動態(tài)頻譜接入中基于最小貝葉斯風險的穩(wěn)健頻譜預測
doi: 10.11999/JEIT170519
基金項目:
國家自然科學基金(61471395, 61471392, 61301161)，江蘇省自然科學基金(BK20141070)
Minimum Bayesian Risk Based Robust Spectrum Prediction in Dynamic Spectrum Access
-
(School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China)
Funds:
The National Natural Science Foundation of China (61471395, 61471392, 61301161), The Natural Science Foundation of Jiangsu Province (BK20141070)
-
摘要: 針對頻譜感知錯誤累積造成頻譜預測性能惡化問題，該文提出一種基于最小貝葉斯風險的穩(wěn)健頻譜預測策略。分布擬合檢驗表明頻譜預測輸出服從正態(tài)分布，定義頻譜預測輸出的貝葉斯風險函數(shù)，證明使貝葉斯風險函數(shù)最小的頻譜預測輸出判決門限將使頻譜預測的均方誤差最小，求得了使貝葉斯風險最小的最優(yōu)判決門限，構(gòu)建穩(wěn)健頻譜預測策略。仿真結(jié)果表明，與固定判決門限的神經(jīng)網(wǎng)絡頻譜預測相比，穩(wěn)健頻譜預測策略改進了頻譜感知錯誤下的頻譜預測性能，改善了非授權(quán)用戶的動態(tài)頻譜接入性能。
-
關鍵詞:
- 動態(tài)頻譜接入 /
- 穩(wěn)健頻譜預測 /
- 神經(jīng)網(wǎng)絡 /
- 貝葉斯風險 /
- 預測準確率
Abstract: The accumulation of miss detection and false alarm in spectrum sensing leads to the persistently decreasing of prediction accuracy in spectrum prediction. This paper takes neural network based spectrum prediction for example, and presents a minimum Bayesian Risk based spectrum prediction to solve this problem. The distribution fitting shows that the prediction output follows the normal distribution. The expectation of prediction mean square error is defined as the Bayesian Risk, and the optimal detection threshold of the prediction output is derived through minimizing the Bayesian Risk. Through this method, the prediction accuracy is insensitive to the spectrum sensing errors. Compared with the traditional spectrum prediction with fixed detection thresholds, simulation results demonstrate the robust spectrum prediction keeps the prediction accuracy stable, and improve the performance in dynamic spectrum access. -
NING Guoqin and NINTANAVONGSA P. Time prediction based spectrum usage detection in centralized cognitive radio networks[C]. 2012 IEEE Wireless Communications and Networking Conference (WCNC), Shanghai, China, 2012: 300-305. doi: 10.1109/WCNC.2012.6214320. WANG Tan, LI Gen, and DING Jiaxin. 5G Spectrum: Is China ready?[J]. IEEE Communications Magazine, 2015, 53(7): 58-65. doi: 10.1109/MCOM.2015.7158266. SONG Yi and XIE Jiang. ProSpect: A proactive spectrum handoff framework for cognitive radio ad hoc networks without common control channel[J]. IEEE Transactions on Mobile Computing, 2012, 11(7): 1127-1139. doi: 10.1109/ TMC.2011.140. GUAN Quansheng, RICHARD Y F, and JIANG Shengming. Prediction-based topology control and routing in cognitive radio mobile Ad hoc networks[J]. IEEE Transactions on Vehicular Technology, 2011, 59(9): 4443-4452. doi: 10.1109/ TVT.2010.2069105 GHOSH A and SARKAR S. Quality-sensitive price competition in secondary market spectrum oligopoly-single location game[J]. IEEE/ACM Transactions on Networking, 2016, 24(3): 1894-1907. doi: 10.1109/TNET.2015.2440422. GHOSH A, SARKAR S, and BERRY R. The value of side information in secondary spectrum markets[J]. IEEE Journal on Selected Areas in Communications, 2017, 35(1): 6-19. doi: 10.1109/JSAC.2016.2632579. KUMAR A, SINGH S, and ZHENG Haitao. Reliable open spectrum communications through proactive spectrum access[C]. 2006 International Workshop on Technology and Policy for Accessing Spectrum (TAPAS), Boston, American, 2006: 5. doi: 10.1145/1234388.1234393. ELTOM H, KANDEEPAN S, LIANG Yingchang, et al. HMM based cooperative spectrum occupancy prediction using hard fusion[C]. 2016 IEEE International Conference on Communications Workshops (ICC), Kuala Lumpur, Malaysia, 2016: 669-675. doi: 10.1109/ICCW.2016.7503864. TUMULURU V, WANG Ping, and NIYATO D. Channel status prediction for cognitive radio networks[J]. Wireless Communications and Mobile Computing, 2012, 12(10): 862-874. doi: 10.1002/wcm.1017. SU Jinzhao and WU Wei. Wireless spectrum prediction model based on time series analysis method[C]. 2009 ACM workshop on Cognitive radio networks, Beijing, China (CoRoNET), Beijing, China, 2009: 61-66. doi: 10.1145/ 1614235.1614250. WEN Zheng, LUO Tian, and WANG Xiang. Autoregressive spectrum holes prediction model for cognitive radio systems[C]. 2008 IEEE Communications Workshop (ICC), Beijing, China, 2008: 154-157. doi: 10.1109/ICCW.2008.34. TANG Mengyun, DING Guoru, WU Qihui, et al. A joint tensor completion and prediction scheme for multi- dimensional spectrum map construction[J]. IEEE Access, 2016(4): 8044-8052. doi: 10.1109/ACCESS.2016.2627243. RAZALI N M and WAH Y B. Power comparisons of shapiro- wilk, kolmogorov-smirnov, lilliefors and anderson-darling tests[J]. Journal of Statistical Modeling and Analytics, 2011, 2(1): 21-33. BAO Yong. On sample skewness and kurtosis[J]. Econometric Reviews, 2013, 32(4): 415-448. doi: 10.1080/07474938.2012. 690665. MASONTA M, MZYECE M, and NTLATLAPA N. Spectrum decision in cognitive radio networks: A survey[J]. IEEE Communications Surveys Tutorials, 2013, 15(3): 1088-1107. doi: 10.1109/SURV.2012.111412.00160. -
計量
- 文章訪問數(shù): 1281
- HTML全文瀏覽量: 153
- PDF下載量: 185
- 被引次數(shù): 0
下載:
