of the proposed defending mechanisms is verified in IEEE standard power system testcase. Our results here will be useful in the security upgrade projects of large-scale electrical power system towards smart power grids.

The normal operation of electricity market requires accurate state estimation of the power grids. However, recent research shows that carefully synthesized false-data injection attacks can easily introduce errors to state estimates without being detected by the current security systems. In this paper, we analyze and formulate an effective false-data injection attack to control real-time electricity price at any tagged bus. It is observed that an adversary capable of false-data injection attack can induce false real-time electricity price by fabricating biased transmission congestion pattern. From the strategic level, we propose a simple algorithm that finds the effective congestion pattern with minor distortion to the normal system operation. For practical implementation, load redistribution attack, a special false-data injection attack which produces biased load estimates, is used to achieve the derived congestion pattern. We also propose a cost-aware neighborhood load redistribution attack, which only compromises limited measurements around the tagged bus. From theory to practice, our results here reveal the potential cyber vulnerabilities of current electricity market and would help to build more secure smart power grid in the future.

We present a novel physical layer encryption (PLE) scheme that randomizes the radio signals using a discrete Fourier transform (DFT) based encryption algorithm. For any baseband signaling method, we show that perfect secrecy is asymptotically achievable with the proposed DFT-based encryption method when the signal block length (N) approaches infinity. For practical systems with finite N, we also show that the proposed encryption method can transmit at a secrecy rate close to the main channel's achievable data rate. In this sense, transmission privacy is achieved without compromising the capability of the communication channel. Besides, the proposed encryption method can hide the transmission data rate and is immune to all existing upper-layer attacks. The performance advantages of the proposed DFT-based encryption method is verified through comparisons against other existing PLE methods.

In this paper, we study energy-efficient transmissions for multiple-input multiple-output (MIMO) cognitive radio (CR) networks in which the secondary unlicensed users coexist with the primary licensed users. We want to optimize the time allocations and beamforming vectors for the secondary users (SUs), in order to minimize the energy consumption of the SUs while satisfying the SUs' rate requirements and the primary receivers' interference constraints. Compared with the tradition MIMO networks, the challenge here is that the SUs may not always be able to obtain the channel state information (CSI) to the primary receivers. We are interested in two different scenarios. The first is when the SUs have the luxury of knowing the CSI to the primary receivers, and the second is when the SUs do not have such an luxury. The corresponding optimization formulations involve joint time scheduling and beamforming, which are non-convex and are complicated to solve. Fortunately, we show that when the SUs are not able to obtain the CSI, the optimal time allocation and the optimal beamforming vectors can be found very efficiently in polynomial-time through a proper decomposition. When the SUs have perfect knowledge about the CSI, we show that the optimal solutions can still be obtained in polynomial time when the secondary system is under-utilized. If the traffic load to the secondary system is heavy, we propose a polynomial-time heuristic to generate a near-optimal solution. The simulation results show that our proposed energy-optimal-transmission algorithms can achieve an energy-saving of 30% to 91%, compared with the simplistic maximum-rate transmission policy, depending on the secondary system's traffic load.

This paper proposes a real-time pricing scheme that reduces the peak-to-average load ratio through demand response management in smart grid systems. The proposed scheme solves a two-stage optimization problem. On one hand, each user reacts to prices announced by the retailer and maximizes its payoff, which is the difference between its quality-of-usage and the payment to the retailer. On the other hand, the retailer designs the real-time prices in response to the forecasted user reactions to maximize its profit. In particular, each user computes its optimal energy consumption either in closed forms or through an efficient iterative algorithm as a function of the prices. At the retailer side, we develop a Simulated-Annealing-based Price Control (SAPC) algorithm to solve the non-convex price optimization problem. In terms of practical implementation, the users and the retailer interact with each other via a limited number of message exchanges to find the optimal prices. By doing so, the retailer can overcome the uncertainty of users' responses, and users can determine their energy usage based on the actual prices to be used. Our simulation results show that the proposed real-time pricing scheme can effectively shave the energy usage peaks, reduce the retailer's cost, and improve the payoffs of the users.

Being the predominant air interface of next-generation wireless standards, orthogonal frequency division multiple access (OFDMA) is well known for its flexibility in allocating subcarriers to different mobile users according to their different fast channel variations. Numerous research studies have demonstrated that OFDMA can bring substantial capacity gain when the subcarriers are optimally allocated. Nonetheless, practical systems can hardly afford optimal subcarrier allocation, because frequent re-optimization performed at the same timescale as fast fading variation would lead to excessively high computational and signaling costs. As a result, most practical systems settle for low-complexity schemes that operate far from the optimum, thus making them unable to enjoy the large capacity gain predicted by theoretical studies. To address this problem, we propose a novel alternative, termed the slow adaptive OFDMA, to drastically reduce the computational and signaling costs. The proposed scheme adapts subcarrier allocation at a much slower timescale than that of channel fading variation, yet achieves similar system capacity and quality of service (QoS) levels as the optimal fast adaptive OFDMA. Moreover, it possesses several attractive features. First, neither prediction of channel state information nor specification of channel fading distribution is needed for subcarrier allocation. As such, the algorithm is robust against any mismatch between actual channel state/distributional information and the one assumed. Secondly, although the optimization problem arising from our proposed scheme is non-convex in general, based on recent advances in chance-constrained optimization, we show that it can be approximated by a certain linear program with provable performance guarantees. In particular, we only need to handle an optimization problem that has the same structure as the fast adaptive OFDMA problem, yet we are able to enjoy lower computational and signaling costs. Last but not the least, instead of relying on standard but abstract linear program solvers such as the interior-point method to solve the aforementioned linear program, we can exploit its special structure and design a provably efficient algorithm for it. The proposed algorithm not only has a transparent engineering interpretation but is also easy to implement at the base stations of practical systems.

In wireless location-aware networks, mobile nodes (agents) typically obtain their positions using the range measurements to the nodes with known positions. Transmit power allocation not only affects network lifetime and throughput, but also determines localization accuracy. In this paper, we present an optimization framework for robust power allocation in network localization with imperfect knowledge of network parameters. In particular, we formulate power allocation problems to minimize localization errors for a given power budget and show that such formulations can be solved via conic programming. Moreover, we design a distributed power allocation algorithm that allows parallel computation among agents. The simulation results show that the proposed schemes significantly outperform uniform power allocation, and the robust schemes outperform their non-robust counterparts when the network parameters are subject to uncertainty .

Multicast/broadcast is regarded as an efficient technique for wireless cellular networks to transmit a large volume of common data to multiple mobile users simultaneously. To guarantee the quality of service for each mobile user in such single-hop multicasting, the base-station transmitter usually adapts its data rate to the worst channel condition among all users in a multicast group. On one hand, increasing the number of users in a multicast group leads to a more efficient utilization of spectrum bandwidth, as users in the same group can be served together. On the other hand, too many users in a group may lead to unacceptably low data rate at which the base station can transmit. Hence, a natural question that arises is how to efficiently and fairly transmit to a large number of users requiring the same message. This paper endeavors to answer this question by studying the problem of multicasting over multicarriers in wireless orthogonal frequency division multiplexing (OFDM) cellular systems. Using a unified utility maximization framework, we investigate this problem in two typical scenarios: namely, when users experience roughly equal path losses and when they experience different path losses, respectively. Through theoretical analysis, we obtain optimal multicast schemes satisfying various throughput-fairness requirements in these two cases. In particular, we show that the conventional multicast scheme is optimal in the equal-path-loss case regardless of the utility function adopted. When users experience different path losses, the group multicast scheme, which divides the users almost equally into many multicast groups and multicasts to different groups of users over nonoverlapping subcarriers, is optimal .

Exponential backoff (EB) is a widely adopted collision resolution mechanism in many popular random-access networks including Ethernet and wireless LAN (WLAN). The prominence of EB is primarily attributed to its asymptotic throughput stability, which ensures a non-zero throughput even when the number of users in the network goes to infinity. Recent studies, however, show that EB is fundamentally unsuitable for applications that are sensitive to large delay and delay jitters, as it induces divergent second- and higher-order moments of medium access delay. Essentially, the medium access delay follows a power law distribution, a subclass of heavy-tailed distribution. To understand and alleviate the issue, this paper systematically analyzes the tail delay distribution of general backoff functions, with EB being a special case. In particular, we establish a tradeoff between the tail decaying rate of medium access delay distribution and the stability of throughput. To be more specific, convergent delay moments are attainable only when the backoff functions g(k) grow slower than exponential functions, i.e., when g(k)∈ o(r^k) for all r>1. On the other hand, non-zero asymptotic throughput is attainable only when backoff functions grow at least as fast as an exponential function, i.e., g(k)∈Ω(r^k) for some r>1. This implies that bounded delay moments and stable throughput cannot be achieved at the same time. For practical implementation, we show that polynomial backoff (PB), where g(k) is a polynomial that grows slower than exponential functions, obtains finite delay moments and good throughput performance at the same time within a practical range of user population. This makes PB a better alternative than EB for multimedia applications with stringent delay requirements.

