2019/04/08 by Nguyễn Văn Huynh, Diep N. Nguyen, Van Huynh, Nguyen +5 · 1 citation
Computer Science · Engineering · #Artificial Intelligence (cs.AI) #Energy Harvesting in Wireless Networks #FOS: Computer and information sciences #Networking and Internet Architecture (cs.NI) #Security in Wireless Sensor Networks #Wireless Communication Security Techniques
paper · pdf · doi:10.48550/arxiv.1904.03897
openalex publication_date 2019/04/08 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
With conventional anti-jamming solutions like frequency hopping or spread\nspectrum, legitimate transceivers often tend to "escape" or "hide" themselves\nfrom jammers. These reactive anti-jamming approaches are constrained by the\nlack of timely knowledge of jamming attacks. Bringing together the latest\nadvances in neural network architectures and ambient backscattering\ncommunications, this work allows wireless nodes to effectively "face" the\njammer by first learning its jamming strategy, then adapting the rate or\ntransmitting information right on the jamming signal. Specifically, to deal\nwith unknown jamming attacks, existing work often relies on reinforcement\nlearning algorithms, e.g., Q-learning. However, the Q-learning algorithm is\nnotorious for its slow convergence to the optimal policy, especially when the\nsystem state and action spaces are large. This makes the Q-learning algorithm\npragmatically inapplicable. To overcome this problem, we design a novel deep\nreinforcement learning algorithm using the recent dueling neural network\narchitecture. Our proposed algorithm allows the transmitter to effectively\nlearn about the jammer and attain the optimal countermeasures thousand times\nfaster than that of the conventional Q-learning algorithm. Through extensive\nsimulation results, we show that our design (using ambient backscattering and\nthe deep dueling neural network architecture) can improve the average\nthroughput by up to 426% and reduce the packet loss by 24%. By augmenting the\nambient backscattering capability on devices and using our algorithm, it is\ninteresting to observe that the (successful) transmission rate increases with\nthe jamming power. Our proposed solution can find its applications in both\ncivil (e.g., ultra-reliable and low-latency communications or URLLC) and\nmilitary scenarios (to combat both inadvertent and deliberate jamming).\n