Abstract: A new structure of nonlinear equalizer-neural Chebyshev orthogonal polynomial adaptive equalizerwasproposed. It is composed ofa nonlinearneuraladaptive filterand aChebyshev orthogonal polynomial filter. The corresponding adaptive algorithm was derived based on the steepest descent method. Simulation results show that the proposed equalizer has less mean square error (MSE ) comparedwith the neural adaptive equalizer and the SVE ( second-order volterra equalizer ) in linear channe,l and that itcan achieve similargood performances as the 3-layered neuralnetwork trainedwith BP algorithm in the nonlinear channelwith a simpler structure.