Heterogeneous cellular network (Hetnets), where various classes of low power base stations (BS) are underlaid in a macro-cellular network, is a promising technique for future green communications. These new types of BSs can achieve substantial improvement in spectrum-efficiency and energy-efficiency via cell splitting. However, mobile stations perceive different channel gains to different base stations. Therefore, it is important to associate a mobile station with the right BS so as to achieve a good communication quality. Oftentimes, the already-challenging BS association problem is further complicated by the need of transmission power control, which is an essential component to manage co-channel interference in many wireless communications systems. Despite its importance, the joint BS association and power control (JBAPC) problem has remained largely unsolved, mainly due to its non-convex and combinatorial nature that makes the global optimal solution difficult to obtain. This paper aims to circumvent this difficulty by proposing a novel algorithm based on Benders' Decomposition to solve the non-convex JBAPC problem efficiently and optimally. In particular, we endeavor to maximize the system revenue and meanwhile associate every served mobile station with the right BS with the minimum total transmission power. We first propose a single-stage formulation that captures the two objectives simultaneously. The problem is then transformed in a way that can be efficiently solved using the proposed joint BS Association and poweR coNtrol algorithm (referred to as BARN) that is derived from classical Benders' Decomposition. Finally, we derive a closed-form analytical formula to characterize the effect of the termination criterion of the algorithm on the gap between the obtained solution and the optimal one. For practical implementation, we further propose an Accelerated BARN (A-BARN) algorithm that can significantly reduce the computational time. By carefully choosing the termination criterion, both BARN and A-BARN are guaranteed to converge to the global optimal solution.

In wireless cellular systems, it is an important and challenging task to reliably multicast to numerous users that require the same contents at one transmission. In this paper, we propose a network coding based multicast scheme for wireless cellular OFDM systems. The base station encodes source packets with linear network coding and multicasts the coded packets to the target users over multicarriers. Thus, the users can correctly recover the source message as long as they successfully receive a certain number of packets. The reliability of coded wireless multicast is characterized by the user outage probability, which can be greatly reduced by efficiently exploiting frequency diversity gain via network coding. We show that the BS shall adjust the data transmission rate per carrier to strike a good balance between the reliability at each subcarrier and the redundancy among coded packets. It is also found that full diversity gain and no diversity gain should be exploited in the high and low SNR regimes, respectively. Simulation results also reveal that network coding generally provides great advantage for reliable wireless multicasting to a large number of users.

The IEEE 802.11 wireless local area network (WLAN) standard was originally designed for best-effort services, targeting at providing high throughput and throughput fairness. However, high system throughput does not necessarily translate to good delay performance. Recent studies show that exponential backoff, the key collision avoidance mechanism in distributed coordination function (DCF) of 802.11, is fundamentally defected in the sense that it induces divergent moments of medium access delay. Essentially, the medium access delay follows a power law distribution, a subclass of heavy tailed distribution. With practical system configurations, the delay variance can easily approach infinity, which translates to service starvation of some users and eventually leads to severe unfairness among users. In this paper, we show that the power law delay distribution can be mitigated if exponential backoff is replaced by a polynomial backoff mechanism. Through rigorous analysis, we find that all delay moments are finite with polynomial backoff, thus fundamentally solving the problem of starvation and unfairness. In addition, polynomial backoff yields a similarly high throughput as exponential backoff with a practical network size when the order of the polynomial backoff function is set reasonably. In this sense, we argue that polynomial backoff is a better alternative than exponential backoff in IEEE 802.11 DCF, especially when there are increasingly more broadband multimedia traffics with stringent delay requirements in the network.

In wireless localization systems, mobile nodes (agents) typically obtain their positions through ranging with respect to fixed infrastructure (anchors). Transmission power allocation not only affects network lifetime, throughput and interference, but also determines the localization accuracy. In this paper, we develop a robust anchor power allocation strategy to combat imperfect network topology parameters. We formulate the problem to minimize the squared position error bound (SPEB), which characterizes the fundamental limit of localization accuracy, and show that such formulation can be efficiently solved via semidefinite programming (SDP). The simulation results show that the proposed robust scheme significantly outperforms both non-robust scheme and uniform power allocation.

The authors propose a cross-layer approach, which exploits distributed beamforming in the physical layer and token passing in the media access control (MAC) layer, to improve quality of service (QoS) for secondary users (SUs) with bursty traffics in cognitive relay systems. In this scheme, source-to-destination transmissions are relayed by some SU nodes, which can form a distributed beamformer to forward messages in busy timeslots while completely eliminating interference to primary users (PUs). In contrast with previous cognitive relaying protocols, this scheme can utilise more spectrum resources, namely idle timeslots (or temporal spectrum holes) as well as busy timeslots (or spatial spectrum holes). Based on a token passing mechanism, an opportunistic scheduling protocol is then developed to dynamically balance available spectrum holes between the source and the relays, and hence adapts to bursty arrival of secondary traffics and random presence of PUs. By formulating a tandem queueing analytical framework, the performance of the proposed scheme is then analysed using a multi-dimensional Markov chain model. Numerical results demonstrate that the proposed scheme can achieve significant QoS gains over conventional cognitive relaying protocols that utilise only idle timeslots.

Cognitive Radio (CR) can significantly improve the utilization of the precious radio spectrum by allowing Secondary Users (SUs) to borrow the licensed spectrum if they do not cause harmful interference to Primary Users (PUs). As a wireless technology, CR confronts the challenges of wireless channels inevitably and thus wishes to employ node cooperation to achieve spatial diversity gain. However, conventional cooperative diversity technologies require two idle timeslots for each transmission. This implies two temporal spectrum holes are needed for each transmission when the technologies are applied to CR Networks (CRNs). This can cause severe delay, as temporal spectrum holes are only available from time to time in CRNs. In this paper, we present a cross-layer approach, where cooperative beamforming is adopted to forward messages in busy timeslots without causing interference to PUs, so as to achieve cooperative diversity gain and improve Quality of Service (QoS) for SUs without consuming additional idle timeslots or temporal spectrum holes. In the physical layer, the beamforming weight vector and the cooperative diversity gain are obtained using a geometric approach. The MAC layer of the cooperative communication in CRNs can be modeled by a tandem queue, where the source queue is the bottleneck. Therefore, we propose an optimal opportunistic priority scheduling scheme in the MAC layer, the timeout probability of which is obtained using an absorbing Markov chain. A cross-layer optimization of the transmission rate is then carried out to jointly reduce the timeout and outage probabilities. Its significant QoS gain is demonstrated by simulations.

Transmit power control in wireless networks has long been recognized as an effective mechanism to mitigate co-channel interference. Due to the highly non-convex nature, optimal power control is known to be difficult to achieve if a system utility is to be maximized. In our earlier paper , we have proposed a centralized optimal power control algorithm that obtains the global optimal solution for both concave and non-concave system utility functions. A question remained unanswered is whether such global optimal solution can be achieved in a distributed manner. This paper addresses the question by developing a Gibbs Sampling based Asynchronous distributed power control algorithm (referred to as GLAD). The proposed algorithm quickly converges to the global optimal solution regardless of the concavity, differentiability and monotonicity of the utility function. To further enhance the practicality of the algorithm, this paper proposes two variants of the GLAD algorithm, namely I-GLAD and NI-GLAD, to reduce message passing in two dimensions of communication complexity, i.e., time and space. In particular, I-GLAD, where the prefix "I" stands for Infrequent message passing, reduces the "time overhead" of message passing. The convergence of I-GLAD can be proved regardless of the reduction in the message passing rate. Meanwhile, NI-GLAD, where the prefix "N" stands for Neighborhood message passing, restricts the computation overhead related to message passing to a small neighborhood space. Our results show that the optimality of the solution obtained by NI-GLAD depends on the selection of the neighborhood size.

There have been increasing interests in integrating cooperative diversity into Cognitive Radios (CRs). However, conventional cooperative diversity protocols require at least two randomly available idle timeslots or temporal spectrum holes for one transmission, thus leading to limited throughput and/or large latency. In this paper, we propose a novel cross-layer approach for efficient scheduling in CR systems with bursty secondary traffics. Specifically, cooperative beamforming is exploited for Secondary Users (SUs) to access busy timeslots or spatial spectrum holes without causing interference to primary users. We first propose a basic cooperative beaMforming and Automatic repeat request aided oppoRtunistic speCtrum scHeduling (MARCH) scheme to balance available spectrum resources, namely temporal and spatial spectrum holes, between the source and the relays. To analyze the proposed scheme, we develop a tandem queuing framework, which captures bursty traffic arrival, dynamic availability of spectrum holes, and time-varying channel fading. The stable throughput region and the average delay are characterized using a structured matrix-analytical method. We then obtain delay optimal scheduling schemes for various scenarios by jointly optimizing the scheduling parameters. Finally, we propose a modified scheme, MARCH-IR, which combines MARCH with Incremental Relay selection to further improve the system performance. Simulation results reveal that the proposed schemes provide significant Quality of Service (QoS) gains over conventional scheduling schemes that access only temporal spectrum holes.

