Fine-tuning (FT) large language models (LLMs) is crucial for adapting general-purpose models to specific tasks, enhancing accuracy and relevance with minimal resources. To further enhance generalization ability while reducing training costs, this paper proposes Federated LoRA with Dropout (FedLoDrop), a new framework that applies dropout to the rows and columns of the trainable matrix in Federated LoRA. A generalization error bound and convergence analysis under sparsity regularization are obtained, which elucidate the fundamental trade-off between underfitting and overfitting. The error bound reveals that a higher dropout rate increases model sparsity, thereby lowering the upper bound of pointwise hypothesis stability (PHS). While this reduces the gap between empirical and generalization errors, it also incurs a higher empirical error, which, together with the gap, determines the overall generalization error. On the other hand, though dropout reduces communication costs, deploying FedLoDrop at the network edge still faces challenges due to limited network resources. To address this issue, an optimization problem is formulated to minimize the upper bound of the generalization error, by jointly optimizing the dropout rate and resource allocation subject to the latency and per-device energy consumption constraints. To solve this problem, a branch-and-bound (B&B)-based method is proposed to obtain its globally optimal solution. Moreover, to reduce the high computational complexity of the B&B-based method, a penalized successive convex approximation (P-SCA)-based algorithm is proposed to efficiently obtain its high-quality suboptimal solution. Finally, numerical results demonstrate the effectiveness of the proposed approach in mitigating overfitting and improving the generalization capability.

Wireless sensing has recently found widespread applications in diverse environments, including homes, offices, and public spaces. By analyzing patterns in channel state information (CSI), it is possible to infer human actions for tasks such as person identification, gesture recognition, and fall detection. However, CSI is highly sensitive to environmental changes, where even minor alterations can significantly distort the CSI patterns. This sensitivity often leads to performance degradation or outright failure when applying wireless sensing models trained in one environment to another. To address this challenge, Domain Alignment Learning (DAL) has been widely adopted for cross-domain classification tasks, as it focuses on aligning the global distributions of the source and target domains in feature space. Despite its popularity, DAL often neglects inter-category relationships, which can lead to misalignment between categories across domains, even when global alignment is achieved. To overcome these limitations, we propose K-Nearest Neighbors Maximum Mean Discrepancy (KNN-MMD), a novel few-shot method for cross-domain wireless sensing. Our approach begins by constructing a “help set” using K-Nearest Neighbors (KNN) from the target domain, enabling local alignment between the source and target domains within each category using Maximum Mean Discrepancy (MMD). Additionally, we address a key instability issue commonly observed in cross-domain methods, where model performance fluctuates sharply between epochs. Further, most existing methods struggle to determine an optimal stopping point during training due to the absence of labeled data from the target domain. Our method resolves this by excluding the support set from the target domain during training and employing it as a validation set to determine the stopping criterion. We evaluate the effectiveness of the proposed method across several cross-domain Wi-Fi sensing tasks, including gesture recognition, person identification, fall detection, and action recognition, using both a public dataset and a self-collected dataset. In a one-shot scenario, our method achieves accuracy rates of 93.26%, 81.84%, 77.62%, and 75.30% for the respective tasks. The dataset and code are publicly available athttps://github.com/RS2002/KNN-MMD.

With the prevalence of emerging artificial intelligence services in next-generation wireless edge networks, Split Federated Learning (SFL), which divides a learning model into server-side and client-side models, has emerged as an appealing technology to deal with the heavy computational burden for network edge clients. However, existing SFL frameworks would frequently upload smashed data and download gradients between the server and each client, leading to severe communication overheads. To address this issue, this work proposes a novel communication-and-computation efficient SFL framework, which allows dynamic model splitting (server- and client-side model cutting point selection) and broadcasting of aggregated smashed data gradients. We theoretically analyze the impact of the cutting point selection on the convergence rate, revealing that model splitting with a smaller client-side model size leads to a better convergence performance and vise versa. Based on the above insights, we formulate an optimization problem to minimize the model convergence rate and latency under the consideration of data privacy via a joint Cutting point selection, Communication and Computation resource allocation (CCC) strategy. To deal with the proposed mixed integer nonlinear programming optimization problem, we develop an algorithm by integrating the Double Deep Q-learning Network (DDQN) with convex optimization methods. Extensive experiments validate our theoretical analyses across various datasets, and the numerical results demonstrate the effectiveness and superiority of the proposed communication-efficient SFL compared with existing schemes, including parallel split learning and traditional SFL mechanisms.

The widely adopted extremely large-scale multiple-input multiple-output (XL-MIMO) wideband systems are fundamentally constrained by their two-dimensional planar coverage, primarily due to the conventional use of uniform linear arrays (ULAs). This limitation significantly degrades the quality of service in practical three-dimensional (3D) deployment scenarios. Furthermore, XL-MIMO wideband systems inherently suffer from severe beam squint effects caused by non-negligible signal propagation delays, and the narrow beamwidth results in substantial beam training overhead. To address these challenges, we propose a novel cost-efficient frequency-dependent beamforming framework capable of supporting 3D spatially distributed users by leveraging a full-dimensional uniform planar array (UPA) architecture. Based on this design, we introduce a controllable 3D beam squint scheme via a partially-connected time-delay network, alongside a low-dimensional user-centric spatial-frequency scanning codebook to enable rapid and robust beam training. Numerical results demonstrate that the proposed scheme achieves accurate 3D angle estimation with significantly reduced training overhead during the initial access phase. Additionally, the proposed method can maintain high angle estimation accuracy in poor channel conditions.

This paper introduces the Federated LoRA with Dropout (FedLoDrop) framework, designed to enhance generalization performance for downstream tasks at the network edge while simultaneously reducing overhead. Within this framework, we derive a generalization error bound under sparsity regularization, elucidating the theoretical principles that balance underfitting and overfitting. Our analysis shows that a higher dropout rate increases sparsity, lowering the Pointwise Hypothesis Stability (PHS) upper bound and narrowing the gap between empirical and generalization errors. However, this also leads to a higher empirical error, which, together with the gap, contributes to the total generalization error. Consequently, we formulate an optimization problem that jointly considers dropout rate and resource allocation, aiming to minimize the upper bound of the generalization error. Finally, numerical results demonstrate the effectiveness of the proposed approach in mitigating overfitting and enhancing generalization capabilities.

Simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) are widely adopted in integrated sensing and communication (ISAC) systems to sense users in non-line-of-sight (NLoS) scenarios. However, the uniform spacing of elements in the uniform linear array (ULA) imposes fundamental limitations on angular resolution. To address the challenge, we propose replacing the ULA with a coprime array in the sensor module of the STAR-RIS to enhance high-accuracy direction-of-arrival (DOA) estimation. Furthermore, we derive the Cramér-Rao bound (CRB) of the proposed architecture, and a joint optimization problem is formulated to minimize the CRB. To solve the problem, we introduce Schur complement to transform the non-convex problem into a Fisher information matrix (FIM) optimization problem, and an efficient dual-loop iterative algorithm based on the penalty dual decomposition method is adopted to resolve it. Numerical results demonstrate that the proposed coprime array structure can achieve higher theoretical sensing accuracy compared with the traditional ULA structure.

