This paper proposes a vital-sign monitoring system based on near-field WBAN, with integrated sensing, communication, and powering. A two-layer communication medium composed of air and human tissue is established to model both in vitro and in vivo environments. In the in vivo scenario, the propagation, reflection, and scattering of electromagnetic signals are described for vital sign detection. Conversely, in the in vitro setting, a multi-antenna access point (AP) operates in the near-field regime to transmit wireless energy to a wearable vital-sign sensor node, collects vital-sign data to the AP via backscatter communication, and senses the sensor’s position based on the echo signals. We formulate a multi-stage stochastic optimization problem that jointly optimizes the AP’s transmission strategy, time-slot allocation, and beamforming, incorporating the age of information (AoI) to ensure timely data transmission. Since the information-theoretic limit of the monitoring task can be characterized by mutual information (MI), it is adopted as the optimization metric. The objective is to maximize MI for vital-sign monitoring under constraints on communication rate, wireless power transfer, and position sensing accuracy. To solve the resulting joint optimization problem, we use a Lyapunov optimization framework to transform the long-term AoI constraint into a tractable per-slot control form. Building on this formulation, we propose the JO-VSM algorithm, which employs a block coordinate descent (BCD) method to decouple and solve the coupled optimization variables within each slot. Simulation results demonstrate that the proposed JO-VSM algorithm can effectively balance vital-sign monitoring performance, communication rate, and position sensing accuracy, ensuring information freshness and robustness, as indicated by stable AoI convergence over time-slot evolution.

The rise of privacy-preserving applications, such as medical diagnostics and the Metaverse, highlights the importance of federated learning (FL) for distributed model training at the wireless edge. These applications often rely on multimodal data (e.g., text, images, audio), necessitating advances in multimodal federated learning (MMFL). However, MMFL faces challenges like energy efficiency, multimodal fusion, and heterogeneity. To address these, a scheduling and fusion-based MMFL framework (SFMMFL) is proposed that focuses on improving both the scheduling mechanism and aggregation strategy. To improve the training performance under energy constraint, a Lyapunov-based scheduling algorithm is proposed, in which long-term optimization is transformed into immediate optimization. After that, to tackle the issue of model separation caused by multimodal datasets, a multimodal model aggregation strategy based on Knowledge Distillation (KD) is introduced for multimodal fusion. Convergence analysis proves its feasibility, and simulation results demonstrate that it can achieve faster and more stable convergence performance while improving model training accuracy. Specifically, our proposed SFMMFL can lower the energy consumption of the system by about$20\%$for computing and$16.67\%$for transmission.

Federated learning (FL) is a decentralized approach for collaborative model training on edge devices. This distributed method of model training offers advantages in privacy, security, regulatory compliance, and cost efficiency. Our emphasis in this research lies in addressing statistical complexity in FL, especially when the data stored locally across devices is not identically and independently distributed (non-IID). We have observed an accuracy reduction of up to approximately 10%–30%, particularly in skewed scenarios where each edge device trains with only 1 class of data. This reduction is attributed to weight divergence, quantified using the Euclidean distance between device-level class distributions and the population distribution, resulting in a bias term$(\delta _{k})$. As a solution, we present a method to improve convergence in FL by creating a global subset of data on the server and dynamically distributing it across devices using a dynamic data queue-driven FL (DDFL). Next, we leverage Data Entropy metrics to observe the process during each training round and enable reasonable device selection for aggregation. Furthermore, we provide a convergence analysis of our proposed DDFL to justify their viability in practical FL scenarios, aiming for better device selection, a non-suboptimal global model, and faster convergence. We observe that our approach results in a substantial accuracy boost of approximately 5% for the MNIST dataset, around 18% for CIFAR-10, and 20% for CIFAR-100 with a 10% global subset of data, outperforming the state-of-the-art (SOTA) aggregation algorithms.

As we transition from Narrow Artificial Intelligence towards Artificial Super Intelligence, users are increasingly concerned about their privacy and the trustworthiness of machine learning (ML) technology. A common denominator for the metrics of trustworthiness is the quantification of uncertainty inherent in DL algorithms, and specifically in the model parameters, input data, and model predictions. One of the common approaches to address privacy-related issues in DL is to adopt distributed learning such as federated learning (FL), where private raw data is not shared among users. Despite the privacy-preserving mechanisms in FL, it still faces challenges in trustworthiness. Specifically, the malicious users, during training, can systematically create malicious model parameters to compromise the models’ predictive and generative capabilities, resulting in high uncertainty about their reliability. To demonstrate malicious behaviour, we propose a novel model poisoning attack method named Delphi which aims to maximise the uncertainty of the global model output. We achieve this by taking advantage of the relationship between the uncertainty and the model parameters of the first hidden layer of the local model. Delphi employs two types of optimisation, Bayesian Optimisation and Least Squares Trust Region, to search for the optimal poisoned model parameters, named as Delphi-BO and Delphi-LSTR. We quantify the uncertainty using the KL Divergence to minimise the distance of the predictive probability distribution towards an uncertain distribution of model output. Furthermore, we establish a mathematical proof for the attack effectiveness demonstrated in FL. Numerical results demonstrate that Delphi-BO induces a higher amount of uncertainty than Delphi-LSTR highlighting vulnerability of FL systems to model poisoning attacks.

