Swarm intelligence RFID network planning using multi-antenna readers for asset tracking in hospital environments

Abstract

Radio frequency identification (RFID) systems have been widely used for the inventory management and asset tracking in hospital industry. Although the total cost of an RFID system relies mostly on the number of readers, existing RFID network planning (RNP) techniques focused on single-port readers. This paper introduces a swarm intelligence RFID network planning model with multi-antenna readers (named RNP-MAR) to efficiently track medical assets in hospital environments. Our motivation is to utilize multi-port RFID readers, in order to minimize the total cost of the RFID network, and simultaneously maximizing the network coverage. To achieve this purpose, we consider three reader types with one, two, and four antenna ports. Moreover, different antenna models with different coverage characteristics are utilized to be allocated on the different ports of RFID readers. The aim of the RNP-MAR is not only to determine the optimal placements of the RFID readers in the hospital plan, but also to design the multi-antenna readers and their antennas. To efficiently solve the RNP-MAR model, a combined global-local linear-nonlinear metaheuristic based on whale optimization algorithm and simulated annealing (named WOA-SA) is presented. The multi-objective function of the WOA-SA is to maximize the network coverage, while minimizing the total cost, interference, collision, and power consumption. Simulation results show that the proposed method provides 39.57% saving (on average) in the total cost of the RFID network through efficiently utilizing multi-antenna RFID readers.