Abstract With the development of wireless communications, heterogeneous ultra‐dense networks (HUDNs) have emerged to meet the requirements of massive connectivity, high data rate, and low latency in the 5G era. Nevertheless, HUDN usually leads to a high‐complexity and non‐convex NP‐hard energy‐efficient resource allocation problem. Therefore, A novel heterogeneous Graph neural network (GNN) with high‐dimensional computation structure (namely OverGNN) is proposed for the power allocation problem in this work. Particularly, OverGNN enabled nodes directly interact with high‐order neighbours and extract abundant graph topological information, which can facilitate effective feature aggregation among nodes as well as alleviate the over‐smoothing problem. Based on this fact, an efficient message passing scheme for user equipments under the same base station is developed to approximate the optimal power allocation strategy for maximizing system energy efficiency. In addition, an unsupervised approach is proposed to train the GNN model that can reduce the cost of dataset collection and enhance the scalability of the proposed method. Numerical results verify the effectiveness of the proposed OverGNN and demonstrate its advantages over the benchmarks.

Federated edge learning (FEL) is regarded as a promising distributed machine learning paradigm to reduce transmission latency and resources as well as preserve raw data privacy by collaboratively training local deep learning models across multiple edge devices. However, with the development of artificial intelligence (AI) technologies, the size of neural network models grows exponentially with their parameters to meet variable application requirements, which poses significant challenges to the computation, communication, and memory abilities of edge devices. Existing designs typically focus on either communication or computation efficiency without caring each device’s memory ability. To deal with the above issues, we first introduce a novel model sparse update enabled tiny FEL (TinyFEL) architecture, which terminates the backpropagation early in local model training processes. Therefore, the proposed TinyFEL can reduce local memory occupation and lessen the communication-and-computation burden. Furthermore, we propose a parameter splitting mechanism instead of transmitting the full model, only a part of updated layers of parameters is transmitted for aggregation, which significantly reduced the communication overheads. Thereafter, we develop a communication and computation latency minimization problem to accelerate the training of TinyFEL. To this end, we theoretically analyze the convergence performance of TinyFEL, which unveils the mathematical relationship among sparse update ratio assignment, device selection, and learning performance. Then, a joint sparse update ratio assignment, device selection, and resource allocation strategy is introduced based on the alternating direction method of multipliers (ADMMs) and block coordinate descent (BCD) algorithms. Numerical results indicate that our proposed TinyFEL can reduce training memory occupation by over 40% than the traditional FEL at the cost of negligible accuracy loss.

Edge artificial intelligence (AI) is an emerging solution for pervasive intelligence service in future 6G networks, by learning machine learning (ML) models at network edge. Edge AI typically consists of three processes: sensing, communication, and computation (SC²). Edge devices first collect data samples through the sensing process, then train local models individually through the computation process, and finally update the local models periodically through the communication process to obtain the global model. Federated edge learning (FEEL) is particularly attractive for Edge AI due to its collaborative ML framework and privacy-enhancing feature. However, the research FEEL with SC2 integration remains an open question. On the one hand, there is still a lack of theoretical insight into the learning performance that is jointly influenced by the processes of SC2. On the other hand, the performance evaluation is another challenge for the proposed SC2-FEEL, which further poses the difficulties in design of efficient resource allocation. To address these issues, an SC2 integrated FEEL (SC2-FEEL) is investigated in this article, where the processes of SC2 are jointly considered and the over-the-air computation (AirComp) technique is employed for a communication-efficient model aggregation. First, theoretical analyses are conducted, which reveals both the sample sensing strategy and the AirComp-induced communication error significant affect the learning performance of SC2-FEEL. Then, we further formulate a latency and energy consumption minimization problem with learning performance guaranteed based on the theoretical results, which is mixed integer nonlinear programming (MINLP) and dynamic programming. To deal with this problem, we propose a joint SC2 resource optimization strategy with low complexity based on the block coordinate update and Lyapunov optimization framework. Extensive simulation results are provided to validate our theoretical analysis, and demonstrate the effectiveness of developed algorithm.

In recent years, Wi-Fi sensing has garnered significant attention due to its numerous benefits, such as privacy protection, low cost, and penetration ability. Extensive research has been conducted in this field, focusing on areas, such as gesture recognition, people identification, and fall detection. However, many data-driven methods encounter challenges related to domain shift, where the model fails to perform well in environments different from the training data. One major factor contributing to this issue is the limited availability of Wi-Fi sensing datasets, which makes models learn excessive irrelevant information and over-fit to the training set. Unfortunately, collecting large-scale Wi-Fi sensing datasets across diverse scenarios is a challenging task. To address this problem, we propose CrossFi, a siamese network-based approach that excels in both in-domain scenario and cross-domain scenario, including few-shot, zero-shot scenarios, and even works in few-shot new-class scenario where testing set contains new categories. The core component of CrossFi is a sample-similarity calculation network called CSi-Net, which improves the structure of the siamese network by using an attention mechanism to capture similarity information, instead of simply calculating the distance or cosine similarity. Based on it, we develop an extra Weight-Net that can generate a template for each class, so that our CrossFi can work in different scenarios. Experimental results demonstrate that our CrossFi achieves state-of-the-art performance across various scenarios. In gesture recognition task, our CrossFi achieves an accuracy of 98.17% in in-domain scenario, 91.72% in one-shot cross-domain scenario, 64.81% in zero-shot cross-domain scenario, and 84.75% in one-shot new-class scenario. The code for our model is publicly available athttps://github.com/RS2002/CrossFi.

The massive-antenna wideband millimeter wave (mmWave)/terahertz (THz) systems inevitably suffer from a severe beam split effect due to the non-negligible signal propagation delays, which dramatically reduces communication efficiency. Nevertheless, if the wideband split effect is properly utilized, it can also bring benefits via sensing split directions for channel training. Hence, this paper proposes a novel wideband beam alignment framework with true-time-delayer (TTD) modules, which can fully exploit the controllable split beams for efficient angle-of-arrivals (AoAs) estimation. Moreover, we develop a hierarchical posterior matching (PM) enabled wideband beam alignment approach, which proactively configures the split beams to accelerate the estimation of AoAs posterior probability distributions. To deal with the computational complexity of the predesigned codebook and the insensitivity of the Gaussian distribution assumption in PM, we further introduce a low-complex and high-flexible wideband beam alignment approach based on a deep unfolding mechanism. Numerical results verify that: 1) The proposed framework can significantly improve the AoAs estimation accuracy at the cost of the same pilot overheads. 2) The proposed low-complexity deep unfolding approach outperforms the conventional PM mechanism even in low signal-to-noise-ratio (SNR) scenarios.

Federated multi-task learning (FMTL) is a promising technology to tackle one of the most severe non-independent and identically distributed (non-IID) data challenge in federated learning (FL), which treats each client as a single task and learns personalized models by exploiting task correlations. However, the transmission of individual task models generally results in a significant amount of communication overhead compared with global model broadcasting. Furthermore, related works mainly focus on FMTLs with default and static relationships among tasks, which obliterates the non-IID data characteristic. To address these issues, we propose a novel Clustered FMTL mechanism via Sparse Sharing (FedSS). Specifically, we introduce an iterative model pruning approach that trains customized client models to deal with the non-IID issue. Thereafter, we divide clients into different tasks according to their model similarities to promote communication efficiency. Based on clustered tasks, we introduce a sparse sharing mechanism that allows clients to share model parameters dynamically among different tasks to further boost the training performance. On the other aspect, the infertile communication resources would degrade the FMTL performance by restricting the personalized model transmissions. Hence, we first theoretically analyze the convergence performance of the proposed FedSS, which quantitatively unveils the relationship between the local model training performance and communication resources. Thereafter, we formulate a communication-and-computation efficient optimization problem via a joint sparsity ratio assignment and bandwidth allocation strategy. Closed-form expressions for the optimal sparsity ratio and bandwidth allocation are derived based on Lyapunov optimization and block coordinate update (BCU) algorithms. Numerical results illustrate that the proposed FedSS outperforms the benchmarks, and achieves an efficient communication and computation performance.

