Shell partition-based joint blind equalization for QAM systems

For quadrature amplitude modulation (QAM) systems, the joint blind equalization (JBE) algorithm, concatenation of the constant modulus algorithm (CMA) and the decision-directed (DD) algorithm, has been widely used. After initial equalization is accomplished by the CMA, the DD algorithm follows for fine equalization. This paper proposes a shell partition-based CMA (SPCMA), a multi-modulus CMA based on the equalizer output power, in which decision boundaries are obtained by maximum likelihood (ML) estimation. Unlike the DD algorithm, the SPCMA is independent of the carrier phase error and yields better results than the CMA, showing the optimum performance for QAM constellations. The proposed SPCMA is applied to the JBE algorithm, resulting in the JBE-SPCMA, in which the SPCMA follows the CMA. To compensate for the carrier phase error, the shell partition-based hybrid algorithm (SPHA) combining the SPCMA and a shell partition-based generalized Sato (1975) algorithm (SPGSA) is proposed and it is combined with the JBE, generating the JBE-SPHA. Computer simulation results for 16- and 64-QAM show the effectiveness of the proposed JBE algorithms