王波 劉德亮
摘 要:針對近場源波達方向(DOA)和距離的聯(lián)合估計問題,提出一種近場迭代自適應(yīng)算法(NF-IAA)。首先通過劃分二維網(wǎng)格表示出近場區(qū)域內(nèi)信源所有可能的位置,每個位置都看作存在一個潛在的信源入射到陣列上,表示出陣列輸出的數(shù)據(jù)模型;然后通過循環(huán)迭代利用上一次譜估計的結(jié)果構(gòu)建信號的協(xié)方差矩陣,將協(xié)方差矩陣的逆作為加權(quán)矩陣估計出每個位置對應(yīng)的潛在信源能量;最后繪制出三維能量譜圖,由于只有真實存在的信源能量不為0,因此譜峰對應(yīng)的位置即為真實存在信源的位置。仿真實驗表明在10個快拍條件下,NF-IAA的DOA分辨概率達到了90%,而二維多重信號分類(2D-MUSIC)算法只有40%;當(dāng)快拍數(shù)降至2時,2D-MUSIC算法已經(jīng)失效,而NF-IAA仍然能很好地分辨出3個入射信源并且準(zhǔn)確地估計出位置參數(shù)。隨著快拍數(shù)和信噪比(SNR)的增加,NF-IAA的估計性能一直優(yōu)于2D-MUSIC。實驗結(jié)果表明,NF-IAA具備少快拍條件下高精度、高分辨地估計近場源二維位置參數(shù)的能力。
關(guān)鍵詞:迭代自適應(yīng)方法;加權(quán)最小二乘法;二維參數(shù)估計;近場源;陣列信號處理
中圖分類號: TN911.6
文獻標(biāo)志碼:A
Abstract: A Near-Field Iterative Adaptive Approach (NF-IAA) was proposed for the joint estimation of Direction Of Arrival (DOA) and range of near-field sources. Firstly, all possible source locations in the neaar field region were represented by dividing two-dimensional grids. Each location was considered to have a potential incident source mapping to an array, indicating the output data model of the array. Then, through the loop iteration, the signal covariance matrix was constructed by using the previous spectral estimation results, and the inverse of the covariance matrix was used as the weighting matrix to estimate the energy of the potential sources corresponding to each location. Finally, the three-dimensional energy spectrum was figured. Since only the energy of real existing source is not 0, the angles and distances corresponding to the peaks are the two-dimensional location parameters of real existing sources. Simulation experimental results show that the DOA resolution probability of the proposed NF-IAA reaches 90% with 10 snapshots, while the DOA resolution probablity of Two-Dimension Multiple Signal Classification (2D-MUSIC) algorithm is only 40%. When the number of snapshots is reduced to 2, 2D-MUSIC algorithm has failed, but NF-IAA can still distinguish 3 incident sources and accurately estimate the two-dimension location parameters. As the number of snapshots and Signal-to-Noise Ratio (SNR) increase, NF-IAA always performs better than 2D-MUSIC. The experimental results show that NF-IAA has the ability to estimate the two-dimensional location parameters of near-field sources with high precision and high resolution when the number of snapshots is low.
