The intelligent development of high-speed railways (HSRs) necessitates support for various services to ensure safe and reliable train operations while providing high-quality travel experiences for passengers. Network slicing presents a promising solution via isolated and service-specific radio resource management. However, meeting heterogeneous quality of service (QoS) requirements in HSR communications is particularly challenging due to imperfect channel state information (CSI) caused by high-speed mobility. In this work, we investigate a slicing puncture strategy in a moving network with an imperfect train-to-ground downlink, supporting passenger entertainment and safety-related services. The system includes two transmission links: outboard and inboard. Given the impact of high-speed mobility, we characterize the statistical probability distribution of the actual CSI and model the average transmission rates in the outboard link. We aim to minimize the system Resource Block (RB) and power in the above two links while satisfying the diverse QoS requirements of services. Since the resource minimization problem is a mixed-integer nonlinear programming, we decompose it into three subproblems: RB allocation, power allocation, and slicing puncture optimization. A Speed-Aware Resource allocation and Slicing puncture (SA-RS) algorithm is proposed. Specifically, analytical expressions are derived for RB allocation, and a bisection-based algorithm is designed for power allocation. Moreover, the slicing puncture strategy is obtained using a genetic-based algorithm. Simulation results demonstrate that the proposed strategy can improve the system performance compared with other baseline schemes under imperfect CSI.

Deep learning-based joint source-channel coding (DeepJSCC) has shown significant benefits in emerging semantic and task-oriented communications, providing a promising solution for reducing latency and bandwidth requirements in next-generation mobile networks. However, its deployment on resource-constrained devices is limited by model complexity. Devices with varying computational capacities require models of distinct architectures and complexity levels, motivating the design of a cross-architecture model compression scheme for DeepJSCC. In this paper, we propose a cross-architecture knowledge distillation framework called CAKDJSCC for heterogeneous DeepJSCC models. Specifically, we design a teaching assistant network with feature fusion modules (FFMs) that dynamically perceive architecture gaps between teacher and student models, thereby generating student-adaptive feature representations to alleviate feature space misalignment caused by architectural inconsistencies. In addition, we introduce a conditional information bottleneck (CIB) loss to optimize the distillation process, which prevents students from overfitting to teacher-specific inductive biases while enhancing knowledge transfer efficiency in cross-architecture scenarios. Extensive experiments demonstrate that our approach significantly improves the student model's reconstruction accuracy and perceptual quality without increasing the inference latency while minimizing the performance degradation during model compression.

In recent years, Deep Joint Source-Channel Coding (DeepJSCC) has demonstrated superior performance over traditional digital schemes in wireless image transmission tasks. However, existing DeepJSCC approaches often overlook three critical challenges in broadcast communication scenarios: bandwidth heterogeneity among users, dynamic bandwidth variations experienced by individual users, and users diverse requirements. To address these issues, we propose a scalable DeepJSCC framework tailored for broadcast communications. This framework enables users to adaptively intercept an appropriate amount of encoded data according to their bandwidth conditions or specific requirements, thereby reconstructing images with corresponding quality. Specifically, the transmitter generates a multi-layered codestream, consisting of multiple base layers for image reconstruction at different resolutions, as well as enhancement layers built upon each base layer to improve visual quality. At the receiver, users can selectively intercept the base layer stream corresponding to their desired resolution, and further receive parts of the associated enhancement layer stream. This design allows for flexible and quality-adaptive image reconstruction based on the amount of data received. Extensive experiments demonstrate that the proposed scheme consistently achieves high-quality reconstruction across various resolutions and under diverse interception conditions.

Recently, federated learning (FL) has attracted much attention as a promising decentralized machine learning method that provides privacy and low latency. However, the communication bottleneck is still a problem that needs to be solved to effectively deploy FL on wireless networks. In this article, we aim to minimize the total convergence time of FL by sparsifying and quantizing local model parameters before uplink transmission. More specifically, we first present the convergence analysis of the FL algorithm with random sparsification and quantization, revealing the impact of compression error on the convergence speed. Then, we jointly optimize the computation, communication resources and the number of quantization bits, sparsity to minimize the total convergence time, subject to the energy and compression error requirements derived from the convergence analysis. By simulating the impact of different compression errors on model accuracy, we reveal that the low-precision updates do not inherently yield a better balance between efficiency and accuracy than the high-precision updates. Furthermore, compared with the equal resource allocation schemes and the unilateral compression optimization schemes on four different data distributions, the proposed scheme has faster convergence speed and less total convergence time.