Communication and energy efficiencies are two crucial objectives in the pursuit of edge intelligence in 6G networks, and become increasingly important given the prevalence of large model training. Existing designs typically focus on either communication efficiency or energy efficiency due to the fact that improving one objective generally comes at the expense of the other. Over-the-air federated learning (OTA-FL) has recently emerged as a promising approach to enhance both efficiencies through an integrated communication and computation design. Nevertheless, most previous studies on OTA-FL only consider scenarios where the dataset for the entire FL procedure is collected and available prior to training. In real-world applications, devices continuously collect new data in an online manner. This underscores the significance of sample collection through sensing in a practical FL pipeline. We propose to integrate sensing with communication and computation into a joint design to further boost the communication-and-energy efficiencies of OTA-FL. Specifically, we consider a training latency and energy consumption minimization problem with performance guarantees. To this end, we first derive an average training error (ATE) metric to quantify convergence performance. Then, a joint sensing, communication and computation resource allocation strategy is developed based on a deep reinforcement learning (DRL) algorithm that nests convex optimization with a deep Q-network. Extensive experiments are conducted to validate our theoretical analysis, and demonstrate the effectiveness of the proposed design for communication-and-energy efficient FL.

Integrated sensing and communication (ISAC) within the unlicensed millimeter-wave (mmWave) frequency bands has been emerged as a pivotal technology in the next generation wireless communication era. However, the interference management issue between sensing and communication becomes much severe due to the absence of centralized scheduling function of the widely existed WiGig networks with the IEEE 802.11ay protocol in the unlicensed mmWave bands. In this way, we aim to investigate an efficient ISAC scheme for the promising WiGig networks via embedding radar pulses into the IEEE 802.11ay beamforming training (BFT) period. Particularly, we transmit both radar pulses and communication signals by exploiting the sector-level sweep of the WiGig network. Since there is a trade-off between radar detection/sensing and mmWave communication under diverse resource assignments, we propose a joint BFT and energy allocation strategy to find an achievable balance. Numerical results validate the effectiveness of the proposed scheme.

The effectiveness of Millimeter Wave full-duplex (FD) Integrated Access and Backhaul (IAB) system relies on high-dimensional channel estimation with high computational complexity. To avoid high-overhead pilot training, we propose a novel low-complexity end-to-end (E2E) hybrid beamforming strategy for FD mmwave IAB systems using implicit channel state information (CSI). Particularly, the IAB node first dynamically senses spatial channels, where an sensing Transformer block is introduced to actively design the sensing vector. The active sensing strategy can effectively handle the sequential pilot observations with an arbitrary input length. Capitalizing on the implicit channel features extracted by the Transformer, a hybrid beamforming neural network (HBFnet) is further exploited to design the hybrid precoder/combiner of IAB node, thus efficiently mitigating SI while compensating channel fading. Simulation results demonstrate that the proposed scheme outperforms the benchmarks, especially with low pilot overheads.

The massive-antenna millimeter wave (mmWave)/ terahertz (THz) system inevitably suffers from a severe beam split effect, which will significantly decrease array gains and communication efficiency. However, the wideband split effect can be beneficial for channel training by sensing from splitting directions. This paper proposes a novel wideband beam alignment framework in True-Time-Delayers (TTDs) enabled hybrid beamforming systems, which can fully exploit the controllable beam split effect for efficient angle-of-arrival (AoA) estimation. A splitting-sensing enabled wideband posterior matching algorithm is developed, which proactively controls the steering direction of the splitting-sensing beams at each pilot training slot to accelerate the AoA estimation process. Numerical results verify that the proposed algorithm can significantly improve the AoA estimation accuracy at the cost of the same pilot overheads.

This article investigates an integrated sensing-communication-computation (ISCC)-based multiview-multitask (MVMT) edge artificial intelligence inference system. Each device senses a narrow view of a target area and processes the echo signal to generate real-time sensory data. An edge server receives and combines multiple views of data from multiple devices to complete several downstream inference tasks. Compared with existing designs where dedicated sensory data are obtained, transmitted, and processed for each task, this ISCC-based MVMT framework enjoys reduced costs of sensing, on-device computation, and communication overhead due to data sharing among different tasks. The challenges of improving all tasks’ inference accuracy lie in the tight coupling of sensing, communication, and computation among different devices and sensory view competition among different tasks. These two challenges intertwine, making the multitask optimization problem mixed-integer nonconvex programming. To tackle this problem, we propose a joint device scheduling and resource allocation (JDSRA) scheme, which alternatively solves a subproblem of joint device scheduling and time allocation and a subproblem of resource allocation till convergence. Particularly, in addition to a dynamic-programming-based optimal device scheduling algorithm, a low-complexity suboptimal algorithm is proposed based on sorting a derived closed-form indicator, which represents the increase of all tasks’ inference accuracy per time unit consumption. Besides, a low-complexity optimal resource allocation algorithm is proposed by parallelly solving multiple simple convex subproblems. Numerical results based on jointly completing three tasks of human motion recognition, human height recognition, and localization in smart home scenarios are conducted to verify the performance of our proposed schemes.

Metaverse has emerged as a revolutionary technique for transforming the way people interact with digital content, which relies on a distributed computing and communication infrastructure, encompassing terminal users, edge servers, and cloud servers. However, the rapid evolution of the Metaverse presents challenges that surpass the capabilities of existing communication and network infrastructures, particularly on network bandwidth and latency. Additionally, human experience becomes a critical factor in this domain. Therefore, we introduce a human-aware hierarchical software defined network (SDN) architecture consisting of a Metaverse cloud layer, a mobile edge computing (MEC) server empowered edge layer, and a distributed terminal layer. Each MEC server dynamically controls a multi-antenna base station (BS) and several reconfigurable intelligent surfaces (RISs) according to the terminal immersive experience requirements in real-time. To overcome the bandwidth limitation, we propose a novel smart reconfigurable spatial reuse new radio in unlicensed spectrum (NR-U) framework, which can realize customizable communications through flexibly and coordinately reconfiguring beams among the coordination between BSs and RISs. The objective function is formulated as a Lyapunov optimization based decentralized partially-observable Markov decision process (Dec-POMDP) problem to maximize the spectral efficiency while guaranteeing the latency and reliability requirements in Metaverse, via a joint user selection, phase-shift control, and beam coordination strategy. To solve the above non-convex, strongly coupled, and mixed integer nonlinear programming (MINLP), we propose a novel multi-agent hierarchical deep reinforcement learning (MAHDRL) algorithm that integrates deep Q-network (DQN) to solve discrete problems, deep deterministic policy gradient (DDPG) to solve continuous problems, and mixing network to capture complex interactions between multiple agents. Numerical results demonstrate the effectiveness of the proposed algorithm and verify the performance improvements compared to traditional multi-agent deep reinforcement learning (MADRL) algorithms.

In this paper, a cloud radio access network (Cloud-RAN) based collaborative edge AI inference architecture is proposed. Specifically, geographically distributed devices capture real-time noise-corrupted sensory data samples and extract the noisy local feature vectors, which are then aggregated at each remote radio head (RRH) to suppress sensing noise. To realize efficient uplink feature aggregation, we allow each RRH receives local feature vectors from all devices over the same resource blocks simultaneously by leveraging an over-the-air computation (AirComp) technique. Thereafter, these aggregated feature vectors are quantized and transmitted to a central processor (CP) for further aggregation and downstream inference tasks. Our aim in this work is to maximize the inference accuracy via a surrogate accuracy metric called discriminant gain, which measures the discernibility of different classes in the feature space. The key challenges lie on simultaneously suppressing the coupled sensing noise, AirComp distortion caused by hostile wireless channels, and the quantization error resulting from the limited capacity of fronthaul links. To address these challenges, this work proposes a joint transmit precoding, receive beamforming, and quantization error control scheme to enhance the inference accuracy. Extensive numerical experiments demonstrate the effectiveness and superiority of our proposed optimization algorithm compared to various baselines.

