Reinforcement learning-based method and system for UAV-assisted IoT anti-hostile interference

Abstract

The invention discloses an anti-hostile interference method based on reinforcement learning based on UAV-assisted Internet of Things, comprising: establishing an attack-defense Stackelberg game model against hostile interference, wherein ground sensor nodes, UAVs and intelligent jamming machines are the three participants in the game. The game equilibrium point and its existence conditions in the anti-hostile interference attack and defense Stackelberg game model are derived, and the game equilibrium point includes the optimal interference power of the intelligent jammer, the optimal transmission power of the UAV, and the Optimal moving distance and optimal transmit power of ground sensor nodes; under the condition of unknown interference model, WoLF‑PHC algorithm is introduced to dynamically optimize the transmit power of ground sensor nodes, the transmit power of UAV and the movement trajectory of UAV. The invention discloses a method and system for anti-hostile interference based on reinforcement learning based on drone-assisted Internet of Things, which can resist intelligent interference with variable interference signal strength by adjusting the trajectory or transmission power of the drone in time.

Description

technical field [0001] The invention relates to the technical field of the Internet of Things, in particular to a method and system for anti-hostile interference based on the reinforcement learning of the drone-assisted Internet of Things. Background technique [0002] Using the controllable mobility of unmanned aerial vehicles (UAVs), the limited coverage of the Internet of Things (IoTs) can be solved, and the combination of IoTs and UAVs can realize more diverse IoT applications . In some places with complex geographical environments, such as disaster areas, highways, and downtowns, the direct link from ground sensor nodes (GSN) to base stations (BS) in IoT devices is damaged, and UAV can be used as a Relay to assist the smooth communication of the ground network. However, the broadcast nature of radio propagation makes UAV communication vulnerable to hostile jamming attacks, which can affect normal communication once the wireless link is jammed. [0003] Existing unman...

AI Technical Summary

Problems solved by technology

However, the broadcast nature of radio propagation makes UAV communication vulnerable to hostile jamming attacks. Once the wireless link is interfered, normal communication will be affected

[0003] Existing unmanned aerial vehicles (UAV) anti-hostile jamming schemes usually use a single flight trajectory or frequency hopping strategies, which can resist attacks with fixed jamming power, but are resistant to intelligent jammers (Jammers) with variable jamming signal strength. ) at a loss

Secondly, the existing scheme assumes that the UAV has known information about the changes in the external environment. Due to the time-varying nature of the wireless channel, the high mobility of the UAV, and the variable interference intensity at any time, the UAV cannot accurately obtain information such as the current channel gain or interference signal strength.

Correspondingly, UAVs cannot adjust their trajectory or transmission power in time to resist intelligent interference

Images