In multi-access wireless networks, transmission scheduling is a key component that determines the efficiency and fairness of wireless spectrum allocation. At one extreme, greedy opportunistic scheduling that allocates airtime to the user with the largest instantaneous channel gain achieves the optimal spectrum efficiency and transmission reliability but the poorest user-level fairness. At the other extreme, fixed TDMA scheduling achieves the fairest airtime allocation but the lowest spectrum efficiency and transmission reliability. To balance the two competing objectives, extensive research efforts have been spent on designing opportunistic scheduling schemes to reach certain tradeoff points between the two extremes by tuning the greediness in scheduling policy. In this paper and in contrast to the conventional wisdom, we find that in relay-assisted MIMO cellular networks, being greedy in user scheduling is unnecessary since it does not directly translate to larger diversity gain. When each mobile user has no less antennas than the base station, even fixed TDMA achieves the optimal diversity gain that is otherwise achievable by greedy opportunistic scheduling. In addition, by incorporating very limited opportunism, a simple TDMA-based scheme, named relaxed-TDMA, asymptotically achieves the same optimal system reliability in terms of outage probability as greedy opportunistic scheduling. This reveals a surprising fact: transmission reliability and user fairness are not necessarily contradicting each other in relay-assisted systems. They can be both achieved by the simple TDMA schemes. For practical implementations, we further propose a fully distributed algorithm to implement the relaxed-TDMA scheme. Our results here may find applications in the design of next-generation wireless communication systems with relay architectures such as LTE-advanced and WiMAX.

In this paper, frame error rate (FER) performance of enhanced verification-based decoding algorithm (EVA) is investigated for packet-based low-density parity-check (LDPC) codes over binary symmetric channel (BSC). By taking the false verification into account, a recursive statistical model is proposed to analyze the FER performance based on the packet-level analysis. Simulation results demonstrate that the proposed model works for various packet sizes and channel parameters of practical interest with different verification constraints.

This paper investigates the number of multipath components in realistic wireless indoor channels. A measurement campaign was performed to collect channel realizations in the Stata Center on the MIT campus. We develop an algorithm with low computational complexity to extract multipath components from the measured waveforms. Based on the results obtained from the measured data, we quantify the behavior of the number of paths with respect to both center frequency and bandwidth.

Many future wireless applications rely on the availability of position information for mobile wireless nodes (agents). Such information can be obtained through ranging and communication between agents and fixed infrastructure (anchors). Since the transmission power of the anchors affects network lifetime, throughput, and interference, in this paper we will investigate the problem of power allocation among anchors. We start with a formulation of minimizing the squared position error bound (SPEB), which characterizes the limits of location accuracy. However, the problem is difficult to solve due to its non-convexity. To address this issue, we leverage the insights obtained from the geometric interpretation of localization information and formulate the objective to maximum directional position error bound (DPEB). We first solve the problem for the single-agent case, and then extend the algorithm for the multiple-agent case. The simulation results show that the proposed scheme achieves close-to-optimal solution but with much lower computational complexity.

In this paper, we propose an opportunistic scheduling scheme to serve bursty traffics in cognitive radios, where cooperative beamforming is exploited to access busy timeslots or spatial spectrum holes to forward messages without causing interference to primary users. Specifically, based on cooperative beamforming in the physical layer and automatic repeat request for error recovery in the link layer, our proposed scheme strives to balance available spectrum resources, namely temporal and spatial spectrum holes, between the source and the relays so as to greatly reduce the average delay. To analyze the proposed scheme, we then develop a tandem queueing analytical framework, which captures bursty traffic arrival, dynamic availability of spectrum holes, and time-varying channel fading. By modelling it with a multi-dimensional Markov chain, the average delay is derived using a structured matrix-analytical method. Finally, we obtain delay optimal scheduling schemes by jointly optimizing the scheduling parameters. Simulation results reveal that the proposed scheme provides significant quality of service gains over conventional scheduling schemes that access only temporal spectrum holes.

We propose a cooperative beamforming aided incremental relaying scheme to improve spectrum efficiency of cognitive radio systems. In this scheme, the source and relays can utilize cooperative beamforming to activate packet retransmission in busy timeslots or spatial spectrum holes, if the destination fails to receive the packets transmitted from the source. Therefore, cooperative diversity gain is obtained without consuming extra idle timeslots. Given a packet loss constraint, our proposed scheme endeavors to further improve the system throughput by dynamically adjusting the maximum number of retransmissions. We derive the average throughput of the proposed scheme and obtain the maximum throughput by optimizing the scheduling parameters. Theoretical and simulation results reveal that the proposed scheme obtains a significant throughput gain compared to direct transmission as well as conventional incremental relaying schemes utilizing idle timeslots only.

In cognitive radio (CR) networks with multiple-input multiple-output (MIMO) links, secondary users (SUs) can exploit "spectrum holes" in the space domain to access the spectrum allocated to a primary system. However, they need to suppress the interference caused to primary users (PUs), as the secondary system should be transparent to the primary system. In this paper, we study the optimal secondary-link beamforming pattern that balances between the SU's throughput and the interference it causes to PUs. In particular, we aim to maximize the throughput of the SU, while keeping the interference temperature at the primary receivers below a certain threshold. Unlike traditional MIMO systems, SUs may not have the luxury of knowing the channel state information (CSI) on the links to PUs. This presents a key challenge for a secondary transmitter to steer interference away from primary receivers. In this paper, we consider three scenarios, namely when the secondary transmitter has complete, partial, or no knowledge about the channels to the primary receivers. In particular, when complete CSI is not available, the interference-temperature constraints are to be satisfied with high probability, thus resulting in chance constraints that are typically hard to deal with. Our contribution is fourfold. First, by analyzing the distributional characteristics of MIMO channels, we propose a unified homogeneous quadratically constrained quadratic program (QCQP) formulation that can be applied to all three scenarios, in which different levels of CSI knowledge give rise to either deterministic or probabilistic interference-temperature constraints. The homogeneous QCQP formulation, though non-convex, is amenable to semidefinite programming (SDP) relaxation methods. Secondly, we show that the SDP relaxation admits no gap when the number of primary links is no larger than two. A polynomial-time algorithm is presented to compute the optimal solution to the QCQP problem efficiently. Thirdly, we propose a randomized polynomial-time algorithm for constructing a near-optimal solution to the QCQP problem when there are more than two primary links. Finally, we show that when the secondary transmitter has no CSI on the links to primary receivers, the optimal solution to the QCQP problem can be found by a simple matrix eigenvalue-eigenvector computation, which can be done much more efficiently than solving the QCQP directly.

Opportunistic scheduling has been recognized as an effective method that significantly outperforms fixed TDMA scheduling in both channel capacity and communication reliability. Nevertheless the superior system performance comes at a price of high computational complexity, large signaling overhead, and unfairness among users. In this paper, we find that in relay-assisted next-generation wireless networks, fixed TDMA scheduling achieves the same diversity gain as opportunistic scheduling, as long as the best relay is carefully selected. Moreover, by introducing very limited opportunism, TDMA scheme yields the same optimal outage probability attained by full scale opportunistic scheduling at high SNR. In other words, we can safely enjoy the advantages of opportunistic scheduling without suffering its drawbacks. We also propose a simple distributed algorithm to implement our scheme. Our results here may find wide application in next-generation wireless communication systems, as relay-assisted cellular system is one of the major architectures in 4G wireless system included LTE or WiMAX.

Recent work has shown that distributed (or cooperative) beamforming can achieve cooperative gain, such as throughput gain and diversity gain, with no need for extra spectral holes in Cognitive Wireless Networks (CWNs). However, how to efficiently schedule cooperative beamforming to improve the quality of service of unlicensed secondary users has not been well addressed. In this paper, a simple opportunistic scheduling scheme is proposed to serve delay-sensitive traffics in CWNs with cooperative beamforming. After the probabilities of outages due to channel fading and random appearance of primary users are analyzed, respectively, the overall outage probability of our scheme is minimized by optimizing the scheduling parameter. The optimal scheduling scheme is then derived for the high-SNR regime. Simulation results show that compared to conventional schemes without cooperative beamforming, our scheduling scheme can significantly lower down the probability of message transmission failure within a given time period.

We consider a distributed power control algorithm for infrastructureless ad hoc wireless networks, where each link distributively and asynchronously updates its transmission power with limited message passing among links. This algorithm provably converges to the set of global optimal solutions despite the non-convexity of the power control problem. In contrast with existing distributed power control algorithms, our algorithm makes no stringent assumptions on the system utility functions. In particular, the utility function is allowed to be concave or non-concave, differentiable or non-differentiable, continuous or discontinuous, and monotonic or non-monotonic.

