Research On The Uncertain RFID Event Model Calculation Method Of Probability

In recent years, radio-frequency identification (RFID), a non-contact automatic identification and data acquisition technology, has been widely acknowledged. With the extensive application of military system, intelligent building system and supply chain system, it is becoming more and more the focus of concern. RFID data has an extremely important characteristic that is uncertainty. The expression method of uncertain events commonly uses probability value to show the possibilities of their occurrence, therefore, the specific forms of the uncertain events are called the probability events. The events vary and their solutions vary, therefore how to calculate the compound event probability which is effectively composted by basic events is a difficult point. How to extract the useful data from RFID data and to semantically transform then to determine the occurrence of events is the research emphasis of this paper.Through analysis processing the amount of basic events, to extract the useful compound event which users interest in. This paper goes on a further research on the detection of RFID events with uncertain semantic.In the base of the uncertain event model, presenting the probability calculation method of the uncertain event model. Bringing about discussion from the view of the relations between events, it puts forward the three kinds of relations between events, that is, independent relation, conditional relation and temporal relation. It introduces the constructor—operators of events; it classifies the operators into atomic operators and non-atomic operators, and it puts forward the calculation method of the atomic operator and the non-atomic operator, and finally it discusses the operation laws of events operators. It introduces the solving methods of the probability of basic events, to deal with compound events respectively according to the three relations between the probabilities of events. For compound events between which exist independent relations, according to its semantic adopt corresponding operation compound event structure, direct use operators computing to solve it. For temporal relations, refers to in a period of time, the same label in the same reader range, during this period, the condition of label has changed. It should adopt Markov chain models to calculate probability. When working out the probability of the compound events, how to judge whether the compound events occur, this paper introduces a confidence interval algorithm, According to the needs of the users judge the credibility of the compound event .when the work-out probability values the confidence interval, we can assure the compound events has happen.