Research On Identification Efficiency Of RFID System In Large-scale Tag Environment

Radio frequency identification(RFID) is one of the key technologies in Internet of Things(IoT), and it realizes wireless non-contact information identification through radio frequency communication. With the continuous development of RFID technology, its application scope is becoming more and more widely. However, the problems existing in RFID technology are increasingly obvious. For example, when a reader is identifying a number of tags, the problem of tag collision comes out and also the capture effect problem resulting from the cause that the different distance between the reader and the tags appears and so on. The phenomena of that the reader identifies a tag mistakenly or not seamlessly will apparently lower the identification efficiency of RFID system. The key to improve the identification efficiency of RFID system is to solve the problem of tag collision. Especially, improving the identification efficiency of RFID system by taking advantage of the capture effect has become a research hotpot in the field of RFID technology.Aim at handling with the tag collision in RFID system, a series of algorithms based on Aloha are emphatically analyzed in this paper. Then some existing tag estimate algorithms are listed and their disadvantages are pointed out. Finally, a modified DFSA algorithm, based on the theory of dynamic frame slot Aloha(DFSA) algorithm, is proposed to be applied to improve the identification efficiency of RFID system in large-scale tag environment.(1) In large-scale tag environment, the number of tags in the magnetic field of the reader may be greater than an initial frame length. Then the settings of initial frame length in conventional algorithms are not suitable in this situation, which will influence the setting of optimal frame length by increasing the error of tag estimation, and eventually lower the identification efficiency of RFID system. To ensure the reasonability of setting of initial frame length, a modified adjustment of initial frame length is proposed in the paper, which adjusts the initial frame length by judging the status of the first several slots.(2) The algorithms based on Aloha need to set the optimal frame length by estimating the present tags in the magnetic field of a reader. Therefore, it is essential to improve the accuracy of tag estimation for the setting of the optimal frame length and the improvement of the identification efficiency. Because of the existence of capture effect, some existing algorithms do not have commendable estimation performance. The situation that the capture effect has an influence on tag estimation is well considered in this paper, which is used to improve the efficiency of identifying tags. Minimum mean square error(MMSE) method is applied to the tag estimation, which estimates the number of tags and the probability of capture effect simultaneously. Furthermore, according to the Bayes estimation theory, three different forms of Bayes estimation expressions are respectively deduced by using different cost functions. Moreover, the expression of optimal frame length is further deduced in isometric time slot situation, and then the optimal frame length can be set by the estimation values for improving the identification efficiency.