A New Approach of Spectrum Compression based on Morphology Filter

Abstract

Aiming at solving the problem of transmitting and storing radio spectrum data in large volume, this paper proposes a new method in data compression and refactoring. Based on the research of mathematical morphology, the method designs a morphology filter by which spectrum form could be determined by extraction of its features according to structural element. Advantage of this method lies in significantly increasing the compressing ratio while maintaining good performance in signal-noise separation and signal feature reservation, realized by extracting signal and noise respectively, and using appropriate structural element to match the form of noise. Being tested in experiment, this method has been proved to be effective in compressing radio spectrum data, which provides a new approach in its transmission and storage preprocessing.