Research On Real-Time Estimation Algorithm Of Labels For Large Scale RFID Systems

The Internet of Things is a network that realizes the connection between objects and objects.In order to realize the interconnection of everything and provide more convenient services,it involves automatic identification technology,and Radio Frequency Identification(RFID)technology is one of key technology.Radio frequency identification technology is widely used in many fields such as urban logistics transportation,public security,agricultural forestry and medical treatment.As the number of markers increases,it is necessary for real-time,fast and accurate updating of the number of abnormal items in the warehouse for large-scale storage in order to timely update the replenishment and prevent theft.The tag number estimation algorithm,it can be further applied to missing tag search,unknown tag identification,and tag information collection.In response to the above problems,the main work of this paper is as follows:(1)A tag estimation algorithm based on slot pairing in a large-scale RFID system is proposed.A large number of unused time slots wastes exist in the traditional estimation algorithm.This paper uses the time slot pairing method to provide a second selection opportunity for the tag after the tag selects the time slot for the first time.A tag that is expected to be a collision time slot is paired with an empty time slot in a time slot pairing phase to generate more expected useful time slots,thereby constructing a single time slot to communicate with the reader.Subsequently,the algorithm uses the indication vector to obtain the real-time slot state.In the subsequent operation,the reader only needs to monitor and record the number indicated in the vector,and finally use the value of the collected indication vector to simultaneously estimate the existing label and the missing tag.The simulation results show that the algorithm can accurately estimate the number of existing tags and the number of missing tags in a short time.The algorithm execution time is significantly faster than other existing algorithms and has good scalability.(2)A real-time monitoring algorithm for tag quantities in large-scale RFID systems is proposed.For privacy protection policies,the algorithm does not require tags to send their IDs,but uses the RN16 random sequence to check the status of the time slots so that the reader does not need to read the tag ID information.For a large amount of data,a sampling method is employed to reduce the number of time slots to be detected.For the estimation accuracy,the inverse function is used to obtain the estimation expression and the method of estimating the averaging multiple times.For the real-time problem,the time slot situation at different moments is compared to update the system's leaving tag and the newly added number in real time.The simulation results show the superiority of the algorithm.The execution time of the algorithm is significantly less than that of other existing algorithms under the 95% reliability requirement.Under the 98% reliability requirement,the algorithm execution time increase is slow.With the wide application of RFID,there are massive tags in the system.The tag number estimation algorithm can work alone and can be embedded in other algorithms,and the efficient and fast estimation of the number of abnormal tags needs further research.