With the advent of the 5G and 6G eras and the explosive growth of mobile users, machine learning (ML) is increasingly used for extracting important information from a large amount of generated data and making intelligent decisions for complex environments. Especially, distributed ML techniques are getting more attention to enable training ML models in a distributed manner by exploiting distributed computational resources at the network edge. Federated learning (FL) as a classical distributed learning approach can not only protect data privacy but also reduce communication overhead. However, it requires synchrony among users, which is hard to satisfy due to the heterogeneity of the wireless networks. Hence, we first propose a clustering-based semi-asynchronous Online FL with AoU-based local update (CSAOFL-ALU) with importance-based user clustering and AsynFL-ALU-based local update. After that, the BS aggregates the cluster model of each cluster with synchronous FL. We also provide mathematical convergence analysis of the CSAOFL-ALU algorithm. The results show that the global model convergence rate is inversely proportional to the users’ AoU, at the same time, the convergence bound of the global loss function is inversely proportional to the size and the importance of the user dataset. The experiments are conducted on the non-IID MINST dataset. Numerical results demonstrate that the proposed AsynFL-ALU with priority-based user scheduling achieves better learning performance than fully AsynFL, and converges faster than the baseline user scheduling schemes. The CSAOFL-ALU converges faster with less communication time than the baseline algorithms and increases the fairness of user participation.

Metaverse is envisioned to be a human-centric framework, and provide a new concept of living by offering comprehensively immersive experience for users in education, medicine and entertainment domain. Since a large amount of private data is generated at each user for accessing Metaverse, the emerging federated learning (FL) provides an effective solution to address the potential privacy leakage of data sharing by adopting the mechanism of local training and global model aggregation. However, the model aggregation is susceptible to adversarial poisoning attacks. This imposes critical issues for the privacy-preserving mechanism in Metaverse. In this research, we develop two poisoning attacks in order to emulate the behaviour of adversaries possibly existing in practical Metaverse scenarios. First, we develop a data poisoning attack using Bayesian optimisation to search for the optimal parameters of generating adversarial examples to conduct reversed adversarial training. Second, we develop a model poisoning attack where we apply layer optimisation using Bayesian optimisation to search the optimal weights for the convolutional layer in order to induce uncertainty in the classification. Numerical results show that both attack schemes can cause attacks that can not be recognised by the FL server, and layer optimisation is a stronger poisoning attack.

Non-orthogonal multiple access (NOMA)-aided mobile edge computing (MEC) system can enhance the spectral-efficiency with massive tasks offloading. However, with more dynamic devices and the uncontrollable stochastic channel environment, it is even desirable to deploy appealing technique, i.e., intelligent reflecting surfaces (IRS), in the MEC system to flexibly adjust the communication environment and improve the system energy-efficiency. In this paper, we investigate the joint offloading, communication and computation resource allocation for IRS-assisted NOMA-aided MEC system. We firstly formulate a mixed integer energy-efficiency maximization problem with the system queue stability constraint. We then propose a Het-erogeneous Multi-agent Lyapunov-function-based Mixed Integer Deep Deterministic Policy Gradient (HMA-LMIDDPG) algorithm which is based on the multi-agent reinforcement learning (MARL) framework with homogeneous edge devices (EDs) and heterogeneous base station (BS) as heterogeneous multi-agent. Numerical results show that our proposed algorithms can achieve superior energy-efficiency performance to the benchmark algorithms while maintaining the queue stability.

In this paper, we propose an opportunistic scheme for the transmission of model updates from Federated Learning (FL) clients to the server, where clients are wireless mobile users. This proposal aims to opportunistically take advantage of the proximity of users to the base station or the general condition of the wireless transmission channel, rather than traditional synchronous transmission. In this scheme, during the training, intermediate model parameters are uploaded to the server, opportunistically and based on the wireless channel condition. Then, the proactively-transmitted model updates are used for the global aggregation if the final local model updates are delayed. We apply this novel model transmission scheme to one of our previous work, which is a hybrid split and federated learning (HSFL) framework for UAVs. Simulation results confirm the superiority of using proactive transmission over the conventional asynchronous aggregation scheme for the staled model by obtaining higher accuracy and more stable training performance. Test accuracy increases by up to 13.47% with just one round of extra transmission.

Cellular-connected unmanned aerial vehicle (UAV) with flexible deployment is foreseen to be a major part of the sixth generation (6G) networks. The UAVs connected to the base station (BS), as aerial users (UEs), could exploit machine learning (ML) algorithms to provide a wide range of advanced applications, like object detection and video tracking. Conventionally, the ML model training is performed at the BS, known as centralized learning (CL), which causes high communication overhead due to the transmission of large datasets, and potential concerns about UE privacy. To address this, distributed learning algorithms, including federated learning (FL) and split learning (SL), were proposed to train the ML models in a distributed manner via only sharing model parameters. FL requires higher computational resource on the UE side than SL, while SL has larger communication overhead when the local dataset is large. To effectively train an ML model considering the diversity of UEs with different computational capabilities and channel conditions, we first propose a novel distributed learning architecture, a hybrid split and federated learning (HSFL) algorithm by reaping the parallel model training mechanism of FL and the model splitting structure of SL. We then provide its convergence analysis under non-independent and identically distributed (non-IID) data with random UE selection scheme. By conducting experiments on training two ML models, Net and AlexNet, in wireless UAV networks, our results demonstrate that the HSFL algorithm achieves higher learning accuracy than FL and less communication overhead than SL under IID and non-IID data, and the learning accuracy of HSFL algorithm increases with the increasing number of the split training UEs. We further propose a Multi-Arm Bandit (MAB) based best channel (BC) and best 2-norm (BN2) (MAB-BC-BN2) UE selection scheme to select the UEs with better wireless channel quality and larger local model updates for model training in each round. Numerical results demonstrate it achieves higher learning accuracy than BC, MAB-BC and MAB-BN2 UE selection scheme under non-IID, Dirichlet-nonIID and Dirichlet-Imbalanced data.

By employing powerful edge servers for data processing, mobile edge computing (MEC) has been recognized as a promising technology to support emerging computation-intensive applications. Besides, non-orthogonal multiple access (NOMA)-aided MEC system can further enhance the spectral efficiency with massive tasks offloading. However, with more dynamic devices brought online and the uncontrollable stochastic channel environment, it is even desirable to deploy appealing technique, i.e., intelligent reflecting surfaces (IRS), in the MEC system to flexibly tune the communication environment and improve the system energy efficiency. In this paper, we investigate the joint offloading, communication and computation resource allocation for the IRS-assisted NOMA MEC system. We first formulate a mixed integer energy efficiency maximization problem with system queue stability constraint. We then propose the Lyapunov-function-based Mixed Integer Deep Deterministic Policy Gradient (LMIDDPG) algorithm which is based on the centralized reinforcement learning (RL) framework. To be specific, we design the mixed integer action space mapping which contains both continuous mapping and integer mapping. Moreover, the award function is defined as the upper-bound of the Lyapunov drift-plus-penalty function. To enable end devices (EDs) to choose actions independently at the execution stage, we further propose the Heterogeneous Multi-agent LMIDDPG (HMA-LMIDDPG) algorithm based on distributed RL framework with homogeneous EDs and heterogeneous base station (BS) as heterogeneous multi-agent. Numerical results show that our proposed algorithms can achieve superior energy efficiency performance to the benchmark algorithms while maintaining the queue stability. Specially, the distributed structure HMA-LMIDDPG can acquire more energy efficiency gain than the centralized structure LMIDDPG.

The ever-growing use of unmanned aerial vehicles (UAVs) as aerial users is becoming a major part of the sixth generation (6G) networks, which could provide various applications, like object detection and video surveillance, by exploiting machine learning (ML) algorithms. However, the training of conventional centralized ML algorithms causes high communication overhead due to the transmission of large datasets and may reveal user privacy. Hence, distributed learning algorithms, including federated learning (FL) and split learning (SL), are proposed to train ML models in a distributed manner via sharing model parameters rather than raw data. Due to the different learning structures, they have different communication and learning efficiency. We propose a new distributed learning architecture, namely hybrid split and federated learning (HSFL), by adopting the parallel model training mechanism of FL and the network splitting structure of SL. Through the simulations in wireless UAV networks, the HSFL algorithm is demonstrated to have higher learning accuracy than FL and less communication overhead than SL under non-IID data. We further propose a Multi-Arm Bandit (MAB) based best channel (BC) and best 2-norm (BN2) (MAB-BC-BN2) UE selection scheme to select the UEs with better channel quality and larger local model updates in each round. Numerical results demonstrate it achieves higher learning accuracy than the benchmark schemes, BC, MAB-BC, and MAB-BN2 UE selection schemes.

