Conventional microwave orientation methods based on the array antenna require high element isolation in most cases. In our work, a neural network-based microwave orientation method is proposed to avoid the interferences caused by the mutual couplings between elements. An end-to-end orientation prototype with a strongly coupled array, phase detectors and a non-distortion CMOS video camera is designed and the real-time target source location can be determined by the back-propagation neural network (BPNN) based on the information of phase differences. Simulation and experimental results verify the effectiveness of the proposed orientation method.
|State||Published - Jan 2020|