Jamming mitigation in cognitive radio networks using a modified Q-learning algorithm

The jamming attack is one of the most severe threats in cognitive radio networks, because it can lead to network degradation and even denial of service. However, a cognitive radio can exploit its ability of dynamic spectrum access and its learning capabilities to avoid jammed channels. In this paper, we study how Q-learning can be used to learn the jammer strategy in order to pro-actively avoid jammed channels. The problem with Q-learning is that it needs a long training period to learn the behavior of the jammer. To address the above concern, we take advantage of the wideband spectrum sensing capabilities of the cognitive radio to speed up the learning process and we make advantage of the already learned information to minimize the number of collisions with the jammer during training. The effectiveness of this modified algorithm is evaluated by simulations in the presence of different jamming strategies and the simulation results are compared to the original Q-learning algorithm applied to the same scenarios.