Blind channel identification: Subspace tracking method without rank estimation

Subspace (SS) methods are an effective approach for blind channel identific ation. However, thses methods also have two major disadvantages: 1) They re quire accurate channel length estimation and/or rank estimation of the corr elation matrix, which is difficult with noisy channels, and 2) they require a large amount of computation for the singular value decomposition (SVD), which makes it inconvenient for adaptive implementation. Although many adap tive subspace tracking algorithms can be applied, the computational complex ity is still O(m(3)), where m is the data vector length. In this paper, we introduce new recursive subspace algorithms using ULV updating and successi ve cancellation techniques. The new algorithms do not need to estimate the rank of the correlation matrix. Furthermore, the channel length can be over estimated initially and be recovered at the end by a successive cancellatio n procedure, which leads to more convenient implementations. The adaptive a lgorithm has computations Of O(m(2)) in each recursion. The new methods can be applied to either the single user or the multiuser cases. Simulations d emonstrate their good performance.