The deployment of Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surfaces (STAR-RIS) has proven to be an effective means to extend coverage and improve wireless signal quality. STAR-RIS in wireless networks for aided Unmanned Aerial Vehicle (UAV) communications enables a significant boost in network capacity and the provision of virtual line-of-sight links to efficiently meet the quality-of-service (QoS) requirements of user equipment (UE). Accordingly, this paper proposes a novel STAR-RIS-aided multi-UAV communication framework to exploit energy efficiency and total throughput maximally. We formulate the long-term optimization problem as a decentralized, partially observed Markov decision process (DEC-POMDP). Then, we formulate the discrete association scheduling problem as a non-cooperative theoretical game and propose the UA-CFG algorithm to realize the UE association scheme that converges to a Nash equilibrium (NE). Then, a multi-agent reinforcement learning (MARL) method with well-established robustness is devised to continuously optimize the trajectories and energetic consumption of UAVs through centralized training and distributed implementation. Experimental results reveal that the performance of the proposed algorithm is considerable compared to other traditional schemes.

Networks in the 5G and beyond era can use software-defined networks (SDN) to achieve network slicing (NS), so as to meet the extremely diverse service requirements of diverse applications in the Internet of Vehicles (IoV). However, the flow fluctuations in the highly dynamic IoV make it difficult to provide reliable, flexible, and scalable services for the IoV by the SDN control plane. Careful SDN controller placement can be a feasible solution to achieve its robustness and flexibility to deal with the changes in network status. Thus, this paper studies a dynamic controller placement problem to improve the performance of IoV services. To be specific, a hierarchical SDN control plane for the IoV is considered with the SDN controllers placed at the edge of networks. Under this architecture, we model the dynamic controller placement by Markov Decision Process (MDP). To efficiently solve the formulated NP-hard problems, we develop an algorithm based on Deep Deterministic Policy Gradient (DDPG) because of its advantages in solving the problem with multi-dimensional action and large solution space. Further, we incorporate a random process into the action selection strategy of DDPG to prevent it from getting trapped in local optimum. Simulation results show that the proposed DDPG-based controller placement approach can adapt to a highly dynamic IoV environment with outstanding performance.

Cloud gaming represents a major part of contemporary gaming. To boost the Quality-of-Experience (QoE) of cloud gaming, the integration of Dynamic Adaptive Video Encoding (DAVE) with Multi-access Edge Computing (MEC) has become the natural candidate owing to its flexibility and reliable transmission support for real-time interactions. However, as multiple gamers compete for limited resources to achieve personalized QoE, such as ultra-high video quality and ultra-low latency, how to support efficient edge resource optimization is a fundamental and important problem. Furthermore, determining the optimal game video encoding configuration in real-time poses significant challenges, especially when lacking the information on future video and edge network resources. To address these key issues, we jointly optimize the video encoding as well as computing and communication resource allocation by active mutual adaptation of video coding configurations and physical resources in a Software Defined Networking (SDN)-assisted edge network. This eliminates the performance bottleneck caused by decoupling optimization of coding parameter configuration and physical resource allocation. The SDN-assisted edge network architecture supports efficient on-demand resource management, provides global network information, and meets the stringent time-varying game requests. Due to the significant time scale difference between video chunk and physical resource block, we propose a novel Asynchronous Decision-Making Multi Agent Proximal Policy Optimization algorithm (AD-MAPPO), which can address the credit assignment problem with a single agent. It can also adapt to the highly dynamic cloud gaming environment without prior knowledge and a deterministic environmental model. Extensive experimentation based on real cloud gaming datasets convincingly demonstrates that our approach can significantly enhance the overall QoE of gamers.

Deep learning techniques, particularly those based on Wi-Fi fingerprinting, have become increasingly prevalent in the field of indoor localization. These methods typically require specialized training for specific environments and often lack adaptability to changes in indoor settings. In contrast, this study introduces an indoor localization approach based on few-shot regression. The aim is to enable the model to rapidly adapt to new indoor environments using a limited number of labeled Wi-Fi Received Signal Strength Indicator (RSSI) samples. This research treats indoor location prediction as a regression problem, initially pretrain the model on a Wi-Fi dataset from a source domain and establishing a general mapping relationship between Wi-Fi signals and locations using the concept of basis functions. Subsequently, the model is fine-tuned with a small set of Wi-Fi samples from the target domain to learn specific weights. This process of transferring the model from the source to the target domain aids in achieving accurate localization in new and constantly changing environments. Experimental results demonstrate the method’s superior performance in localization accuracy, showing a 57.9% improvement over few-shot classification, a 13% improvement over KNN and a 11.1% improvement over SAE-CNN.

