Optimum one-bit block quantizing
D. Esteban, J. Menez, J. P. Temine · The Journal of the Acoustical Society of America · 1976
An iterative method is described which allows determination of the optimum parameters of a first-order one-bit quantizer in the least-mean-square error sense. Those parameters are computed from a finite sequence of a block of samples. The reference first-order equation being nonlinear, an iterative substitution method is used to obtain the optimum parameters. The basic formula and the procedure are independent of the chosen type of coding scheme and can be applied indifferently to either one-bit PCM or delta modulation or adaptive delta modulation or any method supplying a binary quantizer sequence (for instance, Viterbi algorithm). This method has been applied with success in the area of voice encoding at low bit rate. Experimental simulations will be played during the presentation.