A recursive algorithm for blind identificaiton and equalization based on second-order statistics
Feng Da · Journal of Xidian University · 2003
A new recursive algorithm for blind identificaiton and equalization is proposed, which exploits the cyclostationarity of oversampled communication signals to achieve identification and equalization of possibly nonminimum phase channels. Compared with other blind equalization algorithms which only exploit one or two eigenmatrices, this algorithm utilizes a set of eigenmatrices for estimation so that the performance is improved. By solving the cost function to obtain th best solution, the recursive algorithm is given which can obtain all the eigenvectors one by one. Further, the analytic solution is provided. Simulations show that the algorithm here is effective even under the condition of the low signaltonoise ratio (SNR) which can overcome the shortcomings of conventional algorithms. In addition, our algorithm has low complexity of computation.