Abstract:

A novel constant modulus blind equalization algorithm (CMA) with accelerated convergence is proposed. It is based on the concepts of setmembership and datareusing. Applying the idea of affine projection, the algorithm includes many constraint sets, whose number is equal to the length of datareusing. Compared with basic CMA, the convergence rate of the proposed algorithm is significantly faster; meanwhile, the excess mean square error can be maintained in a relatively low level because of the idea of setmembership. Deriving details of the algorithm are presented and its performance is analyzed theoretically. The convergence rate and stational error of the proposed algorithm are compared with those of other constant modulus algorithms by computer simulation.

Key words: communications, set membership, affine projection, data reusing, CMA