The space-air-ground integrated network (SAGIN) has drawn increasing attention for its benefits, such as wide coverage, high throughput for 5G and 6G communications. As one of the links, space-air communications between multiple unmanned aerial vehicles (UAVs) and Ka-band orbiting low earth orbit (LEO) satellites face a crucial challenge in tracking the 3D dynamic channel information. This paper exploits a statistical dynamic channel model called the multi-dimensional Markov model (MD-MM), which investigates the more realistic spatial and temporal correlation in the sparse UAVs-satellite channel. Specifically, the spatial and temporal probabilistic relationships of multi-user (MU) hidden support vector, single-user (SU) joint hidden support vector, and SU hidden value vector are investigated. The specific transition probabilities that connect the SU and MU hidden support vector for both azimuth and elevation directions are defined. Moreover, based on the proposed MD-MM, we derive a novel multi-dimensional dynamic turbo approximate message passing (MD-DTAMP) algorithm for tracking the 3D dynamic channel in multiple UAVs systems. Furthermore, we also develop a gradient update scheme to recursively find the azimuth and elevation offset for 3D off-grid estimation. Numerical results verify that the proposed algorithm shows superior 3D channel tracking performance with smaller pilot overhead and comparable complexity.

To boost the performance of wireless communication networks, unmanned aerial vehicles (UAVs) aided communications have drawn dramatically attention due to their flexibility in establishing the line of sight (LoS) communications. However, with the blockage in the complex urban environment, and due to the movement of UAVs and mobile users, the directional paths can be occasionally blocked by trees and high-rise buildings. Intelligent reflection surfaces (IRSs) that can reflect signals to generate virtual LoS paths are capable of providing stable communications and serving wider coverage. This is the first paper that exploits a three-dimensional geometry dynamic channel model in IRS- assisted UAV-enabled communication system. Moreover, we develop a novel deep learning based channel tracking algorithm consisting of two modules: channel pre-estimation and channel tracking. A deep neural network with off-line training is designed for denoising in the pre-estimation module. Moreover, for channel tracking, a stacked bi-directional long short term memory (Stacked Bi-LSTM) is developed based on a framework that can trace back historical time sequence together with bidirectional structure over multiple stacked layers. Simulations have shown that the proposed channel tracking algorithm requires fewer epochs to convergence compared to benchmark algorithms. It also demonstrates that the proposed algorithm is superior to different benchmarks with small pilot overheads and comparable computation complexity.

In this article, we investigate resource allocation with edge computing in Internet-of-Things (IoT) networks via machine learning approaches. Edge computing is playing a promising role in IoT networks by providing computing capabilities close to users. However, the massive number of users in IoT networks requires sufficient spectrum resource to transmit their computation tasks to an edge server, while the IoT users were developed to have more powerful computation ability recently, which makes it possible for them to execute some tasks locally. Then, the design of computation task offloading policies for such IoT edge computing systems remains challenging. In this article, centralized user clustering is explored to group the IoT users into different clusters according to users' priorities. The cluster with the highest priority is assigned to offload computation tasks and executed at the edge server, while the lowest priority cluster executes computation tasks locally. For the other clusters, the design of distributed task offloading policies for the IoT users is modeled by a Markov decision process, where each IoT user is considered as an agent which makes a series of decisions on task offloading by minimizing the system cost based on the environment dynamics. To deal with the curse of high dimensionality, we use a deep Q-network to learn the optimal policy in which deep neural network is used to approximate the Q-function in Q-learning. Simulations show that users are grouped into clusters with optimal number of clusters. Moreover, our proposed computation offloading algorithm outperforms the other baseline schemes under the same system costs.

The space-air-ground integrated network (SAGIN) aims to provide seamless wide-area connections, high throughput and strong resilience for 5G and beyond communications. Acting as a crucial link segment of the SAGIN, unmanned aerial vehicle (UAV)-satellite communication has drawn much attention. However, it is a key challenge to track dynamic channel information due to the low earth orbit (LEO) satellite orbiting and three-dimensional (3D) UAV trajectory. In this paper, we explore the 3D channel tracking for a Ka-band UAV-satellite communication system. We firstly propose a statistical dynamic channel model called 3D two-dimensional Markov model (3D-2D-MM) for the UAV-satellite communication system by exploiting the probabilistic insight relationship of both hidden value vector and joint hidden support vector. Specifically, for the joint hidden support vector, we consider a more realistic 3D support vector in both azimuth and elevation direction. Moreover, the spatial sparsity structure and the time-varying probabilistic relationship between degree patterns named the spatial and temporal correlation, respectively, are studied for each direction. Furthermore, we derive a novel 3D dynamic turbo approximate message passing (3D-DTAMP) algorithm to recursively track the dynamic channel with the 3D-2D-MM priors. Numerical results show that our proposed algorithm achieves superior channel tracking performance to the state-of-the-art algorithms with lower pilot overhead and comparable complexity.

