Title: LDSAE: LeNet deep stacked autoencoder for secure systems to mitigate the errors of jamming attacks in cognitive radio networks

Authors: Chandrakant Athavale Chhaya; K.P. Patil

Addresses: Sinhgad Academy of Engineering, Pune, 411048, India ' Department of E&TC, Sinhgad Academy of Engineering, Pune, 411048, India

Abstract: A hybrid network system for mitigating errors due to jamming attacks in cognitive radio networks (CRNs) is named LeNet deep stacked autoencoder (LDSAE) and is developed. In this exploration, the sensing stage and decision-making are considered. The sensing unit is composed of four steps. First, the detected signal is forwarded to filtering progression. Here, BPF is utilised to filter the detected signal. The filtered signal is squared in the second phase. Third, signal samples are combined and jamming attacks occur by including false energy levels. Last, the attack is maliciously affecting the FC decision in the fourth step. On the other hand, FC initiated the decision-making and also recognised jamming attacks that affect the link amidst PU and SN in decision-making stage and it is accomplished by employing LDSAE-based trust model where the proposed module differentiates the malicious and selfish users. The analytic measures of LDSAE gained 79.40%, 79.90%, and 78.40%.

Keywords: CRNs; cognitive radio networks; FC; fusion center; band pass filter; LeNet; DSAE; deep stacked auto encoder.

DOI: 10.1504/IJNVO.2024.142242

International Journal of Networking and Virtual Organisations, 2024 Vol.31 No.2, pp.127 - 146

Received: 10 Feb 2024
Accepted: 21 Jun 2024
Published online: 15 Oct 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article