| 1. |
- Qi, Nan, et al.
(författare)
-
A Learning-Based Spectrum Access Stackelberg Game : Friendly Jammer-Assisted Communication Confrontation
- 2021
-
Ingår i: IEEE Transactions on Vehicular Technology. - : IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC. - 0018-9545 .- 1939-9359. ; 70:1, s. 700-713
-
Tidskriftsartikel (refereegranskat)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.
|
|
| 2. |
- Qi, Nan, et al.
(författare)
-
Two Birds With One Stone : Simultaneous Jamming and Eavesdropping With the Bayesian-Stackelberg Game
- 2021
-
Ingår i: IEEE Transactions on Communications. - : Institute of Electrical and Electronics Engineers (IEEE). - 0090-6778 .- 1558-0857. ; 69:12, s. 8013-8027
-
Tidskriftsartikel (refereegranskat)abstract
- In adversarial scenarios, it is crucial to timely monitor what tactical messages that opponent transmitters are sending to intended receiver(s), and disrupt the transmissions immediately if in need. The issue becomes more challenging in face of an intelligent transmitter. To address the above-stated issue, a full-duplex (FD) technique is utilized to enable simultaneous jamming and eavesdropping (SJE) at a friendly jammer node. In particular, the "Two Birds with One Stone" strategy is utilized at the jammer node to realize effective rate degradation and information eavesdropping. A confrontation game between an intelligence-empowered FD jammer and its opponent is investigated. Specifically, to capture their adversarial relationship in an environment with incomplete information, a power-domain Bayesian-Stackelberg game is proposed. The existence of a Stackelberg equilibrium (SE) power solution is proved. The semi-closed-form solutions of SE are derived, which are proved to be asymptotically optimal (have a gap of less than 1% with the exact utility), and improves the jammer node 10% utility compared with the Nash equilibrium. Additionally, the SJE strategy outperforms the half-duplex (HD) and other benchmark schemes.
|
|
| 3. |
- Wang, Shuqi, et al.
(författare)
-
Trajectory Planning for UAV-Assisted Data Collection in IoT Network: A Double Deep Q Network Approach
- 2024
-
Ingår i: Electronics. - : Multidisciplinary Digital Publishing Institute (MDPI). - 2079-9292. ; 13:8
-
Tidskriftsartikel (refereegranskat)abstract
- Unmanned aerial vehicles (UAVs) are becoming increasingly valuable as a new type of mobile communication device and autonomous decision-making device in many application areas, including the Internet of Things (IoT). UAVs have advantages over other stationary devices in terms of high flexibility. However, a UAV, as a mobile device, still faces some challenges in optimizing its trajectory for data collection. Firstly, the high complexity of the movement action and state space of the UAV’s 3D trajectory is not negligible. Secondly, in unknown urban environments, a UAV must avoid obstacles accurately in order to ensure a safe flight. Furthermore, without a priori wireless channel characterization and ground device locations, a UAV must reliably and safely complete the data collection from the ground devices under the threat of unknown interference. All of these require the proposing of intelligent and automatic onboard trajectory optimization techniques. This paper transforms the trajectory optimization problem into a Markov decision process (MDP), and deep reinforcement learning (DRL) is applied to the data collection scenario. Specifically, the double deep Q-network (DDQN) algorithm is designed to address intelligent UAV trajectory planning that enables energy-efficient and safe data collection. Compared with the traditional algorithm, the DDQN algorithm is much better than the traditional Q-Learning algorithm, and the training time of the network is shorter than that of the deep Q-network (DQN) algorithm.
|
|
