On December 10, the Network and Cloud Computing Team from Tianjin University made significant strides in the realm of wireless sensing. Leveraging the common Wi-Fi signals found in homes, the team can decipher alterations in human activities to discern individuals' needs. This enables smart homes to offer tailored services without the necessity for vocal instructions. The pertinent findings have been published in the esteemed international journal IMWUT.

Presently, smart homes predominantly depend on user inputs or specialized sensors, posing challenges in seamlessly and consistently comprehending user requirements. To overcome this, the team has devised a novel, high-precision sensing application that eliminates the need for users to don any devices. By scrutinizing the minute fluctuations in Wi-Fi signals induced by human actions, it can accurately gauge a person's whereabouts, condition, and conduct, and subsequently guide smart home services accordingly.

Transitioning laboratory applications to real-world homes necessitates overcoming two primary hurdles: firstly, the cumbersome nature of system deployment, and secondly, the imprecise sensing resulting from the intricate home environment. To surmount the "deployment difficulty," the team harnessed the routine cleaning and mapping capabilities of a robotic vacuum cleaner. This approach enabled the simultaneous creation of a physical space map of the home and a Wi-Fi signal map, automatically pinpointing and recording device locations with an impressive precision of 0.1 meters. Users are merely required to initiate a single cleaning session to complete the system's initialization.

To combat the issue of "low accuracy," the team discarded traditional idealized presumptions and, for the first time, formulated a new theoretical model specifically designed for real-world home settings. This model comprehends the propagation dynamics of Wi-Fi signals in convoluted environments, thereby achieving stable and dependable high-precision sensing.

This accomplishment was co-guided by Associate Professor Tong Xinyu and Professor Li Keqiu from the School of Computer Science and Technology at Tianjin University, along with Assistant Researcher Chen Sheng from the School of Cybersecurity. Doctoral candidates Tan Renrui and Meng Xuanqi are credited as the primary authors. Tong Xinyu remarked that this achievement paves the way for wireless sensing technology to permeate countless households, offering pivotal technological backing for the construction of intelligent and considerate living spaces.