Corrections to "Robust Detection in DS-CDMA"

A number of printing errors have appeared in [1] some of which were typos in the original manuscript, and others were introduced during the typesetting. Certain equations appeared only in half, while others were repeated twice. The authors had not had a chance to correct the galley proofs due to their absence from their contact addresses, and unfortunately, these errors went unnoticed until recently. Below we supply the corrections. In defining Huber’s mixture classes in Section II, F0 and F1 should have the respective outlier densities be denoted by h0 and h1. The new outlier densities fh ; , i = 0; 1, represent the aggregate of hi(x) and the probability density function (pdf) of m K (incorrectly called the joint densities in [1]), and have nonzero means. In Section IV, the Gaussian mixture model was considered. Suppose that the outlier pdf g(x) is Gaussian with zero-mean and variance 2 n, where 1. Let fg ; (x) = fh ; (x)jh (x)=g (x), i = 0; 1, which is Gaussian with mean and variance 2 o = 2 n+ 2 . Below we rewrite the equations (17)–(20) in Sections IV–V of [1], with the typos and other problems removed.