Satellite-terrestrial integrated network (STIN) has been recognized as a promising paradigm to provide ubiquitous and reliable coverage for billions of devices over the world. Integrated sensing and communication (ISAC) can achieve higher spectrum resource utilization efficiency, reduce the hardware size and lighten the payload of satellites for STIN. Multi-access mobile edge computing (MEC) leverages distributed edge servers to alleviate the computational burden for sensing data processing on the satellites. In this paper, we propose multi-access MEC empowered ISAC for STIN. Specifically, a group of low earth orbit (LEO) satellites perform radar sensing operations with optimized scheduling. While a portion of the acquired sensing data undergoes onboard processing at the satellites, the remaining part is processed remotely on multiple terrestrial edge servers. We formulate an optimization problem which concurrently optimizes the following strategy variables: the sensing scheduling, the beamforming for offloading transmission, the beamforming for radar sensing, the duration for sensing and data offloading, the offloaded workload and the computing capacity allocation of each edge server. Notwithstanding the non-convex nature of the formulated optimization problem, we develop a hierarchical decomposition algorithm for achieving the solution efficiently. Extensive numerical simulations confirm the superior performance of our proposed multi-access MEC-enabled ISAC framework in STIN scenarios while simultaneously verifying the efficiency of our optimization algorithm.

Integrated sensing and communication (ISAC) networks constitute a crucial paradigm for facilitating numerous advanced services in future wireless networks. This paper investigates the joint bandwidth and power allocations for device-to-device (D2D) communications aided ISAC in which D2D pairs complete their data transmissions by using the bandwidth allocated by the base station (BS) while providing sensing services for the BS. To this end, we formulate a joint optimization of bandwidth allocation and power allocations for both the target sensing and data transmission of each D2D pair, with the objective of maximizing a system-wise gain that accounts for both performances of target sensing and D2D data transmission. Despite the formulated optimization problem is strictly non-convex, we develop an efficient algorithm based on Lagrangian duality and sequential convex programming for solving it. Simulation results demonstrate that our proposed D2D communications aided ISAC is both accurate and efficient over several benchmark schemes.

Integrated sensing and communication (ISAC) provides a spectrum-efficient approach for simultaneously enabling reliable data transmission and high-quality sensing. This paper investigates an ISAC-enabled multi-device cooperative sensing system in which the devices perform cooperative sensing towards multiple targets in a time-division manner. Within the allocated time, each device senses the targets and transmits data to the base station simultaneously via ISAC. To investigate this problem, we formulate a joint optimization of the beamforming for both sensing and transmission as well as the time allocation for different devices, aiming at maximizing the total throughput of the devices while guaranteeing the multi-target sensing quality, the cooperative sensing requirement and the fairness in data transmission. To tackle the non-convexity of the formulated problem, we first decompose the problem into a beamforming subproblem and a time allocation subproblem. Subsequently, we transform the beamforming subproblem into a tractable form. We then analyze the feature of the optimal time allocation in the time allocation subproblem while providing its semi-analytical expression, based on which we further propose an efficient algorithm to solve the original problem. Simulation results validate the effectiveness of our algorithm and the performance advantages of our fairness-aware ISAC-enabled cooperative sensing in improving both throughput and cooperative sensing accuracy.

Integrated sensing, communications and computing (ISCC) system has been emerged as a crucial paradigm for addressing the growing demand of emerging wireless applications that require both ultra-reliable low-latency computing and high-precision sensing. In this paper, we investigate a non-orthogonal multiple access (NOMA)-assisted integrated sensing and two-tier task offloading (ISTTO) system in which the multi-functional access point (AP) provides task offloading services for a group of edge computing users via NOMA while performing sensing towards a target. To balance the utilization of the computing resources across different tiers, the AP can further offload part of the received workloads to a group of cloudlet servers. To investigate this problem, we formulate a joint optimization of the AP’s transmit beamforming, the two-tier dedicated sensing signals, the two-tier computation offloading strategies and the associated allocations of the communication and computing resources, with the objective of minimizing the total energy consumption, while guaranteeing the required sensing performance over the total duration. Although the formulated joint optimization problem is strictly non-convex, we identify the features of its solutions and exploit a decomposition-based framework for solving it. Numerical results validate the accuracy and effectiveness of our proposed algorithm and show the performance advantages of our NOMA-assisted ISTTO scheme. Compared with several benchmark schemes, our NOMA-assisted ISTTO scheme achieves better performances in both sensing and task offloading, while suppressing the interference from undesired directions.

