eduzhai > Applied Sciences > Engineering >

Resource Allocation in Uplink NOMA-IoT Networks A Reinforcement-Learning Approach

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 32 pages

Abstract: Non-orthogonal multiple access (NOMA) exploits the potential of the powerdomain to enhance the connectivity for the Internet of Things (IoT). Due totime-varying communication channels, dynamic user clustering is a promisingmethod to increase the throughput of NOMA-IoT networks. This paper develops anintelligent resource allocation scheme for uplink NOMA-IoT communications. Tomaximise the average performance of sum rates, this work designs an efficientoptimization approach based on two reinforcement learning algorithms, namelydeep reinforcement learning (DRL) and SARSA-learning. For light traffic,SARSA-learning is used to explore the safest resource allocation policy withlow cost. For heavy traffic, DRL is used to handle traffic-introduced hugevariables. With the aid of the considered approach, this work addresses twomain problems of fair resource allocation in NOMA techniques: 1) allocatingusers dynamically and 2) balancing resource blocks and network traffic. Weanalytically demonstrate that the rate of convergence is inversely proportionalto network sizes. Numerical results show that: 1) Compared with the optimalbenchmark scheme, the proposed DRL and SARSA-learning algorithms have lowercomplexity with acceptable accuracy and 2) NOMA-enabled IoT networks outperformthe conventional orthogonal multiple access based IoT networks in terms ofsystem throughput.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...