eduzhai > Applied Sciences > Engineering >

Deep Reinforcement Learning for Dynamic Spectrum Sensing and Aggregation in Multi-Channel Wireless Networks

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: In this paper, the problem of dynamic spectrum sensing and aggregation isinvestigated in a wireless network containing N correlated channels, wherethese channels are occupied or vacant following an unknown joint 2-state Markovmodel. At each time slot, a single cognitive user with certain bandwidthrequirement either stays idle or selects a segment comprising C (C < N)contiguous channels to sense. Then, the vacant channels in the selected segmentwill be aggregated for satisfying the user requirement. The user receives abinary feedback signal indicating whether the transmission is successful or not(i.e., ACK signal) after each transmission, and makes next decision based onthe sensing channel states. Here, we aim to find a policy that can maximize thenumber of successful transmissions without interrupting the primary users(PUs). The problem can be considered as a partially observable Markov decisionprocess (POMDP) due to without full observation of system environment. Weimplement a Deep Q-Network (DQN) to address the challenge of unknown systemdynamics and computational expenses. The performance of DQN, Q-Learning, andthe Improvident Policy with known system dynamics is evaluated throughsimulations. The simulation results show that DQN can achieve near-optimalperformance among different system scenarios only based on partial observationsand ACK signals.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×