This paper focuses on spectrum sharing in heterogeneous wireless networks, where nodes with different Media Access Control (MAC) protocols to transmit data packets to a common access point over a shared wireless channel. While previous studies have proposed Deep Reinforcement Learning (DRL)-based multiple access protocols tailored to specific scenarios, these approaches are limited by their inability to generalize across diverse environments, often requiring time-consuming retraining. To address this issue, we introduce Generalizable Multiple Access (GMA), a novel Meta-Reinforcement Learning (meta-RL)-based MAC protocol designed for rapid adaptation across heterogeneous network environments. GMA leverages a context-based meta-RL approach with Mixture of Experts (MoE) to improve representation learning, enhancing latent information extraction. By learning a meta-policy during training, GMA enables fast adaptation to different and previously unknown environments, without prior knowledge of the specific MAC protocols in use. Simulation results demonstrate that, although the GMA protocol experiences a slight performance drop compared to baseline methods in training environments, it achieves faster convergence and higher performance in new, unseen environments.

Recent popularization of the Internet of Vehicles (IoVs) and vehicles-to-everything (V2X) enables the emergence of real-time vehicular applications, posing challenges to resourcelimited vehicles. Toward this end, vehicle edge computing (VEC) has been proposed to alleviate the computational burden on vehicles by leveraging resources from roadside units (RSUs) and VEC servers. While existing works mainly focus on the task requirement for either vehicles or RSUs, the joint task offloading for both V2X and RSUs-to-everything (R2X) has not been fully studied. In this paper, we aim at optimizing the task offloading strategies for both vehicles and RSUs, and adopt a multi-hop task offloading manner to fully utilize the VEC network resources. This problem introduces a severe state-action space shift issue with varying dimensions and representation, which poses challenges for conventional DRL approaches. To address it, we propose a Bidirectional Encoder Representations from Transformers (Bert)-based matching Q-network (BMQN) algorithm. First, we design the BMQN model to efficiently capture correlations among all vehicles and RSUs through bidirectional attention. Then, we propose type-embedded grouped attention and available action embedding to mitigate the overfitting sequence length issue, thereby enhancing generalization capacity. Moreover, we propose to address the state-action space shift issue through a matching-based manner, which can significantly enhance the task offloading ability by matching the states among devices. Simulation results demonstrate that: 1) the BMQN can achieve much better performance than other approaches in scenarios comprising various numbers of vehicles and RSUs as well as diverse road lengths; 2) the BMQN has sufficient generalization capacity to adapt to inexperienced scenarios through matching-based architecture and available action embedding.

This paper explores the challenges of user equipment (UE) association and downlink power control in scalable cell-free massive multiple-input multiple-output (MIMO) systems. Initially, we develop a scalable UE association algorithm, which ensures that each UE is connected to only a small set of access points (APs). This approach effectively reduces both fronthaul requirements and the computational load on the APs. The algorithm employs a competition-based mechanism to ensure that APs efficiently distribute workloads while satisfying the quality of service (QoS) requirements of UEs, preventing them from losing network connectivity. Second, we introduce a distributed downlink power control method based on deep neural networks (DNNs) to improve the long-term downlink spectral efficiency (SE) of the entire network. This method relies solely on locally collected large-scale fading information as DNN input, making it adaptable to dynamic scenarios involving varying numbers of associated UEs. To further enhance the training efficiency of the DNN, we design a distributed training framework that fully leverages the computational resources of distributed edge processors (EPs). Simulation results show that the proposed UE association algorithm and downlink power control method exhibit significant advantages over the benchmarks.

This study investigates patient's perspective on the use of AI in healthcare and the role of Explainable AI in this context. Through a co-creative workshop with six participants from diverse disciplines, we investigated the impact of transparency on trust. The findings highlight parallels between AI and doctors as “black boxes,” the complexity of informed consent and the importance of emotional safety. This work serves as a starting point for ongoing research that engages diverse stakeholder groups, to ensure the development of usercentered XAI solutions that can be effectively implemented in clinical practice.

This paper considers downlink power allocation and beamforming (PABF) for integrated sensing and communications (ISAC) against multipath interference. Yet, ISAC-oriented PABF is of great difficulty, due to its parameter-coupling structure and non-convex problem nature. A novel PABF method is proposed to address this issue. Firstly, in order to handle its complex problem structure, the PABF problem is divided into three subproblems, where communication-end beamformer, sensing-end beamformer and multipath power vector are decoupled. Secondly, structured models of the complex problem are extracted to address the non-convexity challenge. An efficient alternating optimization-based PABF algorithm with closed-form iterations is obtained. At the sensing receiver end, our PABF method can focus beams at the line-of-sight direction, while form null beams at reflection directions for suppressing multipath interference. Simultaneously, at the communication transceiver ends, it can smartly adjust beam gains and transmitting power over multiple paths to maximize the communication performance while ensuring a promised sensing performance. The proposed PABF algorithm can strike an on-demand communication and sensing performance tradeoff, via adjusting the sensing performance requirement. We have verified the efficiency of our PABF method by numerical simulations.

