Renato Candido, Magno T. M. Silva, Vítor H. Nascimento. Uma combinação afim de dois equalizadores autodidatas adaptados com o CMA. In: Simpósio Brasileiro de Telecomunicações – SBrT’2008, 2008, Rio de Janeiro. Anais do SBrT 2008, 2008, pp.1–6 (available only in portuguese).
Recently, an affine combination of two least mean-square (LMS) adaptive filters was proposed and its transient performance analyzed. This method combines linearly the outputs of two LMS filters operating in parallel with different step-sizes. The purpose of the combination is to obtain an LMS adaptive filter with fast convergence and reduced steady-state mean-square error. Since the mixing parameter is not restricted to the interval [0, 1], this method is a generalization of the convex combination of two LMS filters. In order to extend this scheme to blind equalization, we propose, in this paper, an affine combination of two constant modulus algorithms (CMA). We obtain analytical expressions for the mixing parameter and for the excess mean-square error at steady-state for stationary and non-stationary environments. Furthermore, we propose two stochastic algorithms to adapt the mixing parameter. Good agreement between analytical and simulation results is observed.
Blind equalization, Affine combination, Constant modulus algorithm, Steady-state analysis.