Roadside 3D object detection is crucial for vehicle infrastructure cooperation systems. Due to the distinctive placement of roadside LiDAR, the distribution patterns of roadside point clouds and vehicle-side point clouds differ. In roadside point clouds, the proportion of foreground points in each instance is lower, leading to a notable decline in accuracy when using current sampling methods because of an unguided down-sampling strategy. To address this issue, this paper proposes a point-based single-stage 3D object detector called RDSSD for 3D object detection in roadside scenes. The paper designs a class-guided sampling strategy to efficiently select foreground points associated with potential objects. Furthermore, a task-oriented candidate prediction approach is introduced to generate candidate points that accurately represent the local scene from sampled key points. The experimental results on the DAIR-V2X-I have demonstrated that our method achieves the best detection performance with minimal computational cost.

Due to the unstable and complex social network environment, the sole user–item interaction data become insufficient for generating precise recommendations. However, too much emphasis on user–item interactions prevents the discovery of internal connections among them, such as trustworthy user relations. In this work, we have integrated the collaborative and the sequential relations into an end-to-end graph neural network (GNN) simultaneously and proposed a novel framework, namely multirelational collaborative filtering (MRCF), to explore the evolutional social relations. MRCF mainly consists of two components: relational GNN (RGNN) and simple dot-product attention (SDPA), where RGNN is used to capture not only the collaborative but also the sequential relationship from reliable user–item historical interactions through the graph representation, while SDPA can further concentrate on the dominated interaction sequences between users and items. Moreover, a negative sampling method based on user interest is proposed to help train our model. Extensive experiments on three real-world datasets show that the proposed model performs competitively with other state-of-the-art methods in CF.

The Internet of Things (IoT) mainly consists of a large number of Internet-connected devices. The proliferation of untrusted third-party IoT applications has led to an increase in IoT-based malware attacks. In addition, it is infeasible for the IoT devices to support the sophisticated detection systems due to the restricted resources. Edge computing is considered to be promising. It provides solutions to the data security and privacy leakage brought by untrusted third-party IoT applications. In this article, an intelligent trusted and secure edge computing (ITEC) system is proposed for IoT malware detection. In this system, a signature-based preidentification mechanism is built for matching and identifying the malicious behaviors of untrusted third-party IoT applications. A delay strategy is then embedded into the risk detection engine in order to “buy time” for threat analysis and rate-limit the impact of suspicious third-party IoT applications in the system. We conduct extensive experiments to verify the effectiveness of the ITEC system and show that we can achieve accuracies of up to 98.52%.

Although collaborative 3D perception has successfully improved detection performance by sharing LIDAR information among multiple agents, its impact under adverse weather is under poor investigation. It is non-trivial to reduce the noise effect in the multi-agent system, as each agent may generate defective feature representations with aleatoric uncertainty, and such uncertainty will be further amplified in the collaborative stage due to deterministic collaboration models. To mitigate the negative effects of weather noise on the collaborative framework, we proposed a method called Co-Denoising, which incorporates a two-stage denoising approach within the intermediate collaborative framework. In our method, a sampling-based noise filtering is first performed at each agent to make a coarse denoising. Then, during the collaboration stage, the global feature representations are expanded through Bayesian neural networks to improve the robustness against environmental noise. The extensive experiments on sunny and rainy datasets have indicated the proposed collaborative perception method can significantly reduce performance degradation under adverse weather.