Integrated sensing and communication (ISAC) is a promising paradigm for supporting emerging wireless services and applications that require both high-throughput data transmission and accurate environment sensing. In this paper, we investigate the energy-efficient channel sharing aided ISAC with sensing scheduling, in which the ISAC base station (BS) can simultaneously sense multiple targets by reusing the channel of conventional cellular users. To investigate this problem, we formulate a joint optimization of the multi-target sensing scheduling, the BS’s transmitting beamforming, and its receiving beamforming for each sensing target, with the objective of maximizing the energy efficiency for radar sensing while guaranteeing each cellular user’s throughput requirement. Despite that the formulated joint optimization problem is strictly non-convex, we exploit a framework of alternating optimization and propose the corresponding algorithms for solving the problem. Specifically, we address the fractional structure of the objective function by utilizing Dinkelbach’s method. Then, we identify the convexity of the problem after semidefinite relaxation and obtain the beamforming by utilizing the Lagrange duality. Furthermore, we formulate the sensing scheduling problem as a matching game and solve it by adopting the swap matching. Numerical results validate the effectiveness of our proposed algorithms compared to some benchmark algorithms and show the performance advantage of our channel sharing aided ISAC in comparison with different schemes.

Integrated sensing and communication (ISAC) provides an emerging paradigm for enabling a variety of next-generation wireless services and applications. Due to the limited computation resources on ISAC devices and the latency as well as the reliability requirements, we propose a paradigm of mobile edge computing (MEC) aided ISAC with short-packet transmissions, where multiple ISAC devices adopt short-packet transmissions to offload their sensed radar data to an edge-server for analysis. We adopt the mutual information to measure the performance of radar sensing and quantify the reliability and latency performances for analyzing the radar-data via edge computing. We formulate an energy minimization problem that jointly optimizes the size of each short packet, the duration of each short packet, the computing-capacity allocations of edge-server, the beamforming of the radar sensing and the offloading transmission, while providing guaranteed performances for the radar sensing, the latency for radar-data analysis, and the reliability of offloading transmission. We identify the hierarchical structure of the formulated problem and divide the problem into three subproblems. For both the bottom-layer problem optimizing the computing-capacity allocations of the edge-server and the middle-layer problem optimizing the size of each short packet and the duration of each short packet, we derive their solutions analytically. Finally, for the top-layer problem optimizing the beamforming of the radar sensing and the offloading transmission, we transform it into a difference of convex (DC) problem which can be efficiently solved. We show the performance advantages of our proposed scheme. The simulation results show that our proposed algorithm can outperform the benchmark algorithms.

The rapid development of communication and computing techniques enables unmanned aerial vehicles (UAVs) to provide reliable and cost-effective wireless communication and computing services from the air. Compared to the conventional fixed infrastructure, UAVs have attractive attributes, such as high flexibility and operability, and, as a result, on-demand line-of-sight connection links. Therefore, UAV-assisted wireless networks have been envisioned as a promising paradigm to achieve enhanced coverage and connectivity for future wireless communications. Meanwhile, achieving high levels of energy efficiency, sensing, communication, and computing capacities, and security and privacy are critical to the success of UAV-assisted wireless networks. In order to improve the performance of UAV-assisted wireless networks, some frameworks and mechanisms have been developed in the past few years. In this article, we provide a comprehensive survey of these developments. Specifically, we conduct a brief overview for the architecture of UAV-assisted wireless networks from four domains (i.e., framework-related, technology-related, challenge-related, and solution-related) and four aspects (i.e., sensing-related, communication-related, computing-related, and application-related). Then, the integrated sensing, communication, and computing for UAV-assisted wireless networks is introduced, followed by the characteristics and requirements. We also provide the implementation and applications of UAV-assisted wireless networks. Next, we discuss the challenges and the state-of-the-art solutions for UAV-assisted wireless networks. Finally, the advanced technologies for UAV-assisted communication and computing networks are exploited, followed by the potential research directions.