Cognitive radio (CR) is a powerful solution that can significantly improve the utilization of the precious limited radio spectrum. It allows secondary users (SUs) to opportunistically access spectral holes of the licensed spectrum without causing harmful interference to primary users (PUs). However, waiting for idle timeslots may induce very poor quality of service (QoS) for SUs. To alleviate this, an opportunistic relaying protocol exploiting distributed beamforming and token passing is proposed in this paper. We consider a cognitive radio network (CRN), where SUs constitute a two-hop relaying network. Specifically, a distributed beamforming method is applied to enable concurrent transmissions of PUs and SUs, thereby improving the opportunistic spectrum access. Our protocol applies a token passing mechanism in the MAC layer to dynamically balance transmission opportunities between two hops, and hence adapts to the random packet arrival and PUs' presence. We shall formulate a Markov chain to analyze the performance of this protocol. Numerical results show that our proposed protocol can significantly improve QoS of SUs in terms of the packet-loss rate and average delay, compared to conventional relaying protocols that utilize only silent timeslots.

With the rapid proliferation of broadband wireless services, it is of paramount importance to understand how fast data can be sent through a wireless local area network (WLAN). Thanks to a large body of research following the seminal work of Bianchi, WLAN throughput under saturated traffic condition has been well understood. By contrast, prior investigations on throughput performance under unsaturated traffic condition was largely based on phenomenological observations, which lead to a common misconception that WLAN can support a traffic load as high as saturation throughput, if not higher, under nonsaturation condition. In this paper, we show through rigorous analysis that this misconception may result in unacceptable quality of service: mean packet delay and delay jitter may approach infinity even when the traffic load is far below the saturation throughput. Hence, saturation throughput is not a sound measure of WLAN capacity under nonsaturation condition. To bridge the gap, we define safe-bounded-mean-delay (SBMD) throughput and safe-bounded-delay-jitter (SBDJ) throughput that reflect the actual network capacity users can enjoy when they require finite mean delay and delay jitter, respectively. Our earlier work proved that in a WLAN with multi-packet reception (MPR) capability, saturation throughput scales superlinearly with the MPR capability of the network. This paper extends the investigation to the nonsaturation case and shows that superlinear scaling also holds for SBMD and SBDJ throughputs. Our results here complete the demonstration of MPR as a powerful capacity-enhancement technique for WLAN under both saturation and nonsaturation conditions.

The rapid proliferation of wireless sensor networks has stimulated enormous research efforts that aim to maximize the lifetime of battery-powered sensor nodes and, by extension, the overall network lifetime. Most work in this field can be divided into two equally important threads, namely (i) energy-efficient routing that balances traffic load across the network according to energy-related metrics and (ii) sleep scheduling that reduces energy cost due to idle listening by providing periodic sleep cycles for sensor nodes. To date, these two threads are pursued separately in the literature, leading to designs that optimize one component assuming the other is pre-determined. Such designs give rise to practical difficulty in determining the appropriate routing and sleep scheduling schemes in the real deployment of sensor networks, as neither component can be optimized without pre-fixing the other one. This paper endeavors to address the lack of a joint routing-and-sleep-scheduling scheme in the literature by incorporating the design of the two components into one optimization framework. Notably, joint routing-and-sleep-scheduling by itself is a non-convex optimization problem, which is difficult to solve. We tackle the problem by transforming it into an equivalent Signomial Program (SP) through relaxing the flow conservation constraints. The SP problem is then solved by an iterative Geometric Programming (IGP) method, yielding an near optimal routing-and-sleep-scheduling scheme that maximizes network lifetime. To the best of our knowledge, this is the first attempt to obtain the optimal joint routing-and-sleep-scheduling strategy for wireless sensor networks. The near optimal solution provided by this work opens up new possibilities for designing practical and heuristic schemes targeting the same problem, for now the performance of any new heuristics can be easily evaluated by using the proposed near optimal scheme as a benchmark.

In interference-limited wireless networks where simultaneous transmissions on nearby links heavily interfere with each other, power control alone is not sufficient to eliminate strong levels of interference between close-by links. In this case, scheduling, which allows close-by links to take turns to be active, plays a crucial rule for achieving high system performance. Joint power control and scheduling that maximizes the system utility has long been a challenging problem. The complicated coupling between the signal-to-interference ratio of concurrently active links as well as the flexibility to vary power allocation over time gives rise to a series of non-convex optimization problems, for which the global optimal solution is hard to obtain. This paper is a first attempt to solve the non-convex joint power control and scheduling problems efficiently in a global optimal manner. In particular, it is the monotonicity rather than the convexity of the problem that we exploit to devise an efficient algorithm, referred to as S-MAPEL, to obtain the global optimal solution. To further reduce the complexity, we propose an accelerated algorithm, referred to as A-S-MAPEL, based on the inherent symmetry of the optimal solution. The optimal joint-power-control-andscheduling solution obtained by the proposed algorithms serves as a useful benchmark for evaluating other existing schemes. With the help of this benchmark, we find that on-off scheduling is of much practical value in terms of system utility maximization if "off-the-shelf" wireless devices are to be used.

Multi-packet reception (MPR) has been recognized as a powerful capacity-enhancement technique for randomaccess wireless local area networks (WLANs). As is common with all random access protocols, the wireless channel is often under-utilized in MPR WLANs. In this paper, we propose a novel multi-round contention random-access protocol to address this problem. This work complements the existing randomaccess methods that are based on single-round contention. In the proposed scheme, stations are given multiple chances to contend for the channel until there are a sufficient number of ¿winning¿ stations that can share the MPR channel for data packet transmission. The key issue here is the identification of the optimal time to stop the contention process and start data transmission. The solution corresponds to finding a desired tradeoff between channel utilization and contention overhead. In this paper, we conduct a rigorous analysis to characterize the optimal strategy using the theory of optimal stopping. An interesting result is that the optimal stopping strategy is a simple threshold-based rule, which stops the contention process as soon as the total number of winning stations exceeds a certain threshold. Compared with the conventional single-round contention protocol, the multi-round contention scheme significantly enhances channel utilization when the MPR capability of the channel is small to medium. Meanwhile, the scheme automatically falls back to single-round contention when the MPR capability is very large, in which case the throughput penalty due to random access is already small even with single-round contention.

In multi-cell networks where mobile stations perceive different channel gains to different base stations (BS), it is critical to associate a mobile station with the proper BS to maintain good communication quality with limited bandwidth resources. Oftentimes, the already-challenging BS association problem is further complicated by the need of transmission power control, which is an essential component to manage co-channel interference in many wireless communications systems. Despite its importance, joint BS association and power control (BAPC) problem has remained largely open, mainly due to its non-convex nature that makes the global optimal solution difficult to obtain. In this paper, we propose a novel algorithm, referred to as BARN, to solve the joint BAPC problem efficiently and optimally in the sense that the number of mobile stations in service is maximized and the total transmission power is minimized in the same time. In particular, we first propose a single-stage formulation that captures the two objectives simultaneously. Then, the problem is transformed in a way that can be efficiently solved using the BARN algorithm that is derived from the standard Benders' Decomposition. Finally, we derive a close-form analytical formula to characterize the effect of the termination criterion of the algorithm on the gap between the obtained solution and the optimal one. By carefully choosing the termination rule, the BARN algorithm can always converge to the global optimal solution.

Cognitive radio is a powerful solution that can significantly improve the utilization of the precious limited radio spectrum. It allows secondary users (SUs) to opportunistically access spectral holes in the licensed spectrum without causing harmful interference to primary users (PUs). However, the secondary communication opportunity becomes extremely poor when primary systems are heavily loaded. In this paper, a distributed beamforming method is proposed to allow concurrent transmissions of PUs and SUs, thereby improving the opportunistic spectrum access. Specifically, a SU source broadcasts a message to a set of cognitive users, which can serve as a set of relays, when PUs are absent. The relays that correctly decode the message will create a distributed beamformer to forward the message to the SU destination without causing any interference irrespective of whether PUs are silent or not. To achieve this, we use the method of orthogonal projection to obtain the beamforming weight vector. In addition, we derive the distribution of the received signal power at the SU destination, based on which the average outage probability of our proposed scheme is analyzed when PUs' occupation changes fast. Theoretical and numerical results reveal that the spatial diversity order of this scheme equals the number of SU relays minus that of primary receivers. Furthermore, numerical results show that the outage probability of this scheme outperforms other schemes that access the spectrum only when PUs are absent.

Fueled by the promises of high spectral efficiency, adaptive OFDMA has attracted enormous research interests over the last decade. The significant capacity gain of adaptive OFDMA comes from fast adaptation of resource allocation in response to instantaneous channel conditions. Despite years of efforts to improve the practicality of adaptive OFDMA, such promising technique is still far from real implementation due to the prohibitively high computational complexity and excessive control overhead. This paper is an endeavor to address the problem by proposing a slow adaptation scheme, where resource allocation is adapted on a much slower time scale than the fluctuation of wireless channel fading. Specifically, the slow adaptive OFDMA is formulated into a stochastic programming problem, which adapts resource allocation according to the channel statistics within an adaptation window rather than according to instantaneous channel conditions. By tuning the length of the adaptation window, we could engineer a desirable tradeoff between spectral efficiency and computational complexity. Furthermore, the proposed scheme can be modified to accommodate inelastic traffics. The modification, referred to as "safe" slow adaptation, ensures worst-case data rates to all users. In this work, safe slow adaptation is formulated into a conic linear program, which is efficiently solved via interior-point methods. Through extensive simulations, we show that the proposed schemes drastically reduce the computational complexity and control overheads, while achieving satisfactorily high spectral efficiency and QoS provisioning as their fast-adaptation counterpart does with a much higher cost.

