| 講演抄録/キーワード |
| 講演名 |
2013-03-01 14:00
A Reinforcement Learning Based Sensing Policy for Cognitive Radio Systems ○Fereidoun H. Panahi・Tomoaki Ohtsuki(Keio Univ.) RCS2012-363 |
| 抄録 |
(和) |
Spectrum sensing is a critical function for enabling dynamic spectrum access (DSA) in wireless networks that utilize cognitive radio (CR). In DSA networks, an unlicensed secondary user can gain access to a licensed spectrum band as long as it does not cause harmful interference to primary user (PU). As a result, an efficient sensing scheme is essential for the secondary user in making sound opportunistic spectrum access decisions in a cognitive radio network. In this report, we present a learning based scheme for spectrum (channel) sensing in CR network. Specifically, we formulate the channel sensing problem as a partially observable Markov decision process (POMDP), where the most likely channel state is derived by a learning process called Fuzzy Q-Learning (FQL). Then, we use Baum-Welch Algorithm (BWA) for Markov sources in noise to estimate channel state transition probabilities of the PU which are used in the POMDP framework. |
| (英) |
Spectrum sensing is a critical function for enabling dynamic spectrum access (DSA) in wireless networks that utilize cognitive radio (CR). In DSA networks, an unlicensed secondary user can gain access to a licensed spectrum band as long as it does not cause harmful interference to primary user (PU). As a result, an efficient sensing scheme is essential for the secondary user in making sound opportunistic spectrum access decisions in a cognitive radio network. In this report, we present a learning based scheme for spectrum (channel) sensing in CR network. Specifically, we formulate the channel sensing problem as a partially observable Markov decision process (POMDP), where the most likely channel state is derived by a learning process called Fuzzy Q-Learning (FQL). Then, we use Baum-Welch Algorithm (BWA) for Markov sources in noise to estimate channel state transition probabilities of the PU which are used in the POMDP framework. |
| キーワード |
(和) |
Cognitive Radio (CR) / Reinforcement learning (RL) / Baum-Welch Algorithm (BWA) / / / / / |
| (英) |
Cognitive Radio (CR) / Reinforcement learning (RL) / Baum-Welch Algorithm (BWA) / / / / / |
| 文献情報 |
信学技報, vol. 112, no. 443, RCS2012-363, pp. 471-476, 2013年2月. |
| 資料番号 |
RCS2012-363 |
| 発行日 |
2013-02-20 (RCS) |
| ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
| PDFダウンロード |
RCS2012-363 |