Optimal scheduling-based RFID reader-to-reader collision avoidance method using artificial immune system

In radio frequency identification (RFID) systems, communication signals from one desired reader is subject to interference from the other adjacent readers operating at the same time, so the reader-to-reader collision problem occurs. Many RFID reader collision avoidance methods have been developed, such as coverage-based methods, control mechanism-based methods and scheduling-based methods. In the scheduling-based methods, how to allocate frequency channels and time slots for the RFID reader network is emphasized. In this case, the RFID reader collision avoidance problem is transferred as an optimal scheduling problem, which can be solved by analytical methods and intelligent algorithms. Artificial Immune System (AIS) optimization is an emerging heuristic method derived from the human immune system. Due to its powerful global searching capability, AIS has been widely applied to scientific and engineering problems. This paper attempts to formulate the reader-to-reader collision problem (R2RCP) and its scheduling-based reader-to-reader collision avoidance model (R2RCAM), and proposes an improved AIS optimization for resource allocation (RA-AIS) in R2RCAM. Within the proposed RA-AIS optimization, the candidate antibody is constructed by using frequency channels and time slots, and in the mutation phase, the candidate antibody evolves dynamically according to its corresponding readers’ interfering power. The proposed RA-AIS optimization is examined on a series of numerical experiments to evaluate the effects of time slots, frequency channels, and transmitting power. Moreover, a group of comparative experiments are also arranged. The experimental results demonstrate that the proposed RA-AIS optimization is an effective method for R2RCAM, and performs better in searching the maximum interrogation area than other algorithms, such as random method (RM), genetic algorithm (GA), particle swarm optimization (PSO) and the canonical artificial immune system optimization (opt-aiNet).