Cost and Lightweight Modeling Analysis of RFID Authentication Protocols in Resource Constraint Internet of Things

Abstract

Internet of Things (IoT) is a pervasive environment to interconnect the things like: smart objects, devices etc. in a structure like internet. Things can be interconnected in IoT if these are uniquely addressable and identifiable. Radio Frequency Identification (RFID) is one the important radio frequency based addressing scheme in IoT. Major security challenge in resource constraint RFID networks is how to achieve traditional CIA security i.e. Confidentiality, Integrity and Authentication. Computational and communication costs for Lightweight Mutual Authentication Protocol (LMAP), RFID mutual Authentication Protocol with Permutation (RAPP) and kazahaya authentication protocols are analyzed. These authentication protocols are modeled to analyze the delays using lightweight modeling language. Delay analysis is performed using alloy model over LMAP, RAPP and kazahaya authentication protocols where one datacenter (DC) is connected to different number of readers (1,5 or 10) with connectivity to 1, 5 or 25 tags associated with reader and its results show that for LMAP delay varies from 30-156 msec, for RAPP from 31-188 while for kazahaya from 61-374 msec. Further, performance of RFID authentication protocols is analyzed for group construction through more than one DC (1,5 or 10) with different number of readers (10, 50 or 100) and tags associated with these readers (50, 500, 1000) and results show that DC based binary tree topology with LMAP authentication protocol is having a minimum delay for 50 or 100 readers. Other authentication protocols fail to give authentication results because of large delays in the network. Thus, RAPP and Kazahaya are not suitable for scenarios where there is large amount of increase in number of tags or readers.

Availability Aspects Through Optimization Techniques Based Outlier Detection Mechanism in Wireless and Mobile Networks

“…9 shows comparative throughput analysis of the proposed protocol with Kumar et al protocol [1]. Results show that the proposed protocol is having better throughput compared to Kumar et al protocol [46][47][48][49][50][51][52][53] because of connected hierarchical network construction and consideration of small-scale (50 to 500 nodes) to large scale (3000 to 5000 nodes) networks. Fig.…”