Existing edge inference methods only consider one paradigm, i.e., one of on-device inference, on-server inference, or edge-device cooperative inference. Each paradigm has its pros and cons as well as dominant application scopes. For example, the on-device paradigm is the best choice when the inference task is not computationally intensive, the on-server paradigm is suitable if the communication capacity is strong, and the edge-device cooperative mode should be selected in the scenario of weak on-device communication and computation. However, each paradigm suffers from poor performance if deployed outside of its application scope, thus leading to limited potential and flexibility. This paper proposes an edge AI inference framework, which makes the first attempt to jointly consider the three modes for making full use of their benefits. In addition, sensing for data acquisition is enabled at both the edge server and the device. This can effectively improve the inference accuracy with rich information on the target area from two different views. On the other hand, energy cost minimization turns out to be a key target all over the world and a significant issue in wireless networks. To this end, we target minimizing the system energy cost under a given inference accuracy guarantee and other network resource constraints, by coordinating sensing, communication, and computation in different modes. By optimally solving the optimization problem, an integrated sensing-communication-computation (ISCC) based task-oriented mode selection scheme is proposed. A practical ISCC platform is built and extensive experiments are conducted to verify our theoretical analysis.

This paper focuses on the mobility prediction problem of specific occupational groups that rely on mobile devices, predicting the mobilities of these groups using data shared across different platforms. While the mobility patterns of general populations have been studied extensively, predicting the movements of specific occupational groups like ride-hailing drivers, who frequently move over long distances, requires a more nuanced approach. This paper introduces FedOGM, a federated learning framework designed to predict the mobility of specific occupational groups while preserving user privacy. The framework utilizes a dynamic bidirectional graph attention network model DyBGAT for predicting the edge weights of an occupational dynamic Origin-Destination graph, representing the mobility behavior. In order to address key challenges in federated learning, FedOGM integrates By leveraging multi-source data from users’ mobile devices and extracting unique occupational features and occupational features into the framework, such as applications usage duration feature, location switching feature, and call duration feature. Assisted by these personalized occupational features, we have devised a client selection algorithm, generated occupational dynamic OD graphs to diminish communication overhead, and proposed a group-level FedAvg aggregation method. Experiments conducted on real-world datasets of ride-hailing drivers, ride-hailing truck drivers, and real estate salespersons, including scenarios with false data, confirm the efficacy of the proposed methods in both mobility prediction and privacy protection.

Federated multi-task learning (FMTL) is a promising technology to deal with the severe data heterogeneity issue in federated learning (FL), where each client learns individual models locally and the server extracts similar model parameters from the tasks to keep personalization for models of clients. Hence, it is essential to precisely extract the model parameters shared among tasks. On the other aspect, the limitation of communication resources would also restrict the model transmission, and thus influence the FMTL performance. To address the above issues, we propose a novel FMTL with Sparse Sharing (FedSS) mechanism that allows clients to share model parameters dynamically according to diversified model structures under limited communication resources. Particularly, we present an adaptive quantization approach for task relevance, which serves as a metric to evaluate the extent of model sharing across tasks. The objective function is formulated to minimize the model transmission latency while ensure the FMTL learning performance via a joint bandwidth allocation and client selection strategy. Closed-form expressions for the optimal client selection and bandwidth allocation are derived based on a alternating direction method of multipliers (ADMM) algorithm. Numerical results show that the proposed FedSS outperforms the benchmarks, and achieves efficient communication performance.

With the rapid growth of Internet of Things (IoT), smart home develops rapidly in these years, which could assist people who need family medical support. It could integrate health care with ambient assisted living (AAL) technologies and provide activities of daily life (ADLs) to the people who need care. This paper proposes a smart home and cross-cloud-and-edge computing based nursing system (NS). In general, a good NS requires low latency, high stability, and the real-time analysis and response, where the conventional centralized cloud computing based approaches cannot meet those requirements very well. To this end, we introduce a novel distributed joint edge-cloud structure to better satisfy these requirements. Moreover, to deal with the security and privacy issues, we introduce the blockchain to verify the identity of data exchanging and differential-privacy (DP) in the NS to protect the healthcare takers’ data privacy. In a word, we propose a privacy-preserving context-aware multi-armed bandit based online learning approach for edge-cloud-enabled NS via blockchain in IoTs. Additionally, our system with a novel top-down expanding tree based structure can support dynamically increasing health care datasets. Extensive experimental and numerical results demonstrate our solution can achieve accurate recommendation results with sublinear regret performance.

A multi-cell cluster-free NOMA framework is proposed, where both intra-cell and inter-cell interference are jointly mitigated via flexible cluster-free successive interference cancellation (SIC) and coordinated beamforming design. The joint design problem is formulated to maximize the system sum rate while satisfying the SIC decoding requirements and users’ minimum data rate requirements. To address this highly complex and coupling non-convex mixed integer nonlinear programming (MINLP), a novel distributed auto-learning graph neural network (AutoGNN) architecture is proposed to alleviate the overwhelming information exchange burdens among base stations (BSs). The proposed AutoGNN can train the GNN model weights whilst automatically optimizing the GNN architecture, namely the GNN network depth and message embedding sizes, to achieve communication-efficient distributed scheduling. Based on the proposed architecture, a bi-level AutoGNN learning algorithm is further developed to efficiently approximate the hypergradient in model training. It is theoretically proved that the proposed bi-level AutoGNN learning algorithm can converge to a stationary point. Numerical results reveal that: 1) the proposed cluster-free NOMA framework outperforms the conventional cluster-based NOMA framework in the multi-cell scenario; and 2) the proposed AutoGNN architecture significantly reduces the computation and communication overheads compared to the conventional convex optimization-based methods and the conventional GNNs with fixed architectures.

A generalized downlink multi-antenna non-orthogonal multiple access (NOMA) transmission framework is proposed with the novel concept of cluster-free successive interference cancellation (SIC). In contrast to conventional NOMA approaches, where SIC is successively carried out within the same cluster, the key idea is that the SIC can be flexibly implemented between any arbitrary users to achieve efficient interference elimination. Based on the proposed framework, a sum rate maximization problem is formulated for jointly optimizing the transmit beamforming and the SIC operations between users, subject to the SIC decoding conditions and users’ minimal data rate requirements. To tackle this highly-coupled mixed-integer nonlinear programming problem, an alternating direction method of multipliers-successive convex approximation (ADMM-SCA) algorithm is developed. The original problem is first reformulated into a tractable biconvex augmented Lagrangian (AL) problem by handling the non-convex terms via SCA. Then, this AL problem is decomposed into two subproblems that are iteratively solved by the ADMM to obtain the stationary solution. Furthermore, to reduce the computational complexity and alleviate the parameter initialization sensitivity of ADMM-SCA, a Matching-SCA algorithm is proposed. The intractable binary SIC operations are solved through an extended many-to-many matching, which is jointly combined with an SCA process to optimize the transmit beamforming. The proposed Matching-SCA can converge to an enhanced exchange-stable matching that guarantees the local optimality. Numerical results demonstrate that: i) the proposed Matching-SCA algorithm achieves comparable performance and a faster convergence compared to ADMM-SCA; ii) the proposed generalized framework realizes scenario-adaptive communications and outperforms traditional multi-antenna NOMA approaches in various communication regimes.

