Multiple Tags Ambient Backscatter Communications Systems Using Beamforming and Machine Learning

We investigate novel signal processing techniques tailored to improve the performance of an ambient backscatter based multiple tags communication system. We consider the amalgamation of beamforming (BF) and machine learning (Mal) techniques to optimize signal reception under various decoding options. The convention is to decode tag signals modulated using the RF source transmissions but it is susceptible to interference in the presence of multiple tags. We show that our proposed techniques have the ability to outperform the traditional Minimum Mean Square Error (MMSE) detection method, the MaL enhancement of MMSE detection method, while able to decode multiple tag signals simultaneously as opposed to the traditional one tag at a time scheme.