一種用于壓縮感知理論的投影矩陣優(yōu)化算法
doi: 10.11999/JEIT141450
-
2.
(南京理工大學(xué)機(jī)械工程學(xué)院 南京 210094) ②(東華理工大學(xué)電子與機(jī)械工程學(xué)院 撫州 344000)
基金項(xiàng)目:
國(guó)家自然科學(xué)基金(61161010, 11265001)和高等學(xué)校博士學(xué)科點(diǎn)專(zhuān)項(xiàng)科研基金(20133219110027)
Novel Optimization Method for ProjectionMatrix in Compress Sensing Theory
-
2.
(School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China)
-
摘要: 考慮到投影矩陣對(duì)壓縮感知(CS)算法性能的影響,該文提出一種優(yōu)化投影矩陣的算法。該方法提出可導(dǎo)的閾值函數(shù),通過(guò)收縮Gram矩陣非對(duì)角元的方法壓縮投影矩陣和稀疏字典的相關(guān)系數(shù),引入基于沃爾夫條件(Wolfes conditions)的梯度下降法求解最佳投影矩陣,達(dá)到提高投影矩陣優(yōu)化算法穩(wěn)定度和重構(gòu)信號(hào)精度的目的。通過(guò)基追蹤(BP)算法和正交匹配追蹤(OMP)算法求解l0優(yōu)化問(wèn)題,用壓縮感知方法實(shí)現(xiàn)隨機(jī)稀疏向量、小波測(cè)試信號(hào)和圖像信號(hào)的感知和重構(gòu)。仿真實(shí)驗(yàn)表明,該文提出的投影矩陣優(yōu)化算法能較大地提高重構(gòu)信號(hào)的精度。Abstract: Considering the influence of the projection matrix on Compressed Censing (CS), a novel method is proposed to optimize the projection matrix. In order to improve the signals reconstruction precise and the stability of the optimization algorithm of the projection matrix, the proposed method adopts a differentiable threshold function to shrink the off-diagonal items of a Gram matrix corresponding to the mutual coherence between the projection matrix and sparse dictionary, and introduces a gradient descent approach based on the Wolfs-conditions to solve the optimization projection matrix. The Basis-Pursuit (BP) algorithm and the Orthogonal Matching Pursuit (OMP) algorithm are applied to find the solution of the minimuml0-norm optimization issue and the compressed sensing are utilized to sense and reconstruct the random vectors, wavelets noise test signals and pictures. The results of the simulation show the proposed method based on the projection matrix optimization is able to improve the quality of the reconstruction performance.
-
Donoho D L, Elad M, and Temlyakov V N. Stable recovery of sparse overcomplete representations in the presence of noise[J]. IEEE Transactions on Information Theory, 2006, 52(1): 6-18. Candes E J, Romberg J K, and Tao T. Stable signal recovery from incomplete and inaccurate measurements[J]. Communications on Pure and Applied Mathematics, 2006, 59(8): 1207-1223 Candes E J and Tao T. Near-optimal signal recovery from random projections: universal encoding strategies[J]. IEEE Transactions on Information Theory, 2006, 52(12): 5406-5425. 鄭紅, 李振. 壓縮感知理論投影矩陣優(yōu)化方法綜述[J]. 數(shù)據(jù)采集與處理, 2014, 52(1): 43-53. Zheng Hong and Li Zhen. Survey on optimization methods for projection matrix in compress sensing theory[J]. Journal of Data Acquisition and Processing, 2014, 52(1): 43-53. 戴瓊海, 付長(zhǎng)軍, 季向陽(yáng). 壓縮感知研究[J]. 計(jì)算機(jī)學(xué)報(bào), 2011, 34(3): 425-434. Dai Qiong-hai, Fu Chang-jun, and Ji Xiang-yang. Research on compressed sensing[J]. Chinese Journal of Computers, 2011, 34(3): 425-434. Elad M. Optimized projections for compressed sensing[J]. IEEE Transactions on Signal Processing, 2007, 55(12): 5695-5703. Abolghasemi V, Ferdowsi S, and Sanei S. A gradient-based alternating minimization approach for optimization of the measurement matrix in compressive sensing[J]. Signal Processing, 2012, 92(3): 999-1009. 李佳, 王強(qiáng), 沈毅, 等. 壓縮感知中測(cè)量矩陣與重建算法的協(xié)同構(gòu)造[J]. 電子學(xué)報(bào), 2013, 41(1): 29-34. Li Jia, Wang Qiang, Shen Yi, et al.. Collaborative construction of measurement matrix and reconstruction algorithm in compressive sensing[J]. Acta Electronica Sinica, 2013, 41(1): 29-34. Zhang Qi-heng, Fu Yu-li, Li Hai-feng, et al.. Optimized projection matrix for compressed sensing[J]. Circuit System Signal Processing, 2014, 33(5): 1627-1636. Xu Jian-ping, Pi Yi-ming, and Cao Zong-jie. Optimized projection matrix for compressive sensing[J]. EURASIP Journal on Advances in Signal Processing, 2010, DOI: 10.1155/2010/560349. 林波, 張?jiān)鲚x, 朱炬波. 基于壓縮感知的DOA估計(jì)稀疏化模型與性能分析[J]. 電子與信息學(xué)報(bào), 2014, 36(3): 589-594. Lin Bo, Zhang Zeng-hui, and Zhu Ju-bo. Sparsity model and performance analysis of DOA estimation with compressive sensing[J]. Journal of Electronics Information Technology, 2014, 36(3): 589-594. Donoho D L. For most large underdetermined systems of linear equations the minimal l1-norm solution is also the sparsest solution[J]. Communications on Pure and Applied Mathematics, 2006, 59(6): 797-829. Donoho D L and Stark P B. Uncertainty principles and signal recovery[J]. SIAM Journal on Applied Mathematics, 1989, 49(3): 906-931. Donoho D L and Elad M. Optimally sparse representation in general (nonorthogonal) dictionaries via minimization[J]. Proceedings of the National Academy of Science, 2003, 100(5): 2197-2202. Jorge N and Wright S J. Numerical Optimization Theoretical and Practical Aspects[M]. 2nd Edition, New York: Springer- Verlag Berlin and Heidelberg GmbH Co. K, 2006: 30-60. -
計(jì)量
- 文章訪(fǎng)問(wèn)數(shù): 1562
- HTML全文瀏覽量: 129
- PDF下載量: 700
- 被引次數(shù): 0