Efficient nonlinear blind interference cancellation for DS/CDMA

Multiple-access interference (MAI) in direct-sequence code-division multiple-access can be cancelled blindly using minimum mean output energy adaptation. The original MMOE scheme is linear. Because optimal detectors are nonlinear, a more general framework for MMOE detection is presented. A nonlinear MMOE scheme based on the Volterra expansion has high complexity and poor transient performance, so new nonlinear schemes are presented. Pattern classification principles motivate using the hyperbolic tangent function in nonlinear schemes. An efficient hybrid scheme is proposed that combines the advantages of the linear and hyperbolic tangent methods into a single detector.