Research On RFID Adaptive Multi-tree Anti-collision Algorithm In IoT

Radio Frequency Identification(RFID),as a contackless and low-cost technology,has gained great popularity in Internet of Things(Io T)applications.Physical objects labeled by electronic tags can achieve connection with Io T by using RFID applications to identify tags and collect information.Multitask recognizing is a typical advantage of RFID technology,which also brings some critical problems.When multiple tags respond to RFID reader's command simultaneously,the data sent by different tags will collide,leading to the failure of identification.Therefore,it is of vital importance for the development of RFID technology in Io T to establish an efficient anti-collision mechanism to solve the tag collision issue.This thesis mainly studies tree-based deterministic anti-collision algorithms and puts forward some improvement and optimization to the problems of existed tree-based anti-collision algorithms that they cost too many timeslots,too much data and too deep search depth under dense tag environment.The main contributions of the thesis are summarized as follows:(1)The thesis researches the key technology of RFID systems and introduces its composition and basic principles.Also,the thesis briefly introduces the challenges faced by RFID systems such as anti-collision,localization,data privacy and safety.(2)Traditional RFID anti-collision algorithms are intensively studied.The basic principle of anti-collision algorithms based on ALOHA protocol and tree structure are expounded in detail.Meanwhile,the thesis analyzes the performance of the algorithms and summarizes the advantages and disadvantages.(3)An adaptive multidimensional collision tree(AMCT)algorithm is proposed to solve the problem that binary tree-based algorithms suffer too deep search depth compared to algorithms based on multidimensional tree structure,which,on the other hand,generate too many idle timeslots.AMCT algorithm introduces adaptive division mechanism to select bifurcation diagrams and proposes a prefix predication mechanism for multi-tree structure so as to decrease the number of timeslots,eliminate the idle timeslots and increase the system performance.The simulation results prove that AMCT can be an efficient solution for RFID tag collision problem in Io T in terms of number of collision timeslots,number of total timeslots and system efficiency.(4)An adaptive multi-tree anti-collision algorithm based on bit-locking and backtracking(ABLB)is proposed.Considering the improvement of energy consumption,system efficiency and adaptability to different tag environment,ABLB algorithm proposes a new bit-detecting mechanism based on the combination of bit-locking and backtracking strategy and multi-tree structure.Additionally,ABLB algorithm utilizes the number of collided tags to select the search mode to eliminate idle timeslots and reduce transmitted bits and energy.The simulation results prove that ABLB algorithm can effectively reduce the number of timeslots and the transmitted bits,decrease the energy consumption and achieve great improvement in system efficiency.