In this paper, we propose a hierarchical scheduling method for enhanced mobile broadband and ultra-reliable low-latency communication slices in mobile networks with non-stationary traffic arrivals, where resource allocation and traffic scheduling are decoupled into intra-slice and inter-slice processes operating on different timescales. An effective bandwidth estimation algorithm is developed, dynamically triggered by traffic fluctuations to determine transmission rates that ensure ultra-reliable delay guarantees under non-stationary arrivals. The intra-slice scheduling is formulated as a generalized assignment problem, for which we design an efficient algorithm that maximizes throughput under transmission rate constraints and guarantees a provable approximation ratio. We propose an inter-slice online resource allocation algorithm to enhance long-term system utility and establish that it admits a sublinear dynamic regret bound in piecewise-stationary environments. Numerical results demonstrate that the proposed method outperforms alternatives, improving system utility while satisfying slice requirements.

Mission-critical applications require deterministic services for coexisting periodic and bursty flows. While the original cyclic queuing and forwarding (CQF) bounds delay via a dual-queue ping-pong buffering mechanism, its rigid architecture suffers from limited scalability and poor adaptability to traffic bursts. This paper introduces multi-queue CQF—a multi-queue-per-port architecture enabling fine-grained flow isolation—and a co-optimization framework for routing and scheduling with provable guarantees in offline and online regimes. We develop a primal-dual approximation scheme that efficiently determines the maximum schedulable flow set under resource constraints and an online scheduler that ensures high delivery success with only slight capacity augmentation. Numerical evaluations demonstrate that our solutions maintain high schedulability and robustness across diverse optical network configurations.

Channel estimation (CE) is a critical component in the massive multi-input multi-output (MIMO) communication systems. Compared with conventional CE algorithms, deep learning (DL)-based approach becomes a promising alternative, due to its capability of offering enhanced performance and robustness across diverse scenarios. However, efficient DL-based CE algorithms have two key properties that make them challenging for implementation in existing architectures at the edge side: the diversity of deep neural networks (DNNs) and CE strategies, and the involvements of multiple computation-intensive tasks that compass conventional signal processing, artificial intelligence (AI) inference, and online learning. To address these challenges, a domain-specific processor based on an extended RISC-V instruction set architecture (ISA) is proposed to perform these DL-based CE algorithms. First, a dedicated RISC-V ISA extension is developed to support all essential operations required by a DL-based CE algorithm, such as matrix inversion, in a flexible manner. Building on the customized ISA extension, a highly adaptable and scalable RISC-V processor is developed, featuring scalar and vector posit arithmetic units to alleviate high computational and memory demands of DNNs during both inference and training phase. Additionally, a coarse-grained matrix accelerator is integrated to expedite various matrix operations ensuring high throughput. In this way, both high flexibility and computational efficiency are achieved. Finally, our processor is implemented on a TSMC 28-nm technology. Implementation results show that the processor achieves a speedup of$5.16\sim 6.80\times $for all matrix operations compared with the state-of-the-art work. Moreover, the proposed processor provides an area efficiency improvement of$1.61\times $and an energy efficiency enhancement of$6.6\sim 15.4\times $compared to the open-source vector processor Ara. Notably, this work is the first RISC-V domain-specific processor tailored for diverse DL-based CE algorithms.

We investigate the opportunistic channel access problem in a centralized cognitive radio network, where the cognitive base station (CBS) periodically detects spectrum holes in the licensed network and coordinates the unlicensed secondary users to utilize the idle channels. Existing spectrum access mechanisms typically rest on the assumption that the spectrum environment is known in advance, i.e., licensed channel states exhibit either stochastic or adversarial variations across different time instances and locations. These approaches address the spectrum access problem from the perspective of online learning, which relies heavily on the prior knowledge of network parameters such as the time horizon and the user activities to tune hyperparameters. In this paper, we tackle the multiuser channel access task by formulating it as a combinatorial multi-armed bandit problem in an unknown environment, where we propose an online mirror descent-based channel access method for the CBS that does not require any prior knowledge of the environment. Our proposed method adaptively adjusts the probability of secondary users accessing the licensed channels based on the historical transmission feedback, achieving order-optimal performance in both stochastic and adversarial environments. Numerical results validate the theoretical analysis and also demonstrate that the proposed method outperforms others under various network settings.

Introducing the serverless computing paradigm to network edge offers a promising solution to the emerging latency-critical cloud services. Unique features such as highly-variable execution times and unpredictable arrival patterns of the serverless functions pose fundamental challenges to achieve such goals. In this paper, we propose an online algorithm for joint function scheduling and routing, aiming to maximize the rewards of deadline-constrained serverless functions. The algorithm dynamically swaps the scheduling orders by exploiting the actual realizations of the stochastic execution times, and enables splittable routing taking into account the interval estimates of the latest function starting times. We prove that the proposed algorithm is $\frac{1}{{1 - \delta }}\mathcal{O}(\log (\Lambda \mathcal{P}))$-competitive against δ-risk non-anticipatory offline optimum, where parameters Λ and $\mathcal{P}$ capture the instance pattern. Numerical results demonstrate that our proposed scheme outperforms the state-of-the-art ones.

The burgeoning demands of modern applications highlight the imperative of supporting deterministic, ultrareliable, and low-latency services in mobile networks, however, ensuring deterministic delays always faces significant challenges due to the inherent unpredictability of the time-varying channel conditions and the bursty data traffic. In this paper, we investigate online joint user and resource scheduling in unknown dynamic environments, which yields a nested mixed-integer nonlinear programming problem. To deal with such a formidably complex optimization task, we propose a drift-plus-penalty scheme for the higher-level user selection and devise a low-complexity dual method with slight performance degradation for the lower-level resource allocation. Our approach accommodates diverse user deadlines and establishes a tunable tradeoff between power consumption and queue blockage. We provide analysis on convergence and performance bounds, and numerical results demonstrate the effectiveness and the efficiency of our proposal.

The paradigm of mobile edge computing (MEC) has emerged as a promising solution to the increasing demand of time-sensitive applications, where tasks generated by users are offloaded to proximate edge clouds for low-latency execution. Due to the online nature of task generation and edge capacity bottleneck, a fundamental challenge for the MEC network is how to optimally schedule the tasks and resources in face of uncertain future arrivals. To this end, we propose a competitive algorithm named EdgeOPT for online parallel task scheduling aiming to maximize the cumulative reward of completed tasks subject to their hard deadlines. The algorithm leverages an adaptive threshold structure at each server to schedule tasks with different demand patterns based on the status of the system and all active users, while incorporating a subroutine for efficient resource allocations. We prove a bounded competitive ratio for EdgeOPT when scheduling monolithic tasks, and propose its extended version to schedule chains of dependent functions. We conduct extensive experiments to demonstrate the effectiveness and superiority of our proposal compared to all the baselines.

Deterministic mobile networks are essential for advanced applications that demand strict quality of service (QoS) assurances under limited resource availability. Though network slicing can optimize average performance metrics to offer best-effort services, it often fails to meet the high-reliability requirements of deterministic communication scenarios. In this paper, we introduce a novel QoS-guaranteed inter-slice radio resource allocation scheme for mobile networks to deliver deterministic services over the long term. First, we develop an analytical martingale-based stochastic network calculus framework, which yields stochastic bounds for transmission delays and queue backlogs across various traffic arrival patterns. These bounds produce robust interval estimations that guide resource allocation decisions, effectively addressing channel variability and long-tail QoS effects. Then, an efficient resource allocation algorithm is proposed to approach the derived performance bounds while ensuring fairness across different radio slices with diverse QoS needs. The framework also incorporates an adaptive traffic predictor, enabling our algorithm to track and respond to network dynamics. Numerical results demonstrate that our proposed scheme achieves a promising trade-off between resource utilization and QoS guarantees.

Network function virtualization aims to deploy virtual network functions (VNFs) on commercial off-the-shelf hardware, allowing for efficient implementation of network services by scheduling various VNFs. However, the requirements of deterministic networking poses higher demands on VNF scheduling to achieve guaranteed latency performance. In this paper, we study the online VNF scheduling problem with the objective of minimizing the system cost regarding to unfinished services while improving the determinism of latency. The formulated optimization task yields a mixed integer program problem and we propose an online VNF scheduling algorithm to tackle it, which transforms the end-to-end latency constraint into deadlines for each VNF and adopts an online primal-dual method to generate scheduling decisions. Numerical results show that the proposed scheduling algorithm outperforms two traditional rule-based scheduling methods in terms of latency and acceptance ratio.

