Table 1 Summary of existing studies.

Ref.	Components of security framework	AI-based intrusion detection	TM	ECC	TC	TF	TP	TA	TU	RE	RTP	Technique / Algorithm used	Key findings	Strengths	Weaknesses
¹³	√	√	x	x	√	x	√	x	√	x	x	CNN-IDS with static rule-based trust	Improved anomaly detection; 91% Acc	Effective IDS; low FP	No adaptive trust; higher latency
²⁹	√	√	√	x	√	√	x	x	√	x	√	LSTM + Bayesian TMS	Dynamic trust updates; 92% Acc	Resilient to insider threats	Moderate CPU use; complex update cycle
³⁰	√	x	√	x	√	√	√	x	x	√	x	Hybrid RNN-TMS model	Better recall in dynamic nodes	Fast adaptation to new threats	No ECC support; high energy use
¹⁴	x	√	√	x	x	x	√	√	√	x	√	Blockchain + Reputation Trust	Secure trust ledger; 90% Acc	Tamper-Proof; decentralized	High Computation; latency overhead
³¹	√	√	x	√	√	√	x	√	x	√	x	CNN-IDS + ECC auth	93% Acc; low overhead	Low-cost crypto integration	No trust management module
¹⁵	x	√	√	√	x	√	√	x	√	x	√	ECC + TMS integration	94% Acc; high reliability	Balanced efficiency	Lacks IDS adaptability
³²	√	x	√	x	√	√	x	x	√	x	√	Autoencoder anomaly detection	Detects zero-day attacks	High sensitivity	High false positives
³³	√	√	x	√	√	x	√	x	√	√	x	CNN + ECC framework	92% Acc; secure channel	Lightweight; fast auth	Limited scalability
³⁴	x	x	√	x	x	√	√	x	x	√	√	Reputation-based trust model	Dynamic reputation updates	Low computation cost	No IDS/ECC support
³⁵	√	√	√	x	√	√	x	√	√	x	√	Federated DL IDS	94% Acc; privacy preserved	Scalable; data-local	Communication overhead

Quick links

Search