Automated RFID network planning with site-specific stochastic modeling and particle Swarm Optimization

In this paper, a solution for automated passive RFID network planning is proposed. The proposed scheme comprises two parts; i) a fast site specific stochastic propagation model that extracts the probabilities of successful identification for any possible reader-antenna configuration, ii) a particle swarm optimization (PSO) algorithm that selects a subset of the antenna-configurations to be installed so that a given cost function is satisfied. In contrast to prior art, the combinatorial performance of all reader antennas is evaluated at each tag location; this revealed that good identification-performance is recorded at overlapping regions, where no single reader-antenna would operate adequately. As a result, the final solution includes less equipment, reducing the cost of the network. Results presented herein demonstrate that nearly half the number of antennas are needed for the same problem, compared to prior art.