In this work, the problem of training the secure federated learning (FL) algorithm over a multicell wireless network is investigated. FL is indeed a learning method that can protect users’ privacy, but it has also been shown to be vulnerable to gradient leakage attacks, which can leak users’ private data. Therefore, we propose a defense method to prevent gradient leakage attacks by uploading the selected layers to attend global model updates. Moreover, we add the differential privacy (DP) noise to the model during the local training to strengthen the defense further. Furthermore, to minimize the total privacy leakage for users, we formulate an optimization problem that minimizes this leakage by jointly optimizing resource block (RB) allocation, layer selection, and DP noise. To find a locally optimal solution to this problem, we divide it into two subproblems: 1) initially solving for RB allocation and 2) layer selection using successive convex approximation (SCA) for convex approximation, followed by optimizing DP noise. The simulation results demonstrate that our proposed optimization algorithm outperforms the other two benchmarks.

This article introduces an approach to harness digital twin (DT) technology in the realm of integrated sensing and communications (ISACs) in sixth-generation (6G) Internet of Everything (IoE) applications. We consider moving targets in a vehicular network and use DT to track and predict the motion of the vehicles. After predicting the location of the vehicle at the next time slot, the DT designs the assignment and beamforming for each vehicle. The real-time sensing information is then utilized to update and refine the DT, enabling further processing and decision making. In the DT, an extended Kalman filter (EKF) is used for the precise motion prediction. This model incorporates a dynamic Kalman gain, which is updated at each time slot based on the received echo signals. The state representation encompasses both the vehicle motion information and the error matrix, with the posterior Cramér-Rao bound (PCRB) employed to assess sensing accuracy. We consider a network with two roadside units (RSUs), and the vehicles need to be allocated to one of them. To optimize the overall transmission rate while maintaining acceptable sensing accuracy, an optimization problem is formulated. Since, it is generally hard to solve the original problem, the Lagrange multipliers and fractional programming are employed to simplify this optimization problem. To solve the simplified problem, this article introduces both the greedy and heuristic algorithms by optimizing both the vehicle assignments and predictive beamforming. The optimized results are then transferred back to the real space for ISAC applications. Recognizing the computational complexity of the greedy and heuristic algorithms, a bidirectional long short-term memory (LSTM)-based recurrent neural network (RNN) is proposed for efficient beamforming design within the DT. Simulation results demonstrate the effectiveness of the DT-based ISAC network. Notably, the LSTM-based RNN method achieves similar transmission rates as the heuristic algorithm but with significantly reduced computational complexity.

5G new radio (NR) technology is introduced to satisfy more demanding services. Ultra Reliable Low Latency Communication (URLLC) requires very low delay compared with the previous techniques. This is hard to achieve when hybrid automatic repeat request (HARQ) is applied and especially when the feedback channel is erroneous. In this work, we consider various delay components in incremental redundancy (IR) HARQ systems and minimize the average delay by applying asymmetric feedback detection (AFD) and find the optimal transmission length for each transmission attempt. A M/G/1 queuing model is used in this work to analyze the queuing delay in 5G NR when there are multiple uses in the system. Numerical results show that significant performance gains and lower outage probability can be achieved by applying AFD.

In wireless communication networks, relays are required when the quality of the direct link between the source and the destination is not high enough to support reliable transmission because of long distances or obstacles. Selecting the proper relay node (RN) to support hybrid automatic repeat request (HARQ) is of great importance in such a relay-assisted network. Different from previous works, whether to participate in the transmission is determined by each RN itself in this work, thus reducing the overhead. As RNs do not need to obtain channel state information about the whole network, there is no significant overhead in the system. Using the numbers of transmission attempts required by both channels calculated from the obtained channel state information, each RN sets a timer and forwards the packet when time-out occurs. Simulation results show that our proposed method significantly improves the performance of the system. When the channels are of relatively high quality, the performance of our method is close to the optimal relay selection which requires full information about the network.

This work considers downlink incremental redundancy Hybrid Automatic Repeat Request (IR-HARQ) over unreliable feedback channels. Since the impact of positive feedback (i.e., ACK) error is smaller than that of negative feedback (i.e., NACK) error, an asymmetric feedback detection scheme is proposed to protect NACK and further reduce the outage probability. We formulate the HARQ process as a Markov Decision Process (MDP) model to adapt to the transmission rate of each transmission attempt without enriched feedback and additional feedback cost. We aim to optimize the performance of HARQ process under certain outage probability requirements by finding optimal asymmetric detection thresholds. Numerical results obtained on the downlink Rayleigh fading channel and 5G new radio (NR) PUCCH feedback channel show that by applying asymmetric feedback detection and adaptive rate allocation, higher throughput can be achieved under outage probability limitations.

For crowded and hotspot area, unmanned aerial vehicles (UAVs) are usually deployed to increase the coverage rate. In the considered model, there are three types of services for UAV assisted communication: control message, non-realtime communication, and real-time communication, which can cover most of the actual demands of users in a UAV assisted communication system. A bandwidth allocation problem is considered to minimize the total energy consumption of this system while satisfying the requirements. Two techniques are introduced to enhance the performance of the system. The first method is to categorize the ground users into multiple user groups and offer each group a unique RF channel with different bandwidth. The second method is to deploy more than one UAVs in the system. Bandwidth optimization in each scheme is proved to be a convex problem. Simulation results show the superiority of the proposed schemes in terms of energy consumption.