Subspace methods in adaptive filtering and system identification

The purpose of the proposed research is to develop a class of subspace system identification methods based on RLS lattice filters. This provides an alternative way of doing subspace methods, which leads to faster algorithms suitable for real-time system identification. The comparisons with some existing subspace methods will be made in order to get a picture of the comparisons among these methods. Parallel computing and real-time implementation will also be discussed in the research.