A new method for Golomb-Rice parameter estimation
S. Domnic · 2017 IEEE International Conference on Microwaves, Antennas, Communications and Electronic Systems (COMCAS) · 2017
Golomb-Rice coding is a simple and fast data compression technique. It has been used in many areas such as image compression, video compression, audio compression, wireless sensor networks and others. Since Golomb-Rice coding is a parameterized (k) algorithm, its performance depends on the optimal code parameter k* for the given sequence of samples/integers. In this paper, two simple procedures are proposed to estimate the optimal code parameter k* of Golomb-Rice coding for encoding the given sequence of samples efficiently. It is shown that Golomb-Rice optimum code parameter selected under the first method is always within the upper bound and lower bound calculated from mean value of the source and the additional cost due to suboptimum code parameter under the second method is not more than one bit per sample. However, it minimizes the number of computations required by first method. It is also experimentally observed that the parameter estimated by the first method is same as the actual Golomb-Rice optimum code parameter.