Early attack detection and resolution in sensor nodes to improve IoT security

The expanding Internet of Things (IoT) and the sensor network subsystem they are bound to collaboratively contribute to the growing smart technology ecosystem that uses wireless sensor networks presenting the fastest growing attack surface for malware. The smart technology ecosystem integrates processes in the home consumer sector, industry, and economic environments through sensors and wireless sensor networks (WSN). The system is exposed to an expanded attack surface that introduces new attack vectors which are exploited in novel ways for which new mitigative measures are needed. The review aims to find out how latest technologies already known to be effective for attack detection in other realms can be utilised for early attack detection on sensor nodes deployed in WSN and IoST. A review of recent systematically compiled articles on effective detection of attacks on sensors is done focusing on usage of new technologies and their effectiveness. The work showed that early attack detection is feasible and effective. Designs that use contractual models built through integration of AI, ML, or block chain with sensor node sourced data for training and the use of established data sets succeeded in the tasked detection functions. The research intended to find how detecting attacks early in IoT sensor nodes could be used in an advantageous way to mitigate against attacks on the IoT ecosystem. The literature shows there are process with that capability that can still be enhanced to perform better.