The N -level uniform quantizer on [-c,c] plus the assignment of y_{0}\deg = -(a _{s}+ c)/2 and y_{N+1}\deg = (a_{s}+ c)/2 to signal values falling in the saturation regions [-a_{s},- c) and (c,a_{s}] , respectively, is shown to be the minimax (N + 2) -level quantizer with a nonsaturating input range [-c,c] . The performance criterion considered is the mean weighted quantization error and the input signals are only required to be amplitude bounded by \pm a_{s} where a_{s} > c > 0 . The worst case input signal marginal probability distributions are shown to be discrete. From the derivation of this result, the minimax error can be computed. An example is given which illustrates the performance of the minimax quantizer for several input ranges against different input signal probability distributions.
[1]
Bernard M. Smith.
Instantaneous companding of quantized signals
,
1957
.
[2]
James L. Flanagan,et al.
Adaptive quantization in differential PCM coding of speech
,
1973
.
[3]
Allen Gersho,et al.
Theory of an adaptive quantizer
,
1973,
CDC 1973.
[4]
Thomas J. Goblick,et al.
Analog source digitization: A comparison of theory and practice (Corresp.)
,
1967,
IEEE Trans. Inf. Theory.
[5]
Joel Max,et al.
Quantizing for minimum distortion
,
1960,
IRE Trans. Inf. Theory.
[6]
J. Bruce.
On the optimum quantization of stationary signals.
,
1964
.
[7]
N. Jayant.
Adaptive quantization with a one-word memory
,
1973
.