Anti-Jamming Resource-Allocation Method in the EH-CIoT Network through LWDDPG Algorithm.

Area of Science:

• Wireless Communication Networks
• Cybersecurity
• Resource Management

Background:

• Energy-Harvesting Cognitive Internet of Things (EH-CIoT) networks face throughput degradation due to jamming attacks.
• Existing methods lack adaptability to unknown jammer strategies and primary user activity.

Purpose of the Study:

• To develop an anti-jamming resource allocation method for EH-CIoT networks.
• To maximize the Long-Term Throughput (LTT) despite jamming and energy constraints.

Main Methods:

• Modeling the problem as a Markov Decision Process (MDP) without prior knowledge.
• Designing a two-dimensional reward function incorporating throughput and energy rewards.
• Proposing a Linearly Weighted Deep Deterministic Policy Gradient (LWDDPG) algorithm for resource allocation.

Main Results:

• The LWDDPG algorithm effectively allocates channels, power, and work modes to Secondary Users (SUs).
• The method enables SUs to transmit on unjammed channels and harvest radio frequency energy.
• Simulation results validate the proposed method's superiority over traditional approaches against multiple jamming attacks.

Conclusions:

• The developed anti-jamming strategy enhances LTT in EH-CIoT networks.
• The LWDDPG algorithm provides an effective solution for resource allocation in dynamic and adversarial environments.
• The approach balances throughput maximization with energy harvesting for sustainable network operation.