Inter symbol interference cancellation for a mobile cellular receiver using lower and higher order statistics

A code division multiple access (CDMA) signal which is subjected to multipath and other causes of intersymbol interference (ISI) must function in a low SNR environment by virtue of being a spread spectrum signal. In such an environment conventional ISI cancellation techniques, such as least mean square (LMS) equalisation, will not operate satisfactorily. Therefore an equalisation scheme which is insensitive to extremely high levels of Gaussian noise must be developed. In this paper such a scheme which employs both second-order and higher order statistics and exploits the characteristics of CDMA channels has been developed. The algorithm is effective against both mean square error (MSE) and ISI as it employs a cost function which includes both second-order and 4th-order cumulant terms.