Research On RFID Anti-collision Algorithm

RFID is the abbreviation of Radio Frequency Identification, its Chinese name is radio frequency identification technology, which is the application of radio wave technology automatically identify individual items collectively. In internet of Things, start the RFID system, firstly reader send request command, and in the readable range all the tags can receive the request and also sent EPC code to reader, as each tag's serial number is unique, so the data reception process is bound to collide, so how to design safe and efficient anti-collision algorithm, making the reader not repeat and do not miss to read each label is the implementation of RFID systems must address the key technology, the industry has been a hot research issue on it.There are two types of traditional anti-collision algorithm one is the ALOHA algorithm and slot ALOHA algorithm, both algorithms are waiting to solve the problem of collisions by delay, the channel utilization was 18.4% and 36.8%, and with the label increase its performance sharp deterioration. The other is Xu Li Xiang, Lan Wei Yun designed dynamic binary search algorithm, such algorithm reducing collision bit until the elimination of collision problem, the performance will not worsen with the increase of N value, search one tag from the N- single tags average search Log₂N+1 times, identification of all N tags need N*(Log₂N+1) times. However, the dynamic binary collision algorithm has its flaws, After each time it read a label, the reader re-send all '1 'in the access code, which makes the system reads the tags in the course of the previous query information is not obtained to make full use, resulting in subsequent queries have to repeat the previous work, this not only increase the cost time of the system, the system also increases the data redundancy.More detailed analysis in the text describes the RFID technology, research status, and the composition of radio frequency identification system and work flow; more in-depth analysis and comparison of the current widespread use of the advantages and disadvantages of two types of side collision algorithm, on the basis that, point out the lack of dynamic binary arithmetic, and gives an improved method. Improved dynamic binary collision algorithm is when the system discover collision then set the high collision location, respectively, '0 'or '1', leave the high bit the same, so that get two serial number a1, a2, and a1 as the next query code, a2 into stack. After reading a tag remove the top element from the stack as a new query code to work, and so on, until the stack is empty, all the labels well read. The benefit of this is to take advantage of the previous query results obtained in the course (the data in the stack), greatly improving search efficiency, by Matlab software simulation show that the improved algorithm to recognize N tags need to search 2N-1 times, identify the average search 2 times to read a single tag, the channel utilization rate steady at 50% or more.