A vector quantizer (VQ) is a mapping of input vectors to one of a finite collection of predetermined codevectors, where the set of all codevectors is called codebook. The most widely used technique for designing vector quantizers is the Generalized Lloyd algorithm or a tree-structured variant known as LBG algorithm. A VQ can also be designated as a Self-Organizing Map (SOM) using Kohonen learning. Robustness against channel-errors, when using VQ in a communication system, is an important issue which has attracted attention last years. Basically, likely errors induced by the channel should not cause major errors in terms of the reconstructed signal. This is accomplished by arranging the VQ such that codewords having small Hamming distances in terms of the channel codes also have reconstruction points with small signal distances. In this paper, we introduce a new VQ based on a self-organizing pseudo-map (SOPM) which uses Hamming distance as the criterion of neighborhood. Therefore, we have realized a pseudo-Gray coding during the process of VQ design. All the results have indicated that SOPM gives the best results compared to SOM and LBG. In the same time, a relatively small differences were noticed between the resulting distortions of the 3 VQs.