Key words: iterative adaptive approach; weighted least square; two-dimensional parameter estimation; near-field source; array signal processing
0 引言
陣列信號處理技術(shù)近幾十年來在射電天文、無線通信、地震勘測、雷達探測、水下定位等領(lǐng)域發(fā)揮了重要作用。根據(jù)信號源傳播到接收陣列的距離,陣列表面到2D2/λ的空間范圍稱為近場,大于2D2/λ的空間范圍稱為遠場,D、λ分別指陣列孔徑和工作波長。對遠場源來說,陣列接收到的信號可以近似看成平行波,只需要估計出一維波達方向(Direction Of Arrival, DOA)就能確定信號源的位置。對近場源來說,由于信號波前的固有曲率不能忽略,因此需要同時估計DOA和距離二維參數(shù)才能確定信號的位置。
針對近場源參數(shù)估計問題,Huang等[1]提出的二維多重信號分類(Two-Dimensional Multiple Signal Classification, 2D-MUSIC)算法作為經(jīng)典的子空間類算法具有超分辨率,但是需要大量的快拍數(shù)據(jù)才能保證對樣本協(xié)方差矩陣進行特征分解時信號子空間與噪聲子空間不發(fā)生混疊;在少快拍條件下,算法性能驟降甚至失效。近年來,基于稀疏重構(gòu)的近場源參數(shù)估計方法[2-5]成為研究的熱點。梁國龍等[2]通過構(gòu)造虛擬遠場接收陣列把近場二維參數(shù)估計問題轉(zhuǎn)化為兩個基于l1范數(shù)一維稀疏信號恢復(fù)問題,具有較優(yōu)的估計性能,但該算法損失了一半的陣列孔徑。Hu等[3-4]利用接收信號協(xié)方差矩陣反對角元素的稀疏表示分步實現(xiàn)了波達方向和距離的稀疏估計。文獻[3]的方法比文獻[2]的方法具有更低的計算復(fù)雜度,而且同樣陣元數(shù)條件下可以檢測更多的信源數(shù),但是需要額外選擇正則化參數(shù)。文獻[5]基于文獻[3]中參數(shù)分離的思想,利用陣列的對稱性先基于加權(quán)l(xiāng)1范數(shù)優(yōu)化估計出DOA,再利用稀疏重構(gòu)的思想估計距離。與文獻[3]的方法相比,文獻[5]提出的方法具有更好的估計性能,但是正則化參數(shù)的選取依舊對估計結(jié)果有較大的影響。
與基于統(tǒng)計理論的子空間類算法相比,文獻[2-5]中的方法主要通過求解l1范數(shù)約束優(yōu)化問題實現(xiàn)參數(shù)估計,不需要直接對樣本協(xié)方差矩陣進行特征分解,因此一定程度上降低了對快拍數(shù)的要求,但是仍然不能達到只利用少量快拍實現(xiàn)對近場源二維參數(shù)進行高精度、高分辨的估計。在一些信號不能長時間穩(wěn)定或快速時變的應(yīng)用場景,如水下信號處理、城市無線通信、高速目標(biāo)追蹤、跳頻通信等,大量的快拍數(shù)據(jù)會導(dǎo)致采樣時間過長、與真實樣本的誤差增大或運算速度降低,只利用少量快拍甚至單快拍實現(xiàn)對近場源高精度、高分辨的定位具有重要意義。
Stoica等提出的估計幅度和相位的迭代自適應(yīng)方法(Iterative Adaptive Approach for Amplitude and Phase Estimation, IAA-APES)[6-7]、基于協(xié)方差稀疏迭代的估計方法(Sparse Iterative Covariance-based Estimation Method, SPICE)[8-9]、通過迭代最小化的稀疏學(xué)習(xí)算法(Sparse Learning via Iterative Minimization, SLIM)[10]具備在少量甚至一個快拍數(shù)據(jù)的情況下高分辨率DOA估計的能力,但是只適用于遠場窄帶信號。
為了能在少快拍條件下高精度估計近場源參數(shù),本文提出一種近場迭代自適應(yīng)算法(Near-Field Iterative Adaptive Approach, NF-IAA)。首先基于加權(quán)最小二乘法估計出入射信源的能量,然后通過循環(huán)迭代對估計結(jié)果進行更新直至收斂,最后繪制出三維能量譜圖,譜峰對應(yīng)的DOA和距離即為入射信源的二維位置參數(shù)。仿真實驗表明在只有2個快拍的情況下,所提算法經(jīng)過適量迭代就能實現(xiàn)對入射近場源高精度、高分辨的估計。
4 結(jié)語
本文針對少快拍近場源二維參數(shù)聯(lián)合估計問題提出一種近場迭代自適應(yīng)算法。首先通過劃分二維平面網(wǎng)格表示出區(qū)域內(nèi)所有潛在信源的位置;然后通過循環(huán)迭代估計出所有潛在信源的能量;迭代收斂時,真實存在信源位置的能量遠遠大于其他位置的能量,因此譜峰的位置即為真實入射信源的位置。該算法相比2D-MUSIC算法能夠得到更為尖銳的空間譜、更低的旁瓣水平和更小的估計誤差,特別是在少快拍條件下具有優(yōu)越的估計性能。仿真結(jié)果驗證了本文算法的有效性。同時需要指出,網(wǎng)格劃分得越精細估計結(jié)果越精確,但同時會增加算法的計算復(fù)雜度,下一步應(yīng)考慮如何在提高估計精度的同時降低算法的計算復(fù)雜度。
參考文獻:
[1] HUANG Y-D, BARKAT M. Near-field multiple sources localization by passive sensor array [J]. IEEE Transactions on Antennas and Propagation, 1991, 39(7): 968-975.