Multi-link operation (MLO) is regarded as one of the most disruptive features in the upcoming IEEE 802.11be standard, known as Wi-Fi 7. However, the performance characterization of heterogeneous multi-link IEEE 802.11be networks, which consist of Multi-Link Devices (MLDs) and legacy Single-Link Devices (SLDs), remains largely unknown. The challenge originates from the lack of proper modeling of multi-link channel access schemes. In this paper, a novel model is established to study the throughput optimization of heterogeneous two-link IEEE 802.11be networks. MLDs adopt one representative synchronous multi-link channel access scheme with the primary channel. Based on the proposed model, explicit expressions of throughput of MLDs and legacy SLDs are both characterized and verified by simulation results. The network throughput is further maximized by optimally choosing the transmission probabilities of SLDs and MLDs. The analysis shows that MLO can enable MLDs to achieve higher device throughput than SLDs, yet the maximum network throughput of heterogeneous networks decreases compared to homogeneous networks composed solely of MLDs or legacy SLDs.

Millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) is one of the most promising technologies from 5G-based Internet of Things (IoT) to future wireless communication-based IoT, which usually relies on beamforming codebooks for data transmission. However, traditional codebooks often consist of numerous narrow beams, which causes substantial training overhead. Although centralized machine learning-based methods can address this issue to some extent, they overlook minority IoT devices scattered across various areas, which is vital for the coverage equity of the environmental adaptive codebook and the optimal average achievable rate. To circumvent the problem, we propose a distributed learning (DL) framework for codebook design in mmWave massive MIMO systems with uneven user distribution. Specifically, the user channel set is first divided into subsets by pre-classification based on the power responses of the featured combining vectors from different subregions. Then, a novel DL architecture processes these subsets, each assigned to different baseband processing boards (BPBs) in building baseband units, alleviating the centralized machine learning burden on the active antenna unit (AAU) or its directly connected BPB. Meanwhile, the current algorithms lack the hardware perspective or only implement the inference stage of the model. Thus, we deploy an FPGA-adapted DL-based codebook training prototype that runs on FPGA, which fully explores the “Backward-While-Forward" strategy for data reuse in the forward and backward passes. Simulation validates the effectiveness of distributed learning. Notably, the FPGA implementation on the embedded-level board outperforms consumer-grade CPU and GPU in terms of both latency and energy efficiency.

Industrial Cyber-Physical Systems (ICPS) have emerged as a critical component in the industrial domain. To facilitate seamless collaboration among massive devices, cloud-edge computing architectures have emerged as a key enabler for ICPS, leveraging distributed intelligence to orchestrate devices and computational tasks. In cloud-edge computing, efficient task offloading and resource management are essential for optimizing task performance and reducing energy costs. However, conventional centralized resource management strategies struggle to satisfy the real-time, adaptability, and performance demands of dynamic ICPS systems. Industrial Cyber-Physical Systems (ICPS) have emerged as a critical component in the industrial domain. To facilitate seamless collaboration among massive devices, cloud-edge computing architectures have emerged as a key enabler for ICPS, leveraging distributed intelligence to orchestrate devices and computational tasks. In cloud-edge computing, efficient task offloading and resource management are essential for optimizing task performance and reducing energy costs. However, conventional centralized resource management strategies struggle to satisfy the real-time, adaptability, and performance demands of dynamic ICPS systems. In this paper, we propose the Distributed Transformer-based Actor-Critic (DTAC) algorithm to jointly determine task offloading and resource management decisions in cloud-edge computing networks, particularly for delay-sensitive applications in ICPS. The DTAC algorithm integrates the powerful transformer model with the popular actor-critic architecture to address the challenge of a hybrid high-dimensional action space. We first train a centralized model to learn coordination among user equipments (UEs) and then introduce a decentralized transfer learning (TL) approach to efficiently adapt the centralized model into the DTAC framework. Using the DTAC model, each UE can independently manage its local resources based solely on local information, avoiding the significant signaling overhead inherent in centralized approaches. Simulation results demonstrate that DTAC not only outperforms other MARL and TL schemes in both small-and large-scale scenarios, but also exhibits strong generalization capabilities in inexperienced settings. Furthermore, DTAC and decentralized TL approaches significantly reduce training costs by 73% compared to other methods, making them more practical for ICPS deployment.

How to effectively support massive access and data transmission in Internet of Things scenarios has been a long-standing and critical issue for various wireless communication networks. To address this issue, a flexible and efficient medium access control protocol is the key. In this paper, we propose Slotted Aloha with Successive Transmission (SAST) scheme, in which upon the successful transmission of the Head-of-Line (HoL) packet, the node delivers the remaining packets with probability 1 until the buffer is cleared or a collision occurs, thereby capitalizing on immediate channel availability. By formulating vacation queuing models of both node and channel, the access/data throughput and access/data delay are explicitly characterized and optimized by properly choosing the transmission probability of the HoL packet. Our analysis reveals that the maximum data throughput of SAST scheme is 0.5, higher than$e^{-1}$in classic slotted Aloha. The practical insights of the analysis are also demonstrated by taking the example of 2-step Small Data Transmission (SDT) random access in 5G. It is shown that the SAST scheme can be seamlessly implemented into 5G and the comparison with 2-step SDT random access reveals that SAST can improve the throughput performance while significantly reduce the signaling overhead, nearly halved in the saturated case and up to 70% reduction in the unsaturated case.

