A Novel Moore-Penrose-inverse-matrix-based Data Compression Method
Xin Liu, Yanxia Liang · 2021 IEEE 3rd International Conference on Civil Aviation Safety and Information Technology (ICCASIT) · 2021
Data is growing explosively in digital world, which requires efficient technique to store and transmit data. Because of limited resources, we need data compression (DC) techniques to minimize the size of data being stored or communicated. A Moore-Penrose-inverse-matrix-based compression method is proposed in order to compress data easily and simply. We use a non-square matrix with more columns than rows to compress data in matrix. Taking random binary data as an example, we use random matrix as compression matrix. Simulation results show that the error rate get lower as the compression ratio gets close to 1. Moreover, the error rate is neither relevant with the size of data nor with that of compress matrix. Most importantly, this method is more suitable to the situation with data and compression matrix subjected to uniform distribution.