一種基于KL分離度的改進(jìn)矩陣CFAR檢測(cè)方法
doi: 10.11999/JEIT150711 cstr: 32379.14.JEIT150711
基金項(xiàng)目:
國(guó)家自然科學(xué)基金(61179014),青年科學(xué)基金項(xiàng)目(61302193)
An Improved Matrix CFAR Detection Method Base on KL Divergence
Funds:
The National Natural Science Foundation of China (61179014), Youth Science Fund Project (61302193)
-
摘要: 矩陣CFAR檢測(cè)器是根據(jù)信息幾何理論提出的,但其恒虛警特性沒(méi)有從理論上得到分析,且檢測(cè)性能也有待進(jìn)一步提高。該文首先根據(jù)矩陣流形上正態(tài)律的概念從理論上推導(dǎo)了矩陣CFAR檢測(cè)器的恒虛警特性,并在此基礎(chǔ)上,利用積累性能更好的KLD(Kullback-Leibler Divergence)代替測(cè)地線距離,提出了一種改進(jìn)的矩陣CFAR檢測(cè)器。最后通過(guò)仿真實(shí)驗(yàn)驗(yàn)證了改進(jìn)方法具有更好的檢測(cè)性能。
-
關(guān)鍵詞:
- 信息幾何 /
- 恒虛警檢測(cè) /
- 統(tǒng)計(jì)流形 /
- 測(cè)地線距離 /
- Kullback-Leibler分離度(相對(duì)熵)
Abstract: The matrix CFAR detector is proposed according to information geometry theory, but its constant false alarm property is not analysed, and the matrix CFARs detection performance still needs to be improved. Firstly, the matrix CFARs constant false alarm property is analysed according to the normal law on matrix manifold, on this basis an improved matrix CFAR detector is proposed with replacing the geodesic distance with KULLBACK-LEIBLER Divergence (KLD). Finally, simulation experiments verify that the improved method has better detection performance. -
RAO C. Information and accuracy attainable in the estimation of statistical parameters[J]. Bulletin of the Calcutta Mathematical Society, 1945, 37(2): 81-91. CHENTSOV N N. Statistical Decision Rules and Optimal Inference[M]. Rhode Island, USA: American Mathematical Society, 1982: 55-60. EFRON B. Defining the curvature of a statistical problem (with applications to second order efficiency)[J]. The Annals of Statistics, 1975, 3(6): 1189-1242. AMARI S and NAGAOKA H. Methods of Information Geometry[M]. New York: Oxford University Press, 2000: 87-96. MARCELO P, Hadj B, and STEVE M.Exploiting information geometry to improve the convergence properties of variational active contours[J]. IEEE Jounal of Selected Topics in Signal Processing, 2013, 7(4): 700-707. 趙興剛, 王首勇. 雷達(dá)目標(biāo)檢測(cè)的信息幾何方法[J]. 信號(hào)處理, 2015, 31(6): 631-637. ZHAO Xinggang and WANG Shouyong. Information geometry method of radar target detection[J]. Signal Processing, 2015, 31(6): 631-637. BARBARESCO F. Radar detection for non-stationary time-Doppler signal based on Frchet distance of geodesic curves on covariance matrix information geometry manifold [C]. 14th International Radar Symposium, German, 2013: 307-312. LE Y, ARNAUDON M, and BARBARESCO F. Riemannian median, geometry of covariance matrices and radar target detection[C]. 7th European Radar Conference, Paris, 2010: 415-418. BARBARESCO F. Radar detection using Siegel distance between autoregressive processes, application to HF and X-band radar[C]. IEEE Radar Conference, Rome, Italy, 2008: 523-530. 孫華飛, 彭林玉, 張真寧. 信息幾何及其應(yīng)用[J]. 數(shù)學(xué)進(jìn)展, 2011, 40(3): 257-269. SUN Huafei, PENG Linyu and ZHANG Zhenning. Inforamtion geometry and its application[J]. Advances in Mathematics, 2011, 40(3): 257-269. AMARI S. Information geometry of positive measures and positive-definite matrices: decomposable dually flat structure [J]. Entopy, 2014, 16(5): 2131-2145. BARBARESCO F. Information Intrinsic Geometry Flows[C]. MaxEnt06 Conference, Paris, 2006: 133-146. LENGLET M and ROUSSON R. Statistics on the manifold of multivariate normal distributions: Theory and application to diffusion tensor MRI processing[J]. Journal of Mathematical Imaging and Vision, 2006, 25(3): 423-444. 鄭作虎, 王首勇. 基于Alpha穩(wěn)定分布雜波模型的雷達(dá)目標(biāo)檢測(cè)方法[J]. 電子與信息學(xué)報(bào), 2014, 36(12): 2963-2968. doi: 10.3724/SP.J.1146.2014.00072. ZHENG Zuohu and WANG Shouyong. Radar target detection method based on the alpha-stable distribution clutter model[J]. Journal of Electronics Information Technology, 2014, 36(12): 2963-2968. doi: 10.3724/SP. J.1146.2014.00072. KULLBACK S. Information Theory and Statistics[M]. New York: Dover Publications, 1968: 162-188. 嚴(yán)俊坤, 劉紅亮, 戴奉周, 等. 一種具有恒虛警性質(zhì)的檢測(cè)跟蹤聯(lián)合處理算法[J]. 電子與信息學(xué)報(bào), 2014, 36(11): 2666-2671. doi: 10.3724/SP.J.1146.2013.01925. YAN Junkun, LIU Hongliang, and DAI Fengzhou. Joint detection and tracking processing algorithm with constant false alarm rate property[J]. Journal of Electronics Information Technology, 2014, 36(11): 2666-2671. doi: 10.3724/SP.J.1146.2013.01925. -
計(jì)量
- 文章訪問(wèn)數(shù): 1461
- HTML全文瀏覽量: 172
- PDF下載量: 433
- 被引次數(shù): 0