Spatial algorithms for blind channel estimation

Of the different information sources used by blind algorithms, explicit use of finite alphabet (FA) data is more recent than either the statistical data embedded at the source or the algebraic structure present in the channel. In channels that can be modeled using finite impulse response (FIR) structures, the FA property results in the received vector set being clustered around theoretical centers. These centers are the result of the convolution of the channel matrix with a given transmitter symbol constellation. Defined as spatial structure in this paper, they provide sufficient information for blind estimation of channel coefficients. Here, we introduce two spatial tools, the primary and secondary clustering algorithms capable of processing the information structures described above. Then, using these two tools, we present the channel estimation by difference set (CEDS) algorithm for the estimating channel impulse response coefficients.