Compression of Vehicle-Driving Data by Means of Orthogonal Bases

Abstract

The paper deals with application of orthogonal basesin signal approximation with the aim of data compression in avehicle driving simulator. Three different bases are tested: DiscreteFourier Basis, Discrete Cosine Basis, and Slepian Basis.Quality of signal approximation error is assessed in terms of globalsquared errors. Thus obtained numerical results suggest thatSlepian Basis affords the sparsest representation of signals testedin this study. Therefore, a considerable reduction of requiredmemory can be accomplished.