Recursive consistent estimation with bounded noise

Estimation problems with bounded, uniformly distributed noise arise natural ly in reconstruction problems from over complete linear expansions with sub tractive dithered quantization. We present a simple recursive algorithm far such bounded-noise estimation problems. The mean-square error (MSE) of the algorithm is "almost" O(1/n(2)), where n is the number of samples. This ra te is faster than the O(1/n) MSE obtained by standard recursive least squar es estimation and is optimal to within a constant factor.