The Research On RFID Anti-collision Algorithm

The Radio Frequency Identification (RFID) is an automatic identification technology which works with non-contact communicates.Because of its advantages such as high accuracy, large storge, strong durability, diffcult to counterfeit, high intelligence and so on.It has been widely used in many fields, like society, economy and defence.With the rapid development of RFID technology, it is more necessary that a reader identifies a large number of labels efficiently and accurately in a limited time.The singal collision label is the key to influence the performance of the RFID system.In order to solve the problem,this dissertation ventures an improved algorithm based on some basic anti-collision algorithms, the main research work is as follow:(1)It goes into the existing basic anti-collision algorithms, and analyzes their application background and their merits and demerits.(2)It modifies the back-off binary-tree search algorithm. The introduction of the collision mechanism and probability mechanism makes it possible select the position of the reader divides. It is divides according to the probability of collision position of the"0"and"1".We can acquire a improved algorithm by combining this mechanism and the binary back-off algorithm.Through theoretical analysis and system simulation with Matlab,the improved algorithm can be drawn to help reduce the number of searches and the times of the search,thereby improves system performance.This algorithm especially suits the occasions of a large number of tags.(3)This dissertation looks into the future of the anti-collision algorithm of RFID system for future research.