A Comparative Study of JADE and SAGE Algorithms for Joint Multipath Parameter Estimation

In this paper, we investigate the performance of two popular multi-path parameter estimation schemes, eigen-decomposition based SI-JADE and maximum-likelihood based SAGE. An investigation of errors in the estimation of directions (DOA) and times (TOA) of arrival is presented using Monte Carlo simulations. We demonstrate that a significant improvement in DOA estimation is obtained as expected using both methods when the antenna size is increased, but there is also a smaller improvement in TOA estimation. We also show that the SAGE algorithm is more robust against under-estimating the number of signal paths in a multi-path environment.