ANN Based Hybrid RSSI-TDOA DOA Estimation and Beamforming

Abstract

AbstractSmart antenna system comprises of an antenna array interfaced with sophisticated electronic signal processing system that performs direction of arrival (DOA) estimation of the received signals and beamforming to form beams in the directions of the signals of interest (SOI) and nulls or no beams in the directions of the unwanted interferers. In this paper, the problem of DOA evaluation using uniform linear antenna arrays, with reduced computational complexity and cost is modeled by artificial neural network (ANN) and is followed by beamforming. The work consists of two main parts of DOA evaluation and beamforming for system of eight element Uniform Linear Array (ULA). This paper presents a hardware system incorporating three transmitters and eight element ULA interfaced with the hardware system. To perform DOA estimation artificial neural network optimized hybrid RSSI-TDOA model is devised to optimize the angles of arrival. Further, these DOA values are received and processed by RLS beamforming algorithm implemented in MATLAB. The MATLAB simulation results show that beams are formed in the directions of desired signals of interest and a perfect null is formed in the direction of the unwanted interferer and get steered dynamically as per the corresponding positions of the three transmitters in real time.KeywordsDOA (Direction of Arrival)Uniform Linear Array (ULA)Signals of Interest (SOI)BeamformingArtificial Neural Network (ANN)RSSI (Received Signal Strength Indicator)TDOA (Time Difference of Arrival)RLS (Recursive Least Square)