This paper presents WirelessGPT, a generative foundation model designed for multi-task learning in integrated sensing and communication (ISAC) systems. Built upon large-scale heterogeneous wireless datasets including Traciverse, Sensiverse, and DeepMIMO, WirelessGPT learns universal spatio-temporal-frequency representations through self-supervised pretraining with masked channel token prediction. The proposed architecture introduces a multi-scale patch embedding module to capture both local and global channel features, and a triple-axis attention encoder to jointly model temporal, spatial, and frequency-domain dependencies. After pretraining, the model can be efficiently fine-tuned via lightweight adapters for diverse downstream tasks such as channel estimation, channel prediction, human activity recognition, environment reconstruction, and object tracking. Experimental results show that WirelessGPT achieves superior accuracy and generalization under limited labeled data and dynamic ISAC conditions, outperforming traditional and task-specific models in low-SNR and high-mobility scenarios while maintaining efficient inference suitable for edge deployment. By unifying communication and sensing functionalities within a single generative backbone, WirelessGPT establishes a scalable paradigm for AI-native 6G systems, enabling shared representations that support heterogeneous wireless tasks.

Integrated sensing and communication (ISAC) has emerged as a key enabling technology for 6G networks. This paper addresses the joint beamforming design challenge in ISAC systems to prevent sensing information leakage to legitimate communication users. We propose a novel optimization framework that leverages semidefinite relaxation (SDR) and alternating optimization (AO) techniques to jointly design the base station beamforming vectors and the phase shift matrix of the RIS deployed at the radar target. The proposed approach ensures the satisfaction of communication SINR requirements while effectively suppressing the eavesdropping capability of sensing eavesdroppers. Simulation results demonstrate that our method achieves near-complete eavesdropping elimination compared to RIS-free and random-phase RIS configurations, enabling secure decoupling of sensing and communication functionalities.

In this paper, we introduce a layer-wise compensated aggregation algorithm designed for federated learning in dynamic environments. To reduce data transmission and maintain model continuity, our method implements a hierarchical update strategy that avoids artificially modular partition of neural networks and thus more flexible in practice. Specifically, weight layers closer to the output, which significantly impact model performance, receive direct updates. Conversely, layers further from the output with minimal influence, are updated using gradient compensation from the previous iteration. To ensure consistency across the model, we compensate these less influential weight layers with residual values from earlier rounds, thus enhancing integration and continuity. Experimental results from two datasets confirm that our algorithm not only secures stable convergence but also matches the generalization performance of traditional federated averaging, even under conditions with dropout rates as high as 90%.

In smart ocean Internet of Things (IoT) systems, autonomous underwater vehicles (AUVs) are responsible for underwater information collection. Due to the nature of the medium, the acoustic communications for AUVs are of low bandwidth and adverse environmental conditions causing severe transmission problems. In order to realize cross-media transmission from AUVs to the offshore platform, unmanned surface vehicles (USVs) have been suggested to forward the collected information in a coordinated manner. Against this backdrop, this contribution develops a cooperative USV-to-USV (U2U) cross-media cooperative communications scheme. Then, we formulate a rate maximization problem with the objective of optimizing the reputation-aware USV selection strategy. Furthermore, to characterize the impact of the mobility of AUVs/USVs, a long-term dynamic process is constructed. Meanwhile, we also develop an efficient algorithm which transforms the reputation-aided dynamic USVs selection problem into the infinite-time horizon average one restricted by time average rate constraints in the collection of penalty processes with the help of the Lyapunov optimization framework and drift-plus-penalty method. Finally, numerical results are presented to validate the convergence behavior and the performance for the designed dynamic USVs selection algorithm.

In this article, we investigate the jointly optimized resource allocation with hybrid multiple access in energy-efficient industrial Internet of Things (IIoT), where some devices (e.g., those for critical control devices) have higher transmission priority and stable energy supply while some devices (e.g., those for comprehensive sensors) may not. We consider a system model supporting wireless powered IIoT devices, with certain user terminal as a potential relay for the transmission between a hybrid access point and another user terminal. Constrained by the limited energy storage, the user needs to harvest energy before relaying and only the harvested energy is utilized for the following transmission. We propose a collaborative orthogonal and nonorthogonal multiple access protocol where two cooperation schemes with and without decoding the relay message are applied. Jointly considering time sharing in the transmission process, power splitting for simultaneous wireless information and power transfer, and transmit power allocation at the cooperative user, the achievable rate regions under the Rayleigh fading channel model are derived. Based on which, an optimization problem on resource allocation strategies is formulated and discussed. Both analytical and numerical results are provided, illustrating the impact of user geometry on the achievable rates as well as the optimal resource allocation with different cooperative strategies applied in different use cases. Aiming to enhance resource utilization, energy-efficient cooperation enables the combination of various transmission modes and networking classes in large scale networks, as well as a better use of ambient radio frequency signals for wireless powered transmissions.

With the expansion of the IoT, it is important to optimize available bandwidth to reliably support edge to device communications. Thus we propose a wireless network where each edge server communicates with its end devices using its wireless band as a primary channel, assisted by a secondary edge server that can relay communications via its own wireless band as a secondary channel. The network can optimize capacity by balancing load between primary and secondary wireless bands, and we analyze the geometry of achievable rate regions, depending on the state of bands modeled as Rayleigh fading channels. The allocation of a connection to the primary or secondary band is formulated as an optimization problem which is then solved, and illustrated with numerical examples.