As a key infrastructural technology, Industrial Internet of Things (IIoT) and its related techniques have emerged in the age of Industrial Internet. Among them, an increasing popular and attractive federated edge learning (FEL) mechanism, which performs data analysis and inference at the edge devices distributedly, and aggregates local FEL units at a centralized controller, is introduced to meet the stringent data privacy and low-latency requirements for high-stake IIoT devices. Due to the bandwidth limitation, only parts of the IIoT devices can be selected to transmit their local FEL models to the centralized controller at each learning step. However, the centralized controller prefers to collect all the local FEL models to generate the global FL model since each IIoT device has a differential data set. Existing works mainly focus on selecting an appropriate subset of IIoT devices through advanced scheduling mechanisms without extending the resource bandwidth. However, the new radio in unlicensed spectrum (NR-U) technology in the 5G network opens up new possibilities for FEL since it is a privately owned network with fruitful bandwidth resources. We thus propose a novel communication-efficient FEL mechanism for NR-U-based IIoT networks, which aims to select data importance IIoT devices for local training under relatively sufficient unlicensed resources. The objective function is formulated as a tradeoff between total FEL data importance and the transmission latency via joint learning, device selection, and resource management scheduling, which is a mixed-integer nonlinear programming (MINLP). To deal with this problem, an alternating direction method of multipliers and block coordinate update (ADMM-BCU) algorithm with low computational complexity has been used. Closed-form expressions for both optimal device selection and resource management are derived, which highlighted significant insights. Numerical results demonstrate the algorithmic advantages and structural benefits of the proposed strategies.

The sixth-generation (6G) era is expected to provide even higher levels of massive connectivity/Internet of Things (IoT), tremendous data rate, and low latency than the 5G communication. Therefore, the future 6G wireless networks would confront with much more severe spectrum scarcity problems, which make the new radio in unlicensed spectrum (NR-U) technology attractive. Meanwhile, the high densely deployed massive IoT devices lead to fearful intercell and intracell interference issues. To address these challenges, a coordinated multipoint (CoMP) technology can be adopted. In this work, we introduce a CoMP-based NR-U network, particularly for 6G-enabled massive IoT scenarios. Under the proposed network, we introduce a spatial listen-before-talk (LBT) scheme to control mobile network operators (MNOs) to coexist with the incumbent WiFi devices orthogonally, which can significantly improve the spectrum efficiency. To suppress the strong co-channel interference among multiple MNOs, we investigate a joint beamforming coordination and user selection problem, which is NP-hard. To deal with this problem, we first adopt a fractional programming and an integer replacement method to transform the objective problem into a tractable one. Then, we introduce a semidistributed alternating direction method of multipliers and block coordinate update (ADMM-BCU) algorithm to find a suboptimal solution, which only requires limited information exchange via a CoMP server and can help reduce energy consumption. Theoretical analysis and numerical results demonstrate the effectiveness of the proposed algorithm for large-scale 6G IoT scenarios.

With the accelerated development of immersive applications and the explosive increment of internet-of-things (IoT) terminals, 6G would introduce terahertz (THz) massive multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) technologies to meet the ultra-high-speed data rate and massive connectivity requirements. Nevertheless, the unreliability of THz transmissions and the extreme heterogeneity of device requirements pose critical challenges for practical applications. To address these challenges, we propose a novel smart reconfigurable THz MIMO-NOMA framework, which can realize customizable and intelligent communications by flexibly and coordinately reconfiguring hybrid beams through the cooperation between access points (APs) and reconfigurable intelligent surfaces (RISs). The optimization problem is formulated as a decentralized partially-observable Markov decision process (Dec-POMDP) to maximize the network energy efficiency, while guaranteeing the diversified users’ performance, via a joint RIS element selection, coordinated discrete phase-shift control, and power allocation strategy. To solve the above non-convex, strongly coupled, and highly complex mixed integer nonlinear programming (MINLP) problem, we propose a novel multi-agent deep reinforcement learning (MADRL) algorithm, namelygraph-embedded value-decomposition actor-critic (GE-VDAC), that embeds the interaction information of agents, and learns a locally optimal solution through a distributed policy. Numerical results demonstrate that the proposed algorithm achieves highly customized communications and outperforms traditional MADRL algorithms.

Convolutional neural networks (CNNs) have become the critical technology to realize face detection and face recognition in the face tracking (FT) system. However, traditional CNNs usually have nontrivial computational time and high energy consumption, making them inappropriate to be deployed in the large-scale time-sensitive FT system. To address this challenge, we design an artificial intelligence and Internet of Things (AIoT) empowered edge-cloud collaborative computing (ECCC) system based on the energy-efficient field-programmable gate array (FPGA)-based CNN accelerators for the purpose of realizing a low-latency and low-power FT. First, we present the AIoT-empowered ECCC system architecture, which consists of an intelligent computing subsystem, an Internet-of-Things (IoT) subsystem, an edge-cloud collaborative subsystem, and an application subsystem. In what follows, we investigate the enabling technologies for these subsystems. Thereafter, we develop an FPGA-based hardware accelerator dedicated to the compact MobileNet CNN by using the hardware design techniques, such as systolic array, matrix tiling, fixed-point precision, and parallelism. Furthermore, we integrate the FPGA accelerators with CPUs and GPUs to build a context-aware CPU/GPU/FPGA heterogeneous computing system. Finally, we implement a delay-aware energy-efficient scheduling algorithm dedicated to this heterogeneous system. With the above hardware and software codesign mechanism, the energy cost and execution time of CNNs can be decreased significantly. The real-world experiments on the CPU/GPU/FPGA-based ECCC system proved the effectiveness of the proposed schemes in reducing the latency and improving the power efficiency of the FT system.

Millimeter wave (mmWave) communication is a promising New Radio in Unlicensed (NR-U) technology to meet with the ever-increasing data rate and connectivity requirements in future wireless networks. However, the development of NR-U networks should consider the coexistence with the incumbent Wireless Gigabit (WiGig) networks. In this paper, we introduce a novel multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) based mmWave NR-U and WiGig coexistence network for uplink transmission. Our aim for the proposed coexistence network is to maximize the spectral efficiency while ensuring the strict NR-U delay requirement and the WiGig transmission performance in real time environments. A joint user grouping, hybrid beam coordination and power control strategy is proposed, which is formulated as a Lyapunov optimization based mixed-integer nonlinear programming (MINLP) with unit-modulus and nonconvex coupling constraints. Hence, we introduce a penalty dual decomposition (PDD) framework, which first transfers the formulated MINLP into a tractable augmented Lagrangian (AL) problem. Thereafter, we integrate both convex-concave procedure (CCCP) and inexact block coordinate update (BCU) methods to approximately decompose the AL problem into multiple nested convex subproblems, which can be iteratively solved under the PDD framework. Numerical results illustrate the performance improvement ability of the proposed strategy, as well as demonstrating the effectiveness to guarantee the NR-U traffic delay and WiGig network performance.