The spectral efficiency in cell-free massive MIMO with grant-free random access under Ricean fading, utilizing a linear maximal-ratio combining detector, is the focus of this paper. Firstly, explicit expressions for the effective signal-to-interference-plus-noise ratio (SINR) are introduced, using the least squares (LS) and minimum mean squared error (MMSE) estimation methods, valid for varying Ricean K-factor and for varying numbers of access point antennas M. Furthermore, we drive the asymptotic SINR expressions under extreme conditions of infinite Ricean K-factor, infinite M, and the appropriate scaling of the users’ power. The corresponding analysis shows that as M approaches infinity, the sum uplink rate converges to $\log _{2} M$, regardless of the Ricean K-factor. Particularly, in the case of LS estimation, reducing each user’s transmit power by $1 / \sqrt{M}$ maintains the rate, converging to a definitive value irrespective of the Ricean K-factor. In the case of MMSE estimation, power is able to be scaled down by $1 / \sqrt{M}$ to achieve it when Ricean K-factor is zero, whereas the power of nonzero Ricean K-factor necessitates a reduction by $1 / M$. Finally, simulations confirm all theoretical outcomes.

We consider the age of information (AoI) evaluation in an Aloha-based random access network powered by energy harvesters. We derive a closed-form expression for the average AoI with general energy buffer capacity. The average AoI is then optimized by adjusting the update rate. The results indicate that when the sum of the energy arrival rate of all nodes is greater than or equal to one, the optimal average AoI in the Aloha network is equivalent to that in a network without energy constraints, by setting the update rate to one divided by the total number of nodes. The optimized average AoI then grows linearly with the number of nodes. Otherwise, a degradation of the optimal average AoI emerges, and the update rate should be tuned to be higher than the energy arrival rate.

Millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) technology represents a promising technology in wireless communication. This technology relies on beamforming codebooks for initial access and transmission. However, conventional codebooks comprise a multitude of single-lobe narrow beams, resulting in redundant beams that may never be utilized in beam training. While centralized machine learning methods can partially address the concern of redundancy, they tend to overlook the presence of minority users scattered across diverse regions. The equitable coverage of environmental adaptive codebooks depends on addressing this issue. Hence, we devise a distributed learning (DL) framework for codebook design, which is tailored for scenarios with uneven user distribution and fully exploits the decentralized and online learning features of DL. Our approach begins by segmenting the user channels into various subsets through a pre-classification process. Then, we introduce a novel DL architecture designed to process the subsets that are assigned to individual user equipments (UEs). Each UE then generates a phase shift matrix that contributes to the concatenation-based global aggregation in the base station. The simulation results confirm the effectiveness of DL in improving the performance of mmWave massive MIMO systems in scenarios with unevenly distributed users.

Mobile-edge Computing (MEC) has now emerged as a complement to cloud computing, providing computational capacity for the resources-constrained edge devices. Recently, intelligent computation offloading and cache placement stands as effective approaches to enhance the performance of dynamic MEC networks. In this paper, we propose an online centralized joint resource management approach, named Transformer-based Actor-Critic (TAC), to minimize the task execution time subject to resource constraints. We decouple this mixed-integer non-linear programming (MINLP) problem into a non-convex offloading decision part and a convex joint resources allocation part, and propose the TAC approach to address the non-convex task offloading problem with low computational complexity. In the joint resources management problem, the high-dimensional state-action space is addressed by the transformer-based actor-critic architecture. Through the proposed TAC, the joint cache, communication and computation resource management can be obtained without the knowledge of future task arrivals. Simulation results demonstrate that the TAC can save 48.4% average task execution time with only 2.3% additional computation delay compared to Random with lowest computational complexity. In particular, it further demonstrates great generalization ability to enhance the performance in untrained scenarios.

Multi-link operation is regarded as a crucial feature in the upcoming WiFi 7 networks, which allows a single multi-link device (MLD) to make concurrent data transmissions over multiple links. To facilitate synchronous multi-link channel access, IEEE 802.11 Task Group BE has proposed various channel access schemes, such as Longest Backoff (LB) access and Shortest Backoff (SB) access. However, the coexisting performance of WiFi 7 networks with multiple channel access schemes remains largely unexplored. In this paper, we develop an analytical model to evaluate the data rate and mean access delay performance of a multi-link WiFi 7 network with two types of devices adopting LB and SB, respectively, each employing different initial backoff window sizes. The ratio of device data rates between LB-MLDs and SB-MLDs is inversely correlated with the number of links, and the ratio of their initial backoff window sizes, indicating potential unfairness if the backoff parameters are not appropriately chosen. The optimal initial backoff window sizes to maximize the network sum rate and minimize the mean access delay under a given data rate ratio are further derived and verified by simulation results. The maximum network sum rate scales with the number of links, and is independent of the target fairness requirement or number of devices. Conversely, the minimum mean access delay for each type of devices, is strongly influenced by the target fairness requirement, and shows a linear increase with the network size.

