SCHEISB: Design of a high efficiency IoMT security model based on sharded chains using bio-inspired optimizations

Abstract

Securing Internet-of-Medical Things (IoMT) devices is a multidomain task that involves analysis of attack patterns from data & control packets. Existing security models either work on pre-emptive analysis, real-time analysis or post-processing analysis of different packet types. These operations add to the complexity of data processing, which reduces Quality of Service (QoS) levels of IoMT deployments. To overcome these issues, this paper proposes design of a sharded chain based high efficiency IoMT security model that uses bioinspired optimizations. The proposed model initially deploys a novel single chained Proof-of-Medical-Trust (PoMT) based consensus technique, which assists in identification of high-trust miners. The QoS levels are continuously monitored for each block request, and a Teacher–Learner based Optimization (TLbO) model is deployed to make sharding decisions. These decisions are reconfirmed by a Particle Swarm Optimization (PSO) model, which assists in estimation of shard length under different attack types. These attacks include Spoofing, Spying, and Flooding attacks, which affects the authenticity of data samples from IoT devices under real-time scenarios. The PSO Model also recommends splitting & merging decisions based on the context of IoMT deployments. The PoMT consensus is modified to support sharded chains, and assists in reducing mining delays without compromising miner trust levels. The modified PoMT consensus is capable of selecting overlapping miner nodes, while incorporating load balancing for optimal energy consumption during the mining process. Due to these enhancements, the proposed model is able to reduce mining delay by 8.5%, while reducing energy consumption by 3.2%, and reducing access delay by 4.9% under real-time clinical scenarios.