Driven by the rapid technological advances, industrial Internet of Things (IIoT) has recently been embraced to enhance autonomous industrial processes. Since a huge diverse traffic would be generated by IIoT, the industrial processes would meet the challenges of spectrum scarcity and on-demand service requirements. Millimeter wave (mmW) and pervasive edge computing (PEC) technologies in 5G communication are available to deal with these requirements. In this article, a novel dual-band framework that integrates both mmW and microwave (μW) networks in PEC environment has been proposed, which locally performs joint resource allocation and power assignment over mmW and μW to meet IIoT devices' specific requirements. To consider the new prominent figure of merit in IIoT scenario, the scheduling problem is formulated as an optimization problem to minimize the IIoT energy consumption in real-time environment. A Lyapunov optimization technique has been applied for the objective function with low complexity and rapid convergence. To solve the NP-hard Lyapunov algorithm, we introduce a block coordinate descent method that decompose the Lyapunov problem into two nested subproblems over the mmW and μW networks. An initialization-free semidistributed scheme is proposed in mmW PECs, which not only requires little information exchange via the μW network but also achieves the global optimal solution. Numerical results are shown to demonstrate the effectiveness of our proposed algorithms and confirm our theoretical analyses.

Given the abundance of unlicensed spectrum in 5 GHz band, licensed-assisted-access (LAA) technology presents an efficient and simple approach to alleviate the spectrum crunch in wireless networks. The recent proposal of IEEE 802.11ax as an advanced WiFi standard to accommodate more high rate connections motivates the investigation with respect to the feasibility and benefits of LAA in coexistence with unlicensed user access under this new WiFi protocol. In this article, we first introduce an enhanced LAA and WiFi coexistence mechanism based on spatial multi-stream transmission within IEEE 802.11ax access. We then derive a stream selection and user replacement strategy based on the proposed synchronous coexistence scheme to improve LAA performance without sacrificing WiFi throughput. We further present a Lyapunov algorithm to optimize the LAA access intensity level with rapid convergence and low-complexity. Numerical results demonstrate the effectiveness of the proposed algorithms, and the mutual benefits of the proposed coexistence framework to both LAA and WiFi users.

In this paper, inspired by the minority game (MG), we propose a distributed user association mechanism for the heterogenous networks (HetNets) on unlicensed bands. Our proposal aims to achieve load balance under different resource contention schemes between the LTE-unlicensed (LTE-U) and Wi-Fi networks in a fully distributed fashion. To formulate the user association problem as MG, we first prove that there exists a unique cut-off value in the single-AP scenario for both listen-before-talk and duty cycle muting schemes. Meanwhile, both the pure strategy and the mixed strategy are developed and the Nash equilibria are achieved. We further extend our analysis into the scenario with multiple Wi-Fi access points and prove the existence and uniqueness of the cut-off value set. Numerical results show that the proposed MG-based user association algorithm can achieve load balance and fine spectrum utilization without channel state information (CSI). Some inspiring results are also highlighted through the numerical simulation. The proposed distributed mechanisms not only handle the LTE-U/Wi-Fi selection, but also give fascinating insights into user association and resource allocation in other scenarios of heterogenous networks.

In licensed assisted access (LAA) based small cell networks (SCNs), the small base station (SBS) can reuse the uplink licensed bands with the macro cell while sharing the unlicensed bands with the Wi-Fi networks to improve its throughput. To mitigate the severe co-channel interference to the macro cell and guarantee the harmonious coexistence with the Wi-Fi networks, the spectrum and power should be jointly allocated at the SBSs. Moreover, to overcome the overwhelming signaling overheads introduced by the traditional centralized scheme and adapt to the variable radio environments, an adaptive decentralized scheme is necessary. Therefore, in this paper, an adaptive semi-distributed scheme is proposed to jointly allocate the power and spectrum on both licensed and unlicensed bands, which can achieve the global optimal spectrum efficiency (SE) of the SCNs. The proposed scheme can enable the SBSs to work independently and adaptively without sharing the whole information of the SBSs, but requires some Lagrangian parameters exchange via the coordination of the macro base station (MBS). Theoretical analysis and numerical results are presented to show that the proposed scheme is capable of achieving the optimal SE on both licensed and unlicensed bands adaptively while confining the co-channel interference to the MBS and guaranteeing the fair coexistence with the Wi-Fi network.

This paper proposes an enhanced licensed-assisted-access (eLAA) and WiFi coexistence mechanism for LTE deployment in the unlicensed spectrum. Specifically, we utilize the Clear-to-Send-to-Self (CTS-to-Self) frame to seamlessly embed the eLAA transmission within the WiFi access protocol. Unlike traditional LAA ideas, our proposed eLAA is non-intrusive to WiFi users and beneficial to both eLAA and WiFi users. To fully consider the performance of WiFi users, we propose to insert the CTS-to-Self frame under the carrier sense multiple access with collision avoidance (CSMA/CA) mechanism. For further throughput enhancement, we adopt the Nash bargaining solution (NBS) to develop a fair and optimal unlicensed resource allocation scheme and also develop a low computational complexity algorithm to solve the NBS-based problem. Our numerical results corroborate the analytical results and demonstrate the strength of the proposed mechanism.

Using small cell base station (SBS) with unmanned aerial vehicle (UAV) as a carrier becomes a promising solution for areas with high-density mobile users. On the other hand, 5G network would apply the LTE technology into the unlicensed spectrum, named Licensed-assisted Access (LAA), due to the limitation of licensed band. In this paper, we propose to utilize LAA technology into the UAV to expand available transmission band. By focusing on the transmission experience of very important (VIP) users, we propose an enhanced LAA (eLAA) technology, which integrates LAA into the IEEE 802.11e protocol. Under the proposed UAV enabled eLAA network, our goal is to maximize the energy efficiency of the on-board communication device through a joint trajectory design and resource allocation strategy. The proposed nonlinear fractional problem has been solved by the Dinkelbach-type algorithm and the block coordinate descent (BCD) mechanism. Numerical results demonstrate the effectiveness of our proposed scheme.

New radio in unlicensed spectrum (NR-U) is an exciting evolution of LTE-U/LAA from 4G LTE to 5G NR, which generates an opportunity to alleviate the spectrum crunch in future wireless networks by operating NR in unlicensed spectrum. Due to the openness of unlicensed spectrum, networks with heterogeneous radio access technologies (RATs) will coexist with NR-U, especially for the incumbent WiFi networks. In this paper, we first introduce a NR-U framework based on network slicing and spatial multiplexing, which can help on the management of networks with heterogeneous RATs. We then propose a synchronous RAT for the proposed NR-U network and derive a user group construction strategy to improve performance as well as provide flexibility for both cellular and WiFi users. Numerical results demonstrate the mutual benefits of the proposed RAT to both cellular and WiFi users.

Licensed-assisted-access (LAA) has recently emerged as a heterogeneous network technology to help mitigate the scarcity of licensed spectrum by extending the long-term evolution (LTE) network to unlicensed spectrum. This work proposes an enhanced LAA (eLAA) as a practical technique by exploiting the inherent transmit opportunity (TXOP) reservation via the Clear-to-Send-to-Self (CTS-to-Self) frame in Enhanced Distributed Channel Access (EDCA) to seamlessly integrate eLAA transmission within existing WiFi protocol. Unlike other LAA proposals, our eLAA is non-intrusive to unlicensed WiFi users and beneficial to both eLAA and WiFi networks. To further improve throughput, we analyze and derive an optimized unlicensed resource allocation scheme before deriving several intrinsic properties. Our results demonstrate that tagging CTS-to-Self frames as a higher-priority access category can substantially improve eLAA throughput without seriously degrading the per-user WiFi throughput.

