基于贝叶斯准则的随机共振算法研究

刘书君; 杨婷; 唐明春; 王品; 李勇明

doi:10.11999/JEIT160361

基于贝叶斯准则的随机共振算法研究

doi: 10.11999/JEIT160361

基金项目:

重庆市基础与前沿研究(cstc2016jcyjA0134, cstc2016 jcyjA0043)，国家自然科学基金(61501072, 61301224, 41404027, 61108086, 61471072)，重庆市社会事业与民生保障专项(cstc2016 shmszx40002)，中央高校重点基金(CDJZR155507)

计量
- 文章访问数: 1269
- HTML全文浏览量: 124
- PDF下载量: 459
- 被引次数: 0
出版历程
- 收稿日期: 2016-04-14
- 修回日期: 2016-10-18
- 刊出日期: 2017-02-19

Study on Stochastic Resonance Algorithm Based on Bayes Criterion

Funds:

The Basic and Advanced Research Project in Chongqing (cstc2016jcyjA0134, cstc2016jcyjA0043), The National Natural Science Foundation of China (61501072, 61301224, 41404027, 61108086, 61471072), The Chongqing Social Undertaking and People,s Livelihood Guarantee Science and Technology Innovation Special Foundation (cstc2016shmszx40002), The Fundamental Research Funds for the Central Universities (CDJZR155507)

摘要

摘要: 该文针对二元假设检验问题，首先在贝叶斯准则的基础上，分析了最小化贝叶斯代价所对应的最优噪声，将贝叶斯代价的最小化问题等价为虚警概率和/或检测概率的最优化。其次，在保证一定虚警概率和检测概率的前提下，建立起能同时改善检测概率和虚警概率的模型。然后分别给出当检测概率一定时虚警概率最小和虚警概率一定时检测概率最大这两种极限情况下对应的最优加性噪声，并对其进行线性凸组合以获得模型所需的最优加性噪声，进一步分析并证明了该模型能够成立的充分条件。再次，获得先验概率已知和未知两种情况下最小化贝叶斯代价时所对应的加性噪声，且当先验知识发生改变时，该算法只需调整加性噪声中一个可变参数即可获得相应的最优贝叶斯代价。最后，结合具体的检测问题，通过仿真验证了所提算法的有效性。
- 信号处理 /
- 贝叶斯准则 /
- 噪声增强模型 /
- 加性噪声 /
- 假设检验
Abstract: The optimal noise that minimizes Bayes risk for a binary hypothesis testing problem is analyzed firstly. As a result, the minimization of Bayes risk can be equivalent as the optimization of the detection probability and/or false alarm probability . Secondly, a noise enhanced model, which can increase and decrease simultaneously, is established under the premise of maintaining predefined and . Then the optimal additional noise of this model is obtained by a convex combination of two optimal noises of two limit cases, which are the minimization of with maintaining the predefined and the maximization of with maintaining the predefined , respectively. Furthermore, the sufficient conditions for this model are given. Whats more, the additive noise that minimizes the Bayes risk is determined when the prior probabilities are known or not, and the corresponding additive noise can be obtained by recalculating a parameter only if the prior information changes. Finally, the availability of algorithm is proved through the simulation combined with a specific detection example.
- Signal processing /
- Bayes criterion /
- Noise enhanced model /
- Additive noise /
- Hypothesis testing

HTML全文

参考文献(24)

COHEN L. The history of noise[J]. IEEE Signal Processing Magazine, 2005, 22(6): 20-45. doi: 10.1109/MSP.2005. 1550188.

BENZI R, SUTERA A, and VULPIANI A. The mechanism of stochastic resonance[J]. Journal of Physics A: Mathematical General, 1981, 14(11): L453-L457. doi: 10.1088 /0305-4470/14/11/006.

张雷, 宋爱国. 随机共振在信号处理中应用研究的回顾与展望[J]. 电子学报, 2009, 37(4): 811-818. doi: 10.3321/j.issn: 0372-2112.2009.04.025.

ZHANG Lei and SONG Aiguo. Development and prospect of stochastic resonance in signal processing[J]. Acta Electronica Sinica, 2009, 37(4): 811-818. doi: 10.3321/j.issn:0372-2112. 2009.04.025.

ADDESSOA P, PIERROB V, and FILATRELLA G. Interplay between detection strategies and stochastic resonance properties[J]. Communications in Nonlinear Science Numerical Simulation, 2015, 30(1/3): 15-31. doi: 10.1016/j.cnsns.2015.05.026.

YU Haitao, GUO Xinmeng, WANG Jiang, et al. Adaptive stochastic resonance inself-organized small-world neuronal networks with time delay[J]. Communications in Nonlinear Science Numerical Simulation, 2015, 29(1/3): 346-358. doi: 10.1016/j.cnsns.2015.05.017.

