Intelligent Anti-jamming based on Deep Reinforcement Learning and Transfer Learning

One of the security issues in a wireless network is jamming attacks, where the jammer causes congestion and significant decrement in the network throughput by obstructing channels and disrupting user signals. Recent works have proposed using a deep-reinforcement learning (DRL) model to confront the jamming attacker due to its capability in predicting the jammer decisions and pattern recognition. Training a DRL model from scratch may take a long time. We first propose a recurrent neural network architecture to minimize the number of parameters for training a DRL model. We further propose a transfer learning (TL) approach to enable the DRL agent to learn fast in dynamic wireless networks to confront jamming attacks effectively. To make our proposed TL method adaptive to different network environments, we propose a novel method to quantitatively measure the difference between the source and target domains, using an integrated feature extractor. Afterward, based on the measured difference, we can choose an optimal setting for the TL model. We also show that the proposed TL method can effectively reduce the training time for the DRL model and outperforms other existing TL methods.