An overview of maximum-likelihood based algorithms for estimating multipath parameters

Recently the European research trend has shown an increased interest on the use of maximum-likelihood based algorithms, e.g. the Space-Alternating Generalised Expectation-maximisation (SAGE) algorithm, to estimate multipath parameters from raw measurement data. This process can require considerable processing time and resources, especially when dealing with vast multi-dimensional measurement databases. With the aim of achieving significant timesaving, and reducing both the memory utilisation and processing power of computing resources, different versions of maximum-likelihood based algorithms have been developed. This paper provides a general overview of these algorithms based on different implementation methodologies that can achieve the above objectives successfully, subject to some prerequisites. A number of results based on numerical simulations and real measurement data is also presented