<p>In addition to traditional visual appearance recognition, various other methods are employed for human identity authentication, including fingerprint detection, blood analysis, voice recognition, and gait recognition. However, most contemporary approaches for ascertaining individual identities through the acquisition of personal body data rely on specialized and often expensive equipment, such as cameras and DNA analyzers, and there is a risk of leaking personal body characteristics and other privacy data. Given the widespread availability and affordability of wireless devices, particularly WiFi routers, this paper introduces Wi-GWIR, a pioneering framework that combines off-the-shelf WiFi devices with principles of human gait recognition for indoor human identity recognition. Wi-GWIR addresses the challenges that conventional camera-based facial recognition encounters, such as lighting conditions, viewing angles, obstructions, hardware prerequisites, and privacy concerns, thus demonstrating superior versatility. Wi-GWIR is founded on a novel Channel State Information analysis model, addressing several technical hurdles. Firstly, it captures CSI channel information influenced by individuals’ walking, ensuring a sufficient pool of gait information samples. Secondly, it calculates the instantaneous walking speed of individuals by analyzing the auto-correlation function of the CSI, thereby precisely segmenting the CSI channel waveforms representing distinct individuals. This process effectively extracts the gait feature waveforms from the overall signal. Finally, Wi-GWIR employs a uniquely designed algorithm combined with Dynamic Time Warping to precisely categorize and identify gait shape features. Experimental results affirm that Wi-GWIR can promptly detect the presence of individuals and accurately identify their identities when they move within the experimental space, all without the need for cameras or additional equipment. When tested on 10 users, the average recognition accuracy reached 94.31%.</p>