Linearly constrained interference rejection for improved spread spectrum performance

An adaptive technique is presented that utilizes the known characteristics of the pseudonoise spread spectrum sequence to enhance the detection capabilities diminished by interference excision. The new technique uses a constrained excision to decrease the bit error rate. The formation of constraint surfaces is the basis of the projection technique used in the adaptive update of the excision filter. The constrained updating of the filter coefficients retains the interference rejection properties of the excision filter while decreasing the variance of the decision variable. The filter operates without training bits and requires little additional computational processing cost over a standard adaptive excision filter. Simulation results demonstrate the mean square error improvement obtained by using the constrained excision filter