Deep Learning Based Channel Estimation and Artificial Noise Aided Security for NOMA-QSM System

• Published in: SN computer science Vol. 6; no. 3; p. 244
• Main Authors: Singh, Shekhar Pratap, Pradhan, Pyari Mohan
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 01.03.2025; Springer Nature B.V
• Subjects: Antennas; Codes; Communication; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Cyber-Physical System and Artificial Intelligence: Trends and Challenges; Data Structures and Information Theory; Deep learning; Design; Eavesdropping; Efficiency; Information Systems and Communication Service; Noise levels; Nonorthogonal multiple access; Original Research; Pattern Recognition and Graphics; Quadratures; Receivers & amplifiers; Signal processing; Signal to noise ratio; Simulation; Software Engineering/Programming and Operating Systems; Sums; Transmitters; Vision; Wireless networks; Deep learning; NOMA; Artificial Noise; Secrecy rate; Quadrature spatial modulation; Channel estimation
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-025-03793-w

In order to avoid eavesdropping with no additional hardware, a secure non-orthogonal multiple access based quadrature spatial modulation (NOMA-QSM) scheme is proposed. Artificial noise (AN) is added to the information-carrying signal at the transmitter, causing interference at the eavesdropper while not affecting the legitimate users. A deep learning (DL) based channel estimation is also proposed with the rectified linear unit. Simulation results show that due to the addition of AN, the sum secrecy rate of the system increases to 1.6 bits/s/Hz at a 40 dB signal-to-noise ratio (SNR). The sum secrecy rate of the system increases as the number of transmitting antennas (TAs) increases. The proposed DL-based channel estimation scheme provides comparable performance as that of the minimum mean square estimator at low and high SNR values. At 20 dB SNR, the mean square estimation error for the proposed scheme is approximately -17 dB when two TAs are used. This study also analyzes the effect of imperfect channel state information and imperfect successive interference cancellation on the secrecy rate of users.