Research And Application Of Missing Tag Detection And Cardinality Estimation Method For RFID Systems

Radio Frequency Identification(RFID)technology is a key part of the Internet of Things(IOT),which has been widely applied in the fields of logistics transportation,warehouse management,etc.RFID technology uniquely identifies items in some system through RFID tags,and manages status of items through readers.When some items in one system are lost due to management fault or theft,the normal operation of the system will be affected,and owners of these items will suffer from economic loss.RFID technology provides a solution to this issue by identifying missing tag events in the system.Therefore,it is necessary to study the methods of missing tag detection and cardinality estimation protocols in RFID systems for realistic application scenarios.The main innovations of this dissertation is summarized as follows:1.For items in some RFID systems that are classified by multiple categories,this dissertation studies the missing tag detection protocol for multi-category RFID systems to check whether there is a missing tag event.Category identifiers are used to distinguish tags from distinct categories.In order to eliminate redundant bits in category identifiers sent by tags,"major category" and "minor category" are introduced in this dissertation."major category" and "minor category" are defined according to the number of tags in the category.Following the framework of the framed slotted Aloha protocol,the time frame for tags to send responses is divided into two parts.The tags of the major categories can send responses in any slot in the time frame,and the tags in the minor categories can only send responses in slots of the first part of the time frame.The protocol also proposes a method to dynamically compress bit-length of category identifiers according to the number of active categories during system operation in order to compress redundant bits of current category identifier.Simulation results show that the proposed protocol not only meets the detection accuracy requirement of the system,but also takes less time overhead than other protocols.2.The dissertation studies the problem of missing tag detection in random error channel model.Channel noise will lead traditional missing tag detection protocols to report incorrect missing tag events.That is,an active tag may be reported as a missing tag,which reduces the reliability of the related protocols’ output.The protocol proposed in this dissertation first selects candidate missing tags in the system.Then,the protocol performs several rounds to confirm candidate missing tags’ state.Finally,a candidate missing tag’s state is determined according to the principle of majority vote.In order to reduce the protocol’s execution time and simplify the process to confirm candidate missing tags’ state,the protocol only selects the singleton slots which are only picked by one expected tag.The simulation results show that compared with the traditional schemes,the measures taken by the proposed protocol against channel noise can effectively reduce the ratio of incorrect missing tag events in the protocol’s output.3.The dissertation studies the missing tag cardinality estimation problem under the random error channel model.The status of slots picked by tags in the time frames can reflect statistical characteristics of the size of tag set.By identifying the state of a specific slot,a probability formulation can be established to estimate the number of tags in the system.By combining the bit transmission error rate under the random noise channel model and the position of the first expected non-empty but actually decoded as an idle slot in the response time frame,the number of missing tags can be estimated.Through theoretical analysis,this dissertation gives the estimation formula for the number of missing tags under the ideal channel model and the random error channel model respectively,and discusses the optimization of relevant parameters.The simulation results show that compared with other related protocols,the proposed protocol ensures reliable estimation accuracy under the random error channel model.