Research On RFID Anti-collision Algorithms Based On Binary Tree

Radio Frequency Identification technology (RFID) is a kind of wireless automaticidentification technology. RFID system transmits data and recognizes target byelectromagnetic waves in the air. RFID has many advantages such as large capacity ofstorage, fast read speed, low cost, small energy consumption, and can recognize a pluralityof targets simultaneously. RFID is one of the core technologies of the Internet of things(IoT). In the Internet of things, RFID is mainly responsible for information acquisition andtarget recognition in perception layer. One of the largest technical advantages of RFID ismulti-targets recognition, which leads to a collision problem when multiple targets transmitdata at the same time. Developing a good anti-collision mechanism to reduce or even avoidcollisions is of great significance for improving the overall performance of RFID system.The existing anti-collision technology in RFID system can be mainly divided into twocategories: the probabilistic anti-collision algorithms based on ALOHA and thedeterministic anti-collision algorithms based on binary tree.In this paper, some innovation has been done on the classic anti-collision algorithmsbased on the binary tree. The mainly innovation points are as follows:1. The adaptive collision tree algorithm (ACT) is proposed based on the classiccollision tree algorithm (CT) and its improved algorithm (ICT). In order to solve theproblem of unflexible forking in original algorithms, the proposed new algorithminnovatively introduced the adaptive M-ary forking theory to the original algorithm. Bothperformance analysis and simulation results show that the new algorithm ACT comparedwith ICT can reduce the number of queries by over30%, reduce the quantity oftransmission data by over30%, and increase the system throughput by over50%. The ideaof adaptive forking makes ACT more flexible and efficient compared to ICT.2. The new bit-locking back-off (NBLBO) anti-collision algorithm is proposed basedon the bit-locking back-off (BLBO) anti-collision. In order to solve the problem of notflexible query forking and query process redundancy in original algorithms, the proposednew algorithm innovatively introduced the adaptive M-ary forking theory and theautomatic recognition two ideas to the original algorithm. The simulation results indicatethat the new algorithm NBLBO compared with BLBO can reduce the number of queriesby over60%, reduce the quantity of transmission data by over60%, and increase thesystem throughput by over2times. The reduction of the query number and transmission data quantity, indirectly proves that a reduction of both the system energy consumption andtransmission delay. The idea of adaptive forking and automatic recognition two ideasdecrease both the communication complexity and time complexity of NBLBO greatly. Andwith the increase of the number of tags, the improved effect is more obvious andsignificant.3. In view of that both the two proposed new algorithms include the adaptive forkingand automatic recognition two ideas. The introduction of the two ideas makes newalgorithms outperform the previous CT algorithm and BLBO algorithm in term of querynumber, transmission data, system throughput and other performance indicators.Accordingly, this paper boldly infer that these two ideas can be put together as a generalstrategy to extend to other new deterministic anti-collision algorithms based binary tree toenhance the performance of the original algorithm further.