Channel estimation is crucial to beamforming techniques in directional millimetre wave (mmWave) communications, which is generally designed based on channel state information with the assumption that the channel is static. However, due to the Doppler effect caused by the mobility of the users in highly mobile applications, the mmWave channel is changing rapidly. Spatial channel covariance, defined by long-term statistic information of channels, is a promising solution to reduce channel estimation frequency, and which can be used to design hybrid precoders. In this paper, we investigate compressive sensing based spatial channel covariance estimation for hybrid mmWave multiuser (MU) multiple input multiple output (MIMO) system. The updated sparse Bayesian learning (Updated-SBL) algorithm is proposed which is achieved by reducing the total squared mutual coherence of the sensing matrix in it. Simulations demonstrate that the total squared mutual coherence of the proposed Updated-SBL algorithm is dramatically reduced and the superiority of the proposed algorithm is validated by comparing to the other benchmark methods.

In this paper, we consider resource allocation for edge computing in internet of things (IoT) networks. Specifically, each end device is considered as an agent, which makes its decisions on whether offloading the computation tasks to the edge devices or not. To minimize the long-term weighted sum cost which includes the power consumption and the task execution latency, we consider the channel conditions between the end devices and the gateway, the computation task queue as well as the remaining computation resource of the end devices as the network states. The problem of making a series of decisions at the end devices is modelled as a Markov decision process and solved by the reinforcement learning approach. Therefore, we propose a near optimal task offloading algorithm based on ϵ-greedy Q-learning. Simulations validate the feasibility of our proposed algorithm, which achieves a better trade-off between the power consumption and the task execution latency compared to these of edge computing and local computing modes.

In this paper, we investigate optimal schemes to manage time scheduling of multiple modules, including spectrum sensing, radio frequency (RF) energy harvesting (RFH) and ambient backscatter communication (ABCom) by maximizing data transmission rate in Internet of Things networks. We first detect ambient RF signals with high signal power as the RF resource of RFH and ABCom by using spectrum sensing with energy detection techniques. Specifically, compressive sensing (CS) is adopted to detect the wideband RF signals with improving spectrum sensing efficiency at the same time. We formulate a joint optimization problem to manage time scheduling parameter and power allocation ratio. In addition, we propose to find the threshold of spectrum sensing for ABCom communications by analyzing the outage probability of backscatter communications. Numerical results demonstrate that the optimal schemes using spectrum sensing are achieved with better transmission rates. The designed time scheduling scheme with CS is confirmed to be more efficient, and the superiorities become more obvious with the increase of network operation time. Moreover, the optimal scheduling parameters and power allocation ratios are obtained. Simulations illustrate that the threshold of spectrum sensing for backscatter communications is obtained by analyzing the outage probability of backscatter communications.

In this paper, the efficient resource allocation for the uplink transmission of wireless powered Internet of Things (IoT) networks is investigated. We adopt LoRa technology as an example in the IoT network, but this paper is still suitable for other communication technologies. Allocating limited resources, like spectrum and energy resources, among a massive number of users faces critical challenges. We consider grouping wireless powered IoT users into available channels first and then investigate power allocation for users grouped in the same channel to improve the network throughput. Specifically, the user grouping problem is formulated as a many to one matching game. It is achieved by considering IoT users and channels as selfish players which belong to two disjoint sets. Both selfish players focus on maximizing their own utilities. Then we propose an efficient channel allocation algorithm (ECAA) with low complexity for user grouping. Additionally, a Markov decision process is used to model unpredictable energy arrival and channel conditions uncertainty at each user, and a power allocation algorithm is proposed to maximize the accumulative network throughput over a finite-horizon of time slots. By doing so, we can distribute the channel access and dynamic power allocation local to IoT users. Numerical results demonstrate that our proposed ECAA algorithm achieves near-optimal performance and is superior to random channel assignment, but has much lower computational complexity. Moreover, simulations show that the distributed power allocation policy for each user is obtained with better performance than a centralized offline scheme.