Unmanned Aerial Vehicles (UAVs) can be employed as short-term aerial base stations or as access points for User Equipments (UEs) to communicate with other UEs effectively. However, communication links may be obstructed by buildings, leading to poor data transfer performance and significant energy consumption. Deploying Reconfigurable Intelligent Surfaces (RIS) as part of the UAV-assisted communication system proves to be an effective means to avoid building obstructions and enhance wireless information quality. However, the complexity of communication relationships in multi-UAV systems with RIS-aided communication poses a significant challenge in energy reduction. Therefore, this study investigates a new RIS-aided multi-UAV communication framework for edge computing systems. The system aims to meet the quality-of-service (QoS) for UEs while minimizing the total energy consumption. To optimize the total energy consumption of RIS-aided multi-UAV communication, the impact of communication between multiple UAVs and differences between UE clusters on that system’s performance is also considered. We introduce a Stackelberg game to deal with the communication relationship between multiple UAVs and design a K-means-based clustering algorithm to segment UEs periodically. A model-free deep reinforcement learning algorithm grounded in maximum entropy is proposed to jointly optimize UAV trajectory design, phase shift control, and power allocation to reduce energy consumption further. Experimental results indicate that the system proposed performs favorably concerning both energy consumption and throughput.

Neural Networks’ performance on a sequence of incremental class tasks drops over time for Class Incremental Learning (IL). The gradient-based IL methods can simultaneously adapt to both new and previous tasks by promoting the update of model in the correct direction. However, existing methods simply consider the previous/new task gradients separately. In this paper, we propose Parallel Gradient Blend (PGB) paradigm. On the one hand, PGB uses the gradients generated by mixing previous and new samples in equal proportions with Batch-Normal layers to adjust a reasonable model update direction. By comparing gradient similarities, the model selects either the previous task gradient or mixed gradient to update. On the other hand, PGB uses the sample feature gradient distribution difference to construct a regularized gradient. Finally, we experimentally demonstrate that PGB outperforms state-of-the-art methods on class-IL benchmarks.

In this paper, we study the design of neural network based joint source-channel coding (JSCC) for point-to-point communication of Gauss-Markov sources over the additive white Gaussian noise (AWGN) channel with bandwidth compression. Among the existing deep learning (DL) -based JSCC methods for such sources, the long short-term memory (LSTM) based structure has good performance. However, it takes up huge time and space consumption because of its complex structure. In this work, we propose to adopt the causal convolution and dilated convolution to form our encoder and decoder due to their abilities of effectively extracting the temporal information of sources and their superiority in terms of reducing the time and space consumption. Experimental results show that the proposed model outperforms the traditional JSCC schemes and is comparable to the LSTM-based model in terms of source reconstruction quality. Besides, the proposed model shows a great robustness in the case of channel quality mismatch and correlation coefficient mismatch. Furthermore, our model takes lower time in the test phase and much lower space consumption compared to LSTM-based model.

The integration of Mobile Edge Computing (MEC) and microservice architecture drives the implementation of the sustainable Internet of Vehicles (IoV). The microservice architecture enables the decomposition of a service into multiple independent, fine-grained microservices working independently. With MEC, microservices can be placed on Edge Service Providers (ESPs) dynamically, responding quickly and reducing service latency and resource consumption. However, the burgeoning of IoV leads to high computation and resource overheads, making service resource requirements an imminent issue. What’s more, due to the limited computation power of ESPs, they can only host a few services. Therefore, ESPs should judiciously decide which services to host. In this paper, we propose a Microservice-oriented Service Placement (MOSP) mechanism for MEC-enabled IoV to shorten service latency, reduce high resource consumption levels and guarantee long-term sustainability. Specifically, we formulate the service placement as an integer linear programming program, where service placement decisions are collaboratively optimized among ESPs, aiming to address spatial demand coupling, service heterogeneity, and decentralized coordination in MEC systems. MOSP comprises an upper layer to map the service requests to ESPs and a lower layer to adjust the service placement of ESPs. Evaluation results show that the microservice-oriented service deployment mechanism offers dramatic improvements in terms of resource savings, latency reduction, and service speed.

The International Classification of Diseases (ICD) code is a disease classification method formulated by the World Health Organization(WHO). ICD coding usually requires clinicians to manually allocate ICD codes to clinical documents, which is labor-intensive, expensive, and error-prone. Therefore, many methods have been introduced for automatic ICD coding. However, most of the methods have ignored or cannot combine two essential features well: long-tailed label distribution and label correlation. In this paper, we propose a novel end-to-end Joint Attention Network (JAN) to solve these two problems. JAN includes Document-based attention and Label-based attention to capture semantic information from clinical document text and label description, respectively, which helps solve the classification of dense and sparse data in long-tailed label distribution. Besides, an Adaptive fusion layer and CorNet block are presented to adaptively adjust the weight of these two attentions and exploit label co-occurrence relations, respectively. Experiments on the MIMIC-III and MIMIC-II datasets demonstrate that our proposed JAN outperformed previous state-of-art methods achieving Micro-F1 of 0.553, Micro-AUC of 0.989 and precision at top 8(P@8) of 0.735. Finally, we also provide attention and label correlation visualization to verify the effectiveness of our model and improve the interpretation of our deep learning-based method.