In the IEEE 802.11ax Wireless Local Area Networks (WLANs), Orthogonal Frequency Division Multiple Access (OFDMA) has been applied to enable the high-throughput WLAN amendment. However, with the growth of the number of devices, it is difficult for the Access Point (AP) to schedule uplink transmissions, which calls for an efficient access mechanism in the OFDMA uplink system. Based on Multi-Agent Proximal Policy Optimization (MAPPO), we propose a Mean-Field Multi-Agent Proximal Policy Optimization (MFMAPPO) algorithm to improve the throughput and guarantee the fairness. Motivated by the Mean-Field games (MFGs) theory, a novel global state and action design are proposed to ensure the convergence of MFMAPPO in the massive access scenario. The Multi-Critic Single-Policy (MCSP) architecture is deployed in the proposed MFMAPPO so that each agent can learn the optimal channel access strategy to improve the throughput while satisfying fairness requirement. Extensive simulation experiments are performed to show that the MFMAPPO algorithm 1) has low computational complexity that increases linearly with respect to the number of stations 2) achieves nearly optimal throughput and fairness performance in the massive access scenario, 3) can adapt to various diverse and dynamic traffic conditions without retraining, as well as the traffic condition different from training traffic.

Abstract Deploying networks in unlicensed spectrum has been drawing significant attention, which serves to alleviate the increasing demands in licensed spectrum. However, the network coexistence in unlicensed channel may lead to throughput degradation and unfairness. An appropriate spectrum‐sharing mechanism is therefore of great significance. In this paper, we study the performance limit of two representative mechanisms used in the coexistence with WiFi, including Duty Cycle (DC) and Listen‐Before‐Talk (LBT). In particular, both the throughputs of the coexisting network and WiFi under two mechanisms are derived as explicit expressions of system parameters, based on which the maximum total throughput of the coexisting network and WiFi is characterized under throughput fairness and 3GPP fairness, respectively. A systematic comparison between the optimal throughput performance of DC and LBT is conducted. It is found that if the coexisting network with LBT occupies the channel for a large period each time it successfully accesses the channel, then the maximum total throughput in LBT would be close to that in DC under both throughput fairness and 3GPP fairness. The optimal settings for DC and LBT mechanisms to achieve maximum total throughput are obtained, respectively, which sheds important light on the design of fair and efficient spectrum‐sharing protocols.

Delivering as many data packets as possible and making the life time of the network as long as possible is one fundamental request for battery-driven wireless network design, where sleeping schemes are usually adopted for prolonging the life time, while, at the sacrifice of the throughput performance. For random access networks, fulfilling this fundamental request is rather challenging due to the distributed nature of the access behavior of nodes. This paper considers massive Machine-Type Communication (mMTC) networks where each node adapts the representative random access scheme Aloha and periodical sleeping-awake cycle. We aim to address how to maximize the life-time throughput of each node, i.e., average number of packets each node can successfully deliver during its life time, with a guarantee of targeted life time via optimal selection of the channel access probability and the sleeping ratio of each node. By deriving the explicit expressions of the life time and the life-time throughput of each node and jointly tuning both the channel access probability and the sleeping ratio, we characterize the maximum life-time throughput with targeted life time, and the corresponding optimal settings. The analysis reveals that if only the channel access probability is optimally tuned, then the throughput and life-time throughput cannot be optimized simultaneously when the network becomes saturated with a large packet arrival rate. In contrast, the network would operate at unsaturated conditions via the joint tuning of the access probability and the sleeping ratio. In this case, the maximum life-time throughput always grows with the packet arrival rate. In addition, it is shown that the effect of the life-time constraint becomes significant only when it exceeds a threshold, where maximum life-time throughput will sacrifice for life-time expectation. The analysis sheds important light on the access and sleeping scheme design of practical Aloha-type networks. By taking Narrow Band-IoT with Power Saving Mode (PSM) as an example, extensive simulation results corroborate that with the proposed optimal setting, the life-time throughput could be significantly improved, especially when the life time requirement is demanding, e.g., 10 years without battery replacement.

Carrier Sense Multiple Access (CSMA) has been widely applied to various kinds of wireless networks, such as Wi-Fi, to serve portable devices which are usually greedy in terms of throughput, but with finite battery budget. Accordingly, how to optimize the usage of finite battery budget to get the best possible throughput performance is of great importance. This paper aims to address this issue by focusing on a saturated CSMA network. Explicit expressions of maximum energy efficiency and the corresponding optimal backoff parameter with or without throughput constraint are derived. It is revealed that optimizing the energy efficiency leads to throughput performance degradation. With a stringent throughput constraint, the energy efficiency has to be sacrificed. The energy efficiency and the throughput can be optimized at the same time only in special cases, e.g., the network size is large. The analysis is verified by simulations and sheds important light on performance optimization of practical CSMA-based networks such as Wi-Fi 6 networks.

