IGAA: An Efficient Optimization Technique for RFID Network Topology Design in Internet of Things

Abstract

Most RFID applications in the Internet of Things (IoTs)

"NOP: An Efficient Non-optimization-based Method for RFID Network Topology Design"

“…The fitness function covers six objectives: the overlapping of the reading areas, the number of useless readers, the number of redundant readers, the number of tags located in the overlap reading areas, the number of uncovered tags and the deployment cost. Note that, in the last objective, we take the deployment cost in [10] to replace "the number of readers located out of the deployment area" in [14] (because it is practically unreasonable to deploy readers out of a pre-specified deployment area). Like [10], we also define each objective as f i = 1/(100+|ε i |) in order to remove possible biased effects due to a certain objective, i.e., to be more practical and reasonable.…”

“…The technology has now become one of the principal building blocks of the thriving Internet of Things (IoTs) [6][7][8][9][10], which will soon dominate people's daily life. Most RFID applications in IoTs need to use multiple readers to read the IDs of multiple tags which form the RFID network (for instance, a supermarket will need multiple readers to read the multiple items in an area).…”

“…Unguarded or unplanned reader deployment -e.g., readers are largely or randomly deployed -may incur tremendous deployment cost and jeopardize the performance of an RFID network. This is because when unplanned reader deployment generates over-crowded readers in the network, it can impact the success ratios of tag detection and cause reader-to-tag or reader-to-reader interferences [10][11][12][13]. Therefore, when setting up reader deployment for RFID networks, we must take all of the factors (the reader-to-tag and reader-to-reader interferences, success ratios of tag detection and deployment cost) into careful considerations.…”

“…Established methods for RFID network topology designs are all optimizationbased, utilizing such optimization techniques as the Genetic Algorithms ( GA) [14], Genetic Annealing Algorithms (GAA) [15,10] and Improved Genetic Annealing Algorithms (IGAA) [10]. Optimization-based methods usually take conspicuous complexity in the heuristic search for optimal or sub-optimal solutions.…”

“…Optimization-based methods usually take conspicuous complexity in the heuristic search for optimal or sub-optimal solutions. They may also turn over unfavorable results due to improper utilization or concurrent consideration of the six objectives (the overlapping of the reading areas, the number of useless readers, the number of redundant readers, the number of tags located in the overlap reading areas, the number of uncovered tags and the deployment cost [10,14]) which are covered by the multi-objective fitness function.…”