The informed dynamic scheduling (IDS) strategy decoding algorithms performed exceptionally well for low-density parity-check codes in terms of the error-rate performance. However, the IDS decoding algorithm is greedy because of the unfair computation resources allocation among different variables nodes, which leads to poor convergence performance. In order to reduce the greediness of the IDS algorithm, the tabu search (TS) algorithm is introduced to the dynamic scheduling-based decoding in this paper. In the TS-based dynamic scheduling (TSDS) algorithm, the variable nodes in the Tanner graph are temporarily stored in a tabu list. In the decoding process with the TSDS algorithm, variable nodes stored in the tabu list will not be selected and updated until they are shifted out of the tabu list. Besides, an improved updating order is provided for the TSDS algorithm, by which the computational complexity can be decreased without the loss of error correction performance. Simulation results show that the proposed algorithm outperforms other decoding algorithms of interest in terms of bit error rate and convergence performance over the additive white Gaussian noise channel.

This work introduces an indoor positioning system (IPS) which is a combination of wireless sensor network (WSN) and inertial navigation system (INS) for locating a moving object indoor. Here, WSN is adopted to measure the ranges from the unknown node to those anchor nodes by time of arrival (ToA) method. The core of the INS is the inertial measurement unit (IMU), which consists of accelerometers and gyroscopes. The real time inertial measurements from IMU and the range information by ToA method are both transmitted to processing terminal, where we propose to use three kinds of recursive Bayesian algorithms to make use of data to obtain the location estimations. The experimental results show that even with only two anchor nodes, the estimation accuracy of hybrid method by these three algorithms is higher than both standalone ToA method with 3 anchors and pure inertial solution.

This paper studies zero-delay joint source-channel coding (JSCC) of Gaussian sources over additive white Gaussian noise (AWGN) channels in the Wyner-Ziv scenario, that is, when there is side information about the source available only to the receiver. The proposed encoding scheme is based on hybrid digital-analog (HDA) transmission in a zero-delay fashion. Specifically, to achieve zero-delay, after applying scalar quantization to the source, the properly scaled analog information (quantization error) is superimposed on the scaled digital information (quantized source), and then transmitted. At the receiver side, several reconstruction schemes are analyzed. It is shown that all the schemes, when optimized, are superior to existing uncoded, digital, or hybrid transmission methods. It is also shown that the proposed HDA transmission may result in increased robustness to channel and/or side information mismatch compared to uncoded transmission. Observing that the bandwidth-matched Wyner-Ziv problem is intimately related to the coding scenario without side information but with bandwidth expansion factor 2, the proposed scheme is applied to the latter problem. It is shown that this application results in a better performance than the well-known inverse spiral mapping.

We study the zero-delay joint source-channel coding problem of transmitting a Gaussian source over a Gaussian channel in the presence of side information known only to the receiver. To achieve zero-delay, after applying scalar quantization to the source, the properly scaled analog information, namely the quantization error, is superimposed on the scaled digital information, i.e., the quantized source, and then transmitted. At the decoder, two decoding schemes are proposed, both of which estimate digital component first, followed by the analog component. It is shown that both schemes, when optimized over all related parameters, are superior to pure analog transmission for high enough correlation between source and side information. The robustness of one of the proposed HDA schemes against varying channel and side information conditions is also compared with that of the purely analog scheme.

A predictive scheme for zero-delay Wyner-Ziv coding of sources with memory is proposed and analyzed in the high-resolution regime for Gaussian source-side information pairs. Incorporating the propagation of distortion due to occasional decoding errors into the analysis, the quantizers and first-order filter parameters are simultaneously optimized. The rate-distortion performance of the scheme is compared with those of non-distributed coding and Wyner-Ziv coding with a naive choice of prediction filters (minimizing temporal correlation), as well as the asymptotic Wyner-Ziv rate-distortion function.