Tuning the azimuths and tilts of antennas in an optimal way is crucial for the mmWave mobile communication systems, nevertheless, it always yields an intractable combinatorial optimization problem due to the huge number of possible antenna configurations. In this paper, we propose a scalable reinforcement learning method to deal with this optimization task, which decomposes and distributes the policy learning process by exploiting the interactions among adjacent antennas implicitly. A distributed constraint optimization technique is introduced to coordinate the distributed learning process, which can lead the learning towards desired directions. Experiment results in large-scale scenarios demonstrate that our proposed method yields significant performance improvement in terms of signal-to-interference ratio coverage while ensuring high power coverage.

In this paper, we study the unmanned aerial vehicle (UAV)-assisted emergency communication network, where multiple UAVs are dispatched from emergency centers and cruise the target region to provide service. To ensure timely communications, a strict requirement is that the time spent on the tour of each UAV is no greater than a delay threshold. Our optimization task is to minimize the total cost while satisfying the delay constraints, in which the total cost is a weighted sum of deployment cost and cruise cost. We propose a joint deployment and path planning scheme to address the NP-hard task from a load-balancing perspective. Specifically, we first adopt a continuous approximation method to estimate the number of needed emergency centers and locate them by clustering. Then, we assign the ground users to the emergency centers via loadbalancing region partitioning, based on which the path planning of multiple UAVs can be reduced to independent traveling salesman problems. Numerical results demonstrate that our proposal can complete the delay-bounded mission at the lowest cost as compared with other methods, providing an efficient and cost-effective way for practical emergency communications.

Deterministic communications systems are critical for time-sensitive applications in the Internet of Things, which demand stringent delay requirements under the conditions of limited available resources. Though network slicing can provide best-effort services by focusing on average performance metrics, it generally cannot address the worse-case issues arising from the deterministic communication scenarios. In this paper, we propose an efficient inter-slice radio resource allocation scheme for mobile networks to provide delay-guaranteed services, where a stochastic network calculus model is introduced to analyze the service delay and its variation. We derive a tight and time-invariant upper bound of the delay violation probability, and develop an efficient resource allocation algorithm to meet the stringent delay requirement of different radio slices. Numerical results demonstrate our proposal achieves a promising trade-off between resource utilization and delay.

In this paper, we propose a joint source-channel coding (JSCC) scheme for wireless image transmission, where the encoder and decoder of the transmission system are implemented by convolutional neural networks. A neural architecture search (NAS) method is introduced to find promising network structures aiming at minimizing the distortion of the target image. Experimental results demonstrate that our proposed scheme, referred to as NAS-JSCC, significantly outperforms the conventional manually designed ones in terms of peak-signal-to-noise ratio in nearly all tested signal-to-noise ratio regions.

In this paper, we investigate the problem of distributed dynamic spectrum access with multiple secondary users (SUs), where the availabilities of licensed channels are unknown to SUs. If multiple SUs access the same channel simultaneously, none of them would transmit successfully because of collision. As a result, each SU has to learn unknown channel statistics and coordinate with others based on its local observations. We develop an online learning based channel access method, which identifies the best channel quickly by using Thompson sampling and orthogonalizes SUs on different channels efficiently without prior information. Numerical results show that the proposed method achieves the highest probability of identifying the best channel compared to the existing ones.

Orthogonal frequency-division multiplexing (OFDM) is widely adopted in narrowband Internet of Things (NB-IoT). Nevertheless, the OFDM system is highly sensitive to the impairments caused by imperfect radio-frequency hardwares, which may greatly jeopardize the orthogonality between different subcarriers and degrade the demodulation performance. Though large efforts have been devoted to the hardware impairment estimation, however, it is highly challenging to jointly estimate multiple hardware impairments due to their coupling effects, especially for the NB-IoT systems which usually work in low signal-to-noise ratio (SNR) regions with limited computing and radio resources. In this article, we propose a parallel multitask learning (MTL) estimator to jointly estimate carrier frequency offset and in- and quadrature-phase imbalance in NB-IoT systems. Specifically, MTL is introduced to extract the inherent correlations between different hardware impairments so as to address the coupling effect and average the noise impact. In addition, we propose a parallel structure and a sliding window scheme to reduce the network complexity and decrease the estimation bias. Numerical results show that our proposed parallel MTL estimator can jointly estimate multiple hardware impairments with short pilot sequences, and outperform the conventional methods in terms of estimation accuracy and computation time in the typical SNR regions of the IoT devices.

The automatic recognition of synthetic aperture radar (SAR) targets has been extensively studied in recent years. Specifically, due to the high cost of SAR image acquisition and the low occurrence probability of high-value targets, it is of great importance to identify SAR targets with only a few available images, which is referred to as few-shot SAR target classification. However, most existing solutions straightly adopt meta-learning and transfer learning methods that are originally designed for optical images, which do not consider the unique frequency information of SAR images. In this paper, we propose a novel hybrid classification network that combines both spatial and frequency information for few-shot SAR target classification. Specifically, we first train the proposed network in a source dataset, which contains a large number of related SAR images without the targets of interest. Then, the pre-trained network is fine-tuned on the target dataset consisting of only a few SAR images of interest. Compared with the conventional method based on convolutional neural networks and model-agnostic meta-learning, the proposed method can achieve superior top-1 accuracy in various settings.

In orthogonal frequency division multiplexing (OFDM) systems, carrier frequency offset (CFO) and in- and quadrature-phase (IQ) imbalance are two critical hardware impairments that may lead to severe amplitude and phase mismatches and therefore greatly degrade the demodulation performance. Due to the noise impact and the coupling effects between different hardware impairments, it is considered to be highly challenging to perform joint estimation of CFO and IQ imbalance. In this paper, we propose a novel multi-task learning-based hardware impairment estimation network (HIENet) to simultaneously estimate the CFO and IQ imbalance in OFDM systems. The proposed HIENet can address the noise impact by averaging the task-dependent noise patterns and overcome the coupling effects by extracting correlated features from different estimation tasks. Numerical results show that the proposed HIENet can achieve comparable estimation accuracy to conventional one-shot and iterative methods, while at the same time having the smallest computation time.

With the increasing of the directional antennas in millimeter wave mobile communication systems, tuning the azimuths and downtilts of these antennas in an optimal manner is the core method to enhance the network performance, including the power coverage and the system throughput of the network. Since the formulated optimization problem is generally computational intractable, we introduce a codebook based optimization scheme to handle this difficult task efficiently. Specifically, a codebook with a small set of promising azimuths and downtilts is generated for each base station according to its surrounding radio propagation environment, based on which all the antennas are adjusted to serve the target area with strong enough signals while alleviating the interference to adjacent base stations. Numerical results show that the proposed scheme can yield promising antenna configurations with limited computing resources.

In dynamic spectrum access (DSA), unlicensed secondary users can estimate the idle probability of each primary channel by using historical sensing results and access the channel with the highest idle probability for opportunistic transmission. Most of the existing works assume that each primary channel is associated with a constant idle probability, which can be accurately estimated by sensing the channel multiple times. However, due to the rapid traffic change and irregular user mobility, primary channels can be highly dynamic and the associated idle probability is generally time-varying. In this paper, we consider DSA in non-stationary environments where the idle probabilities of primary channels vary with time. Specifically, we propose a DSA scheme based on the Thompson sampling method with a change-detection technique, which is capable of detecting the variation of channel statistics and adjusting the channel access strategy accordingly. Numerical results show that the proposed algorithm outperforms the existing algorithms in terms of successful transmission ratio in various settings.

Unmanned aerial vehicle (UAV)-assisted communications has emerged as a promising technology in many domains, such as disaster relief and emergency scenarios. However, the limited battery capacity restricts UAVs from performing such persistent missions. In this paper, we consider a general hybrid trajectory planning problem to efficiently provide emergency communications in time-constrained disaster-affected regions, where a ground vehicle carrying backup batteries moves along with the UAV as a “mobile charging platform” to handle the energy issue of the UAV. Our optimization task is to minimize the total cost of the mission. We show that the optimization task is an extension of the traveling salesman problem with soft time window constraints. Due to the NP-hardness of the task, we propose a novel deep reinforcement learning with a sequential model strategy to learn the policy for the UAV's visiting order, based on which the collaborative routes of the UAV and ground vehicle are designed. Numerical results show that our proposed learning-based route planning scheme is effective and efficient.

