Position and Orientation Independent Wireless Gesture Recognition

Gesture recognition through wireless communication signal is a new approach for human-computer interaction, and is an important topic for integrated sensing and communication designs in next-generation mobile communication system. Wireless gesture recognition is mainly realized by analyzing the change of propagation channel, especially the phase change of reflection path. However, the law of phase change is closely related to the position and orientation of the user, as well as the positions of transmitter and receiver. The gesture recognition system trained in a given environment usually does not work well in a new environment. In this paper we propose a position and orientation independent gesture recognition method, where the phase change under any position configuration is transformed to the gesture trajectory in an invariant body coordinate system. The transformation matrix implicitly embodies the impact of position and orientation, and is estimated by a preamble gesture each time reaching a new environment. The method is verified in a prototype system, and the recognition accuracies under different position configurations are tested.