Integrated sensing and communications (ISAC) has become increasingly crucial in next-generation wireless networks. Leveraging the reliable line-of-sight (LoS) links and mobility of unmanned aerial vehicles (UAVs), UAV-assisted ISAC has attracted significant attention. Different from the previous UAV-ISAC scenarios with single target or overlapping users and targets, we investigate ISAC in a more general multi-UAV network with independent multiple communication users and multiple sensing targets, where the UAVs provide downlink communications to the users while sensing the targets. Additionally, we consider the complicated interference management among the UAVs to further enhance the network’s practicality. Such a scenario presents a new challenge for the joint optimization problem in terms of the UAV trajectories, the user association, the target association, and the power control. Furthermore, since the existing single-objective and weighted optimization approaches may result in potential performance loss and optimization biases, we propose a dual-objective model to further optimize ISAC, aiming for a better tradeoff between the communication and sensing performance. Specifically, we propose an efficient sensing and communication dual-objective multi-UAV optimization algorithm (SC-DO-MUOA) to maximize communication rate and simultaneously minimize sensing Cramér-Rao bound (CRB). Simulation results demonstrate that our proposed SC-DO-MUOA outperforms various baselines in both communication and sensing performance.

Motion state sensing is crucial in wireless communication systems employing Integrated Sensing and Communication (ISAC), with accurate target velocity estimation being central to its effectiveness. While waveform design for ISAC signals can enhance the sensing capability, most existing studies focus on single-target scenarios, with limited discussion on multi-target scenarios. This paper proposes a novel waveform design approach aiming for enhancing the multi-target sensing performance, under the prior knowledge that the Doppler difference between two hardly resolvable targets lies in a certain interval. In particular, we use the Ziv-Zakai Bound (ZZB) of Doppler difference to design the waveform for multi-target sensing, which effectively captures the ambiguity phenomenon. We first derive the expression of the Doppler difference ZZB, which reveals the connection between ZZB and the widely-used ambiguity function. Based on this expression, we propose an SNR-adaptive pulse modulation strategy that significantly improves the velocity estimation accuracy for multi-target scenarios. Numerical results demonstrate that the Doppler difference ZZB effectively reflects the multi-target Doppler frequency estimation performance of maximum aposterioriestimators.

Accurate velocity sensing is crucial in Integrated Sensing and Communication (ISAC) systems, while most studies focus on single-target cases with limited attention to multi-target scenarios. This paper proposes a novel waveform design approach that enhances multi-target sensing performance by leveraging the Ziv-Zakai Bound (ZZB) of Doppler difference, effectively capturing the ambiguity phenomenon under the prior knowledge that the Doppler difference between two hardly resolvable targets lies in a certain interval. We first derive the ZZB for Doppler difference, which reveals the connection between ZZB and the widely-used ambiguity function. Based on this, an SNR-adaptive pulse modulation strategy is developed to enhance multi-target sensing accuracy. Numerical results confirm that the proposed Doppler difference ZZB effectively captures the estimation performance of maximum a posteriori estimators in multi-target scenarios.

The growing demands for privacy protection challenge the joint training of one model by leveraging multiple datasets. Federated learning (FL) provides a new way to overcome this challenge and has attracted many research interests, which enables multiple parties to collaboratively train a machine learning model without exchanging their local data. Despite some success, the non-independent and identically distributed (non-IID) data distributions in different parties remain challenging and easily damage the performance of FL methods, specifically for the heterogeneous multimodal data. Existing FL studies on non-IID data settings are often dedicated to the label space, neglecting the non-IID issues in feature space, thus limiting their performance when the parties with non-IID multimodal data. This paper proposes a newFederated learning method viaSelective featureAlignment (FedSea) to align representations across multiple parties in the feature space. FedSea uses a domain adversarial learning framework consisting of an affine-transform-based generator and a gradient-reversal-based client discriminator to perform IID transformation and reduce data source distinguishability, respectively. An attention-based mask module and a feature IID confidence quantification method are introduced to effectively address the diverse feature non-IID levels across multimodal data. Comprehensive experiments are conducted on three widely-used public datasets and one large-scale industrial dataset, showing FedSea has: 1) better performance than state-of-the-art FL methods on both multimodal and single-modal datasets; 2) superior feature alignment ability on non-IID datasets, and 3) good model interpretability.

A graph is an abstract model that represents relations among entities, for example, the interactions between characters in a novel. A background story endows entities and relations with real-world meanings and describes the semantics and context of the abstract model, for example, the actual story that the novel presents. Considering practical experience and prior research, human viewers who are familiar with the background story of a graph and those who do not know the background story may perceive the same graph differently. However, no previous research has adequately addressed this problem. This research article thus presents an evaluation that investigated the effects of background stories on graph perception. Three hypotheses that focused on the role of visual focus areas, graph structure identification, and mental model formation on graph perception were formulated and guided three controlled experiments that evaluated the hypotheses using real-world graphs with background stories. An analysis of the resulting experimental data, which compared the performance of participants who read and did not read the background stories, obtained a set of instructive findings. First, having knowledge about a graph's background story influences participants' focus areas during interactive graph explorations. Second, such knowledge significantly affects one's ability to identify community structures but not high degree and bridge structures. Third, this knowledge influences graph recognition under blurred visual conditions. These findings can bring new considerations to the design of storytelling visualizations and interactive graph explorations.

The sixth-generation (6G) communication requires supporting massive Internet of Things (IoT) devices and extremely differentiated IoT applications for the air–space–ground integrated network. Relying on the aerial superiority, unmanned aerial vehicle (UAV) is capable of acting as an aerial base station (BS) and supporting IoT deployment in remote and disaster areas. A UAV-supported clustered nonorthogonal multiple access (C-NOMA) system is put forward in this article. Specifically, the UAV provides services to IoT terminals as an aerial BS based on the wireless-powered communication (WPC) technique. According to this system, we propose a synergetic scheme for UAV trajectory planning and subslot allocation. Our goal is to maximize the uplink average achievable sum rate of IoT terminals by synergistically planning UAV trajectory and subslot duration, while guaranteeing the uplink achievable sum rate and the UAV mobility constraints. As the formulated problem suffers nonconvexity and complication, an efficient iterative algorithm is proposed to address it. First, for fixed UAV trajectory, all the terminals are clustered and a subslot allocation algorithm based on the Lagrange multiplier and bisection method is proposed. Then, for a fixed clustering state and subslot duration, we optimize the UAV trajectory. Finally, we solve these two subproblems alternatively until the objective function converges. The effectiveness of the proposed scheme in the UAV-supported C-NOMA system is verified by the numerical results.