Network function virtualization redefines a network service as a softwarized chain of virtual network functions (VNFs), which decouples the specific network service from dedicated devices and greatly reduces the hardware cost. The VNFs are generally deployed on common off-the-shelf servers and statistically share computational resources. Due to the inherent integer constraints, the corresponding VNF mapping and scheduling issue is a highly challenging task. In this paper, we consider the mapping and scheduling of VNFs for a given network service, for which a flexible job shop scheduling problem is formulated to optimize the max-min fairness while ensuring the delay requirements of different service chains. Specifically, we propose a deep reinforcement learning method based on offline proximal policy optimization, which dynamically determines the mapping and scheduling decision based on the state of unfinished service chains. The proposed algorithm is scalable to the number of service chains and can be enhanced by Monte Carlo tree search. Numerical results show that the proposed algorithm outperforms the traditional random forest and the greedy algorithms in terms of both service fairness and acceptance ratio.

Due to the flexibility and agility, unmanned aerial vehicle (UAV) is a promising way for gathering data generated by wireless sensor networks. However, the limited battery capacity of the UAV restricts its application on many occasions, e.g., the networks deployed in the wild. In this paper, we propose a cooperative trajectory planning scheme to deal with the energy issue of the UAV, where a truck carrying backup batteries moves along with the UAV acting as a “mobile recharging station”. Our optimization task is to minimize the total mission time for gathering data from all the sensor nodes, which can be achieved by solving two problems: First, we need to divide the entire mission area into multiple subregions so that the UAV can hover over each subregion to collect the data of the sensor nodes through just one taking-off and landing under the constraint of battery capacity; second, we should find out the optimal trajectory of the truck so that the UAV can get to the hovering positions of each subregion from the truck and fly back to it before the battery drains considering the road condition in real world. We introduce an efficient clustering algorithm to partition the area into subregions in a load-balanced way to minimize the number of movements of the UAV. The trajectory planning task is formulated as a coordinated traveling salesman problem, which is solved by a three-step trajectory planning algorithm heuristically, and we also give the analysis of the upper bound and lower bound to demonstrate the performance guarantee. Numerical results show that our proposed scheme provides an effective and cost-efficient way for the data collection of large-scale wireless sensor networks in practical scenarios.

In frequency-division-duplex (FDD) massive MIMO systems, channel state information (CSI) feedback waiting phase does not get fully exploited since base station needs to wait for the CSI feedback before transmitting downlink data. The proportion of the CSI feedback waiting phase during the downlink transmission would be high as the MIMO system scales up, which sacrifices downlink rates of the FDD massive MIMO systems significantly. In this paper, we first present a channel prediction-aided FDD scheme to utilize the idle waiting time efficiently. Then we propose a novel channel prediction method based on Bayesian neural network (BNN), which can handle the uncertainty in a natural manner and learn regularization from data without painstaking manual pre-tuning of network hyperparameters. Numerical results show that our proposed channel prediction-aided FDD scheme can achieve remarkable performance gains in terms of either achievable downlink rates or bit error rate. Moreover, our proposed BNN-based channel predictor is much more effective and robust in contrast to the state-of-the-art channel prediction techniques such as autoregressive model and recurrent neural network.

Network function virtualization aims at deploying network functions on general-purpose hardwares, referred to as virtual network functions (VNFs), rather than the specialized devices. A network service can be implemented by scheduling different VNFs. In this paper, we study the VNF scheduling problem with the objective of satisfying the diversified end-to-end delay requirements while maintaining the fairness among different network services. We formulate the problem as a mixed integer nonlinear program and propose a VNF scheduling algorithm based on deep reinforcement learning, in which proximal policy optimization is adopted to optimize the policy network. Numerical results show that the proposed scheduling algorithm outperforms other canonical ones and the designed policy network can scale to fit different problem sizes.

In this paper, we investigate the network planning problem in mmWave mobile communication systems, where the narrow-beam antennas can adjust azimuths and downtilts of antennas so as to maximize the power coverage of the network, as well as the system throughput. Searching for the optimal configurations of antennas generally yields a combinatorial opti-mization problem, which cannot be addressed even for a medium scale antenna set case. We formulate this optimization task as a finite Markov decision process, and develop a multi-layer Monte Carlo tree search method to produce a promising solution with reasonable complexity, which evaluates the outcome of given azimuth and downtilt settings without acquiring all configurations of antennas. Experiments in a real urban environment show that our proposed scheme outperforms the state-of-the-art algorithms over 10% in terms of system throughput while guaranteeing high power coverage.

In massive multiple-input-multiple-output (MIMO) systems, wireless channels are estimated at a fixed and short time interval for all users, which causes redundancy since most of users in practice might have a considerably larger coherence time than the prescribed time interval of estimation, and consequently wastes considerable signaling resources on channel acquisition. In this paper, we propose a novel channel estimation scheme for time-division-duplex massive MIMO systems, which can fully exploit the redundancy by exploring the temporal channel correlation underlying the channel aging effect. We also derive a rigorous lower bound on the achievable spectral efficiency, and maximize the lower bound to determine the optimal time interval of channel estimation. Numerical results show that the proposed estimation scheme can offer great spectral efficiency gains over the conventional one, and provide insights on how to put into practice the proposed scheme.

Unmanned aerial vehicle (UAV) is a promising way to collect data generated by wireless sensor networks, nevertheless, the battery capacity of the UAV restricts its application on many occasions, e.g., the network deployed in the wild. In this paper, we propose a coordinated route planning scheme to deal with the energy issue of the UAV, where a truck carrying backup batteries moves together with the UAV as a “mobile recharging station”. Our optimization task is to minimize the total mission time for collecting data from all the sensor nodes. We develop an efficient clustering algorithm to divide the entire mission area into multiple subregions in a load-balanced way to minimize the number of movements of the UAV, and formulate the trajectory planning task as a coordinated traveling salesman problem which is heuristically solved by a three-step route planning algorithm. Numerical results show that our proposed scheme provides an effective and cost-efficient way for the data collection of wireless sensor networks in practical application scenarios.

Radio access network (RAN) slicing is one of the key technologies in 5G and beyond mobile networks, where multiple logical RANs, also referred to as RAN slices, are allowed to run on top of the same physical infrastructure so as to provide slice-specific services. Due to the dynamic environments of wireless cells and the diverse requirements of RAN slices, inter-slice radio resource management (IS-RRM) is a highly challenging task. In this paper, we propose a novel online convex optimization (OCO) framework for the IS-RRM, where the instant resource allocation is learned by using historical data revealed from previous allocations. Compared with the existing methods, OCO is an online optimization process that can avoid sophisticated modeling and tuning in highly complicated and dynamic environments. Specifically, a low-complexity online ISRRM algorithm is proposed, which employs multiple expert-algorithms running parallelly to keep track of environmental changes. Simulation results show that the proposed method can provide efficient IS-RRM with a comparable performance to the optimal strategies in hindsight.

Densely deployed base station (BS) network is one of the important technologies for 5G and beyond mobile communication system, which improves the system throughput by deploying a large number of BSs in the service area. However, such a mobile network has to deal with the consequent power issue since the energy consumption of the BSs generally accounts for a substantial part of the whole system. In this paper, we propose an intelligent BS sleeping scheme to reduce the system energy consumption as much as possible with reasonable signaling overhead while guaranteeing the quality of experience of users. First, we introduce a long short term memory learning method to forecast the traffic distribution in the service area, by which we can determine when the BS sleeping operation is triggered; second, we develop an efficient three-step procedure to determine which of the BSs would sleep or be wakened. Experiment results show that our proposed traffic prediction method works quite well in practical scenarios. The prediction error is not more than 10%, and the energy consumption decreases more than 40% in average for a commercial mobile network with 20 BSs.

Radio access network (RAN) slicing is one of the key technologies in 5G and beyond mobile networks, where multiple logical subnets, i.e., RAN slices, are allowed to run on top of the same physical infrastructure so as to provide slice-specific services. Due to the dynamic environments of wireless networks and the diverse requirements of RAN slices, inter-slice radio resource management (IS-RRM) has become a highly challenging task in RAN slicing. In this paper, we propose a novel online convex optimization (OCO) framework for IS-RRM, which directly learns the instant resource allocation from the data revealed by previous allocations, such that sophisticated modeling and parameterization can be avoided in highly complicated and dynamic wireless environments. Specifically, an online IS-RRM scheme that employs multiple expert-algorithms running in parallel is proposed to keep track of the environmental changes and adjust the resource allocation accordingly. Both theoretical analysis and simulation results show that our proposed scheme can guarantee long-term performance comparable to the optimal strategies given in hindsight.

