Dynamic Adjustment Type Anti-collision Algorithm Based Tag Estimation Grouping

The concept of The Internet of things is the earliest by O Kevin shi put forward in 1999, Through the Auto-ID and western began to spread, The concept of TheInternet of things began to widely circulated, Radio Frequency Identification technology (RFID) is considered as the foundation of the Internet of things. On the basis of EPC RFID is becoming the basis of Internet of things is that it is based on the number of items for identification of terminal technology.RFID automatic identification technology is a kind of application of the Internet of things technology of non-contact whichto use the specific identification equipment,and to identify object identification,And then delivered to the background data processing center for processing. With the widely use of RFID technology in more and more people’s lives, The corresponding identification object, such as radio tags also more and more,Would inevitably produce some collision problems in the process of recognition, Therefore,to solve the collision problem has increasingly become one of the main content of people’s research.The paper focuses on one of the key issues in the RFID technology——Collision algorithm for research, to solve the collision problems between the label when identify it in whole big System of the Internet of thing. The paper have introduces the RFID technology and RFID system identification principle at first, and then when introduces the RFID system identification principle,the problem are put forward the problem of tag collision, and for existing ALOHO collision algorithm and binary tree collision algorithm are also introduced accordingly. In Binary tree algorithm we have classify introduced several memory and memoryless algorithm, such as BT, BS, QT, DBS algorithm, etc. In ALOHO algorithm, this paper introduces several kinds of derivative algorithm respectively, Such as ALOHO algorithm, Needle ALOHO algorithm of contention slot, dynamic slot ALOHO algorithm, etc. And according to the different advantages of the algorithm are explained.This paper finally presents an algorithm,the algorithm are grouped on the basis of the tagestimation, Within the group used the dynamic binary search algorithm, estimation algorithm using theSchoute estimation model, when grouping we set a threshold value, And using binary tree depth type grouping method.so that each group of tags is easier to read, and is easy to be grouped, will not lead to tag number too big in each group, and will also not lead to the label number small and cause the waste of time slot. The algorithm havecombines the DFSA grouping the advantages of high efficiency and the characteristics of the processing ability of DBS algorithm.Because of ALOHO algorithm under some special circumstances can lead to certain tags in infinite loop situation, so that the label can’t read, this algorithm combines the certainty of DBS algorithm, solved such problems; This algorithm is combined with the DFSA grouping,solved the DBS algorithm in dealing with a tag because the label too much and deep level of query generation,and led to the decrease of the recognition efficiency.Finally, the simulation experiment shows that, according to the different parameters according to the recognition rate, time delay, system efficiency and collision rate are obtained by simulation experiment. According to the experimental results in obvious lazy recognition efficiency is superior to DFSA algorithm,because the DBS algorithm is used to identify the group reduced the recognition algorithm, depth DBS algorithm, which can clearly see that delay time of the algorithm is better than DBS.As the proposed algorithm set the threshold value, the group has at most 16 labels, so the collision rate is a stable edge labels. According to the efficiency of the system is not the same number of slots is set to set the threshold value of the simulation parameters, the numerical results from the optimal simulation.