| Radio Frequency Identification(RFID)technology,as a contactless automatic identification data communication technology,provides a solid technical support for the development of the Internet of Things.At present,RFID technology has a very wide range of applications in real life,including supply chain management,warehouse management and so on.By attaching electronic tags with unique code to target items,internal information can be obtained quickly and effectively,which greatly improves work efficiency.However,the existence of unknown tags or cloned tags in RFID systems will affect the normal commication process of RFID systems.For example,responses of unknown tags or cloned tags will affect detection of missing tags,which will may lead to economic loss and security risks.On the other hand,there often exists external noise from communication environment,which will affect accuracy of results of RFID tag cardinality estimation.Therefore,this dissertation studies RFID tag cardinality estimation under unreliable communication environment.The main contributions and innovations of this dissertation are as follows:1.Estimate the number of unknown tags in RFID systems over unreliable channels.Existing protocols for estimating the number of unknown tags generally assume that a wireless channel is ideal.In fact,interference from ambient noise affects transmission of signals.In this dissertation,a new unknown tag cardinality estimation scheme,CEUT,is proposed.In the CEUT scheme,a RFID reader first pre-allocates time slots to the known tags and obtains a prediction frame.The RFID reader then collects response frames from all active tags.Based on the channel noise parameter and the number of empty time slots in the prediction frame that are converted into non-empty time slots in the response frame in each round,an estimator of the number of unknown tags is designed.Simulation results show that under the given channel noise parameter,the proposed unknown tag cardinality estimation scheme can reduce the estimation error rate and achieve the given estimation accuracy parameter compared with the related scheme.2.Since RFID systems are subject to cloning attack,a scheme to estimate the number of cloned tags is designed.The cloning tags produced by an attacker will also affect the normal operation of RFID systems,and estimating the number of cloned tags in a RFID system can help to understand security risks of the system.The proposed scheme also takes into account the capture effect on estimating the number of cloned tags,which is more realistic than the previous schemes.Simulation results show that the proposed scheme for estimating the number of cloned tags can achieve the given estimation accuracy under the capture effect.3.When the capture effect and the existence of unknown tags is further considered,a scheme for estimating the number of cloned tags is designed.In a large-scale RFID system,when RFID tags may be cloned,there may be unknown tags in the system,and responses of unknown tags may lead administrators to treat normal tags as cloned tags.The new estimator for the number of cloned tags depends on two statistical parameters:(1)one parameter depends on the number of unknown tags;(2)the other parameter depends on the number of singleton time slots in the prediction frame that are converted to collision time slots in the response frame.According to the given estimation accuracy and capture effect parameters,the cumulative distribution function yielded by simulation results show that the proposed scheme can meet the reliability requirement. |
