MCS: A Distributed Multi-User Channel Selection Algorithm for Cognitive Radio Networks
Multi-arm bandit (MAB) theory have recently been in use to plan choice issues with exploration-exploitation tradeoff. Dynamic channel assignment in cognitive radio (CR) systems is one of critical applications. In this work, we quickly outline MAB issues with its conceivable applications to cognitive...
Uloženo v:
| Vydáno v: | 2019 International Conference on Information Technology (ICIT) s. 47 - 52 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Konferenční příspěvek |
| Jazyk: | angličtina |
| Vydáno: |
IEEE
01.12.2019
|
| Témata: | |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | Multi-arm bandit (MAB) theory have recently been in use to plan choice issues with exploration-exploitation tradeoff. Dynamic channel assignment in cognitive radio (CR) systems is one of critical applications. In this work, we quickly outline MAB issues with its conceivable applications to cognitive radio systems. We first demonstrate a MAB issue where a solitary client either explores a channel to assemble data to improve its present scenario, or exploits the channel officially chosen dependent on the data that it has as of now gathered. At that point we thought about a completely disseminated framework, where there's no involvement of a centralized element, with the qualities of each channel being unidentified and may differ for every client. We at last propose a MCS (Multi-User Channel Selection) Algorithm and perform simulation based on multi-player multiarmed bandit approach for dynamic cognitive Ad-hoc networks and compare the result with the existing Musical Chair (MC) approach. An extensive simulation is performed which proves that the proposed MCS algorithm is better than the existing approach. |
|---|---|
| DOI: | 10.1109/ICIT48102.2019.00015 |