Device-free motion tracking in RF sensor networks using received signal strength（ RSS） is an emerging technology. Since the motion scene to be reconstructed can often be assumed sparse

i. e

it consists only a few targets

the compressed sensing（ CS） technology can be applied. Present target tracking as a signal reconstruction problem of CS and present greedy detection estimation（ GDE） algorithm to resolve target location. Meanwhile

design a feedback tracking method which utilizes prior location informations of targets to reduce amount of measurements in signal reconstruction. Through experiment

it verifies that the GDE can obtain good result of signal construction

in addition

it is more accurate in target tracking application.