Fusion of artificial intelligence and game theory for resource allocation in non‐orthogonal multiple access‐assisted device‐to‐device cooperative communication

• Published in: International journal of communication systems Vol. 36; no. 14
• Main Authors: Rathod, Tejal, Gupta, Rajesh, Kumar Jadav, Nilesh, Tanwar, Sudeep, Alabdulatif, Abdulatif, Sharma, Ravi
• Format: Journal Article
• Language: English
• Published: Chichester Wiley Subscription Services, Inc 25.09.2023
• Subjects: Accuracy; Artificial intelligence; Cellular communication; Communication; cooperative communication; D2D communication; Game theory; Interference; mobile telecommunication systems; NOMA; Nonorthogonal multiple access; Parameters; Performance degradation; Power management; Radio equipment; Reference signals; Resource allocation; Signal strength
• ISSN: 1074-5351; 1099-1131
• DOI: 10.1002/dac.5556

Summary Device‐to‐device (D2D) communication offers a low‐cost paradigm where two devices in close proximity can communicate without needing a base station (BS). It significantly improves radio resource allocation, channel gain, communication latency, and energy efficiency and offers cooperative communication to enhance the weak user's network coverage. The cellular mobile users (CMUs) share the spectral resources (e.g., power, channel, and spectrum) with D2D mobile users (DMUs), improving spectral efficiency. However, the reuse of radio resources causes various interferences, such as intercell and intracell interference, that degrade the performance of overall D2D communication. To overcome the aforementioned issues, this paper presents a fusion of AI and coalition game for secure resource allocation in non‐orthogonal multiple access (NOMA)‐based cooperative D2D communication. Here, NOMA uses the successive interference cancellation (SIC) technique to reduce the severe impact of interference from the D2D systems. Further, we utilized a coalition game theoretic model that efficiently and securely allocates the resources between CMUs and DMUs. However, in the coalition game, all DMUs participate in obtaining resources from CMUs, which increases the computational overhead of the overall system. For that, we employ artificial intelligence (AI) classifiers that bifurcate the DMUs based on their channel quality parameters, such as reference signal received power (RSRP), received signal strength indicator (RSSI), signal‐to‐noise ratio (SNR), and channel quality indicator (CQI). It only forwards the DMUs that have better channel quality parameters into the coalition game, thus reducing the computational overhead of the overall D2D communication. The performance of the proposed scheme is evaluated using various statistical metrics, for example, precision score, accuracy, recall, F1 score, overall sum rate, and secrecy capacity, where an accuracy of 99.38% is achieved while selecting DMUs for D2D communication. We present a coalition game theoretic model for resource allocation in NOMA‐based D2D cooperative communication. We simulated the D2D cooperative communication scenario in the virtual environment of MATLAB. Performance of the proposed scheme is analyzed by considering various performance metrics like sum rate, accuracy score, precision, recall, and F1 score