张海滨, 何清波, 孔凡让. 基于变参数随机共振和归一化变换的时变信号检测与恢复[J]. 电子与信息学报, 2015, 37(9): 2124-2131. doi: 10.11999/JEIT141618.

ZHANG Haibin, HE Qingbo, and KONG Fanrang. Time-varying signal detection and recovery method based on varying parameter stochastic resonance and normalization transformation[J]. Journal of Electronics Information Technology, 2015, 37(9): 2124-2131. doi: 10.11999/ JEIT141618.

侯成郭, 罗柏文, 李地. 线性调频信号的级联随机共振数字化接收[J]. 电子与信息学报, 2015, 37(12): 2866-2871. doi: 10.11999/JEIT141496.

HOU Chengguo, LUO Bowen, and Li Di. Cascaded stochastic resonance for digitized receiving of linear frequency modulation signal[J]. Journal of Electronics Information Technology, 2015, 37(12): 2866-2871. doi: 10.11999/ JEIT141496.

CHEN Hao, VARSHNEY L R, and VARSHNEY P K. Noise-enhanced information systems[J]. Proceeding of the IEEE, 2014, 102(10): 1607-1621. doi: 10.1109/JPROC.2014. 2341554.

LIU Shujun, YANG Ting, and ZHANG Xinzheng. Effects of stochastic resonance for linearquadratic detector[J]. Chaos, Solitons Fractals, 2015, 77(1): 319-331. doi: 10.1016/j. chaos.2015.06.015.

LU Zeqi, CHEN Liqun, MICHAEL J B, et al. Stochastic resonance in a nonlinear mechanical vibration isolation system[J]. Journal of Sound Vibration, 2016, 370: 221-229. doi: 10.1016/j.jsv.2016.01.042.

邓冬虎, 朱小鹏, 张群, 等. 基于随机共振理论的双基ISAR 弱信号提取及成像分析[J]. 电子学报, 2012, 40(9): 1809-1816. doi: 10.3969/j.issn.0372-2112.2012.09.017.

DENG Donghu, ZHU Xiaopeng, ZHANG Qun, et al. Weak signals extraction and imaging analysis in bistatic ISAR systems based on stochastic resonance[J]. Acta Electronica Sinica, 2012, 40(9): 1809-1816. doi: 10.3969/j.issn.0372-2112. 2012.09.017.

MITAIM S and KOSKO B. Adaptive stochastic resonance in noisy neurons based on mutual information[J]. IEEE Transactions on Neural Networks, 2004, 15(6): 1526-1540. doi: 10.1109/TNN.2004.826218.

高锐, 李赞, 吴利平, 等. 低信噪比条件下基于随机共振的感知方法与性能分析[J]. 电子学报, 2013, 41(9): 1672-1679. doi: 10.3969/j.issn.0372-2112.2013.09.002.

GAO Rui, LI Zan, WU Liping, et al. A spectrum sensing method and performance analysis based on stochastic resonance under low SNR[J]. Acta Electronica Sinica, 2013, 41(9): 1672-1679. doi: 10.3969/j.issn.0372-2112.2013.09.002.

KAY S M, MICHELS J H, CHEN Hao, et al. Reducing probability of decision error using stochastic resonance[J]. IEEE Signal Processing Letters, 2009, 13(11): 695-698. doi: 10.1109/LSP.2006.879455.

CHEN Hao, VARSHNEY P K, KAY S M, et al. Theory of the stochastic resonance effect in signal detection: Part I　Fixed detectors[J]. IEEE Transactions on Signal Processing, 2007, 55(7): 3172-3184. doi: 10.1109/TSP.2007.893757.

BAYRAM S, GEZICI S, and VINCENT P H. Noise enhanced hypothesis-testing in the restricted Bayesian framework[J]. IEEE Transactions on Signal Processing, 2010, 58(8): 3972-3989. doi: 10.1109/TSP.2010.2048107.

BAYRAM S and GEZICI S. Noise enhanced M-ary hypothesis-testing in the Minimax framework[C]. The 3rd International Conference on Signal Processing Communication Systems (ICSPCS), Omaha, NE, USA, 2009: 16. doi: 10.1109/ICSPCS.2009.5306400.

盛骤, 谢式千, 潘承毅. 概率论与数理统计[M]. 北京: 高等教育出版社, 2010: 7677.

SHENG Zhou, XIE Shiqian, and PAN Chengyi. Probability and Statistics[M]. Beijing: Higher Education Press, 2010: 7677.

施引文献

资源附件(0)

访问统计