eduzhai > Applied Sciences > Engineering >

Caching Placement and Resource Allocation for Cache-Enabling UAV NOMA Networks

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 15 pages

Abstract: This article investigates the cache-enabling unmanned aerial vehicle (UAV)cellular networks with massive access capability supported by non-orthogonalmultiple access (NOMA). The delivery of a large volume of multimedia contentsfor ground users is assisted by a mobile UAV base station, which caches somepopular contents for wireless backhaul link traffic offloading. Incache-enabling UAV NOMA networks, the caching placement of content cachingphase and radio resource allocation of content delivery phase are crucial fornetwork performance. To cope with the dynamic UAV locations and contentrequests in practical scenarios, we formulate the long-term caching placementand resource allocation optimization problem for content delivery delayminimization as a Markov decision process (MDP). The UAV acts as an agent totake actions for caching placement and resource allocation, which includes theuser scheduling of content requests and the power allocation of NOMA users. Inorder to tackle the MDP, we propose a Q-learning based caching placement andresource allocation algorithm, where the UAV learns and selects action with emph{soft ${ varepsilon}$-greedy} strategy to search for the optimal matchbetween actions and states. Since the action-state table size of Q-learninggrows with the number of states in the dynamic networks, we propose a functionapproximation based algorithm with combination of stochastic gradient descentand deep neural networks, which is suitable for large-scale networks. Finally,the numerical results show that the proposed algorithms provide considerableperformance compared to benchmark algorithms, and obtain a trade-off betweennetwork performance and calculation complexity.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...