With the proliferation of wireless multimedia applications, multicast/broadcast has been recognized as an efficient technique to transmit a large volume of data to multiple mobile stations at the same time. In most multicast systems, the transmitter (e.g. base station) adapts its data rate to the furthest located users, so as to guarantee service quality to as many users as possible. Predictably, the more users in a multicast group, the lower data rate the base station can transmit. On the other hand, grouping more users together leads to a more efficient utilization of spectrum bandwidth, as these users are served simultaneously. This bring the interesting problem that presses for solution: how to group users in a cell into multicast groups and how to allocate a fixed amount of bandwidth resource to the groups, to achieve a good balance between throughput and fairness in multicast systems. In this paper, we formulate the united user grouping and bandwidth allocation strategy into a utility-based optimization problem. One method of signomial programming is used to solve the non-convex optimization problem. Numerical results will show that this suboptimal algorithm performs well even compared to the optimal one. Moreover, through theoretical analysis, we prove that the best user grouping and bandwidth allocation scheme of throughput maximization is to allocate the entire bandwidth to the unique group containing the users located within a ring-shaped region with an optimal outer radius r*.

Maximizing system utility corresponding to different performance measures through power control has been a long standing open problem in interference-limited multiuser multicarrier wireless networks. The complicated coupling between the mutual interference of links on each subcarrier gives rise to a series of non-convex power control optimization problems, for which the global optimal solution is hard to obtain. This paper proposes a novel algorithm, MARL, to efficiently solve the non-convex power control problem in multiuser multicarrier wireless networks. The algorithm is guaranteed to converge to a global optimal solution, as long as the utility function of each link is monotonically increasing with its data rate. The MARL algorithm is designed based on three key observations of the power control problems considered in this paper: (1) the objective function is increasing in (1+SINR) (SINR: signal to interference- plus-noise ratio); (2) the feasible set of the corresponding equivalent reformulated problem is always "normal", although not necessarily convex; and (3) the two former observations imply that the power control problem can be transformed into a monotonic optimization (MO) problem, where the optimal solution always occurs at the upper boundary of the feasible (1+SINR) region. The MARL algorithm finds the desired optimal power control solution by constructing a series of polyblocks that approximate the feasible (1+SINR) region with an increasing precision. Furthermore, by tuning the error tolerance in MARL, we could engineer a desirable tradeoff between optimality and convergence time. MARL provides an important benchmark for performance evaluation of other heuristic algorithms targeting the same problem. With the help of MARL, we evaluate the performance of a state-of-the-art algorithm through extensive simulations.

Due to its simplicity and cost efficiency, wireless local area network (WLAN) enjoys unique advantages in providing high-speed and low-cost wireless services in hot spots and indoor environments. Traditional WLAN medium-access-control (MAC) protocols assume that only one station can transmit at a time: simultaneous transmissions of more than one station cause the destruction of all packets involved. By exploiting recent advances in PHY-layer multiuser detection (MUD) techniques, it is possible for a receiver to receive multiple packets simultaneously. This paper argues that such multipacket reception (MPR) capability can greatly enhance the capacity of future WLANs. In addition, the paper provides the MAC-layer and PHY-layer designs needed to achieve the improved capacity. First, to demonstrate MPR as a powerful capacity-enhancement technique, we prove a "superlinearity” result, which states that the system throughput per unit cost increases as the MPR capability increases. Second, we show that the commonly deployed binary exponential backoff (BEB) algorithm in today's WLAN MAC may not be optimal in an MPR system, and the optimal backoff factor increases with the MPR capability, the number of packets that can be received simultaneously. Third, based on the above insights, we design a joint MAC-PHY layer protocol for an IEEE 802.11-like WLAN that incorporates advanced PHY-layer signal processing techniques to implement MPR.

Fast rate adaptation has long been recognized as an effective way to improve the PHY-layer data rate of wireless networks. However, in random access wireless networks such as IEEE 802.11 wireless LANs, MAC-layer throughput is dominated by stations with the lowest transmission rates, resulting in an underutilization of spectrum bandwidth. In this paper, we propose a novel two-level medium access framework, referred to as Two-Level MAC, to solve the aforementioned problem and to significantly improve system spectrum efficiency through the exploitation of multiuser diversity. The key idea of Two-Level MAC is to introduce a second level of deterministic channel access on top of the traditional IEEE 802.11 DCF protocol. By doing so, higher priority is granted to high-rate stations in a fully distributed manner. Meanwhile, collisions among potential contending stations are drastically reduced. Through analysis, we show how such Two-Level MAC can be optimized to achieve the maximum system throughput. The superiority of the proposed protocol is verified through analyses and extensive simulations.

Achieving weighted throughput maximization (WTM) through power control has been a long standing open problem in interference-limited wireless networks. The complicated coupling between the mutual interferences of links gives rise to a non-convex optimization problem. Previous work has considered the WTM problem in the high signal to interference-and-noise ratio (SINR) regime, where the problem can be approximated and transformed into a convex optimization problem through proper change of variables. In the general SINR regime, however, the approximation and transformation approach does not work. This paper proposes an algorithm, MAPEL, which globally converges to a global optimal solution of the WTM problem in the general SINR regime. The MAPEL algorithm is designed based on three key observations of the WTM problem: (1) the objective function is monotonically increasing in SINR, (2) the objective function can be transformed into a product of exponentiated linear fraction functions, and (3) the feasible set of the equivalent transformed problem is always ldquonormalrdquo, although not necessarily convex. The MAPEL algorithm finds the desired optimal power control solution by constructing a series of polyblocks that approximate the feasible SINR region in an increasing precision. Furthermore, by tuning the approximation factor in MAPEL, we could engineer a desirable tradeoff between optimality and convergence time. MAPEL provides an important benchmark for performance evaluation of other heuristic algorithms targeting the same problem. With the help of MAPEL, we evaluate the performance of several existing algorithms through extensive simulations.

This paper provides a queueing analysis for the slotted Aloha network. We assume the use of an exponential backoff protocol. Most prior work on slotted Aloha focuses on the analysis of its saturation throughput. Good saturation throughput, however, does not automatically translate to good delay performance for the end users. For example, it is well-known that the maximum possible throughput of slotted Aloha with a large number of nodes is e-1=0.3679 . Prior work showed that binary backoff factor of r = 2 can achieve a saturation throughput of 0.3466, which is very close to the e-1. However, this paper shows that if mean queuing delay is to be bounded, then the offered load must be below 0.2158, a drastic 41% drop from e-1. Fortunately, setting r= 1.3757 allows us to achieve bounded-mean-delay throughput of 0.3545, less than 4% lower than e-1. A general conclusion is that the backoff factor r may significantly affect the queuing delay performance. Our analysis provides a framework to set system parameters properly.

With the proliferation of wireless multimedia applications, multicast/broadcast has been recognized as an efficient technique to transmit a large volume of data to multiple mobile stations at the same time. In most multicast systems, the transmitter (e.g., base station) adapts its data rate to the worst channel among all users in the multicast group, so as to guarantee service quality to each user. Predictably, the more users in a multicast group, the lower data rate the base station can transmit. On the other hand, grouping more users together leads to a more efficient utilization of spectrum bandwidth, as these users are served simultaneously. A natural question that arises is how to group users to maximize the throughput of multicast systems, given a fixed amount of bandwidth resource. In this paper, we attempt to answer this important question that has not been addressed before. Through theoretical analysis, we prove that (1) the average throughput increases with the number of users in a multicast group, when the number of subcarriers allocated to a group is proportional to the number of users therein. Moreover, the throughput approaches infinite-bandwidth Gaussian channel capacity when the number of users gets large; (2) the number of users, and hence the number of subcarriers, that is needed for throughput to be arbitrarily close to its asymptotic value increases almost linearly with the transmit SNR. Our analysis is validated through simulations.

This paper considers the problem of link scheduling with non-preemptive constraint in wireless mesh networks. In real-world implementation, there is often a constraint that a link can only transmit once and occupy consecutive time slots during a frame. We refer to it as the non-preemptive constraint. To date, only few scheduling algorithms in the literature has taken such constraint into consideration. In this paper, we show that optimal non-preemptive link scheduling (NPLS) problems are generally NP-hard and are provably harder to solve than link scheduling without such a constraint. To tackle the problem, a low-complexity list link scheduling (LLS) algorithm is proposed to approximate the optimal NPLS. Our analysis shows that with a randomly selected link-ordering list, throughput degradation of LLS compared to the optimal NPLS is bounded even in the worst case. By carefully constructing the link-ordering list, the performance of LLS can be further greatly improved. In this paper, we propose three schemes to construct link-ordering lists. The performance of the proposed schemes is evaluated through simulations.