[2] 梁國龍,韓博,林旺生,等.基于稀疏信號重構(gòu)的近場源定位[J].電子學(xué)報,2014,42(6):1041-1046. (LIANG G L, HAN B, LIN W S, et al. Near-field sources localization based on sparse signal reconstruction [J]. Acta Electronica Sinica, 2014, 42(6): 1041-1046.)
[3] HU K, CHEPURI S P, LEUS G. Near-field source localization using sparse recovery techniques [C] // Proceedings of the 2014 International Conference on Signal Processing and Communications. Piscataway, NJ: IEEE, 2014: 1-5.
[4] HU K, CHEPURI S P, LEUS G. Near-field source localization: Sparse recovery techniques and grid matching [C]// Proceedings of the 2014 IEEE 8th Sensor Array and Multichannel Signal Processing Workshop (SAM). Piscataway, NJ: IEEE, 2014: 369-372.
[5] 李雙,劉曉,胡順仁,等.加權(quán)稀疏信號重構(gòu)的近場源定位方法[J].聲學(xué)技術(shù),2017,36(1):75-80. (LI S, LIU X, HU S R, et al. Near-field source localization based on weighted sparse signal recovery [J]. Technical Acoustics, 2017, 36(1): 75-80.)
[6] YARDIBI T, LI J, STOICA P, et al. Source localization and sensing: a nonparametric iterative adaptive approach based on weighted least squares [J]. IEEE Transactions Aerospace Electronica System, 2010, 46(1): 425-443.
[7] STOICA P, LI J, HE H. Spectral analysis of nonuniformly sampled data: a new approach versus the periodogram [J]. IEEE Transactions on Signal Processing, 2009, 57(3): 843-858.
[8] STOICA P, BABU P, LI J. New method of sparse parameter estimation in separable models and its use for spectral analysis of irregularly sampled data [J]. IEEE Transactions on Signal Processing, 2011, 59(1): 35-47.
[9] STOICA P, BABU P, LI J. SPICE: a sparse covariance-based estimation method for array processing [J]. IEEE Transactions on Signal Processing, 2011, 59(2): 629-638.
[10] TAN X, ROBERTS W, LI J, et al. Sparse learning via iterative minimization with application to MIMO radar imaging [J]. IEEE Transactions on Signal Processing, 2011, 59(3): 1088-1101.
[11] STARER D, NEHORAI A. Passive localization of near-field sources by path following [J]. IEEE Transactions on Signal Processing, 1994, 42(3): 677-680.
[12] STOICA, P, JAKOBSSON A, LI J. Capon, APES and matched-filterbank spectral estimation [J]. Signal Processing, 1998, 66(1): 45-59.
[13] 趙樹杰,趙建勛.信號檢測與估計理論[M].北京:清華大學(xué)出版社,2005:173-180. (ZHAO S J, ZHAO J X. Signal Detection and Estimation Theory [M]. Beijing: Tsinghua University Press, 2005: 173-180.)
[14] STOICA P, MOSES R L. Spectral Analysis of Signals [M]. Upper Saddle River, NJ: Pearson/Prenyice Hall, 2005: 265-283.
[15] SCARGLE J D. Studies in astronomical time series analysis. II-Statistical aspects of spectral analysis of unevenly spaced data[J]. Astrophysical Journal, 1982, 263: 835-853.