MMSE準則下基于玻爾茲曼機的快速重構算法

劉玲君; 謝中華; 馮久超; 楊萃

doi:10.13374/j.issn2095-9389.2017.08.016

MMSE準則下基于玻爾茲曼機的快速重構算法

doi: 10.13374/j.issn2095-9389.2017.08.016

劉玲君^1,2, ,,
謝中華¹,
馮久超¹,
楊萃^1,2

1) 華南理工大學電子與信息學院, 廣州 510641
2) 國家移動超聲探測工程技術研究中心, 廣州 510641

基金項目:

廣東省科技計劃資助項目(2017A020214011)

國家自然科學基金資助項目(61327005,61302120)

中央高校基本科研業務費資助項目(2017MS039)

詳細信息

中圖分類號: TP391
計量
- 文章訪問數: 569
- HTML全文瀏覽量: 284
- PDF下載量: 10
- 被引次數: 0
出版歷程
- 收稿日期: 2016-09-12

Fast recovery algorithm based on Boltzmann machine and MMSE criterion

1) School of Electronic and Information Engineering, South China University of Technology, Guangzhou 510641, China
2) National Engineering Technology Research Center for Mobile Ultrasonic Detection, Guangzhou 510641, China

摘要

摘要: 全連接的玻爾茲曼機模型可全面描述稀疏系數間統計依賴關系,但時間復雜度較高.為了提高基于玻爾茲曼機的貝葉斯匹配追蹤算法(BM-BMP)的重構速度和質量,本文提出一種改進算法.第一,將BM-BMP算法的最大后驗概率(MAP)估計評估值分解為上一次迭代的評估值與增量,使得每次迭代僅需計算增量,極大縮短了計算耗時.第二,利用顯著最大后驗概率估計值平均的方式,有效近似最小均方誤差(MMSE)估計,獲得了更小的重構誤差.實驗結果表明,本文算法比BM-BMP算法的運行時間平均縮短了73.66%,峰值信噪比(PSNR)值平均提高了0.57 dB.
- 稀疏信號重構 /
- 快速貝葉斯匹配追蹤 /
- 玻爾茲曼機 /
- 最小均方誤差
Abstract: Fully connected Boltzmann machine models can be used to provide a comprehensive description of statistical dependencies between sparse coefficients but with high time complexity. To improve the speed and quality of the Boltzmann machine-Bayesian matching pursuit (BM-BMP) method, an improved algorithm was proposed. First, the maximum a posteriori (MAP) estimation of the BM-BMP algorithm is decomposed into its value at the last iteration and an increment; thus, it only needs to calculate the increment in each iteration, which greatly reduces the computational time. Second, by calculating the mean of the significant MAP estimations, an effective approximation is obtained for the minimum mean square error (MMSE) estimation and a smaller reconstruction error is achieved. Compared with the BM-BMP, this method reduces the running time on average by 73.66% while improving the peak signal to noise ratio (PSNR) by 0.57 dB.
- sparse signal reconstruction /
- fast Bayesian matching pursuit /
- Boltzmann machine /
- minimum mean square error (MMSE)

HTML全文

參考文獻(23)

[2]	Xie Z B, Feng J C. KFCE:a dictionary generation algorithm for sparse representation. Signal Process, 2009, 89(10):2072
[2]	Tuna G, Nefzi B, Conte G. Unmanned aerial vehicle-aided communications system for disaster recovery. J Network Comput Appl, 2014, 41:27
[3]	Tropp J, Gilbert A C. Signal recovery from random measurements via orthogonal matching pursuit. IEEE Trans Inf Theory, 2007, 53(12):4655
[3]	García-Laencina P J, Rodríguez-Bermudez G, Roca-Dorda J. Exploring dimensionality reduction of EEG features in motor imagery task classification. Expert Syst Appl, 2014, 41(11):5285
[4]	Dai W, Milenkovic O. Subspace pursuit for compressive sensing signal reconstruction. IEEE Trans Inf Theory, 2009, 55(5):2230
[5]	Ji S H, Xue Y, Carin L. Bayesian compressive sensing. IEEE Trans Signal Process, 2008, 56(6):2346
[6]	Velasco-Álvarez F, Ron-Angevin R, da Silva-Sauer L, et al. Audio-cued motor imagery-based brain-computer interface:navigation through virtual and real environments. Neurocomputing, 2013, 121:89
[6]	Elad M, Yavneh I. A plurality of sparse representations is better than the sparsest one alone. IEEE Trans Inf Theory, 2009, 55(10):4701
[7]	Eldar Y C, Kuppinger P, Bolcskei H. Block-sparse signals:uncertainty relations and efficient recovery. IEEE Trans Signal Process, 2010, 58(6):3042
[7]	Elnady A M, Zhang X, Xiao Z G, et al. A single-session preliminary evaluation of an affordable BCI-controlled arm exoskeleton and motor-proprioception platform. Frontiers Human Neurosci, 2015, 9:168
[8]	Kim B H, Kim M, Jo S. Quadcopter flight control using a low-cost hybrid interface with EEG-based classification and eye tracking. Comput Biol Med, 2014, 51:82
[8]	Baraniuk R G, Cevher V, Duarte M F, et al. Model-based compressive sensing. IEEE Trans Inf Theory, 2010, 56(4):1982
[9]	Zhang Z L, Rao B D. Extension of SBL algorithms for the recovery of block sparse signals with intra-block correlation. IEEE Trans Signal Process, 2013, 61(8):2009
[10]	Fang J, Shen Y N, Li H B, et al. Pattern-coupled sparse bayesian learning for recovery of block-sparse signals. IEEE Trans Signal Process, 2015, 63(2):360
[10]	Li W, He Q C, Fan X M, et al. Evaluation of driver fatigue on two channels of EEG data. Neurosci Lett, 2012, 506(2):235
[11]	Wang Y K, Chen S A, Li C T. An EEG-based brain-computer interface for dual task driving detection. Neurocomputing, 2014, 129:85
[11]	He L H, Carin L. Exploiting structure in wavelet-based bayesian compressive sensing. IEEE Trans Signal Process, 2009, 57(9):3488
[12]	Garrigues P J, Olshausen B A. Learning horizontal connections in a sparse coding model of natural images//Advances in Neural Information Processing Systems (NIPS). Vancouver, 2007:1
[12]	Stock M G, Akita M, Krehbiel P R, et al. Continuous broadband digital interferometry of lightning using a generalized cross-correlation algorithm. J Geophysical Res Atmospheres, 2014, 119(6):3134
[13]	Lliff K W. Maximum likelihood estimation of lift and drag from dynamic aircraft maneuvers. J Aircraft, 1977, 14(12):1175
[13]	Dremeau A, Herzet C, Daudet L. Boltzmann machine and mean-field approximation for structured sparse decompositions. IEEE Trans Signal Process, 2012, 60(7):3425
[14]	Peleg T, Eldar Y, Elad M. Exploiting statistical dependencies in sparse representations for signal recovery. IEEE Trans Signal Process, 2012, 60(5):2286
[15]	Chen C L P, Zhang C Y, Chen L, et al. Fuzzy restricted Boltzmann machine for the enhancement of deep learning. IEEE Trans Fuzzy Systems, 2015, 23(6):2163"[1] Grzonka S, Grisetti G, Burgard W. A fully autonomous indoor quadrotor. IEEE Trans Rob, 2012, 28(1):90