To date, most analysis of WLANs has been focused on their operation under saturation condition. This work is an attempt to understand the fundamental performance of WLANs under unsaturated condition. In particular, we are interested in the delay performance when collisions of packets are resolved by an exponential backoff mechanism. Using a multiple-vacation queueing model, we derive an explicit expression for packet delay distribution. It is found that under some circumstances, mean delay and delay jitter may approach infinity even when the traffic load is way below the saturation throughput. Saturation throughput is therefore not a sound measure of WLAN capacity when the underlying applications are delay sensitive. To bridge the gap, we define safe-bounded-mean-delay (SBMD) throughput and safe-bounded-delay-jitter (SBDJ) throughput that reflect the actual network capacity users can enjoy when they require bounded mean delay and delay jitter, respectively. The analytical model in this paper is general enough to cover both single-packet reception (SPR) and multi-packet reception (MPR) WLANs, as well as carrier-sensing and non-carrier- sensing networks. We show that the SBMD and SBDJ throughputs scale super-linearly with the MPR capability of a network. Together with our earlier work that proves super-linear throughput scaling under saturation condition, our results here complete the demonstration of MPR as a powerful capacity-enhancement technique for both delay-sensitive and delay-tolerant applications.

This paper studies optimal power control for throughput maximization in wireless ad hoc networks. Optimal power control problem in ad hoc networks is known to be non-convex due to the co-channel interference between links. As a result, a global optimal solution is difficult to obtain. Previous work either simplified the problem by assuming that the signal- to-interference-and-noise-radio (SINR) of each and every link is much higher than 1, or settled for suboptimal solutions. In contrast, we propose a novel methodology to compute the global optimal power allocation in a general SINR regime. In particular, we formulate the problem into an equivalent linear multiplicative fractional programming (LMFP). A global optimization algorithm, referred to as LMFP-based power allocation (LBPA) algorithm, is proposed to solve the LMFP with reasonable computational complexity. Our analysis proves that the LBPA algorithm is guaranteed to converge to a global optimal solution. Through extensive simulations, we show that the proposed algorithm significantly improves the throughput of wireless networks compared with existing ones.

Today's IEEE 802.11 Wireless LANs (WLANs) usually support multiple transmission rates, where mobile stations adapt their data rates based on their link adaptation strategies. In this paper we analyze the throughput performance of multi-rate WLANs with link adaptation. Specifically, an analytical model is developed to derive the system throughput of both the basic access and RTS/CTS mechanisms. The proposed model is applicable to general multi-rate WLANs and independent of specific link adaptation strategies adopted. The accuracy of the model is validated by simulations in various network settings. Using the proposed model, we further derive the theoretical throughput limit of common multi-rate IEEE 802.11 WLANs. Interestingly, it is observed that for typical WLAN configurations, standard RTS/CTS mechanism performs closely to the throughput limit. Therefore, in multi-rate WLANs simply using standard RTS/CTS mechanism is preferred for its low complexity and close-to-optimal performance.

In this paper, a cross-layer design framework for multi-input multi-output (MIMO)/orthogonal frequency division multiplexing (OFDM) based wireless local area networks (WLANs) is proposed. In contrast to conventional systems where the medium access control (MAC) and physical (PHY) layers are separately optimized, our proposed methodology jointly designs a multi-packet reception (MPR) based protocol with adaptive resource allocation. Specifically, a realistic collision model is employed by taking into consideration the PHY layer parameters such as channel information, space-time coded beamforming and multiuser detection, as well as sub-carrier, bit, and power allocation. The allocation problem is formulated, so as to maximize the system throughput, subject to the constraints from both the MAC and PHY layers. These constraints depend on the results of access contention, data packetsiquest length, usersiquest spatial correlation and the quality of channel feedback information. An iterative algorithm is then provided to obtain the optimal solution. Simulation results will show that our proposed approach achieves significant improvement in system performance such as average throughput and packet delay, compared with conventional schemes where cross-layer design and optimization is not used.

Conventional 802.11 medium access control (MAC) protocols have been designed separately from the characteristics of the physical layer so as to simplify the analysis, but this hardly optimizes the overall performance from a system point of view. In this paper, we propose a channel state information (CSI) based random access protocol that takes advantage of multi-packet reception (MPR) in IEEE 802.11-like wireless local area networks (WLANs). Specifically, the proposed MAC protocol dynamically adjusts each nodes transmission probability according to the network population, estimated channel condition, as well as the maximum number of packets that can be simultaneously decoded. System throughput is analyzed by taking adaptive modulation and transmission errors into consideration. Based on such analytical result, an optimal transmission policy is derived to maximize the system throughput. In addition, the impact of imperfect CSI on the throughput performance is investigated and the tradeoff between the CSI quality and average SNR at the receiver is demonstrated. It is shown that compared with the schemes using conventional models, our proposed approach can significantly increase the throughput without sacrificing system resources such as bandwidth and transmission energy.

This is Part I of a two-part paper series that studies the use of the proportional fairness (PF) utility function as the basis for resource allocation and scheduling in multichannel multi-rate wireless networks. The contributions of Part I are threefold. (i) We present the fundamental properties and physical/economic interpretation of PF optimality. We show that PF leads to equal airtime allocation to users for the singlechannel case; and equal equivalent airtime allocation to users for the multi-channel case. In addition, we also establish the Pareto efficiency of joint-channel PF optimal solution (the formulation of interest to us in this paper), and its superiority over the individual-channel PF optimal solution in that the individual user throughputs of the former are all equal to or greater than the corresponding user throughputs of the latter. (ii) Second, we derive characteristics of joint-channel PF optimal solutions useful for the construction of PF-optimization algorithms. In particular, we show that a PF solution typically consists of many zero airtime assignments when the difference between the number of users U and the number of channels S, |U - S|, is large. We present several PF-optimization algorithms, including a fast algorithm that is amenable to parallel implementation. (iii) Third, we study the use of PF utility for resource allocation in large-scale WiFi networks consisting of many adjacent wireless LANs. We find that the PF solution simultaneously achieves higher system throughput, better fairness, and lower outage probability with respect to the default solution given by today's 802.11 commercial products. Part II of this paper series extends our investigation to the time-varying-channel case in which the data rates enjoyed by users over the channels vary dynamically over time.

Wireless ad hoc networks are expected to provide broadband services parallel to their wired counterparts in near future. To address this need, MIMO (multiple-input-multipleoutput) antenna techniques hold significant promise. Most previous work on capacity analysis of ad hoc networks is based on an implicit assumption that each link exclusively occupies a geometric area, referred to as exclusion region that characterizes the amount of spatial resource occupied by a link. When multiple antennas are deployed at each node, however, multiple links can transmit in the vicinity of each other simultaneously, as interference can now be suppressed by spatial signal processing. As such, the concept of "exclusion region" no longer applies. In this paper, we investigate link-layer throughput capacity of MIMO ad-hoc networks. In contrast to previous work, the amount of spatial resource occupied by each link is characterized by the actual interference it imposes on other links. To calculate the link-layer capacity, we first derive the probability distribution of post-detection SINR (signal to interference and noise ratio) at a receiver. The result is then used to calculate the number of active links and the corresponding data rates that can be sustained within an area. Our analysis shows that there exists an optimal active-link density that maximizes the link-layer throughput capacity. This will serve as a guideline for the design of medium access protocols for MIMO ad-hoc networks. To the best of knowledge, this paper is the first attempt to characterize the capacity of MIMO ad-hoc networks by considering the actual PHY-layer signal and interference model. The results in this paper pave the way for further study on network-layer transport capacity of ad-hoc networks with MIMO.

This is Part II of a two-part paper series that studies the use of the proportional fairness (PF) utility function as the basis for resource allocation and scheduling in multichannel multi-rate wireless networks. The contributions of Part II are twofold. (i) First, we extend the problem formulation, theoretical results, and algorithms to the case of time-varying channels, where opportunistic resource allocation and scheduling can be exploited to improve system performance. We lay down the theoretical foundation for optimization that "couples" the time-varying characteristic of channels with the requirements of the underlying applications into one consideration. In particular, the extent to which opportunistic optimization is possible is not just a function of how fast the channel characteristics vary, but also a function of the elasticity of the underlying applications for delayed resource allocation. (ii) Second, building upon our theoretical framework and results, we study subcarrier allocation and scheduling in orthogonal frequency division multiplexing (OFDM) cellular wireless networks. We introduce the concept of a W-normalized Doppler frequency to capture the extent to which opportunistic scheduling can be exploited to achieve throughput-fairness performance gain. We show that a "lookback PF" scheduling can strike a good balance between system throughput and fairness while taking the underlying application requirements into account.

