The past decade has demonstrated the potential of human activity recognition (HAR) with WiFi signals owing to non-invasiveness and ubiquity. Previous research has largely concentrated on enhancing precision through sophisticated models. However, the complexity of recognition tasks has been largely neglected. Thus, the performance of the HAR system is markedly diminished when tasked with increasing complexities, such as a larger classification number, the confusion of similar actions, and signal distortion To address this issue, we eliminated conventional convolutional and recurrent backbones and proposed WiTransformer, a novel tactic based on pure Transformers. Nevertheless, Transformer-like models are typically suited to large-scale datasets as pretraining models, according to the experience of the Vision Transformer. Therefore, we adopted the Body-coordinate Velocity Profile, a cross-domain WiFi signal feature derived from the channel state information, to reduce the threshold of the Transformers. Based on this, we propose two modified transformer architectures, united spatiotemporal Transformer (UST) and separated spatiotemporal Transformer (SST) to realize WiFi-based human gesture recognition models with task robustness. SST intuitively extracts spatial and temporal data features using two encoders, respectively. By contrast, UST can extract the same three-dimensional features with only a one-dimensional encoder, owing to its well-designed structure. We evaluated SST and UST on four designed task datasets (TDSs) with varying task complexities. The experimental results demonstrate that UST has achieved recognition accuracy of 86.16% on the most complex task dataset TDSs-22, outperforming the other popular backbones. Simultaneously, the accuracy decreases by at most 3.18% when the task complexity increases from TDSs-6 to TDSs-22, which is 0.14–0.2 times that of others. However, as predicted and analyzed, SST fails because of excessive lack of inductive bias and the limited scale of the training data.

With the compelling popularity of integrated sensing and communication (ISAC), Wi-Fi sensing has drawn increasing attention in recent years. Starting from 2010, Wi-Fi channel state information (CSI)-based wireless sensing has enabled various exciting applications such as indoor localization, target imaging, activity recognition, and vital sign monitoring. In this paper, we retrospect the latest achievements of Wi-Fi sensing using commodity-off-the-shelf (COTS) devices from the past 5 years in detail. Specifically, this paper first presents the background of the CSI signal and related sensing models. Then, recent studies are categorized from two perspectives, i.e., according to their application scenario diversity and the corresponding sensing methodology difference, respectively. Next, this paper points out the challenges faced by Wi-Fi sensing, including domain dependency and sensing range limitation. Finally, three imperative research directions are highlighted, which are critical for realizing more ubiquitous and practical Wi-Fi sensing in real-life applications.

Cross-view geo-localization is a task of matching the same geographic image from different views, e.g., drone and satellite. Due to its GPS-free advantage, cross-view geo-localization is gaining increasing research interest, especially in drone-based localization and navigation applications. In order to guarantee system accuracy, existing methods mainly focused on image augmentation and denoising while still facing performance degradation when extreme weather conditions are considered. In this paper, we propose a robust end-to-end image retrieval framework, AGEN, serving for cross-view geo-localization under extreme weather conditions. Inspired by the strengths of the DINOv2 network, particularly its strong performance in global feature extraction, while acknowledging its limitations in capturing fine-grained details, we integrate the DINOv2 network with the Local Pattern Network (LPN) algorithm module to extract valuable classification features more efficiently. Additionally, to further enhance model robustness, we innovatively introduce an Adaptive Error Control (AEC) module based on fuzzy control to optimize the loss function dynamically. Specifically, by adjusting loss weights adaptively, the AEC module allows the model to better handle complex and challenging scenarios. Experimental results demonstrate that AGEN achieves a Recall@1 accuracy of 91.71% on the University160k-WX dataset under extreme weather conditions. Through extensive experiments on two well-known public datasets, i.e., University-1652 and SUES-200, AGEN achieves state-of-the-art Recall@1 accuracy in both drone-view target localization tasks and drone navigation tasks, outperforming existing models. In particular, on the University-1652 dataset, AGEN reaches 95.43% Recall@1 in the drone-view target localization task, showcasing its superior capability in handling challenging scenarios.

Cross-view geo-localization is a task of matching the same geographic image from differerent views, e.g., drone and satellite. Due to its GPS-free advantage, cross-view geo-localization is gaining increasing research interest, especially in drone-based localization and navigation applications. In order to guarantee system robustness, existing methods mainly focused on image augmentation and denoising, while facing performace degradation when extreme weather considered. In this paper, we propose an end-to-end image retrieval framework, WELN. By integrating the advanced EVA02 netwotk and LPN algorithm, WELN can extract valuable classification features more efficiently even under extreme weather conditions. Additionally, to enhance model robustness, we expand the University-1652 dataset with nine different weather conditions added. Our method achieves state-of-the-art Recall@1 accuracy on University-1652 dataset, with 92.87% for drone-view target localization task and 93.46% for drone navigation task. Besides, we gain the fourth place in the ACMMM24 Multimedia Drone-Satellite Matching Challenge. Our code will be open sourced at https://github.com/koorter/WELN.

With the compelling popularity of integrated sensing and communication (ISAC), WiFi sensing has drawn increasing attention in recent years. Starting from 2010, WiFi CSI-based wireless sensing has enabled various exciting applications such as indoor localization, target imaging, activity recognition and vital sign monitoring. In this paper, we retrospect the up-to-date achievements of WiFi sensing using commodity-off-the-shelf (COTS) devices over the past 5 years in detail. Specifically, this paper first presents the background of CSI signal and related sensing models. Then, recent researches are categorized from two perspectives, namely according to their application scenario diversity and corresponding sensing methodology difference respectively. Next, this paper points out the challenges faced by WiFi sensing including domain dependency and sensing range limitation. Finally, three imperative research directions are highlighted, which are critical for realizing more ubiquitous and practical WiFi sensing in real-life applications.