Solving the RFID network planning problem under the perturbation effect defined by a new probabilistic power-based model

In many areas, the identification and monitoring of objects have become a necessary requirement for people to facilitate their tasks and minimise their intervention, and for this reason, the Internet of Things (IoT) makes it possible to connect physical objects to the internet. One of the important technologies applied in IoT is a radio-frequency identification (RFID), and its deployment has created an RFID network planning problem that can be seen as a non-linear combinatorial NP-hard optimization problem that needs to identify the location of a minimum number of antennas to satisfy full coverage and zero interference. The ultimate goal of this paper is to solve the RNP problem under perturbation by introducing a probabilistic power-based coverage model and by presenting a hybrid approach based on artificial neural networks and the redundant antenna elimination algorithm RAE. The results obtained by our approach, applied to the most commonly used benchmarks by several researchers, show the effectiveness of the proposed approach in dealing with the existence of simulated perturbations in real cases and, furthermore, in generating a perfect deployment of an RFID system.