The deployment of 5G New Radio in unlicensed spectrum is a promising solution to alleviate the spectrum crunch for cellular networks. With the openness of unlicensed spectrum, 5G New Radio Unlicensed (NR-U) will coexist with the incumbent Wi-Fi networks. It is therefore important to study how to maintain harmonious coexistence with the Wi-Fi network. To address this issue, this paper considers two alternative throughput optimization strategies under the 3GPP fairness by adjusting the access parameter: one is to maximize the total throughput of coexisting scenario, and the other is to maximize the throughput of NR-U network. It is shown that the throughput gain of both optimization strategies are related to the initial backoff window size and the network size of Wi-Fi. Moreover, the first strategy can maximize the total throughput yet it may be unfair to the NR-U network while the second strategy can maximize NR-U throughput yet may be harmful to the total throughput. In practical scenario where the IEEE 802.11 EDCA protocol is adopted in Wi-Fi, the performance of NR-U cannot be guaranteed when optimizing the total throughput, and thus optimizing the throughput of NR-U is suggested for fair coexistence.

Multi-link operation is considered to be one of the new key features in the next generation WiFi 7, i.e., IEEE 802.11be. This paper studies the maximum network sum rate of a general M-link 802.11be network with two different synchronous multi-link channel access methods being proposed by Task Group BE, i.e., Longest Backoff and Shortest Backoff. By using a Markov renewal process to model the behavior of each Head-of-Line packet, explicit expressions of the maximum network sum rate and the corresponding optimal initial backoff window sizes are derived, and verified by simulation results. The analysis shows that Longest Backoff and Shortest Backoff achieve an identical maximum network sum rate. However, to achieve the performance limit, the initial backoff window sizes need to be adaptively tuned in a different manner under the two access methods. As the number of links grows, the initial backoff window size with Longest Backoff should be monotonically decreased, while that with Shortest Backoff should be enlarged.

5G New Radio (NR) introduces a new Radio Resource Control (RRC) state, i.e., RRC INACTIVE, for providing the efficient service for massive Machine Type Communications (mMTC). To release the full potential of the new RRC state, it is of great importance to properly model the new RRC state transition process and reveal the effect of system parameters on the network performance. To address the above issue, this paper proposes a novel 5G RRC analytical model based on discrete-time vacation queuing theory, where the time period of the device in RRC INACTIVE state is regarded as the vacation period of the server in the queueing system. By leveraging this novel model, key performance metrics, such as the random access rate and the RRC resource utilization ratio, are explicitly characterized and obtained as functions of system parameters, including packet arrival rate, service rate and inactivity timer. The analysis reveals that to reduce the random access rate, the system should increase the inactivity timer, packet arrival rate or decrease the service rate. On the other hand, to improve the RRC resource utilization ratio, the inactivity timer should be cut down especially when the arrival rate is small or the service rate is large. The analysis is verified by simulations and sheds important light on practical 5G network design for supporting mMTC.

The timeliness of information is of capital importance for numerous Internet of Things (IoT) services. To improve the information freshness in large-scale distributed IoT systems, this paper focuses on the Peak Age of Information (PAoI) optimization of slotted Aloha networks. Specifically, by assuming the first-come-first-served (FCFS) service discipline and Bernoulli packet arrival model, the mean PAoI is characterized and then optimized by either individually tuning the channel access probability or jointly tuning the channel access probability and packet arrival rate of each sensor. The explicit expressions of optimal parameter settings and the corresponding minimum PAoI are obtained, based on which the age-throughput tradeoff is evaluated. The analysis is verified by simulations. It is found that in the massive access scenarios, the minimum PAoI linearly increases with the network scale in both individual optimization and joint optimization cases, while the latter attains a lower increasing rate, better age performance, and less throughput loss.

In large-scale wireless networks, severe interference may incur that leads to the age of information (AoI) degradation. It is therefore important to study how to optimize the AoI performance. This paper focuses on the average AoI minimization in random access Poisson networks. By considering the spatiotemporal interactions amongst the transmitters, an expression of the average AoI is derived, based on which the optimal average AoI and the corresponding optimal packet arrival rate and channel access probability are further characterized. We further compare the average AoI optimization with the peak AoI optimization. The comparison reveals that the optimal channel access probability for the average AoI optimization and the peak AoI optimization are the same. Yet, the optimal packet arrival rate for the average AoI optimization is smaller than that for the peak AoI optimization. The gap enlarges when the node deployment density becomes small.