Integrated sensing and communication (ISAC), which enables the joint radar sensing and data communications, shows its great potential in many intelligent applications. In this article, we investigate the unmanned aerial vehicle (UAV)-aided ISAC with mobile-edge computing (MEC), where the ISAC device deployed on the UAV senses multiple targets with the sensing scheduling and offloads the radar sensing data to the edge-server to train a machine learning model for target recognition. The radar estimation information rate is utilized to measure the radar sensing performance. We aim to minimize a systemwise cost that includes both the UAV’s energy consumption and the data collecting time, while satisfying the requirements on both the model training error and the radar sensing performance. We formulate a joint optimization problem of the sensing scheduling, the number of time-slots, the sensing power, the communication power, and the UAV trajectory. Despite the strict nonconvexity of the formulated problem, we propose an efficient algorithm for solving it. Our algorithm jointly leverages the vertical decomposition that exploits the layered structure of the formulated problem and the horizontal decomposition that utilizes the block coordinate descent (BCD) method. Numerical results are presented to validate the effectiveness of our proposed algorithms and show the performance gain of our proposed scheme.

As a promising framework for distributed machine learning (ML), wireless federated learning (FL) faces the threat of eavesdropping attacks when a trained ML model is sent over a radio channel. To address this threat, we propose channel-sharing-based artificial jamming to increase the secrecy throughput of FL clients (FCs). Specifically, when an FC performs local model training, a selected device such as a sensor node (SN) not involved in the FL opportunistically accesses the FC’s channel to transmit its sensing data. In return, when the FC sends its locally trained model to the FL server (FLS), the selected SN provides artificial jamming to increase the FC’s secrecy throughput. Considering multiple FCs and SNs, we first consider a given pairing of FCs and SNs and optimize the local training time, the model uploading time, and the transmit-power of the FCs to minimize the total latency of FL training. After proving the convexity of this optimization problem, we propose an efficient algorithm to derive the semi-analytical solution. Then, we further investigate the pairing of the FCs and the SNs to minimize a system-wise cost reflecting both energy consumption and latency. The resulting problem is a bicriteria pairing problem, and we propose an efficient algorithm to compute the optimal pairing solution. Numerical results demonstrate the efficiency and performance advantage of our proposed channel-sharing-based approach with artificial jamming in comparison with different benchmark schemes.

The explosively increasing development of marine communication networks will improve the quality of service (QoS) of marine applications (e.g., ocean farm and marine tourism), which has attracted much attention from both academia and industrial in recent years. However, real-time data processing for diverse marine tasks (especially those computing-intensive and latency-sensitive tasks) is still challenging due to the limited marine communication and computing resources. Mobile edge computing (MEC) driven by powerful computing capability is envisioned as a promising solution to address the issue for resource-constrained marine services. In this paper, we propose a hybrid offshore and aerial-based multi-access edge computing scheme in marine communication networks to improve the QoS of marine applications. Specifically, we consider a scenario that both offshore base-station and unmanned aerial vehicles (UAVs) are equipped with edge-servers, and the computation workloads of unmanned surface vehicle (USV) can be simultaneously offloaded to offshore base-station and UAVs via multi-access manner. To minimize the latency of completing USV’s workloads and reduce USV’s energy consumption, we formulate a joint optimization problem to optimize the offloading decision, transmission time, and computing-rate allocation, with the objective ofMinimizing theMaximumWorkloadsLatency (MMWL). Exploiting the features of the formulated problem, we present a layered structure approach and decompose it into three subproblems. We propose efficient algorithms to obtain the optimal solutions and validate the optimality of the proposed algorithms. Finally, we provide simulation results and analysis to demonstrate the effectiveness and efficiency of the proposed scheme and algorithms in comparison with benchmark algorithms.

Digital twin (DT), which integrates physical networks and digital space by using advanced technologies of sensing, communication and computation, has been envisioned as a promising paradigm for improving the quality of service in physical systems. In this paper, we propose a federated learning (FL)-enabled DT system consisting of the physical layer and DT layer. With FL, all wireless devices (WDs) can collaborate to update a universal DT model, after the DT server cluster (DSC) aggregates all the local models sent by the WDs with non-orthogonal multiple access (NOMA). Moreover, an action model based on the DT system is also updated to optimize the operations of WDs. To increase the energy efficiency, we formulate a problem to minimize the cost of the total energy consumption of the system by optimizing the time allocation of local training, uploading the local models, generating the action model as well as broadcasting the action model and DT model. The numerical results validate the effectiveness and efficiency of our proposed algorithm.