Abstract
This paper introduces a method for computing the time-varying convolution in parallel. It discusses the motivations for this approach, detailing the limitations with the current serial implementation. A detailed review of the signal processing involved is presented, describing the time-varying filter as a modification of the time-invariant case. This is followed by description of the parallel method, which is then implemented in the Open Computing Language. An analysis of tests result is provided, detailing the improvements on the existing approach and noting the cases where it is not the most suitable option.
Full Text
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call