摘要
One research crucial to wider adoption of Radio Frequency Identification （RFID） technology is how to efficiently transform sequences of RFID readings into meaningful business events. Contrary to traditional events, RFID readings are usually of high volume and velocity, and have the attributes representing their reading objects, occurrence times and spots. Based on these characteristics and the Non-deterministic Finite Automata （NFA） implementation framework, this paper studies the performance issues of RFID complex event processing and proposes corresponding optimization techniques. Our techniques include： （1） taking advantage of negation events or exclusiveness between events to prune intermediate results, thus reduces memory consumption; （2） with different selectivities of complex events, purposefully reordering the join operations between events to improve overall efficiency, achieve higher stream throughput; （3） utilizing the slot-based or B＋-tree-based approach to optimizing the processing performance with the time window constraint. We present the analytical results of these techniques and validate their effectiveness through experiments.
One research crucial to wider adoption of Radio Frequency Identification （RFID） technology is how to efficiently transform sequences of RFID readings into meaningful business events. Contrary to traditional events, RFID readings are usually of high volume and velocity, and have the attributes representing their reading objects, occurrence times and spots. Based on these characteristics and the Non-deterministic Finite Automata （NFA） implementation framework, this paper studies the performance issues of RFID complex event processing and proposes corresponding optimization techniques. Our techniques include： （1） taking advantage of negation events or exclusiveness between events to prune intermediate results, thus reduces memory consumption; （2） with different selectivities of complex events, purposefully reordering the join operations between events to improve overall efficiency, achieve higher stream throughput; （3） utilizing the slot-based or B＋-tree-based approach to optimizing the processing performance with the time window constraint. We present the analytical results of these techniques and validate their effectiveness through experiments.
基金
supported by the National Natural Science Foundation of China under Grant No.60720106001
