Abstract
Multichannel adaptive equalization (AE) systems require high computational capacity, which constraints their practical implementation. Graphics Processing Units (GPUs) are well known due to their potential for highly parallel data processing. Although the GPUs seem to be suitable platforms for multichannel scenarios, an efficient use of parallel computation in the adaptive filtering context is not straightforward due to the feedback loops. This paper presents a GPU implementation of a multichannel AE system based on the filtered-x LMS algorithm working over a real-time prototype. Details of the parallelization of the algorithm are given. Experimental results are presented to validate and computationally analyze the real-time performance of the AE GPU implementation.
Results show the usefulness of GPUs to develop versatile, scalable and low cost multichannel AE systems.