5G New Radio Unlicensed (5G NR-U) and WiFi are considered to be the two most representative radio access technologies in the newly released 6 GHz unlicensed bands, and thus their efficient and fair coexistence becomes crucial. In this paper, we study the coexistence performance of 5G NR-U and WiFi by accounting the new physical layer (PHY) enhancements in 5G NR including flexible numerologies and mini-slot scheduling. Consider the 3GPP notion of fairness as the requirement, we further study how to maximize the total network effective throughput of the WiFi and NR-U coexisting network. Explicit expressions of the maximum total network effective throughput and the corresponding optimal initial backoff window sizes of WiFi and NR-U nodes are derived, and verified by simulation results. The analysis shows that if the transmission opportunity (TXOP) value of NR-U nodes exceeds a certain threshold, then a win-win coexistence can be achieved, where both the WiFi and 5G NR-U network can perform no worse than the case when two WiFi networks coexist. In this case, the maximum total network effective throughput steadily grows as the time slot length of NR-U nodes decreases, indicating the PHY enhancement in 5G NR can benefit the coexistence performance of 5G NR-U and WiFi in the unlicensed spectrum.

For battery-limited IoT networks, the energy efficiency and Age of Information (AoI) are two key performance metrics. Yet the tradeoff between energy efficiency and AoI remains unclear for large-scale networks since the analysis becomes challenging due to the couple queue problem. This paper aims to address this issue by studying the performance limit of energy efficiency under AoI constraint.Specifically, we evaluate the energy efficiency via the expected number of successfully transmitted packets during each transmitter’s life time for which the explicit expression is derived based on the spatio-temporal analytical framework in [1]. By further taking the AoI constraint into consideration, explicit expressions of the Maximum Expected Number of Successfully Transmitted Packets (MENSTP) and the corresponding channel access probability are obtained. The analysis reveals that if the Power Ratio of the Transmission state and the Waiting state (PRTW) equals one, i.e., the energy consumption per time slot of the transmission state equals to that of the waiting state, then the expected number of successfully transmitted packets during each transmitter’s life time and the peak AoI can be optimized simultaneously; otherwise, the MENSTP declines with a stringent AoI constraint. Moreover, the performance gap enlarges when the PRTW or the node distribution density increases which reveals a crucial tradeoff between the energy efficiency and AoI. It is therefore of importance to properly tuning the channel access probability to strike an optimal energy-age tradeoff in battery-limited large-scale IoT networks.

Timeliness is an emerging requirement for many Internet of Things (IoT) applications. In IoT networks with a large number of nodes, severe interference may incur that leads to Age-of-Information (AoI) degradation. It is, therefore, important to study how to optimize the AoI performance. This article focuses on the AoI minimization in random-access Poisson networks. By considering the spatiotemporal interactions amongst the transmitters, an expression of the peak AoI is derived, based on which the optimal peak AoI and the corresponding optimal packet arrival rate and channel access probability are further characterized. The analysis shows that when the channel access probability (resp., the packet arrival rate) is given, the optimal packet arrival rate (resp., the optimal channel access probability) is equal to one when nodes are sparsely deployed, and decreases as the node deployment density increases. With a joint tuning of these two system parameters, the optimal channel access probability always equals one. Moreover, with the sole tuning of the channel access probability, the optimal peak AoI is improved with a smaller packet arrival rate only when the node deployment density is high. In contrast, a higher channel access probability always improves peak AoI performance when the packet arrival rate is solely tuned. The analysis in this article sheds important light on freshness-aware design for large-scale networks.

To reduce the signaling overhead for sporadic small packet transmission in massive Machine Type Communications (mMTC), Packet-Based Random Access (PBRA) scheme is introduced in 5G system, where devices can transmit data packets in the random access procedure without connection establishment. Yet, even with PBRA, the signaling overhead may surge if the system parameters are configured improperly. This paper aims to address this issue by studying how to tune the Access Class Barring (ACB) factor to maximize the throughput while maintaining the signaling-to-throughput ratio below a certain level. Explicit expressions of maximum throughput and the corresponding optimal ACB factor in saturated and unsaturated cases are derived. It reveals that with a demanding requirement on signaling-to-throughput ratio, the throughput performance has to be sacrificed even with optimal tuning of ACB factor. To boost the throughput performance, the system should either loose the signaling constraint or enlarge the packet length. The analysis is verified by simulations and sheds important light on practical 5G network design for supporting mMTC with PBRA.

This paper focuses on the outage probability of access delay of Machine-to-Machine (M2M) communications in cellular networks, which is an important performance indicator for Ultra-Reliable and Low-Latency Communication (URLLC). Specifically, by deriving the outage probability for given maximum allowable access delay as a function of system parameters, the outage probability is minimized by optimally tuning the Access Class Barring (ACB) factor. For given outage probability bound, the admission control and resource allocation for the random access channel are further discussed, where the maximum number of Machine-Type Devices (MTDs) that can be admitted with given number of preambles and the minimum number of preambles that should be allocated with given network size are obtained. Compared to the standard setting where the ACB factor is fixed, significant gains in outage probability are demonstrated by optimally tuning the ACB factor according to the number of MTDs and the traffic input rate of each MTD. It is also shown that for given required outage probability bound, the optimal tuning of ACB factor enables much more MTDs to be admitted for given preamble resource, and requires much fewer preambles for given network size.

