Analysis and implementation of an arbitrary variable up-sampling algorithm based on lookup table

In digital signal processing, it is difficult to achieve flexible arbitrary sampling rate conversion. As for this problem, a new algorithm is proposed to realize arbitrary variable up-sampling based on lookup table. The algorithm firstly constructs an original filter whose coefficients are stored in ROMs. Then the target filter is decimated from lookup table by mapping sampling rate as addressed. Arbitrary variable up-sampling can be achieved when input signal passes through the target filter after zero-insertion. The paper describes principles of the algorithm and constructs the model, then proves the validity of the scheme, and analyzes the performance as well as the effects of errors. The results show that the proposed algorithm has better performance than conventional algorithm and the effect of errors is negligible. It can implement arbitrary variable upsampling speedily and conveniently within a certain range.