Intelligent Intrusion Detection Based on Federated Learning for Edge-Assisted Internet of Things

Table 3

Comparison of overall classification accuracy between the CL model and the FL model.

Method; (B = 128, E = 1, K = 10, C = 1, η = 1e − 2)	Accuracy (%); (R = 10)	Accuracy (%); (R = 20)	Accuracy (%); (R = 40)
CL-CNN	98.97	99.40	99.65
FedAVG (CNN)	98.13	98.58	98.90
FedACNN (proposed)	98.73	99.02	99.12