The massive machine-type communications (mMTC) is one of the three generic services for 5G. With the connection-based random access (CBRA) scheme, each machine-type device (MTD) establishes a connection with the base station (BS) prior to its data transmission. Due to the explosive growth of the number of MTDs, many MTDs would establish connections with the BS, which necessitates the study on how to efficiently manage the massive connections with MTDs. To address this issue, in this article, we propose a unified utility-based analytical framework for the optimal connection management of mMTC in 5G networks, where the signaling overhead for connection establishment, access delay, and the connection resource utilization ratio is included. Specifically, we first derive key performance metrics, i.e., the mean time length of each connection and resource utilization ratio, as functions of traffic input rate and inactivity timer. By further considering the signaling overheads and access delay of each MTD, the network utility is formulated and maximized by optimally choosing the inactivity timer. We then present a detailed discussion on the effect of system parameters on the optimal inactivity timer and the corresponding maximum network utility. Finally, we extend the analytical framework to the scenario in which the CBRA scheme coexists with the packet-based random access scheme, i.e., transmitting packets in the random access channel without connection establishment. The critical threshold in terms of the traffic input rate is characterized, which sheds important light on the access scheme selection issue.

Slotted Aloha provides a simple way for accommodating the massive access of Machine-to-Machine (M2M) communications. Yet, the delay performance of slotted Aloha has long been observed to significantly deteriorate as the network size grows. It is therefore important to study how to optimize the delay performance of slotted Aloha in a large-scale network. This paper focuses on the optimization of access delay of a buffered slotted Aloha network, where n nodes transmit to a common receiver in fading channels. Specifically, by deriving the closed-form expressions of the network steady-state points in both unsaturated and saturated conditions, the first and second moments of access delay of each packet are obtained as explicit functions of system parameters, and minimized by optimizing the transmission probability of each node. The analysis shows that to achieve the minimum mean access delay, the transmission probability of each node should be reduced as the network size increases, leading to a diminishing node data rate unless the information encoding rate is jointly optimized. The minimum mean access delay for a given data rate requirement is further characterized, and effects of key parameters such as the minimum required data rate for each node, the mean received signal-to-noise ratio of each packet and the number of nodes on the rate-constrained minimum mean access delay are discussed. The practical insights of the analysis are also demonstrated by taking the example of an LTE-M system with smart grid applications.

The massive Machine Type Communications (mMTC) is one of the three generic services to be supported by 5G wireless systems. To fulfill the ever-increasing network access demand from a large number of Machine Type Devices (MTDs), this paper develops a cluster-based group paging scheme. Specifically, with the proposed scheme, MTDs are divided into clusters and the group paging period is decomposed into two parts: intracluster access period and inter-cluster access period. In the intracluster access period, preamble reuse is adopted for facilitating the access request transmissions from MTDs in each cluster to its cluster head. In the inter-cluster access period, only cluster heads send access requests to the base station.To evaluate and optimize the access efficiency of the proposed scheme, the probability of successful access of each MTD is characterized, based on which the maximum probability of successful access and the corresponding optimal number of clusters and optimal length of the intra-cluster access period are obtained as explicit functions of key system parameters including the number of preambles and the number of MTDs. A comparative study of the access efficiency for group paging with clustering and without clustering is conducted, which reveals the critical threshold in terms of the number of MTDs, above which clustering is beneficial. The analysis is verified via extensive simulations. It is shown that the access performance of the proposed scheme significantly outperforms that of the traditional group paging scheme, especially in massive access scenarios.

Mangrove species classification is of particular importance for coastal conservation and restoration. However, it is challenging to distinguish species-level differences with limited training data. In this letter, we propose an object-oriented classification method for mangrove forests by using the hyperspectral image (HSI) and the 3-D Siamese residual network. First, superpixel segmentation is utilized to obtain objects with various shapes and scales. Second, 3-D patches of each object are extracted from the original HSI, and those patches containing training samples are adopted to pairwise train the network. The 3-D spatial pyramid pooling (3-D-SPP) is added in the network to extract features in multiple scales. Finally, the abstract features of test samples are learned by the trained network, and the labels are determined by the nearest neighbor classifier within the metric space. Experiments on real mangrove hyperspectral data demonstrate the effectiveness of the proposed method in species classification of mangroves.

Group paging is a baseline solution proposed by the long-term evolution (LTE) standardization body for supporting machine-to-machine (M2M) communications in the current-generation and the next-generation cellular networks. Yet, in conventional group paging scheme, upon the reception of paging message, all machine-type devices (MTDs) in a group will simultaneously access the base station, leading to severe network congestion and intolerably low access efficiency. To handle this issue, in this article, we propose a dynamic group paging mechanism, where only the MTDs with packets to send will join the contention process, and the collisions in the random access channel are addressed by the contention resolution scheme. Explicit expressions of key performance measures including the mean access delay of each MTD are derived as functions of the length of waiting period (interval between two consecutive paging periods) TW, where a smaller TWindicates a higher frequency of group paging. It is shown that TWis a key system parameter that determines the crucial tradeoff between the signaling overheads of the system during the paging period and the access delay performance of each MTD. To study how to properly tune the waiting period length, a utility-based analytical framework is established by taking the aforementioned tradeoff into consideration. The optimal waiting period length for maximizing the network utility is derived and verified by simulation results. The analysis in this article reveals that the network should increase the group paging frequency as the traffic becomes heavier or the number of preambles decreases. Providing more preambles can indeed improve the delay performance, while the gain becomes marginal if the number of preambles is large.

