A Learning-Based Spectrum Access Stackelberg Game: Friendly Jammer-Assisted Communication Confrontation
Nan Qi,Wei Wang,Ming Xiao,Luliang Jia,Shi Jin,Qiuming Zhu,Theodoros A. Tsiftsis
DOI: https://doi.org/10.1109/tvt.2021.3049653
IF: 6.8
2021-01-01
IEEE Transactions on Vehicular Technology
Abstract:Defensive and offensive capabilities are both significant in communication confrontation games. By exploiting the above two capabilities, a new confrontation mechanism in the spectrum domain between two opposing teams denoted as the blue team (BT) and red team (RT), is designed. The basic idea is that by sacrificing parts of ally performance to severely deteriorate the opponent side communications. Specifically, a friendly and smart jammer (assuming in the BT) is deployed to weaken opponent (i.e., members in the RT) communications without causing great damages to other BT members, while the smart RT members try to evade the jamming and alleviate mutual interference. The interactions among the friendly jammer and other nodes are modeled as a Stackelberg game, with each player seeking for their respective utility maximization. We prove that each sub-game is an exact potential game. To efficiently search for the equilibrium solutions, a parallel log-linear learning algorithm is proposed, based on which each user intelligently decides their spectrum access policies. Numerical results demonstrate that: 1) RT communications are effectively suppressed; meanwhile, mutual interference among ally BT communication pairs is significantly alleviated; 2) the proposed algorithm achieves a close-to-optimal solution; 3) compared with the current state of solutions, i.e., random selection, stochastic learning automata, our algorithm performs better in terms of both utility and convergence.
telecommunications,engineering, electrical & electronic,transportation science & technology