Efective Object Identification and Association by Varying Coverage Through RFID Power Control

摘要

This paper presents an efective power scheduling strategy for energy efcient multiple objects identification and association. The proposed method can be utilized in many heterogeneous surveillance systems with visual sensors and RFID(radio-frequency identification) readers where energy efciency as well as association rate are critical. Multiple objects positions and trajectory estimates are used to decide the power level of RFID readers. Several key parameters including the time windows and the distance separations are defined in the method in order to minimize the efects of RFID coverage uncertainty. The power cost model is defined and incorporated into the method to minimize energy consumption and to maximize association performance. The proposed method computes the power cost using the range of the outermost position for possible single association and group associations at every sampling time. An RFID reader is activated with the proper coverage range when the power cost for the current time is lower than the power cost for the next time sample. The simplicity of the power cost model relieves the problematic combinatorial comparisons in multiple object cases. The performance comparison simulation with the minimum and maximum energy consumption shows that the proposed method achieves fast single associations with less energy consumption. Finally, the realistic comparison simulation with the fixed range RFID readers demonstrates that the proposed method outperforms the fixed ranges in terms of single association rate and energy consumption.