This is a sequel of our previous work [20] on access throughput optimization of Machine-to-Machine (M2M) communications in Long Term Evolution (LTE) networks. By incorporating a finite data transmission rate, this paper aims to characterize the effect of data transmission on the optimal access performance of Machine-Type Devices (MTDs). Specifically, both the maximum access throughput and the corresponding optimal Access Class Barring (ACB) factor are obtained as explicit functions of the data transmission rate, which show that even with the ACB factor optimally tuned, the access throughput may deteriorate as the number of MTDs increases, and even drop to zero if the data transmission rate is too small. To boost the data transmission rate, more resources should be allocated to data transmission, which, however, leads to fewer chances for access. In light of the tradeoff between the data transmission rate and the access frequency, the time slot length is further optimized for maximizing the normalized maximum access throughput. Simulation results corroborate that by properly choosing the time slot length, substantial gains can be achieved over the default setting in various scenarios.

A key challenge for enabling machine-to-machine (M2M) communications in long-term evolution (LTE) networks is the intolerably low access efficiency in the presence of massive access requests. To address this issue, a new analytical framework is proposed in this paper to optimize the random access performance of the M2M communications in LTE networks. Specifically, a novel double-queue model is established, which can both incorporate the queueing behavior of each machine-type device (MTD) and be scalable in the massive access scenarios. To evaluate the access efficiency, the network throughput is further characterized, and optimized by properly choosing the backoff parameters including the access class barring (ACB) factor and the uniform backoff (UB) window size. The analysis reveals that the maximum network throughput is solely determined by the number of preambles, and can be achieved by either tuning the ACB factor or the UB window size based on statistical information such as the traffic input rate of each MTD. Simulation results corroborate that with the optimal tuning of backoff parameters, the network throughput can remain at the highest level regardless of how many MTDs in the network, and is robust against feedback errors of the traffic input rate and burstiness of data arrivals.

A key challenge for enabling Machine-to-Machine (M2M) communications in Long Term Evolution (LTE) networks is the intolerably low access efficiency in the presence of massive access requests. To address this issue, a new analytical framework is proposed in this paper to optimize the random access performance of M2M communications in LTE networks. Both the maximum network throughput and the corresponding optimal backoff parameters including the Access Class Barring (ACB) factor and the backoff window size are obtained as explicit functions of key system parameters such as the number of preambles, the number of Machine Type Devices (MTDs) and the aggregate input rate. The analysis is verified by simulations and sheds important light on practical network design for supporting massive access of M2M communications in LTE networks.

Spectrum leasing has been widely regarded as one of the most effective ways to improve the utilization of limited spectrum resources. The existing literatures normally regulate secondary users (SUs) to lease unused licensed spectrum from primary users (PUs) for indefinite or predetermined time length, which does not adapt to real-time traffic demands of SUs and may limit the utilization of licensed spectrum channels. In view of this, the present paper proposes a novel traffic-adaptive spectrum leasing scheme, which allows PUs and SUs to negotiate leasing periods with variable time length such that SUs can continuously utilize the leasing channels for transmitting dynamically generated secondary packets until their transmission buffers become empty. Following this scheme, we further formulate the utilities of both PUs and SUs in traffic-adaptive spectrum leasing based on M/M/k queues and derive the unique sub-game perfect equilibrium of this scheme based on the Stackelberg game model. Numerical simulation shows that, compared with the existing spectrum leasing schemes based on cooperative relaying, the proposed scheme can concurrently afford both PN and SN with higher leasing utilities, encourage them to join spectrum leasing, and achieve a better utilization of limited spectrum resources.

To overcome the constraint of spectrum heterogeneity, i.e., different spatial locations may have different available spectrum resources, a cognitive radio ad hoc network (CRAHN) should exchange necessary control information among nodes. To improve the performance of this exchange, the present paper proposes to establish a cluster-based Hamiltonian cycle within the CRAHN to provide an ordered flow of control information among clusters and reduce the collision and delay for control information exchange. Moreover, to offer a better clustering result for this establishment, we also design a novel distributed mechanism for randomly selecting a unique node to collect network information and develop an efficient layered clustering algorithm based on the collected information. Numerical simulation shows that, compared with the existing methods, the proposed collection mechanism is more efficient and incurs less packet collisions in collecting network information, while the proposed clustering algorithm yields a smaller average number of clusters under the condition that each cluster has at least one control channel and helps reducing the overhead of control information exchange.