Spectrum sensing is one of the main components of the cognitive radio (CR) system, based on which secondary users (SUs) can get access to spectrum holes available. On the other hand, a malicious adversary can also attack the primary user (PU) system and the legitimate CR system via spectrum sensing, which can lead to serious security issue for both systems. In this article, we study an online attacking strategy, referred to as PU emulation attacker (PEA), which transmits forged PU signals over available channels to deteriorate the spectrum sensing performance of the SUs. We propose an online learning based attacking scheme for both the single attacker and the multiple-attacker cases, and analyze the regret upper bound of the proposed algorithm. The proposed PEA strategy can work in both stationary and non-stationary CR networks where the statistical characteristics of channels and the access strategy of SUs change over time. Numerical results show that it is more efficient than others in two different performance metrics: successful accesses of SUs and effective attacks of PEAs. Our proposal raises an interesting open question on how to develop CR networks with security guarantee.

Low Earth orbit (LEO) satellites play an important role to realize personal global communication in future mobile communication networks, where terrestrial users can be covered by multiple satellites due to densely deployed satellites in the constellation. Since the speed of LEO satellites is much higher than that of mobile users, it yields a large amount of satellite handovers, which causes heavy signaling overhead. Also, terrestrial users need to compete for satellite channels while they can only obtain partial information of the satellite system from their individual views. Thus, a distributed satellite handover strategy is required to balance satellite load to avoid network congestion, while at the same time, maintain low signalling overhead. In this paper, we propose a novel satellite handover strategy based on multi-agent reinforcement learning that aims to minimize average satellite handovers while satisfying the load constraint of each satellite. Simulation results show that the proposed strategy outperforms the local handover strategies based on basic criteria in terms of average satellite handover and user blocking rate.

Efficient probing spectrum holes is one of the most challenging tasks for the secondary user (SU) in a cognitive radio (CR) network. In this paper, we introduce a novel spectrum sensing framework where the duration for sensing at each time slot is variable. Sensing more channels increases the probability of finding a spectral hole, however, it would spend more time for sensing inevitably, which reduces the time for data transmission at a given time slot. Considering the sensing-transmission tradeoff, the optimization goal of spectrum sensing strategy is set to maximize the expected achievable throughput of the SU, which is formulated as a partially observable Markov decision process (POMDP). Finding an optimal solution to this optimization problem is computationally expensive due to its large state space, as well as large action space. We develop a novel spectrum sensing strategy based on deep reinforcement learning, which converges fast and can deal with complex scenario. Numerical results show that our proposed strategy can improve system throughput significantly.

Antenna selection (AS) is a promising technology that can reduce the implementation complexity and hardware cost of a massive MIMO system, in which a part of the attainable antennas are selected and connected to the radio frequency chains in each time slot. In this article, we propose a low-complexity AS algorithm to maximize the downlink achievable rate in the time-division duplexing massive MIMO system, which is based on online Thompson sampling technique and significantly reduces the pilot overhead required for channel estimation with only partial channel state information. We prove that the distribution-dependent upper bound of the proposed algorithm is sub-linear as a function of time slot by introducing the concept of regret. We also develop three discounting factors to accommodate the large-scale variations across antennas. Numerical results show that the downlink achievable rate can be greatly improved with our proposed scheme as compared to other typical ones.

In this paper, we investigate the power allocation problem with perfect channel state information (CSI) for downlink sum rate maximization in time duplex division massive MIMO systems. We note that the downlink sum rate is generally a non-convex function of the power allocation vector and the corresponding solution space increases exponentially with the number of simultaneous users, which make the optimal power allocation computationally intractable in general. Here, we introduce an online paradigm to achieve a tradeoff between computational complexity and sum rate performance, which utilizes the state-of-art online learning techniques to iteratively update the power allocation according to continuous CSIs. The computational complexity of the proposed algorithm is highly reduced by using the first order optimization technique and the system sum rate is improved by exploiting the time correlation of wireless channels. Simulation results show that the downlink sum rate can be increased by 15%-100% by using our proposed algorithm, compared with the conventional average and pathlossbased power allocation schemes.

Recently, unmaned aerial vehicle (UAV) assisted wireless network has been recognized as a promising technology for post-disaster communications, in which multiple UAVs are dynamically deployed in the air as cellular base stations to provide public wireless connectivities when the ground infrastructure collapses. However, due to the non- uniformity of post-disaster traffic distribution and the large scale of post-disaster area, trajectory planning is considered as a major challenge for UAV- assisted post-disaster communications in many aspects including coverage, energy efficiency and computational complexity. In this paper, we consider a general trajectory planning problem for multi-UAV assisted wireless networks in a post-disaster scenario. We show that the problem can be formulated as a multi-depot vehicle routing problem. Then, we propose two heuristic algorithms that can efficiently utilize the battery of UAVs to improve the coverage performance. Simulation results show that, compared to the intuitive greedy algorithm, the coverage ratio can be improved by 8% and 28%, respectively, by using the proposed algorithms.

In this paper, we consider a massive multiple-input- multiple-output (MIMO) system, where the base station (BS) is equipped with a large number of antennas while serving a much smaller number of users simultaneously. Though massive MIMO systems can provide significant spectral and energy efficiency via simple signal processing, the required channel state information (CSI) overhead is still a huge challenge, especially for the FDD mode. The basic frame structure of the FDD massive MIMO does not fully exploit the CSI feedback waiting phase since the BS needs to wait some time for the CSI feedback sent by users and then transmit data in downlink with the estimated CSI. The proportion of the CSI feedback waiting phase in the downlink transmission would be high as the MIMO scaling up, which reduces downlink rates for the FDD massive MIMO systems to some extent. In this paper, we propose two novel downlink precoding and transmission (DPT) schemes for FDD systems by exploiting the CSI feedback waiting phase. The corresponding performance comparisons between our proposed DPT methods, the contemporary DPT scheme and the ideal DPT one are also provided based on the COST 2100 outdoor channel model. Numerical results show that one of our proposed DPT schemes can achieve higher downlink rates than the contemporary scheme in relative low-mobility scenarios. The other proposed DPT scheme performs much better and is robust to user mobility.

Content delivery networks (CDNs) have been proposed to improve the performance of large-scale content delivery in communication networks, in which content files are dynamically cached in CDN nodes that are close to end-users so as to decrease the transmission latency and traffic redundancy. We investigate a real CDN that is currently utilized by a large social network company in China. We note that there are two important issues that are not well considered in the existing literature. The first is that end-users can usually access multiple CDN nodes with high quality of experience. The second is that the data service price may vary greatly for different regions, which can highly influence the cost of CDNs for large-scale applications. We reconsider the optimal caching and routing problem in CDNs by jointly considering these two practical issues, and propose a joint caching and routing strategy by using the alternating optimization technique. Simulation results show that the proposed algorithm outperforms the current CDN strategy, in which most popular files are cached and end-users are directed to the nearest CDN nodes, by 30% and 12% in terms of latency and data service cost, respectively.

Cognitive radio is an efficient spectrum sharing mechanism to solve the contradiction between spectrum shortage and spectrum underutility, where secondary users (SUs) are allowed to access the spectrum licensed to primary users (PUs) in an opportunistic manner. In cognitive radio networks with multiple access points (APs), due to the information exchange cost and system flexibility, APs may not cooperate with each other and there usually does not exist a central controller in practice. We propose a distributed user association scheme based on multi-agent reinforcement learning to achieve load balancing for cognitive radio networks with multiple independent APs. In our proposed scheme, APs execute reinforcement learning process independently to derive optimal policies on user association. In each iteration, APs make decisions on choosing SUs for association and then SUs choose the optimal AP for association based on the offers of all APs, the behaviors of APs and SUs is modeled as a dynamic matching game. Simulation results show that the proposed multi-agent reinforcement learning approach can highly improve the system performance with excellent robustness, compared to the conventional max-SINR method.

Due to the diversity of traffic scenes and high mobility of vehicles, the propagation environment between moving vehicles and road side access points can be highly dynamic, which causes unstable uplink connectivity and time-varying uplink data rates in vehicular networks. In this paper, we study the uplink performance of heterogenous vehicular networks, which integrate dedicated short-range communications (DSRCs) and Long Term Evolution vehicle-to-everything (LTE V2X) into a single vehicular network to provide high reliability, low latency, and wide area coverage. Here, we adopt a cluster-based approach, in which DSRC and LTE are utilized to provide vehicle-to-vehicle communications between vehicles within a cluster and vehicle-to-infrastructure communications between vehicles and access points, respectively. Specifically, a self-adaptive clustering method is proposed based on the iterative self-organizing data analysis technique algorithm, in which the number of clusters can automatically adjust to the optimal value according to the mobility information. Also, a joint load-bandwidth management scheme is proposed to distribute traffic load and bandwidth resources between DSRC and LTE. Simulation results show that the proposed algorithm outperforms the traditional section-based and ${K}$ -means clustering methods, and a tradeoff between average uplink data rate and signaling overhead can be achieved.