In this paper, an intelligently cooperative communication network with cognitive users is considered, where in a primary system and a secondary system, respectively, a message is communicated to their respective receiver over a packet-based wireless link. The secondary system assists in the transmission of the primary message employing network coding, on the condition of maintaining or improving the primary performance, and is granted limited access to the transmission resources as a reward. The users in both systems exploit their previously received information in encoding and decoding the binary combined packets. Considering the priority of legitimate users, a selective cooperation mechanism is investigated and the system performance based on an optimization problem is analyzed. Both the analytical and numerical results show that the condition for the secondary system accessing the licensed spectrum resource is when the relay link performs better than the direct link of the primary transmission. We also extend the system model into a network with multiple secondary users and propose two relay selection algorithms. Jointly considering the related link qualities, a best relay selection and a best relay group selection algorithm are discussed. Overall, it is found that the throughput performance can be improved with multiple secondary users, especially with more potential users cooperating in the best relay group selection algorithm.

We consider cooperation in downlink non-orthogonal multiple access (NOMA) in a network supporting diverse communication modes. One user (UE2) exists as a potential relay between a base station (BS) and another user (UE1). With relaying the signal for UE1, UE2 obtains the opportunity for its own transmission to UE3 in D2D mode, meanwhile maintaining the transmission efficiency for UE1. On the basis of supporting different communication modes, we propose a NOMA-based cooperation scheme at UE2 to combine the relay message with its own. We derive achievable rate regions for two cases depending on the status of the Rayleigh fading channel of the UE2. We find solutions based on experiments through the transmit power allocation strategy at the UE2 and(/or) the BS. We show the impacts of our cooperative scheme and the corresponding user geometry on the achievable rates and the resource sharing strategies.

We consider optimized cooperation in joint orthogonal multiple access and nonorthogonal multiple access in industrial cognitive networks, in which lots of devices may have to share spectrum and some devices (e.g., those for critical control devices) have higher transmission priority, known as primary users. We consider one secondary transmitter (less important devices) as a potential relay between a primary transmitter and receiver pair. The choice of cooperation scheme differs in terms of use cases. With decode-and-forward relaying, the channel between the primary and secondary users limits the achievable rates especially when it experiences poor channel conditions. To alleviate this problem, we apply analog network coding to directly combine the received primary message for relaying with the secondary message. We find achievable rate regions for these two schemes over Rayleigh fading channels. We then investigate an optimization problem jointly considering orthogonal multiple access and nonorthogonal multiple access, where the secondary rate is maximized under the constraint of maintaining the primary rate. We find both analytical solutions as well as solutions based on experiments through the time sharing strategy between the primary and secondary system and the transmit power allocation strategy at the secondary transmitter. We show the performance improvements of exploiting analog network coding and the impacts of cooperative schemes and user geometry on achievable rates and resource sharing strategies.

We consider a scenario consisting of a primary and a secondary system, each represented by a pair of a transmitter and a receiver. The secondary transmitter assists in the retransmission of the primary message, which prevents the primary performance from being degraded by allowing the secondary system to access the transmission resources. Two network coding schemes applied in retransmission phase are investigated, the stationary network coding (SNC) scheme and the adaptive network coding (ANC) scheme. For each scheme we derive analytical results on packet throughput and infer that the ANC scheme outperforms the SNC scheme. We then provide a numerical performance comparison and a numerical optimization of the secondary packet throughput. Our main result shows cooperation can provide a significant performance improvement through effective network coding.

This paper investigates cooperation for secrecy in cognitive radio networks. In particular, we consider a four-node cognitive scenario where the secondary receiver is treated as a potential eavesdropper with respect to the primary transmission. The cognitive transmitter can help the primary transmission, and it should also ensure that the primary message is not leaked to the secondary user. We consider two cognitive scenarios depending on whether the secondary transmitter knows the primary message or not. In the first case, the secondary transmitter is unaware of the primary transmitter's message and acts as a helping interferer to enhance the secrecy of the primary transmission, whereas in the second case, relaying of the primary message is also within its capabilities. First, we find achievable rate regions for these two scenarios in the case of AWGN channels. We then investigate three different optimization problems: the maximization of the primary rate, the maximization of the secondary rate and the minimization of the secondary transmit power. For these optimization problems, we find closed-form expressions in important special cases. Furthermore, we analyze the cooperation between the primary and secondary transmitters from a game-theoretic perspective. We model their interaction as a Stackelberg game, for which we define and find the Stackelberg equilibrium. Finally, we use numerical examples to illustrate the rate regions, the three optimizations, and the impact of the Stackelberg game on the achievable rates and on the transmission strategies of the secondary transmitter.

This paper explores the trade-off between cooperation and secrecy in cognitive radio networks. We consider a scenario consisting of a primary and a secondary system. In the simplest case, each system is represented by a pair of transmitter and receiver. We assume a secrecy constraint on the transmission in the sense that the message of the primary transmitter has to be concealed from the secondary receiver. Both situations where the secondary transmitter is aware and unaware of the primary message are investigated and compared. In the first case, the secondary transmitter helps by allocating power for jamming, which increases the secrecy of the first message. In the latter case, it can also act as a relay for the primary message, thus improving the reliability of the primary transmission. Furthermore, we extend our results to the scenario where the secondary system comprises multiple receivers. For each case we present achievable rate regions. We then provide numerical illustrations for these rate regions. Our main result is that, in spite of the secrecy constraint, cooperation is beneficial in terms of the achievable rates. In particular, the secondary system can achieve a significant rate without decreasing the primary rate below the benchmark rate achievable without the help of the secondary transmitter. Finally, we investigate the influence of the distances between users on the system's performance.