Millimeter-wave (mmWave) heterogeneous network (HetNet) has been regarded as a promising means to improve the cellular system capacity in the 5G era. In this paper, we investigate the joint user association and resource allocation problem in a multi-band mmWave HetNet where different bands have different propagation characteristics. According to whether a user can transmit on multiple mmWave bands simultaneously, two different access schemes are considered: the single-band access scheme and the multi-band access scheme. For the single-band access scheme, we first find a closed-form expression for the optimal time fraction allocation and then develop an iterative algorithm for joint user association and power allocation based on the Lagrangian dual decomposition methods and the Newton-Raphson method. For the multi-band access scheme, we develop a near-optimal solution based on the Markov approximation framework. Our analytical results reveal that different users can only access at most one band simultaneously although the multi-band access scheme allows a user to transmit on multiple bands. Finally, numerical results demonstrate that the multi-band access scheme performs better than the single-band access scheme, especially in the light load scenario.

LTE-Unlicensed (LTE-U) has been widely considered as one of the most up-to-date promising innovations to achieve more ambitious data rate under a unified networking architecture by extending the LTE network into the bandwidth-rich unlicensed spectrum. In this paper, we study the user association optimization for LTE-U systems with multiple small cell base stations (SBSs) and WiFi access points (APs). Aiming at maximizing the whole throughput of LTE-U systems, a non-convex combinatorial optimization problem is modeled, which is NP-hard. By relaxing the integer variables to continuous variables, an upper-bound algorithm is developed based on the theory of sum-of- ratios optimization. Moreover, a heuristic algorithm based on the student project allocation (SPA) modeling is also developed, whose computational complexity is linear. Numerical results demonstrate the effectiveness and efficiency of our proposed algorithms.

In this paper, we consider multi-homing in terms of the emerging long-term evolution in unlicensed band (LTE-U) technology. Accordingly, the unlicensed band can be simultaneously and dynamically shared by every single user between the two radio access technologies (RATs), WiFi and LTE-U, to boost the system performance and enhance the user experience as well. Because of the additive freedom on unlicensed band, the problem of multi-RAT selection and resource allocation is much more complicated. To realize multi-homing in LTE-U networks, we aim at maximizing the overall throughput of multi-homing users while guaranteeing the quality of service and the power limitation. First, the sufficient conditions for each user to select only one RAT for transmission are developed. Then, since the multi-RAT selection and resource allocation problem is NP-hard, an effective heuristic algorithm is proposed to solve it. Numerical results validate the effectiveness of our proposed algorithm and demonstrate that the proposed mechanism can enhance the system throughput compared with traditional single-homing LTE-U networks.

With the development of the fifth generation mobile communication, long-term evolution in unlicensed spectrum (LTE-U) has been proposed as a promising means to solve the spectrum scarcity problem. In this paper, we investigate the mobile data offloading in a LTE-U system where one LTE small cell base station (SBS) coexists with several WiFi APs. Different from the traditional unidirectional mobile offloading, the proposed bidirectional mobile offloading can improve the system throughput and achieve the load balance among different WiFi APs as well. We first formulate a multi-objective optimization problem (MOOP) to analyze the bidirectional mobile offloading. Then, we propose two different algorithms to achieve the effective solutions to the MOOP. The first one utilizes the Nash bargaining solution (NBS) to obtain the closed-form solution to the optimal unlicensed resource allocation. The second one leverages the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to develop a low-complexity heuristic algorithm. Our proposals are finally validated by numerical simulations.

The employment of Long-Term Evolution (LTE) in unlicensed spectrum, known as LTE-U, can alleviate the spectrum scarcity problem in the 5G networks. With this new technique, the traditional mobile data offloading schemes, which generally offload LTE users to the WiFi network, should be revisited. In this paper, we propose to transfer WiFi users to the LTE-U network and simultaneously allocate some unlicensed spectrum to LTE-U. In this way, a win-win situation could be generated since LTE can achieve better spectrum efficiency than WiFi in the unlicensed spectrum. To facilitate it, three important challenges are addressed in the paper: which WiFi users should be transferred; how many WiFi users need to be transferred; and how much unlicensed resource should be allocated to the LTE network. We utilize the Nash bargaining solution to design fair unlicensed spectrum allocation between WiFi and LTE-U and thereby a win-win strategy is developed, whose performance is demonstrated by numerical simulation.

Long-term evolution in unlicensed spectrum (LTE-U) is an emerging technology for expanding cellular network capacity without additional spectrum cost. This paper investigates effective spectrum sharing for coexisting Wi-Fi and LTE-U services. Based on a novel hyper access point (HAP) we introduced for effectively embedding LTE-U in unlicensed Wi-Fi band, LTE-U can directly take advantage of the Wi-Fi point coordination function protocol. To facilitate the coexistence, our HAP dedicates a contention-free period to LTE-U users and allows a contention period (CP) for traditional Wi-Fi users. We investigate the optimization of joint user association and resource allocation to further improve system throughput and user fairness. We formulate a network utility maximization problem based on the Nash bargaining solution (NBS), for which we derive a closed-form expression for the optimal CP length under a given user association. We analyze this NBS-based utility maximization and the performance of the proposed algorithm under log-normal fading, Rayleigh fading, and Rician fading channel models, respectively. Our numerical results corroborate our analysis and demonstrate effective improvement of the system performance by the proposed HAP algorithm against traditional LTE-U deployment.

To improve system capacity, licensed-assisted access (LAA) has been proposed for long-term evolution (LTE) systems to use unlicensed bands. However, the energy efficiency (EE) of the LTE system may be degraded by LAA since unlicensed bands are generally less energy-efficient than licensed bands. In this paper, we investigate the EE optimization of LAA systems. We first develop a criterion to determine whether unlicensed bands can be leveraged to improve the EE of LAA systems. We prove that unlicensed bands can be used to improve the EE only when the allocated licensed resource blocks (RBs) are not enough. We then investigate joint licensed and unlicensed RB allocation to maximize the EE of each small cell base station (SBS) in a multi-SBS scenario, taking into account fair resource sharing between LTE and WiFi networks. The complete Pareto optimal EE set can be obtained by the weighted Tchebycheff method. We also develop an algorithm to provide fair EE among different SBSs based on the Nash bargaining solution. Numerical results are presented to confirm our analysis and to demonstrate the effectiveness of the proposed algorithms.

Traditional mobile data offloading transfers cellular users to WiFi networks to relieve the cellular system from the pressure of the ever-increasing data traffic load. However, the spectrum utilization of the WiFi network is bound to suffer from potential packet collisions due to its contention-based access protocol, especially when the number of competing WiFi users grows large. To tackle this problem, we propose transferring some WiFi users to be served by the LTE system, in contrast to the traditional mobile data offloading which effectively offloads LTE traffic to the WiFi network. Meanwhile, leveraging the emerging LTE in unlicensed spectrum (LTE-U) technology, some unlicensed spectrum resources may be allocated to the LTE system in compensation for handling more WiFi users. In this way, a win-win situation would be generated since LTE can generally achieve better performance than WiFi due to its capability of centralized co-ordination. To facilitate it, three important challenging issues are addressed in the paper: which WiFi users should be transferred; how many WiFi users need to be transferred; and how much unlicensed resources should be relinquished to the LTE-U network. We investigate three different user transfer schemes according to the availability of channel state information (CSI): the random transfer, the distance-based transfer, and the CSI-based transfer. In each scheme, the minimum required amount of unlicensed resources under a given transferred user number is analyzed. Furthermore, we utilize the Nash bargaining solution (NBS) to develop joint user transfer and unlicensed resource allocation strategy to fulfill the win-win situation for both networks, whose performance is demonstrated by numerical simulation.