Due to the dense deployment of small cells, the number of mobile users served by each access point is decreasing dramatically, which leads to an increasingly dynamic cell load distribution over time and space. In order to match dynamic traffic load with static infrastructure capacity, a variety of load balancing methods have been proposed. However, the existing load balancing approaches either run self-tuning algorithms to optimize local handover parameters, which may not be optimal for the network-wide performance, or formulate a static optimization problem for a ``snapshot" network, which may require a huge amount of handover signaling in dynamic situations. In this paper, we consider the load balancing problem with dynamic traffic models, in which the average throughput over a certain time period is maximized, while the average packet delay is guaranteed to be below a certain threshold. Simulation results show that our proposed user association and resource allocation algorithm can highly increase the average throughput, compared with the baseline algorithm using the maximum SINR association and equal resource allocation.

Dynamic spectrum sharing (DSS) has been envisioned as a promising approach to address the imminent shortage of spectrum resources caused by the exploding growth of wireless traffic. Contrast to the existing DSS approaches which are either based on individual authorization using a geolocation database, such as Licensed Shared Access and Spectrum Access System, or general authorization using spectrum sensing techniques, such as Collective Use of Spectrum, in this paper, we propose a threetier hierarchical model of DSS based on hybrid authorization, in which primary licensees (PLs) and secondary licensees (SLs) register with a geolocation database to guarantee predictable quality-of-service, and tertiary licensees (TLs) opportunistically access the small spectrum holes in time and space to further improve the spectrum usage. We provide a mathematical analysis of the optimal access strategy for each type of licensees in the proposed hierarchical model, such that the weighted sum throughput of SLs and TLs is maximized while the PL throughput is guaranteed. Simulation results show that the proposed hierarchical approach can highly increase the system performance, compared to the current database-driven approaches.

Low-altitude unmanned aerial vehicles (UAVs), which can act as flying base stations, are proposed as a promising paradigm to meet the ever-increasing traffic demand and enhance the coverage of terrestrial mobile communication system. However, how to determine the position and the service region of each UAV is a problem to demand urgent solutions. In this paper, we deal with this burning issue from a load balancing perspective. First, we propose an efficient subregion partition method to make each UAV serve almost equal traffic demand, where we try to minimize the maximum traffic demand of subregions with constraints of the traffic demand and the shape of subregions. Then, we propose a local search procedure to relocate UAVs using backtracking line search algorithm. The service subregions and the positions of UAVs are updated in an iterative manner until the optimum solution is produced. Numerical results show that our proposed strategy can serve more users compared with other ones. Moreover, the traffic distribution among UAVs is more balanced, as well as and the service areas of subregions, indicating the effectiveness and the efficiency of our proposal.

In this paper, we study the resource allocation issues related to OFDM-based multiuser cognitive radio cellular networks, where both of primary network and secondary network are investigated to make an all-round evaluation of the network performance, including the Quality of Experience (QoE) of the primary users (PUs) and the minimal rate requirement of the secondary users (SUs). Since the general resource allocation optimization task we formulate leads to challenging mixed integer nonlinear programming problems, we first propose a heuristic subchannel allocation method to remove the awkward integer constraints. Then an efficient distribution algorithm, which yields nearly linear computational complexity, is developed to work out the optimal power allocation. Simulation results indicate that our proposed resource allocation scheme works better under different performance metrics as compared to others. Moreover, our proposed distribution algorithms converge stably and quickly.

Heterogeneous cellular network (HCN), which generally consists of small cell base stations (SBSs) and macro base stations (MBSs), is proposed as a promising scheme to improve the capacity of the cellular network. However, huge energy consumption by the densely deployed SBSs arises as a challenging problem that should be addressed properly to fulfill the potential of HCNs. In this paper, we aim to minimize the total power consumption of the HCNs by jointly designing energy-efficient user association and SBS switching schemes. We develop an approximation algorithm to solve the intractable user association problem efficiently, based on which an efficient local search procedure is introduced to minimize the total power consumption of the HCN by controlling active/inactive state of each SBS dynamically. Numerical results demonstrate that our proposed method reduces the energy consumption of the HCN significantly as compared to other representative methods.

Heterogeneous cloud radio access network (H-CRAN) is proposed as a cost-effective paradigm to meet the ever-increasing mobile data traffic demand, where the key idea is applying cloud computing technologies in a heterogeneous network (HetNet) to improve both spectral and energy efficiencies of the cellular system. In this paper, we investigate how to provide as many as possible users with QoS-guaranteed services in the H-CRAN. Our optimization task is to maximize the number of users with rate requirements for a given set of access points with bandwidth and transmission power budgets. We develop a novel user association strategy, where we present a Reference Power concept and develop an approximation algorithm to address the intractable user association problem. Numerical results indicate that our proposed user association strategy can not only increase the fully satisfied users significantly as compared to other methods, but also fulfill the capacity potential of the H-CRAN.

Cloud radio access network (C-RAN) is deemed as a promising architecture to meet the exponentially increasing traffic demand in mobile networks, where baseband processing is separated from remote radio heads (RRHs) and performed in a centralized baseband unit (BBU) pool. However, the densely deployed RRHs, as well as the passive optical network which provides high capacity backhauls between the RRHs and the BBU pool, consume a large amount of energy. In this paper, we propose efficient RRH switching schemes to achieve a tradeoff between the system energy saving and the load balance among the RRHs in the C-RAN. We first develop an approximation algorithm to address the intractable user association problem for a given set of RRHs, based on which we introduce efficient local search algorithms to perform RRH selection procedure, which can reduce the load fairness index of the C-RAN by controlling the active/inactive state of each RRH. We also discuss the handover signalling overhead issue and introduce an adaptive trigger mechanism to avoid switching on/off too many RRHs simultaneously so as to keep the signalling overhead of the C-RAN below an acceptable level. Numerical results demonstrate that the proposed RRH switching schemes can improve the system performance of the C-RAN significantly. Moreover, our proposal sheds light on how to design effective and efficient handover schemes for next generation mobile networks.

In this paper, we propose a clustering-based resource allocation (RA) scheme for the multiuser orthogonal frequency division multiplexing (OFDM)-based cognitive radio network, where we aim to maximize the sum capacity of the secondary users (SUs) subject to practical constraints in wireless environment. Our general RA optimization task leads to a challenging mixed integer programming problem that is computationally intractable. We first introduce a simple and efficient clustering method to divide all the SUs into multiple groups based on their mutual interference degrees, where the SUs in different groups can share the same OFDM subchannels to improve spectrum utilization efficiency, while the SUs with heavy mutual interference cluster together in the same group and employ different subchannels to alleviate their mutual interference. Then we develop efficient radio RA algorithms to maximize the sum rate of the SUs in each cluster. A user-oriented subchannel assignment method is presented to remove the awkward integer constraints of the formulated RA problem, followed by a fast power distribution algorithm that can work out optimal solutions with an approximate linear complexity. Simulation results indicate that our proposed RA scheme can improve the throughput of the SUs significantly as compared with other methods. Moreover, our proposed RA algorithms converge stably and quickly.

Since user experience plays a more and more important role in the development of today's communication systems, quality of experience (QoE) becomes a widely used metric, which reflects the subjective experience of end users for wireless services. In this paper, a QoE-driven radio resource allocation scheme is proposed for multi-user orthogonal frequency division multiplexing (OFDM) based cognitive radio (CR) networks. By introducing a QoE-based assessment model, our optimization task is formulated to maximize the overall QoE of the considered system subject to the total transmit power and interference constraints. We first propose a user-oriented subcarrier allocation algorithm. Secondly, for a given subchannel assignment, we propose a fast barrier method to tackle the optimal power allocation problem, where the key is to replace Newton step with complexity O(N3) in the barrier method by a procedure with approximate linear complexity, which is developed by exploiting the structure of the optimization problem. Simulation results validate that our method can always reach a better performance in terms of both single user QoE and the overall QoE with lower complexity.