Conventional 802.11 medium access control (MAC) protocols have been designed separately from the characteristics of the physical layer so as to simplify the analysis, but this hardly optimizes the overall performance from a system point of view. In this paper, we propose a channel state based random access protocol that takes advantage of multi-packet reception (MPR) in IEEE 802.11-like wireless local area networks (WLANs). Specifically, the proposed MAC protocol dynamically adjusts each node's transmission probability according to the network population, current channel condition, as well as the maximum number of packets that can be decoded simultaneously. We shall derive the system throughput by taking adaptive modulation and transmission errors into consideration. An optimal transmission policy is then obtained to achieve the maximum throughput. It is shown that our proposed approach can significantly increase the throughput without sacrificing system resources such as bandwidth and transmission energy, compared to the schemes using conventional models.

MIMO (multiple input multiple output) is capable of offering several-fold increase in spectral efficiency over single-antenna systems through spatial processing. However, current IEEE 802.11 legacy MAC is designed without taking into consideration the interference cancellation capability of MIMO, thereby resulting in suboptimal use of bandwidth spectrum. In this paper, to fully exploit the spatial dimension of freedom offered by MIMO, we propose a methodology that takes into account the spatial correlation between the signal and interference in the design of MIMO ad hoc networks. Through analysis of the effect of spatial correlations on the system throughput, we conclude that it is a crucial factor that cannot be ignored when optimizing the transmission of the networks. Against this backdrop, we propose and investigate a specific MAC protocol for the transmission in a MIMO ad hoc network which 1) allows links to contend for the channel sequentially and transmit data packets simultaneously, yielding more than 35% throughput improvement compared to the system wherein only one link transmits at a time; 2) closely follows the 802.11 DCF (distributed coordination function), allowing simpler implementation as compared to previously proposed MAC protocols.

We first analyze the performance of maximal ratio combining (MRC) scheme for multiple input multiple output (MIMO) in an interference-limited CDMA cellular system. Different from the previous contributions, our work focuses on downlink for the following reasons: first, downlink has always been the bottleneck of capacity due to asymmetry in traffic loads between downlink and uplink; second, the simplicity of MRC MIMO present more attractiveness in downlink for the strict complexity restriction on mobile stations. We analyze the impact of intra-cell interference as well as inter-cell interference on system performance and derive the outage probability and capacity of the system in a closed-form. Then we compare the performance between MRC-MIMO-CDMA and single input single output (SISO) CDMA systems in respect of outage probability and capacity. The results indicate the efficiency predominance of MRC- MIMO-CDMA over SISO and show that employing more transmit antennas at the base station brings on better performance than using more receive antennas.

Fast rate adaptation has been established as an effective way to improve the PHY-layer raw date rate of wireless networks. However, within the current IEEE 802.11 legacy, MAC-layer throughput is dominated by users with the lowest data rates, resulting in underutilization of spectrum bandwidth. In this paper, we propose a novel distributed MAC strategy, referred to as Rate-aware DCF (R-DCF), to leverage the potential of rate adaptation in IEEE 802.11 WLANs. The key feature of R-DCF is that by introducing different mini slots according to the instantaneous channel conditions, only contending stations with the highest data rate can actually access the channel. In this way, the R-DCF protocol not only effectively exploits multi-user diversity in a fully distributed manner but also drastically reduces the loss of throughput due to collisions. Through analysis, we derive a closed-form network throughput expression for R-DCF. Based on the analysis, we further derive the maximal throughput that can be achieved by R-DCF. For practical implementation, an offline adaptive backoff method is developed for R-DCF to achieve a close-to-optimal performance at low runtime complexity. The superiority of R-DCF is proven by extensive analyses and simulations.

Multi-user diversity can be employed in wireless communications to significantly improve system performance by scheduling a channel to the user with the best instantaneous channel condition. Conventional multi-user diversity schemes require all mobile terminals' channel state information to be available at the access point, thereby inducing huge system overhead. In this paper, we propose a user identification approach to significantly reduce system overhead for wireless network with multi-user diversity. The proposed user identification approach is then applied to OFDM systems. We shows that the network throughput upper limit can be significantly improved by using the user identification approach compared with the user polling approach in the literature.

In wireless ad hoc networks, PHY-layer interference is highly dependent on link scheduling schemes, and in turn heavily affects the performance of link scheduling. However, most existing link scheduling schemes ignore such relationship between the PHY and network layers, and hence fail to achieve the optimal end-to-end throughput due to large co-channel interference. This paper proposed a general framework for optimal TDMA (time division multiple access) link scheduling in multi-hop wireless ad hoc networks with general topology. In contrast to existing work, we maximizes the end-to-end throughput by taking into consideration the explicit relationship between the transmission rate of a link and its PHY-layer SINR (signal to interference and noise ratio). In particular, the authors formulate the scheduling problem into a LP (linear programming) problem based on the rate matrices with each entry being a function of SINR. With this formulation, the cross-layer link scheduling problem can be solved in polynomial time. To further reduce the computational complexity the authors proposed an algorithm to effectively reduce the size of the LP problem. Furthermore, to handle large-scale wireless networks, the authors present a decentralized scheduling algorithm that achieves a suboptimal TDMA scheduling solution with dramatically lower computational complexity comparing to the original LP formulation. Numerical results show that the proposed cross-layer link scheduling schemes outperform the existing schemes that assume a simplistic PHY-layer interference model by 59.45%.

Multiuser resource allocation has recently been recognized as an effective methodology for enhancing the power and spectrum efficiency in OFDM (orthogonal frequency division multiplexing) systems. It is, however, not directly applicable to current packet-switched networks because most existing packet- scheduling schemes are based on a single-server model and do not serve multiple users at the same time. In this paper, we propose a cross-layer resource allocation algorithm based on a novel multi-server scheduling framework to achieve overall high system power efficiency in packet-switched OFDM networks. Our contribution is four fold: (i) we propose and analyze a MPGPS (multi-server packetized general processor sharing) service discipline that serves multiple users at the same time and facilitates multiuser resource allocation; (ii) we present a MPGPS-based joint MAC-PHY resource allocation scheme that incorporates packet scheduling, subcarrier allocation, and power allocation in an integrated framework; (iii) by investigating the fundamental tradeoff between multiuser-diversity and queueing performance, we present an A-MPGPS (adaptive MPGPS) service discipline that strikes balance between power efficiency and queueing performance; and (iv) we extend MPGPS to an O-MPGPS (opportunistic MPGPS) service discipline to further enhance the resource utilization efficiency.

This paper studies the use of the proportional fairness (PF) utility function as the basis for capacity allocation and scheduling in multi-channel multi-rate wireless networks. Examples of applications include (i) access point (AP) association and transmission scheduling in large-scale IEEE 802.11 networks; and (ii) subcarrier assignment and transmission scheduling in orthogonal frequency division multiplexing (OFDM) cellular networks. The contributions of this paper are twofold. First, we study the fundamental properties and physical/economic interpretation of PF. We show by general mathematical arguments that PF leads to equal airtime allocation to individual users for the single-channel case; and equal equivalent airtime allocation to individual users for the multi-channel case, where the equivalent airtime enjoyed by a user is a weighted sum of the airtimes enjoyed by the user on all channels, with the weight of a channel being the price or value of that channel. In addition, we establish the Pareto efficiency of PF solutions and derive characteristics of PF solutions that are useful for the construction of PF algorithms. Second, we generate numerical results for application (i) above. We find that the PF solution simultaneously achieves higher system throughput, better fairness, and lower outage probability with respect to the default AP-association and medium access control (MAC) protocol adopted in today's 802.11 commercial products.

Efficient resource management is a major issue in the operation of wireless communication systems given the limited resource availability. In contrast to most existing resource allocation techniques that assume perfect channel state information at the transmitter (CSIT), a novel adaptive resource management algorithm is proposed to optimize the spectral efficiency of a multiuser multi-input multi-output (MIMO)/orthogonal frequency division multiplexing (OFDM) system where only partial channel state information is available. With a space-time-block-coded beamformer employed at each mobile station, the proposed algorithm adaptively allocates subcarrier, bit, and power to realize efficient resource utilization. A distinctive feature of the proposed method is that the allocation strategy is directly related to the users' spatial correlation and the quality of channel feedback information. Numerical results will show that the proposed scheme achieves significant throughput improvement compared to non-adaptive systems. It will be also demonstrated that our proposed method outperforms a scheme which decouples the allocation problem into single-user optimization problems by introducing user-grouping.

The conventional MAC (Medium Access Control) protocols assume that only one packet can be received at a given time. However, with the advent of sophisticated signal processing and antenna array techniques, it is possible to achieve multipacket reception (MPR) in the physical layer (PHY). In this paper, we propose a PHY methodology and the corresponding MAC protocol for MPR in wireless local area networks (WLANs). The proposed MAC protocol closely follows the 802.11 DCF (Distributed Coordination Function) scheme and enables MPR in a distributed manner. For the proposed MPR system, a closed-form expression of the average throughput is derived. Based on the expression, an optimal transmission probability that maximizes the throughput can be attained. In addition, two enhancement schemes are presented to further improve the performance of the MPR protocol. Numerical results show that the proposed MPR system can considerably increase the spectrum efficiency compared to the WLANs with conventional collision models.

