Off-grid compressed sensing for WiFi-based passive radar

Abstract

WiFi signals have been widely used in short-distance wireless communication and thus become a promising option for passive radar applications, where sources of opportunity are exploited in a multi-static system. In the processing of passive radar signals, discrete compressed sensing (CS) techniques have been proved in previous research to be capable of producing better estimation than traditional methods based on correlation and side slope removal. But unstable performance and the need of data-association become remaining problems while the resolution achieved still leaves much to be desired. We introduce an off-grid CS scheme to WiFi-based radar and propose a multi-receiver (SIMO) model, where the positions and speeds of planar objects are directly recovered, to deal with the problems mentioned above, in which case discrete CS requires excessively large space for the storage of the measurement matrix. The simulation result shows its power in overcoming the previous obstacles as well as reaching higher resolution and precision.