Traffic offloading and resource sharing are two common methods for delivering cellular data traffic over unlicensed bands. In this paper, we first develop a hybrid method to take full advantages of both traffic offloading and resource sharing methods, where cellular base stations (BSs) offload traffic to WiFi networks and simultaneously occupy certain number of time slots on unlicensed bands. Then, we analytically compare the cellular throughput of the three methods with the guarantee of WiFi per-user throughput in the single-BS scenario. We find that traffic offloading can achieve better performance than resource sharing when existing WiFi user number is below a threshold and the hybrid method achieves the same performance as the resource sharing method when existing WiFi user number is large enough. In the multi-BS scenario where the coverage of small cells and WiFi access points are mutually overlapped, we consider to maximize the minimum average per-user throughput of each small cell and derive a closed-form expression for the throughput upper bound in each method. Meanwhile, practical traffic offloading and resource sharing algorithms are also developed for the three methods, respectively. Numerical results validate our theoretical analysis and demonstrate the effectiveness of the proposed algorithms as well.

In this paper, two novel mechanisms are developed for the coexistence of cellular and WiFi systems in unlicensed spectrum. In the opportunistic method, the small cell base station opportunistically selects traffic offloading or resource sharing on each WiFi access point (AP). In the hybrid method, the base station simultaneously offloads users and shares the unlicensed spectrum of each AP. The performances of the proposed methods are analyzed and compared. We find that traffic offloading can achieve better performance than resource sharing when the number of existing WiFi users is below a threshold and the hybrid method achieves the same performance as the resource sharing method when existing WiFi user number is large enough. Numerical results are presented to demonstrate the effectiveness of the proposed methods.

Multi-stream aggregation (MSA) allows users to receive data from multiple base stations simultaneously to increase their data rates. In this paper, we propose a joint user association and resource allocation algorithm for MSA systems to achieve energy efficiency (EE) balance among different base stations. The problem is formulated as a non-convex combinatorial sum-of-ratios optimization problem, which is very hard to solve directly. We first relax the combinatorial variables and then transform the problem into a convex optimization problem by the sum-of-ratios algorithm and the successive convex approximation (SCA) method. Based on this, a near-optimal algorithm is developed. Simulation results show that the proposed algorithm can achieve a good performance with a fast convergence speed.

Licensed-assisted access (LAA) has been developed to improve LTE system capacity by using unlicensed bands. However, the energy efficiency (EE) of the LTE system may be degraded by LAA since unlicensed bands are generally less energy-efficient than licensed bands. In this paper, we investigate the EE optimization of LAA systems. We first develop a criterion to determine whether unlicensed bands can be leveraged to improve the EE of LAA systems. We prove that unlicensed bands can be used to improve the EE only when the allocated licensed resource blocks (RBs) are not enough. We then investigate how to jointly allocate licensed and unlicensed RBs to achieve EE fairness among small cell base stations (SBSs), based on the Nash bargaining solution. Numerical results are presented to confirm our analysis and to demonstrate the effectiveness of the proposed algorithm.

When sharing the unlicensed band with Wi-Fi users in unlicensed LTE (U-LTE) systems, the most critical issue is how to guarantee the harmonious coexistence between the two systems. On the other hand, when sharing the licensed band with the micro base station (MBS) in the small cell base station (SBSs), the co-channel interference needs to be properly coordinated. To address these two issues, power control, licensed and unlicensed spectrum allocation are jointly considered at the SBS to maximize the spectrum efficiency while guaranteeing the QoS of small cell users (SUs) and at the same time providing fair resource sharing with Wi-Fi users. The convex optimization method is applied to design the optimal scheme. Then, numerical results are provided to verify the proposed scheme and demonstrate the tradeoff between the Wi-Fi performance and the achievable spectrum efficiency at the SBS.

This paper investigates joint bandwidth and power allocation for energy-efficient uplink communication in cellular networks. Instead of overall system energy efficiency (EE), we focus on maximizing the EE for each individual user while guaranteeing the quality-of-service. Therefore, a multi-objective optimization problem is formulated. To find its optimal solutions, we first utilize two different methods to convert the multi-objective optimization problem into single objective optimization problems. They are the scalarization method to maximize the weighted summation of users' EE and the max-min method to maximize the minimum EE among users. Then, we develop an effective algorithm based on the sum-of-ratios optimization to solve the scalarization problem, and an algorithm based on the generalized fractional programming to solve the max-min optimization. We also discuss a simple scenario with equivalent bandwidth allocation as a benchmark. Numerical results are provided to validate the effectiveness of the proposed algorithms.

In this paper, joint energy-efficient resource allocation for both the base station and users is studied for time division duplex (TDD) systems with carrier aggregation (CA). We aim at balancing the energy efficiency (EE) between downlink and uplink, as well as the EEs among individual users, by joint bandwidth and power allocation on each carrier component (CC). We formulate the optimization problem into maximizing the weighted summation of EEs for the base station and different users, where the weights are used to reflect the levels of importance. The objective function of the problem is a sum of several fractional functions, therefore, nonlinear sum-of-ratios programming needs to be used to solve it, which has not been exploited in resource allocation problems yet. Specifically, a novel transformation is performed to formulate an equivalent but better tractable problem, based on which we develop an iterative algorithm to find the global optimum of the considered problem. Numerical results validate the feasibility, fast convergence, and flexibility of the proposed algorithm in terms of EE balancing.

In this paper, we propose a novel energy-efficient resource allocation method to simultaneously improve both downlink and uplink energy efficiency (EE) for time division duplex (TDD) systems with carrier aggregation (CA). We aim at EE tradeoff between downlink and uplink by optimizing the power and bandwidth allocation on each carrier component (CC) for each user. The objective function is a sum of several fractional functions, therefore, a novel nonlinear sum-of-ratios programming technique is used to solve it. We first transform the problem into an equivalent and better tractable one and then propose an iterative algorithm to find the global optimum solution. Numerical results show that our method can converge with an acceptable number of iterations and achieve flexible EE tradeoff between downlink and uplink.

Sleep mode control is essential to the energy efficiency of small cell networks. However, frequently switching on/off small cell base stations (SBSs) may cause the degradation to the quality‐of‐service of their users and the increase of network operational cost as well. In this study, the authors propose a novel dual‐threshold‐based sleep mode control strategy for small cell networks. The motivation of using dual‐thresholds to control the sleep mode is to minimise the network energy consumption while avoiding the frequent mode transitions of SBSs at the same time. They utilise the Markov chain method to analyse the performance of the proposed strategy. Optimisation problems are formulated to achieve the optimal dual‐thresholds for two different scenarios: the homogeneous threshold scenario in which uniform dual‐thresholds are applied to all SBSs and the heterogeneous threshold scenario where different dual‐thresholds are assigned to SBSs. For the homogeneous threshold scenario, they develop an optimal solution which is based on exhaustive searching. A reinforcement learning‐based algorithm and a heuristic algorithm are proposed for the heterogeneous threshold scenario, respectively. Simulation results are presented to demonstrate the performance of the author's proposed algorithms.

In this paper, we propose a novel power control scheme for small cells deployed within macro cells. Our aim is to find the optimal power level for each small cell according to the energy consumption tradeoff between network and User Equipments (UEs). Two different small cell deployment scenarios are considered: the non-dense scenario and the dense scenario. The multiagent decentralized Reinforcement Learning (RL) technique is applied to deal with the dense deployment scenario where the coverage of different small cells are overlapped. In the proposed multiagent RL algorithm, each small cell is modeled as an agent to learn the optimal policy from interaction with environment to dynamically change its transmit power. Simulation results are presented to validate the proposed method and show that the RL based algorithm could provide a satisfactory performance.