Using RF signals for wireless sensing has gained increasing attention . However, due to the unwanted multi-path fading in uncontrollable radio environments, the accuracy of RF sensing is limited . We propose a deep reinforcementlearning algorithm to jointly compute the optimal beamformer patterns and themapping of the received signals to the sensed outcome . Simulation results verify the effectiveness of the proposed algorithm and show how the sizes of the metasurface and the targetspace influence the sensing accuracy of the sensors . The authors also suggest a deep-learning algorithm for the optimization problem for minimizing thecross-entropy loss of the sensing outcome, and propose a new approach to solving the problem of the problem. The authors conclude that the algorithm is a good candidate to solve the

Author(s) : Jingzhi Hu, Hongliang Zhang, Kaigui Bian, Marco Di Renzo, Zhu Han, Lingyang Song

Links : PDF - Abstract

Code :

https://github.com/google/dopamine


Coursera

Keywords : sensing - algorithm - problem - rf - deep -

Leave a Reply

Your email address will not be published. Required fields are marked *