Performance Evaluation of Combined Methods for the Estimation of Fading Coefficients of Uncorrelated Signal Sources in Multipath Propagation

Abstract

In this study we consider the challenge of estimation of fading coefficients for uncorrelated sources in the presence of multipath propagation. The combination of Joint Approximate Diagonalization of Eigenmatrices (JADE) and some well-known high-resolution algorithms such as Multiple Signal Classification (MUSIC) Minimum norm (Min-norm) and Modified Forward Backward Linear Prediction (MFBLP) are utilized to estimate the fading coefficients of a challenging environment containing many correlated signals. The results are compared with Estimation of Signal Parameters by Rotational Invariance Techniques (ESPRIT) based method in the literature. Several independent trials are performed and it is shown that the combined method of JADE and MFBLP has satisfactory fading coefficient performance even at very low signal-to-noise (SNR) levels and low number of antenna elements and data sample.