Smart grid is characterized by a large number of smart meters (SMs) that exchange huge amounts of data with control center, where an effective communication network is required to guarantee reliable data transmission. In this paper, we introduce software defined network (SDN) technology to the smart grid, which decouples the control plane from the data plane so as to satisfy the communication requirements in the mart grid effectively. Aggregation points (APs) are employed in the data plane to process and forward data between SMs and control center. A general mathematical model is formulated to plan the APs, where we try to minimize the total deployment cost, including the opening expenditure of the APs, the connection cost between SMs and APs, and the connection cost between APs and control center. We present a 5-approximation algorithm to address the generated NP-hard problem, which yields performance-guaranteed solutions. Three representative scenarios are investigated to verity the efficiency of our proposal. Numerical results show that our proposed algorithm has great advantages over other heuristic ones.

Cloud Radio Access Network (C-RAN) is deemed as a promising architecture to enhance the spectrum and energy efficiency of the cellular systems. The key feature of the C-RAN is to separate the baseband processing function from conventional remote radio heads (RRHs) and process the baseband signals in a centralized baseband units (BBUs) pool to obtain statistical multiplexing gain. In this paper, we investigate the BBU pool placement problem in the C-RAN, where we try to minimize the deployment cost of the BBUs while considering the processing capacity of the BBUs, the traffic demands of the RRHs and the signal synchronization among them. Our general problem formulation leads to an NP-hard integer programming optimization task. We propose an efficient local search algorithm to address it. Numerical results show that the proposed algorithm converges quickly. Moreover, it yields a performance-guaranteed placement scheme for the BBU pool in the C-RAN.

Cloud radio access network (C-RAN) is considered as a cost- and energy-efficient architecture to meet the ever-increasing mobile data traffic, where baseband processing is decoupled from the remote radio head (RRH) and conducted in a centralized baseband unit (BBU) pool. Though C-RAN enables efficient resource allocation and reduces network operation cost, the optical transport network power consumption between the RRHs and the BBU pool is usually enormous. In this paper, we focus on the minimization of the total power consumption by jointly considering the transport network power and the transmission power of the RRHs. We formulate a general RRH selection task that falls into the form of capacitated facility location problem (CFLP) and propose an efficient local search algorithm to address it. Numerical results validate that our proposal can significantly reduce the total power consumption of the C-RAN.

Cloud radio access network (C-RAN) is deemed as a promising architecture to meet the exponentially increasing traffic demand in a mobile network. However, it needs a huge amount of transport capacity between the remote radio heads (RRHs) and the baseband unit pool. Though an optical transport network can be employed to support such a massive capacity, it always leads to enormous power consumption that is comparable with the transmission powers of the RRHs, which can significantly decrease the performance of the C-RAN if not well addressed. In this paper, we investigate the power saving issue in the C-RAN, where we attempt to get a power consumption tradeoff between the optical transport network and the RRHs. An RRH selection problem is formulated to achieve this goal, which is based on the traffic density of the service area. Our optimization task is to select a subset of the RRHs to minimize the total power consumption of the C-RAN while satisfying a series of network constraints, including the power and bandwidth budgets of the RRHs, the traffic demands of users, and the spectral efficiency. We develop an efficient local search algorithm, which includes three types of local improvement operations: “add,” “open,” and “close,” to address the intractable optimization task. Numerical results indicate that our proposal can significantly reduce the power consumption of the C-RAN. Moreover, our proposed algorithm converges stably and quickly, indicating that it is promising for applications.

Aggregation Point (AP) plays a vital role in smart grid, which forwards data stream between the client terminals and the control center in smart grid communication system. In this paper, we investigate two representative AP planning models: wired and wireless, and develop efficient algorithms to address the formulated AP planning problems in a cost-efficient way. For the wired case, a 5-approximation algorithm is proposed to minimize the total capital expenditure with the consideration of the installation cost of each AP in the neighborhood area networks (NANs) and the connecting cost between the AP and the home area network (HANs) served by it. For the wireless media-based networking scenario, an O(log W)-approximation algorithm is presented to minimize the total deployment cost of the opening APs under their coverage constraints, where W is the maximum capacity among these APs. Numerical results show that our proposed approximation algorithms have great advantages compared to other heuristic methods.

Heterogeneous cloud radio access networks (H-CRANs) have been proposed as a promising architecture for providing high energy efficiency, high spectral efficiency and high data rate at low cost. In this paper, we develop a novel resource allocation scheme among macro base stations (BSs) and remote radio heads (RRHs) for the H-CRANs. The key idea of our proposal is that we design the coverage regions of macro BSs in a seamless and balanced way. Secondly, we calculate the areas that can not rely on macro BSs for reliable high speed data transmission and allocate resources in these areas among RRHs in the balanced way to alleviate the high transmission pressure on fronthauls between RRHs and the centralized baseband unit (BBU) pool. The proposed resource allocation scheme is triggered when severe disbalance is detected. Numerical results show that our proposal can perform quite well for the data traffic distribution in a real city environment. It provides QoS guaranteed performance with lower capital expenditure (CAPEX) and operating expenditure (OPEX).

Cloud radio access network (CRAN) has been seen as an effective means to address the challenges faced by cellular radio networks, such as high capital expenditure and operating expense, high energy consumption and low spectral efficiency. Especially, CRAN has the potential to equip a cellular network with the load-balancing capability to cope with the non-uniformly distributed traffic in the service area. In this paper, we develop an optimal load balancing scheme for CRAN-based cellular systems by employing an infinite optimization technique. A fairness index is defined to measure the load balancing level of the cellular system and monitored by periodically inspecting the load distribution among all cells. When the fairness index is below a warning threshold, we divide the service zone into compact and connected subregions based on an infinite optimization formulation. Each subregion served by a cell has almost equal area and almost equal throughput requirement. To avoid yielding ill-shaped subregion that is difficult to be covered by a practical cell, a penalty term is introduced to the objective function. Then we update the cell association of each user so that the fairness index return to an acceptable level. Numerical experiments show that our proposal can provide performance-guaranteed load balancing for the cellular network with almost no additional operating expense.

Deploying femtocells underlaying macrocells is a promising way to improve the capacity and enhance the coverage of a cellular system. However, such a heterogeneous network also gives rise to cross-tier and intra-tier interference issue that should be addressed properly in order to acquire the expected performance gain. In this paper, we study the resource allocation (RA) problem in a two-tier Orthogonal Frequency Division Multiplexing Access (OFDM)-based cellular networks, where the femtocells that employ closed access strategy to share subchannels with the macrocells are equipped with cognitive radio (CR) function to identify radio environment so as to choose the subchannels that can yield less interference to the macrocell users in the coverage of the femtocells. Our optimization task is to maximize the sum throughput of the femtocell users under the consideration of imperfect spectrum sensing, while controlling the interference to the MUs under their bearable thresholds in the sense of probability resulting from imperfect channel state information. We introduce a conservative convex approximation to the formulated problem and develop a fast algorithm to solve it by exploiting its structure. Simulation results show our proposed RA scheme can improve the system throughput with almost no changes of the infrastructure of the cellular network.

In this paper, we investigate the energy consumption issue in cognitive radio (CR) networks. We aim to maximize the energy efficiency of the CR network while considering practical restrictions, including the power budget of the system, the interference thresholds of the primary users (PUs), the rate requirements of the secondary users, and the fairness among them. Particularly, due to the lack of explicit support from the PU system, perfect channel state information may not be acquired. Thus, the interference constraint is posed as chance-constrained form and tackled by Bernstein approximation. Then, we convert the optimization task into a quasi-convex problem via relaxing the integer variables, followed by a simple rounding technique to yield feasible subchannels assignment. We derive a fast algorithm to distribute power among subchannels by exploiting the structure of the power-allocation problem. Moreover, we give an efficient heuristic algorithm for subchannels assignment, which reduces the computation load dramatically. Simulation results show that both our proposed resource allocation schemes perform well in practical scenarios. The energy efficiency obtained by the integer subchannels assignment and the fast power distribution achieves more than 98% of the upper bound. On the other hand, the proposed heuristic subchannels assignment with optimal power allocation achieves a good tradeoff between computation complexity and energy efficiency.

