In this overview paper, data-driven learning model-based cooperative localization and location data processing are considered, in line with the emerging machine learning and big data methods. We first review (1) state-of-the-art algorithms in the context of federated learning, (2) two widely used learning models, namely the deep neural network model and the Gaussian process model, and (3) various distributed model hyper-parameter optimization schemes. Then, we demonstrate various practical use cases that are summarized from a mixture of standard, newly published, and unpublished works, which cover a broad range of location services, including collaborative static localization/fingerprinting, indoor target tracking, outdoor navigation using low-sampling GPS, and spatio-temporal wireless traffic data modeling and prediction. Experimental results show that near centralized data fitting- and prediction performance can be achieved by a set of collaborative mobile users running distributed algorithms. All the surveyed use cases fall under our newly proposed \textit{Federated Localization (FedLoc)} framework, which targets on collaboratively building accurate location services without sacrificing user privacy, in particular, sensitive information related to their geographical trajectories. Future research directions are also discussed at the end of this paper.

The recent development of integrated sensing and communications (ISAC) technology offers new opportunities to meet high-throughput and low-latency communication as well as high-resolution localization requirements in vehicular networks. However, considering the limited transmit power of the road site units (RSUs) and the relatively small radar cross section (RCS) of vehicles with random reflection coefficients, the power of echo signals may be too weak to be utilized for effective target detection and tracking. Moreover, high-frequency signals usually suffer from large fading loss when penetrating vehicles, which seriously degrades the communication service quality of users inside vehicles. To handle this issue, we propose a novel sensing-assisted communication scheme by employing an intelligent omni-surface (IOS) on the surface of vehicles to enhance both sensing and communication (S&C) performance. To this end, we first propose a two-stage ISAC protocol, including the joint S&C stage and the communication-only stage, to fulfil more efficient communication performance improvements benefited from sensing. The achievable communication rate maximization problem is formulated by jointly optimizing the transmit beamforming, the IOS phase shifts, and the duration of the joint S&C stage. However, solving this ISAC optimization problem is highly non-trivial since inaccurate estimation and measurement information renders the achievable rate lack of closed-form expression. To handle this issue, we first derive a closed-form expression of the approximate achievable rate under uncertain location information, and then unveil a sufficient and necessary condition for the existence of the joint S&C stage to offer useful insights for practical system design. Moreover, two typical scenarios including interference-limited and noise-limited cases are analyzed to provide a performance bound and a low-complexity algorithm for the considered systems. Finally, simulation results demonstrate the effectiveness of the proposed sensing-assisted communication scheme in achieving a higher achievable rate with lower transmit power requirements.

Due to the features of on-demand deployment and flexible observation, unmanned aerial vehicles (UAVs) are promising for serving as the next-generation aerial sensors by using their onboard sensing devices. Compared to a single UAV with limited sensing coverage and communication capability, multi-UAV cooperation is able to realize more effective sensing and transmission (S&T) services, and delivers the sensory data to the control center more efficiently for further analysis. Nevertheless, most existing works on multi-UAV sensing mainly focus on mutually exclusive task allocation and independent data transmission, which did not fully exploit the benefit of multi-UAV sensing and communication. Motivated by this, we propose a novel multi-UAV cooperative S&T scheme with overlapped sensing task allocation. Although overlapped task allocation may sound counter-intuitive, it can actually foster cooperative transmission among multiple UAVs through a virtual multi-antenna system and thus reduce the overall sensing mission completion time. To obtain the optimal task allocation and transmit power of the proposed scheme, a mission completion time minimization problem is formulated. To solve this problem, a condition that specifies whether it is necessary for the UAVs to perform overlapped sensing is derived. For the cases of overlapped sensing, this time minimization problem is transformed into a monotonic optimization and is solved by the generic Polyblock algorithm. To efficiently evaluate the mission completion time in each iteration of the Polyblock algorithm, new auxiliary variables are introduced to decouple the otherwise sophisticated joint optimization of transmission time and power. While for the degenerated case of non-overlapped sensing, the closed-form expression of the optimal transmission time is derived, which provides insights into the optimal solution and facilitates the design of an efficient double-loop binary search algorithm to optimally solve the degenerated problem. Finally, simulation results demonstrate that the proposed scheme significantly reduces the mission completion time over benchmark schemes.

Recently, many satellites are being launched for providing global internet broadband service to individual consumers. Since satellites and users could move separately, providing seamless connectivity has become one of the most important tasks for mobile satellite networks. Current handover methods are based on either signal strength or service time, however, due to the randomness of user terminal (UT) arrivals and the unbalanced traffic distribution for high-mobility satellite networks, the success rate can hardly be guaranteed. To this end, we propose a QoE-driven intelligent handover mechanism for user-centric mobile satellite networks, through which the access satellites can be selected by predicted service time and communication channel resources. Accordingly, to ensure the service duration, a spatial relationship coupling model is proposed to predict relative motion pattern between UT and satellites; To improve the effectiveness of handover, an available channel estimation model is then developed based on the mobility pattern of adjacent satellites. Finally, reinforcement learning is adopted to maximize the UT's Quality of Experience (QoE) through predicted handover factors. Simulation results show that the proposed handover mechanism outperforms the conventional methods in terms of handover times, handover success rate, and end-to-end delay.

As the application scenarios of remote sensing imagery (RSI) become richer, the task of ship detection from an overhead perspective is of great significance. Compared with traditional methods, the use of deep learning ideas has more prospects. However, the Convolutional Neural Network (CNN) has poor resistance to sample differences in detection tasks, and the huge differences in the image environment, background, and quality of RSIs affect the performance for target detection tasks; on the other hand, upsampling or pooling operations result in the loss of detailed information in the features, and the CNN with outstanding results are often accompanied by a high computation and a large amount of memory storage. Considering the characteristics of ship targets in RSIs, this study proposes a detection framework combining an image enhancement module with a dense feature reuse module: (1) drawing on the ideas of the generative adversarial network (GAN), we designed an image enhancement module driven by object characteristics, which improves the quality of the ship target in the images while augmenting the training set; (2) the intensive feature extraction module was designed to integrate low-level location information and high-level semantic information of different resolutions while minimizing the computation, which can improve the efficiency of feature reuse in the network; (3) we introduced the receptive field expansion module to obtain a wider range of deep semantic information and enhance the ability to extract features of targets were at different sizes. Experiments were carried out on two types of ship datasets, optical RSI and Synthetic Aperture Radar (SAR) images. The proposed framework was implemented on classic detection networks such as You Only Look Once (YOLO) and Mask-RCNN. The experimental results verify the effectiveness of the proposed method.