Vector Space Approach to Narrowband Interference Excision

摘要

To realize the full performance advantage of a pseudonoise spread-spectrum communication system, a receiver must suppress strong narrowband interferers. Conventional frequency-domain excision techniques are complicated by spreading effects due to the sidelobes of the sin(x)/x function. This paper shows that such problems can be avoided by regarding the N samples of a band-limited times series and its finite Fourier transform as a vector in an N-dimensional space. Narrowband interference can be suppressed by projecting the data vector onto the subspace orthogonal to the one that best fits the interference. The theory developed here includes a least squares criterion for finding this subspace. (RH)