Device-to-device (D2D) communication is deemed as a promising technology to improve the spectrum efficiency of the cellular systems. In this paper, we study a resource sharing scheme for the D2D communication underlaying cellular networks, where multiple D2D pairs can share subchannels with multiple cellular users (CUs). Our optimization task is to maximize the sum rate of D2D pairs while satisfying the rate requirements of all CUs. The formulated problem falls naturally into a mixed integer programming form that is intractable. We first develop a subchannel sharing protocol, by which we can determine whether or not a subchannel can be reused by two D2D pairs. Then, we prove that the problem can be well approximated by ignoring the mutual interference among the D2D pairs that share the same subchannels without deteriorating the performance of both the CUs and the D2D pairs. Based on the analysis results, we propose an efficient subchannel allocation scheme by employing a simple greedy strategy, as well as a power distribution algorithm that can work out almost the optimal solution to the original problem. Numerical results show that our proposal can significantly increase spectrum efficiency of the cellular system. Furthermore, our proposed subchannel sharing protocol is effective and efficient for practical communication scenarios.

One fundamental question for wireless power transfer technology is the energy provisioning problem, i.e., how to provide sufficient energy to mobile rechargeable nodes for their continuous operation. Most existing works overlooked the impacts of node speed and battery capacity. However, we find that if the constraints of node speed and battery capacity are considered, the continuous operation of nodes may never be guaranteed, which invalidates the traditional energy provisioning concept. In this paper, we propose a novel metric-Quality of Energy Provisioning (QoEP)-to characterize the expected portion of time that a node sustains normal operation by taking into account node speed and battery capacity. To avoid confining the analysis to a specific mobility model, we study spatial distribution instead. As there exist more than one mobility models corresponding to the same spatial distribution, and different mobility models typically lead to different QoEPs, we investigate upper and lower bounds of QoEP in 1D and 2D cases. We derive tight upper and lower bounds of QoEP for 1D case with a single source, and tight lower bounds and loose upper bounds for general 1D and 2D cases with multiple sources. Finally, we perform extensive simulations to verify our theoretical findings.

Energy efficiency in communication networks has received increasing attention in both industry and academia. In this paper, we investigate the energy saving issue in a heterogeneous network (HetNet), which is introduced to cellular radio networks to improve capacity and enhance coverage. A HetNet consists of base stations (BSs) with different transmission powers, resulting in systematic power control that is more complex than that in the conventional cellular networks. The main difficulty is addressing the mutual interference between different kinds of BSs. In this paper, we try to minimize the power consumption of an OFDM-based HetNet while satisfying all users' rate requirements, as well as considering the inter-cell interference. Our general problem formulation leads to a nonconvex optimization task that is generally hard to tackle. We derive a concave lower bound of user's achievable rate for a given power allocation, based on which an efficient iterative algorithm is developed to solve the formulated problem efficiently. Numerical results show that our proposed resource allocation scheme works well in different scenarios. The energy consumption of the cellular system is reduced dramatically compared to other schemes. Moreover, our proposed algorithm converges quickly and stably, showing great potential for applications.

In this paper, we investigate the Energy Efficiency (EE) - Spectrum Efficiency (SE) tradeoff issue in an OFDM-based cognitive radio (CR) network. A multi-objective resource allocation problem is formulated, where we try to maximize the EE and the SE simultaneously. The Pareto optimal set of the formulated problem is characterized by analyzing the relationship between the EE and the SE. To find a unique globally optimal solution, we proposed a unified EE-SE tradeoff metric, based on which the original optimization task is transformed into a single-objective problem that has a D.C. (Difference of two Convex functions/sets) structure. Then an efficient barrier method is developed, where we speeds up the time-consuming computation of Newton step by exploiting the structure of the D.C. programming problem. Simulation results validate the effectiveness and efficiency of the proposed algorithm. Our general problem formulation sheds some insights on how to design an energy- and spectrum-efficient CR system.

In this paper, we investigate the resource allocation problem in an OFDM-based cognitive radio system with multiple relays. We address the formulated intractable problem by a two-stage procedure which corresponds to two subproblems: subchannel allocation and power distribution. For each subproblem, a fast algorithm is proposed by exploiting its special structure. Our proposed algorithms can achieve performance close to the upper bound. The remarkable characteristic of the proposed algorithm is that the approximate linear complexity is significantly lower than that of standard techniques, making it promising for applications. Simulation results also validate the superior performance of the proposed algorithm.

Device-to-device (D2D) communication, which underlays a cellular network to improve the reuse of spectrum resources, is a promising technique that can increase system capacity, enhance cell coverage and extend battery lifetime of user equipments. In this paper, we study the power allocation for a D2D underlaying cellular network. Our optimization task is to maximize the sum rate of the cellular network subject to the maximum power budgets of both the cellular link and the D2D link. The formulated problem is nonconvex which is generally difficult to obtain the global optimum. We propose a low complexity iterative algorithm to work out promising solutions with reasonable complexity. We develop a linear complexity barrier method to get a concave lower bound of the sum rate, and prove that an improved solution can be obtained after each iteration until the proposed algorithm converges, when the lower bound is tight. Simulation results validate the effectiveness and the efficiency of our proposed algorithm.

Low-power access points, such as pico base stations (BSs), femto BSs, and relays are introduced to the next generation cellular systems to enhance coverage and improve system capacity. Deploying low-power access points to offload the conventional macro BSs is deemed as a spectrum- and cost-efficient way to meet the sharp increase of traffic requirements of cellular networks. However, it also leads to heterogeneous network framework and raises new challenges for cell planning. In this paper, we study the minimum cost cell planning problem in such a heterogeneous network. Our optimization task is to select a subset of candidate sites to lay BSs, including macro BSs, pico BSs and relays, to minimize the total deployment cost while satisfying the rate requirements of the demand nodes (DNs) served by the cellular network. We prove that the general case of the formulated problem is APX-hard, where a DN is constrained to be associated with only one BS. However, if each DN can be served by multiple BSs, which is a reasonable case for practical cellular systems, we show it is not APX-hard and develop an approximation algorithm to work out promising solutions. Our proposed algorithm guarantees an approximation ratio of O(logR) to the global optimum, where R is the maximum achievable capacity of the BSs. Numerical results indicate that our proposal can significantly reduce the deployment cost of the cellular network with given rate requirements of DNs compared to other cell planning schemes.

The power allocation problem in orthogonal frequency division multiplexing‐based cognitive radio (CR) networks with cooperative relays has been investigated here, where both the interference to primary users (PUs) and the power budget of the CR network are considered. The authors try to maximise the overall throughput of the CR network within the given constraints. The coupling variables in the formulated problem make it hard to solve, so an iterative optimisation scheme to find out the optimal solution with a controllable complexity is developed. First, the original problem is decomposed into two subproblems that can be solved independently. A fast barrier method has been employed to work out the optimal solution to one of the subproblems with a complexity of O ( L 2 N ), where L and N are the number of PUs and subcarriers, respectively. Then, an iterative procedure is developed to solve the other subproblem. Numerical results show that the proposed method can significantly increase the throughput of the CR system, comparing with other representative ones. Furthermore, the proposed algorithm gives a general power optimisation framework for CR networks with cooperative relays.

In this paper, we study the power allocation in Orthogonal Frequency Division Multiplexing (OFDM)-based Cognitive Radio (CR) networks with cooperative relay. Since the energy consumption is growing at a staggering rate, green radio, which puts emphasis on the energy-efficiency (EE) in wireless networks, is becoming increasingly important. Therefore, we try to maximize the EE of a relaying CR system, under the consideration of many practical limitations, such as transmission power budget, interference threshold of the primary users and the traffic demand. We first convert the formulated problem into a convex optimization one via its hypogragh form, which can be solved by the barrier method. By exploiting the special structure of the formulated convex optimization problem, we further speed up the computation of Newton step during the barrier method, significantly reducing the complexity of the algorithm. Numerical results validate that our proposal can exploit the overall EE of the CR system, while the algorithm converges efficiently and stably.

In this paper, we investigate the energy consumption issue of Cognitive Radio (CR) systems. We aim to maximize the energy efficiency of the considered CR system with practical constraints, such as the power budget of the CR system, the interference thresholds of the primary users, the minimal throughput requirements and the proportional fairness of the CR users. Since the formulated mixed integer programming task is generally hard to tackle, we relax the original problem and convert it into a quasiconvex one. A bisection-based algorithm is employed to work out the optimal solution in an iterative manner. In each iteration, we develop a fast barrier method to reduce the computational complexity by exploiting the problem's structure. Simulation results show that our proposed method can obtain solutions close to the upper bound with reasonable complexity.

