Development and applied performance analysis of a new coding algorithm for efficient memory usage of RFID tags

RFID is a wireless identification technology used for automatic detection and tracking of objects. The operation of this technology can be summarized as reading the data stored in the RFID tag by the reader through radio frequency signals, then interpreting the data by the controller and performing the specified operations. RFID technologies are used in many different fields such as retail, logistic, health, animal husbandry, education, security, payment and membership systems. The main design features of RFID tags, namely their small physical size, low energy consumption and cost-effective production, result in tags having limited memory capacity. In this study, data compression and decompression operations were performed on the STM32F407VGT6 microcontroller. The RC522 module was used to write compressed data to the tag, which was then read from the tag and decompressed to obtain the original data sequence. The data sequences used for testing were randomly selected from text files in the Calgary corpus. Saving rate, compression-decompression time, write-to-read time, BPC and entropy were used for performance comparison. Huffman coding, Shannon-Fano coding and the developed Base-32 coding were used for compression. The results show that Huffman coding achieves the highest performance with an average saving rate of 36.27%. The Base-32 coding showed the best performance in average compression-decompression times, completing compression in 655.37 ms and decompression in 130.53 ms. The Huffman coding gave the lowest average BPC value with 5.10, while the Shannon-Fano coding gave a very close result with 5.14. The best result was obtained with Huffman encoding in parallel with the savings rate in writing and reading the data to the RFID tag.