Research On Anti-Collision Algorithms For Multi-Tag Identification Rfid Systems

The major advantages of radio frequency identification (RFID) technology are that the tags are not required to touch the reader in the process of identifying and multiple objects can be read simultaneously. However, when two or more tags response a reader simultaneously, communication conflict or data interference will occur, which will reduce the efficiency of the system. In this thesis, a research was made which based on the anti-collision algorithms for dynamic framed slotted ALOHA (DFSA) multi-tag identification of passive RFID system. Besides, to improve the performance of original tag estimation techniques with the sequential weighted least squares algorithms, forgetting factor to enable sequential weighted estimation of tag numbers which was introduced, it gives more weight to the current read cycle data and less weight to the previous data. In this way, the variation in the value of the number of tags estimated in each read cycle will be reduced. Furthermore, we compared the performance of DFSA algorithms before and after the improvement through extensive simulation results and the best forgetting factor were determined. Moreover a low-complexity heuristic stopping criterion was developed in this thesis. At last, simulation results show that our scheme improves the efficiency of the RFID system of tag identification.