WiFi-based identity recognition differs from traditional identification technologies as it is not limited by lighting conditions and does not require dense, specialized sensors or wearable devices. This makes it valuable in modern human–machine interactions. However, the diversity of real-world environmental conditions substantially limits the application of existing WiFi-based identity recognition algorithms, particularly when applied across different environments. As a solution, we introduce the incremental wavelet capsule (IWC) model, which combines a newly designed wavelet convolution layer with a capsule network to accelerate precise feature extraction. We adopt a hybrid incremental learning strategy, solving the catastrophic forgetting1problem in cross-environment tasks and enabling the model to adapt to new environments in the data stream without forgetting the original environment. Furthermore, we developed a customized data augmentation method for WiFi signals, enhancing the model’s adaptability and stability across various environments. Experimental results show that the IWC model achieves an average recognition accuracy of 97.36% across five different environments and maintains an accuracy of 91.5% even when only 5% of the training data from a new environment is used. These findings demonstrate the model’s robust performance and practicality in cross-environment scenarios.

WiFi-based human activity recognition (HAR) plays an essential role in various applications such as security surveillance, health monitoring, and smart home. Existing HAR methods, though yielding promising performance in indoor scenarios, highly depend on a massive labeled dataset for training which is extremely difficult to acquire in practical applications. In this paper, we present an automatic data labeling and HAR system, termed AutoDLAR. Taking a semi-supervised cross-modal learning framework with a hybrid loss function as the core, AutoDLAR transfers rich visual information to automatically label WiFi signals for WiFi-based HAR. Specifically, we devise a lightweight and multi-view WiFi sensing model with a parallel feature embedding method to accurately identify activities and accelerate recognition speed. Then, we exploit the video data to fine-tune a well-established visual HAR model, generating effective pseudo-labels for guiding the WiFi model’s training. We also build a synchronized Video-WiFi dataset with seven types of human activities under different scenarios to enable training and validating the semi-supervised HAR system. Extensive experiments on our collected activity dataset and the emotion recognition benchmark demonstrate that AutoDLAR attains an average accuracy of over 95.89% without manual labeling and only spends the inference time of 3.35 ms, outperforming the state-of-the-art (SOTA) methods.

Gait is regarded as a unique feature for identifying people, and gait recognition is the basis of various customized services of the IoT. Unlike traditional techniques for identifying people, the Wi-Fi-based technique is unconstrained by illumination conditions and such that it eliminates the need for dense, specialized sensors and wearable devices. Although deep learning-based sensing models are conducive to the development of Wi-Fi-based identification, the latter technique relies on a large amount of data and requires a long training time, where this limits the scope of its use for identifying people. In this study, we propose a Wi-Fi sensing model called Wave-CapNet for human identification. We use data processing to eliminate errors in the raw data so that the model can extract the characteristics in channel state information (CSI). We also design a dedicated adaptive wavelet neural network to extract representative features from Wi-Fi signals with only a few epochs of training and a small number of parameters. Experiments show that it can identify human gait with an average accuracy of 99%. Moreover, it can achieve an average accuracy of 95% by using only 10% of the data and fewer than five epochs and outperforms state-of-the-art (SOTA) methods.

WiFi-based Human Activity Recognition (HAR) faces challenges in achieving widespread deployment due to its reliance on massive data and limited scalability. However, the emergence of Few-Shot Learning (FSL) provides opportunities to address this issue. In this paper, we propose a cross-modal meta-learning approach based on Model-Agnostic Meta-Learning (MAML) to enable few-shot WiFi-based HAR. The hypothesis is that models can learn "learning methods" from thousands of diverse image classification tasks and apply them to WiFi-based HAR. By solely leveraging public image and WiFi signal datasets, the proposed approach trains a model capable of recognizing previously unseen activities with only 5 samples per class, achieving an average accuracy of 88.5% over thousands of tests.

WiFi-based human activity recognition has been widely used in many fields such as health diagnosis, intrusion detection and smart home. Most existing recognition methods can achieve a satisfying accuracy only in one domain, but low accuracy occurs when models are trained in source domain but are used in target domain. Meanwhile, considering finetuning network directly is impossible or easy to overfit with limited labeled target data, transfer learning based methods with domain adaptive layers are proposed to solve above problems but just aligning marginal distribution, which may lose massive fine-grained features. Based on this, we present an end-to-end deep subdomain adaptive network based activities recognition (DSANAR) using Channel State Information (CSI) that aligns marginal and matches conditional distribution simultaneously for more fine-grained features in each category of relevant subdomains based on a local maximum mean discrepancy (LMMD). Besides, by using a joint cross-entropy and an adaptive loss as training loss, DSANAR outperforms other state-of-art methods on an autonomous dataset with average 95.6% cross-domain accuracy.

Wireless networks that can exchange any type of data are vulnerable to multiple intrusions and increase potential security risks, so the design of an Intrusion Detection and Prevention System (IDPS) that analyzes the packet features and detects different intruders (i.e., the types of attack) is necessary. Whereas, the existence of redundant and irrelevant features hinders the potential of IDPS. In this paper, we propose TL-IDPS, a Two-Level classification IDPS of wireless network based on optimized features. In the phase of intrusion detection, one-hot method, normalization and correlation estimation are used to mitigate the redundant features. Then, the fuzzy membership function with cuttlefish algorithm maps and consolidates the extracted features and selects optimal features. Based on the optimal features, Di-distance k-nearest neighbor (K-NN) as the first level classify the intruder or non-intruder. Further the type of intruder is identified by deep Q-network. From the result of detected intruders, the further arrival of those intruders is prevented. Experimental results conducted from multiple evaluation metrics using the UNSW-NB15 dataset prove that our proposed TL-IDPS is more effective than existing IDPS methods.

Body movement is one of the most important indicators of sleep quality for elderly people living alone. Body movement is crucial for sleep staging and can be combined with other indicators, such as breathing and heart rate to monitor sleep quality. Nevertheless, traditional sleep monitoring methods are inconvenient and may invade users' privacy. To solve these problems, we propose a contactless body movement recognition (CBMR) method via WiFi signals. First, CBMR uses commercial off-the-shelf WiFi devices to collect channel state information (CSI) data of body movement and segment the CSI data by sliding window. Then, the context information of the segmented CSI data is learned by a bidirectional recurrent neural network (Bi-RNN). Bi-RNN can fuse the forward and backward propagation information at some point, and input it into a deeper independently recurrent neural network (IndRNN) with residual mechanism to extract the deeper features and capture the time dependencies of CSI data. Finally, the type of body movement can be recognized and classified by the softmax function. CBMR can effectively reduce data preprocessing and the delay caused by manually extracting features. The results of an experiment conducted on a complex body movement data set show that our method gives desirable performance and achieves an average accuracy of greater than 93.5%, which implies a prospect application of CBMR.