A collision resolution scheme is essential to the performance of a random-access wireless network. Most schemes employ exponential backoff (EB) to adjust the transmission attempt rate according to the changing traffic intensity. Previous work on exponential backoff was mostly based on the conventional single-packet-reception model where no more than one packet can be successfully received at any one time. In this paper, we analyze the performance of EB based on a multi-packet reception (MPR) model, in which multiple packets can be received successfully at once (i.e., collisions do not occur unless the number of packets transmitted exceeds a threshold that is more than 1). Using a Markov chain model, we derive the throughput expressions for both carrier-sensing and non-carrier-sensing networks with MPR capability under the saturated-traffic condition. We find that the two systems share a number of common performance results. In particular, the state of both systems can be characterized by the same Markov-chain model. The binary exponential backoff (BEB), in which the backoff factor r is set to 2, does not yield the optimum network throughput in both cases. In addition, in both cases, the asymptotic collision probability goes to 1/r and the maximum asymptotic throughput increases roughly linearly with M when the population size approaches infinity. We show how to adjust r to achieve the best throughput performance. Our results show that the optimal r that maximizes the asymptotic throughput increases with M for non-carrier-sensing systems and BEB is close to optimal for carrier-sensing systems. Simulation results validate the accuracy of our theoretical analysis

Adaptive resource management has become a key technique for the next-generation wireless systems to provide desired services at appropriate QoS. Most of the current resource management algorithms are confined to a single layer of the network protocol stack, which leads to an inferior performance. In this paper, we propose a joint MAC-PHY layer resource management algorithm for wireless OFDM networks. The algorithm aims to maximize the system power efficiency given that the fulfillment of the QoS as well as the fairness is guaranteed. We formulate the integrated design of packet scheduling, subcarrier allocation, and power control into a constrained optimization problem. The formulation defines the interaction and cooperation between different components of the algorithm as well as the information exchange between the PHY and MAC layers. A reduced-complexity scheme that solves the optimization problem in polynomial time is then developed based on the special structure of the problem. Analytical and numerical results will show that the proposed algorithm is able to guarantee the same degree of QoS and fairness as the fair queueing systems do in wired channels. Meanwhile, the power efficiency and system performance are significantly improved compared to traditional systems where resources are allocated based on a strict layering architecture

The explosive growth of high-data-rate and multimedia real-time applications imposes new challenges on the design of future wireless communications systems. In this paper, a cross MAC-PHY layer link-adaptive largest-weighted-throughput packet scheduling algorithm is proposed to provide QoS guarantees to real-time traffics over wireless packet-switched networks. The proposed algorithm aims to satisfy the stringent packet delay constraints while obtaining a high system spectral efficiency. The objective is achieved by two interactive components: a MAC service discipline and a PHY rate-and-power adaptation scheme. It is demonstrated in the numerical results that the proposed algorithm significantly improves the system performance in terms of spectral efficiency and packet loss rate, thanks to the successful exploitation of the inherent system diversities in the time, frequency, and multiuser domains.

Most current resource management algorithms are confined to a single layer of the network protocol stack, which leads to an inferior system performance. We propose a joint MAC-PHY layer resource allocation algorithm. The proposed algorithm jointly optimizes the bandwidth and power allocation through an integrated design of packet scheduling, subcarrier allocation, and power control. Analytical and numerical results show that the proposed algorithm is able to provide the same QoS and fairness guarantee as fair queueing systems do in a wired channel. Meanwhile, the power efficiency and system performance are significantly improved compared to traditional systems where resources are allocated based on a strict layering architecture.

Fast adaptive transmission has been recently identified as a key technology for exploiting potential system diversity and improving power-spectral efficiency in wireless communication systems. An adaptive resource-allocation approach, which jointly adapts subcarrier allocation, power distribution, and bit distribution according to instantaneous channel conditions, is proposed for multiuser multiple-input multiple-output (MIMO)/orthogonal frequency-division multiplexing systems. The resultant scheme is able to: 1) optimize the power efficiency; 2) guarantee each user's quality of service requirements, including bit-error rate and data rate; 3) ensure fairness to all the active users; and 4) be applied to systems with various types of multiuser-detection schemes at the receiver. For practical implementation, a reduced-complexity allocation algorithm is developed. This algorithm decouples the complex multiuser joint resource-allocation problem into simple single-user optimization problems by controlling the subcarrier sharing according to the users' spatial separability. Numerical results show that significant power and diversity gains are achievable, compared with nonadaptive systems. It is also demonstrated that the MIMO system is able to multiplex several users without sacrificing antenna diversity by using the proposed algorithm.

In this letter, we propose an adaptive resource-allocation algorithm for multiaccess multiple-input multiple-output/orthogonal frequency-division multiplexing systems. The proposed algorithm endeavors to maximize the system power efficiency, given that the users' quality of service (QoS) requirements, specified by bit-error rate and data rate, are satisfied. Subcarrier allocation, power distribution, and modulation for multiple users are jointly optimized according to users' channel states and QoS requirements. To avoid the joint optimization of resource allocation and beamforming, matched-filter-based receivers are employed at the base station, with cochannel interference being mitigated through dynamic subchannel allocation. A neighborhood search scheme is further developed to obtain a good allocation solution with reasonable computational efforts. Our results show that the proposed algorithm is able to achieve significant enhancement in the system power efficiency due to the successful exploitation of multiuser diversity, as well as channel variations in the time, frequency, and space domains.

As we move towards the next generation wireless communications systems featured by all IP-based protocols and heterogeneous traffic calls, new issues arise in designing resource management algorithms. They include the random packet arrival, limited buffer space, various QoS requirements, and user fairness. In this paper, we propose a cross-layer adaptive resource allocation algorithm for packet-based OFDM systems. This algorithm takes into consideration not only the physical channel conditions but also the random traffic arrival, QoS requirements, and user fairness observed at the link layer. Results show that the proposed algorithm greatly enhances the system spectral efficiency and improves the queuing performance thanks to the successful exploitation of statistical multiplexing flexibility and system diversity in the frequency, time and user domains.

Adaptive resource allocation has been identified as one of the key technologies for providing efficient utilization of the limited power and spectrum in future wireless systems. In this paper, an adaptive resource allocation methodology is proposed for cellular orthogonal frequency division multiplexing systems in a multiuser environment. The proposed method is featured as a low-complexity algorithm that involves not only adaptive modulation, but also adaptive multiple-access control and cell selection. Specifically, a multiuser subcarrier-and-bit loading scheme is developed to maximize the spectral efficiency. Besides, a dynamic cell selection scheme is proposed to deal with the problem of overloading and nonuniform traffic density. Numerical results show that the presented algorithm offers significant improvement in spectral efficiency due to the successful exploitation of channel variation, multiuser diversity, and inter-cell diversity.

Fast adaptive transmission has been recently recognized as an essential method to exploit the system diversity and enhance power/spectral efficiency in time-varying channels. In this paper, we propose adaptive transmission algorithms for multiuser MIMO/OFDM systems. The algorithms adaptively control multiple-access, modulation, and transmit power to exploit the channel variation in space, frequency, and user domains. The ultimate objective is to optimize the overall power efficiency while ensuring the fulfillment of every user's QoS requirements. A major challenge in the problem lies in the optimization complexity due to the presence of non-linear, non-convex constraints. The proposed algorithms successfully reduce the optimization complexity and solve the noise enhancement problem that exists in multiuser systems with non-orthogonal signatures. Numerical results show that significant improvement in power/spectral efficiency is achievable with reasonable computational complexity when the above schemes are applied.

OFDM has recently been considered as one of the main next-generation broadband wireless access solutions because of its robustness over heavily impaired links and its high spectral efficiency. In this paper, OFDM-based adaptive TCM schemes in both single- and multi-user environments are proposed. The adaptive scheme employs the family of pragmatic TCM to combine different coded modulation modes into the one codeword while reducing the hardware complexity. A set of switching levels is analytically derived to determine the appropriate modes for each value of the received SNR. The error and throughput performance of the proposed system is analyzed and it is shown the adaptive single user TCM results in considerable performance improvement. It is also demonstrated that by jointly optimizing the overall spectral efficiency while guaranteeing a minimum data rate and a desired BER level to each user, the spectral or power efficiency can be further improved in the multi-user environment.

A dynamic multiuser subcarrier-and-bit allocation algorithm with low computational complexity for wideband OFDM downlink transmission is proposed to exploit the multiuser diversity in this paper. Our objective is to maximize the overall spectral efficiency in terms of the number of bits per subcarrier while simultaneously satisfying the requirements on each user's data rate, bit error rate (BER), and the total transmit power. Our results show that the proposed dynamic subcarrier-and-bit allocation algorithm offers about 6 dB performance gain over OFDM using adaptive modulation with fixed subcarrier allocation. Since outages may occur when users happen to be located near the boundary of the cell, a novel adaptive cell selection scheme is proposed to further reduce the probability of outage. Our results show that the probability of outage can be reduced by one order of magnitude.