Rejection of Narrow-Band Interferences in PN Spread Spectrum Systems Using an Eigenanalysis Algorithm

A new eigenanalysis based adaptive algorithm is suggested for rejecting narrow-band interferences in spread spectrum communications. The optimal linear interference canceler implemented as a transversal filter is found from the solution of the Wiener-Hopf equations. A different approach is suggested by the eigenanalysis of the data across the filter taps. The spread spectrum signal has a white spectrum, i.e., its energy is uniformly distributed across the eigenvalues of the correlation matrix. The interference, however, has its energy concentrated in just a few large eigenvalues. The corresponding eigenvectors contain all the frequency domain information required to reject the interference. The eigenanalysis based canceler is referred to as an Eigencanceler and is derived as a modified prediction error filter. An adaptive algorithm based on the power method is shown to provide faster convergence than the LMS and RLS algorithms.