Balancing Data Privacy and 5G VNFs Security Monitoring: Federated Learning with CNN + BiLSTM + LSTM Model

Table 11

Proposed model compared with previous works (traditional learning).

Reference	Year	Technique used	Dataset utilized	Accuracy (%)	FPR (%)
Proposed model	2023	CNN-BiLSTM-LSTM	InSDN; CICIDS2017	99.99; 99.68	0.0089; 0.39
[31]	2022	CNN	CICIDS2017	99.0	0.67
[26]	2020	DBN	CICIDS2017	97.73	—
[24]	2021	Hybrid CNN	InSDN	99.28	—
[51]	2021	DFFNN	NSL-KDD; UNSW-NB15	99.0; 98.9	1.0; 1.1
[52]	2022	LSTM-BiLSTM-GRU-BiGRU	NSL-KDD; UNSW-NB15	87.44; 82.46	20.47; 37.61
[53]	2022	MLP	UNSW-NB15	96.7	—