In this paper, we study how to save energy in Heterogeneous Networks (HetNets), which is introduced to the next generation cellular systems. A HetNet based cellular system consists of a mix of macrocells and low power nodes, such as picocells, femtocells and relays, making the systematic power controlling more complex than the conventional ones. The major difficulty for power control in HetNets is the mutual interference among cells with different transmission power. So interference management is very important for the next generation cellular networks. We try to minimize the total power consumption while guaranteeing users' rate requirements, to save energy and keep the user's QoS from degenerating. Our general problem formulation leads to a nonconvex optimization problem which is generally hard to solve. We derive the lower bound of user's achievable rates with given power consumption and develop an efficient iterative algorithm to deal with the intractable optimization task. Numerical results show that our proposed algorithm performs well for practical wireless scenarios.

Relay nodes are introduced to the next generation cellular networks to enhance coverage and improve system capacity, leading to a new radio network planning paradigm. In this paper, we study two planning problems for cellular networks with relay nodes: Minimum cost cell planning and budgeted cell planning. The former is to minimize the total installation cost for opening base stations (BSs), including macro BSs and relay nodes, while satisfying all users' traffic demands. The latter is to maximize the number of users with predefined traffic demands under a given budget. Both of the problems are NP-hard. We present approximation algorithms to work out promising solutions to these problems. For the minimum cost cell planning, we develop an O(logW)-approximation algorithm, where W is the maximum capacity of macro BSs. For the budgeted cell planning, we prove that the problem is NP-hard to approximate and give an e−1 over 3e−1-approximation algorithm for a special case of the problem, which is general enough to meet practical requirements.

Low-power base stations (BSs), such as pico BSs, femto BSs, and relay nodes, are introduced to the heterogeneous cellular networks to enhance coverage and improve system capacity. Compare with macro BS, low-power BS has much lower transmission power, smaller physical size and lower cost. Deploying low-power BSs within the coverage of macro BSs is considered as a cost-efficient way to meet the sharp increase of wireless applications, leading to a new radio network planning paradigm for the next generation cellular networks. In this paper, we study the minimum cost cell planning problem in such heterogeneous networks, where planning task is how to select a subset of possible BS sites, including macro BSs, pico BSs and relay nodes, to minimize the total deployment cost while satisfying all rate requirements of demand nodes. We develop an approximation algorithm to tackle the formulated NP-hard problem, which guarantees an approximation ratio of O(log R) to the optimal solution, where R is the maximal achievable capacity of BSs.

Cognitive Radio (CR) is an attractive technology to deal with current spectrum scarcity problem, while cooperative relay can make distributed receivers benefit from spatial diversity and combat severe fading in wireless environment. CR with cooperative relay is potentially a promising paradigm for developing spectrum-efficient wireless systems. In this paper, we study the resource allocation in Orthogonal Frequency Division Multiplexing (OFDM)-based CR networks with cooperative relays. Since the formulated optimization task defines a mixed integer programming problem that is generally hard to solve, we propose a two-stage method to produce near optimal solutions. Particularly, by jointly considering the Signal-to-Noise Ratios (SNRs) of OFDM subchannels and the interferences introduced to primary users, we propose an efficient subchannel assignment scheme for the CR system, as well as transmission mode selection strategy. Furthermore, we develop a fast algorithm to distribute power among subchannels, which can always work out the optimal power allocation with a reasonable complexity by exploiting the structure of the problem. Numerical results show that our proposal can significantly increase the throughput of the CR system compared with other schemes, and the proposed algorithm converges quickly and stably.

In this paper we study the Resource Allocation (RA) in Orthogonal Frequency Division Multiplexing (OFDM)-based Cognitive Radio (CR) networks, under the consideration of many practical limitations such as imperfect spectrum sensing, limited transmission power, different traffic demands of secondary users, etc. The general RA optimization framework leads to a complex mixed integer programming task which is computationally intractable. We propose to address this hard task in two steps. For the first step, we perform subchannel allocation to satisfy heterogeneous users' rate requirements roughly and remove the intractable integer constraints of the optimization problem. For the second step, we perform power distribution among the OFDM subchannels. By exploiting the problem structure to speedup the Newton step, we propose a barrier-based method which is able to achieve the optimal power distribution with an almost linear complexity, significantly better than the complexity of standard techniques. Moreover, we propose a method which is able to approximate the optimal solution with a constant complexity. Numerical results validate that our proposal exploits the overall capacity of CR systems well subjected to different traffic demands of users and interference constraints with given power budget.

In this paper, we investigate the energy-efficient resource allocation in orthogonal frequency division multiplexing (OFDM)-based cognitive radio (CR) networks, where we try to maximize the system energy-efficiency under the consideration of many practical limitations, such as transmission power budget of the CR system, interference threshold of primary users and traffic demands of secondary users. Our general objective formulation leads to a challenging mixed integer programming problem that is hard to solve. To make it computationally tractable, we employ a time-sharing method to transform it into a non-linear fractional programming problem, which can be further converted into an equivalent convex optimization problem by using its hypogragh form. Based on these transformations, it is possible to obtain (near) optimal solution by standard optimization technique. However, the complexity of the standard technique is too high for this real-time optimization task. By exploiting the structure of the problem extensively, we develop an efficient barrier method to work out the (near) optimal solution with a reasonable complexity, significantly better than the standard technique. Numerical results show that our proposal can maximize the energy efficiency of the CR system, whilst the proposed algorithm performs quickly and stably.

Cognitive radio (CR) network with cooperative relay is potentially a promising technique to solve current spectrum inefficiency and spectrum scarcity problem. In this paper, we study the power allocation in cognitive radio (CR) networks with cooperative relay. The interference to the licensed primary system and the power budget of the CR network are jointly considered. The resulting model is general and the formulated optimization task generally hard to solve. To deal with the intractable coupling variables in the concerned problem, we propose an alternating optimization method, which can work out the optimal solution in an iteration manner and converges rapidly. Experiment results show that our proposed method can significantly increase system capacity, comparing with other representative ones. Furthermore, the proposed method gives a general power optimization framework for CR networks with cooperative relay and can be extended to other scenarios easily.

In this paper we study the resource allocation in OFDM-based cognitive radio (CR) networks, under the consideration of many practical limitations such as imperfect spectrum sensing, limited transmission power, different traffic demands of secondary users, etc. We formulated this general problem as a mixed integer programming task. Considering that this optimization task is computationally intractable, we propose to address it in two steps. For the first step, we perform subchannel allocation to satisfy heterogeneous users' rate requirement roughly and remove the integer constraints of the optimization problem. For the second step, we perform power allocation among the subchannels. By exploiting the problem structure to speedup the Newton step, we propose a Barrier-based method which is able to achieve the optimal power distribution with a complexity of O(N), where N is the number of active OFDM subchannels, significantly better than the complexity of O(N3) of standard techniques. Moreover, we proposed a method which is able to approximate the optimal solution with a constant complexity. Numerical results validate that our proposal exploits the overall capacity of CR systems well subjected to different traffic demands of users.

In this paper we study the resource allocation in multiuser orthogonal frequency division multiplexing (OFDM)-based cognitive radio (CR) networks, where secondary users (SUs) have flexible traffic demands, including heterogeneous real-time (RT) and non-real-time (NRT) services. We try to maximize the sum capacity of the NRT users and maintain the minimal rate requirements of the RT users simultaneously. Additionally, the interference introduced to primary users, which is generated by the access of the SUs, should be kept below a predefined threshold, which makes the optimization task more complex. The contribution of this work is two folds. First, we show that the formulated optimization problem has a special structure which can be exploited to implement a fast barrier method to obtain the optimal solution with a reasonable complexity. Second, we propose an effective measurement criterion to normalize OFDM subchannels' achievable rates, based on which we develop simple but efficient heuristic algorithm for subchannel assignment and power distribution. Simulation results show that our proposed resource allocation schemes work quite well for concerned wireless scenarios. The fast barrier method converges very fast and can always work out the optimal solution, while the heuristic algorithm produces solution close to the optimal with much lower complexity.

Optimum multiuser detection (OMD) in direct-sequence code-division multiple access (DS-CDMA) communication systems is a combinatorial optimization problem and has been proven NP-complete. Many heuristics have been presented to solve this problem, but few of them consider the fitness landscape of OMD carefully. In this paper, we analyze the fitness landscape of OMD, including the neighborhood structure and the distribution of local optima. Numerical results give hints on how to design efficient heuristic algorithms for the problem. A meta-heuristic algorithm considering the analysis results is proposed With a proper local search and a well-chosen perturbation strategy, the proposed algorithm can find the (near) optimal solution rapidly with lower computational complexity. Simulation results show it outperforms other heuristic multiuser detection algorithms when the number of users is large. In the condition of small number of users, it can achieve the bit error rate (BER) bound of OMD.