Multi-hop ML based delay and angle estimation for multipath wideband FH signals

A multi-hop ML (MHML) algorithm for the joint angle and delay estimation of a wideband frequency hopping (FH) source is proposed to estimate multipath delays and directions in the common scenario where coherent multipath propagation exists. By using a multi-hop signal model which incorporates multipath propagation, the likelihood function of the received signal vectors in different frequency hops is formulated. Based on the Newton method, an iterative algorithm that exploits the space-frequency information in the received signals for adaptively estimating the multipath delays and directions is then derived. The use of space-frequency information enables the algorithm to have the capability that the number of the multipath components is not limited by the number of antennas. Simulations are performed to demonstrate the performance and tracking capability of the algorithm.