It is necessary to reduce the configuration bit stream volume if there is a need to write several FPGA configurations to the ROM. This leads to decrease in the topological size of the ROM block on the chip. The article discusses the development of the mechanism of compression and decompression of the FPGA configuration bit stream. It has been shown that the stream of configuration data has some regularities. This is due to the presence of a large number of «forbidden» states in the configuration data set, when inadmissible connections of logical elements are formed. The algorithm of compression and decompression of the FPGA configuration bit stream has been proposed. The algorithm is based on the analysis of the FPGA architecture and the compilation of the conversion table using the prefix code commands. The advantage of the algorithm is the search for the most repetitive combinations based on the analysis of the FPGA architecture, instead of searching for the entropy of a random set of configuration data. Another advantage is the relative simplicity of the decoding algorithm, which saves hardware resources for the implementation of the decompression mechanism.
-
Bibliography link:
Кузьминов И.В., Лосев В.В., Новожилов И.С., Чаплыгин Ю.А. Алгоритм компрессии и декомпрессии потока конфигурационных данных ПЛИС // Изв. вузов. Электроника. – 2019. – Т. 24. – № 6. – С. 610–618. DOI: 10.24151/1561-5405-2019-24-6-610-618
1. Шалтырев В.А., Шалтырев К.А., Шагурин И.И. Динамическая реконфигурация ПЛИС с ис-пользованием сжатых битовых потоков // Изв. вузов. Электроника. – 2009. – №2(76). – С. 43–50.
2. Hemnath P., Prabhu V. Compression of FPGA bitstreams using improved RLE algorithm // 2013
Intern. Conf. on Information Communication and Embedded Systems (ICICES). IEEE. – 2013. – P. 834–839.
3. Hauck S., Wilson W.D. Runlength compression techniques for FPGA configurations // In Proc. IEEE Symp. Field Program. Custom Comput. Mach. IEEE. – 1999. – P. 286–287.
4. Dandalis A., Prasanna V.K. Configuration compression for FPGA-based embedded systems // IEEE Trans. Very Large Scale Integr. (VLSI) Syst. – 2005. – Vol. 13. – Iss. 12. – P. 1394–1398.
5. Rui Yu, Yibin Li. Compression of FPGA bitstreams using a novel bitmask and RLE algorithm // Interna-tional Journal of Advanced Computational Engineering and Networking. – 2015. – Vol. 3. – Iss. 4. – P. 17–22.
6. Марков А.А. Введение в теорию кодирования. – М.: Наука, 1982. – 192 с.
7. Левитин А.В. Жадные методы: алгоритм Хаффмана // Алгоритмы. Введение в разработку и ана-лиз. – М.: Вильямс, 2006. – Гл. 9. – С. 392–398.
8. Choueka Y., Fraenkel A., Klein S., Segal E. Improved hierarchical bit-vector compression in document retrieval systems // Proc. of ACM Conf. on Research and Development in Information Retrieval. – 1986. – P. 88–96.