Comprehensive Assessment of Security Attack Detection Algorithms in Internet of Things

Abstract

The Internet of Things (IoT) is becoming an important part of every business domain including health, education, industrial manufacturing, agriculture and domestic products. These applications of IoT raise concerns over security of IoT system. Several security attacks on IoT systems have been demonstrated in the state of the art. Further multiple attackers attacking IoT system collaboratively is more hazardous and stealthy. This paper reviews and evaluates the state of the art for security attack detection mechanisms on IoT protocol stack. Evaluation of the effectiveness of attack detection mechanism through an extensive simulation study is presented in the paper. Results of simulation study highlight the failure of attack detection mechanisms presented in the state of the art in case of multiple attackers attacking IoT system collaboratively. Simulation study also shows considerable deviations in network parameters in IoT system under attack of multiple collaborative attackers from normal operation. These deviations in parameters can be used for early and effective detection of security attacks on IoT system. This paper highlights need for the design of an early detection algorithm for security attacks on IoT system.