Visible to the public Redundant Dictionary Construction via Genetic Algorithm

TitleRedundant Dictionary Construction via Genetic Algorithm
Publication TypeConference Paper
Year of Publication2018
AuthorsLi, Haipeng, Zheng, Cuie, Zhang, Jucheng
Conference NameProceedings of the 2Nd International Conference on Vision, Image and Signal Processing
PublisherACM
Conference LocationNew York, NY, USA
ISBN Number978-1-4503-6529-1
Keywordsarray signal processing, composability, genetic algorithm, Metrics, pubcrawl, Redundant Dictionary, Resiliency, restricted isometry property, signal processing security
AbstractSparse representation of signals based on redundant dictionary is widely used in array signal processing. In this paper, a redundant dictionary construction method via genetic algorithm (GA) is proposed for array signal processing. The problem is formulated as a dictionary selection problem where the dictionary entries are produced by discretizing the angle space. We apply the orthogonality of the entries to evaluate the dictionary according to the Restricted Isometry Property (RIP). GA is used to discretize the angle space which can make the dictionary more orthogonal. Simulation results show that the proposed method can obtain a better division of angle, improving the orthogonality of dictionary effectively, and is suitable for arbitrary observation space compared with commonly used equal angle division and equal sine division.
URLhttp://doi.acm.org/10.1145/3271553.3271604
DOI10.1145/3271553.3271604
Citation Keyli_redundant_2018