THE WEBER QUANTIZER: PERCEPTUAL CODING FOR NETWORKED TELEPRESENCE AND TELEACTION

We present a theoretical analysis of a perceptual coding approach for networked telepresence and teleaction. Our so called Weber quantizer is based on Weber’s Law and can be used in haptic data communication to eliminate changes which can not be perceived by the human operator. The main advantage of the Weber quantizer is that it minimizes the number of samples or packets to be transmitted. Basic properties like the resulting sample rate and the MSE of the proposed Weber quantizer are derived analytically and proven correct by simulation for the case of a uniformly distributed input sequence. The contributions in this paper provide the basis for the analysis of more realistic signal models and constitute a first step towards the understanding of the relationship between the Weber quantizer, statistical error measures and actual human distortion perception.