利用重要性采样的时差-频差联合估计算法
收稿日期: 2015-12-31
修回日期: 2016-03-15
网络出版日期: 2016-03-21
基金资助
国家自然科学基金(61401469，61501513)；国家“863”计划(2012AA7031015)
TDOA-FDOA joint estimation using importance sampling method
Received date: 2015-12-31
Revised date: 2016-03-15
Online published: 2016-03-21
Supported by
National Natural Science Foundation of China (61401469, 61501513); National High Technology Research and Development Program of China (2012AA7031015)
针对无源定位中参考信号真实值未知的时差（TDOA）-频差（FDOA）联合估计问题，构建了一种新的时差-频差最大似然（ML）估计模型，并采用重要性采样（IS）方法求解似然函数极大值，得到时差-频差联合估计。算法通过生成时差-频差样本，并统计样本加权均值得到估计值，克服了传统互模糊函数（CAF）算法只能得到时域和频域采样间隔整数倍估计值的问题，且不存在期望最大化（EM）等迭代算法的初值依赖和收敛问题。推导了时差-频差联合估计的克拉美罗下界（CRLB），并通过仿真实验表明，算法的计算复杂度适中，估计精度优于CAF算法和EM算法，在不同信噪比条件下估计误差接近CRLB。
赵勇胜 , 赵拥军 , 赵闯 . 利用重要性采样的时差-频差联合估计算法[J]. 航空学报, 2017 , 38(1) : 319994 -319994 . DOI: 10.7527/S1000-6893.2016.0085
To solve the joint estimation of time difference of arrival (TDOA) and frequency difference of arrival (FDOA) in passive location system, where the true value of the reference signal is unknown, a novel maximum likelihood (ML) estimator of TDOA and FDOA is constructed. Then importance sampling (IS) method is applied to find the maximum of likelihood function by generating the samples of TDOA and FDOA. Unlike the cross ambiguity function (CAF) algorithm or the expectation maximization (EM) algorithm, the proposed algorithm can estimate the TDOA and FDOA of non-integer multiple of the sampling interval and has no dependence on the initial estimate. The Cramer Rao lower bound (CRLB) is also derived. Simulation results show that the proposed algorithm outperforms the CAF and EM algorithm with higher accuracy and moderate computational complexity, and approaches the CRLB for different SNR conditions.
[1] LIU J, LI H, HIMED B. On the performance of the cross-correlation detector for passive radar applications[J]. Signal Processing, 2015, 113:32-37.
[2] CORALUPPI S. Multistatic sonar localization[J]. IEEE Journal of Oceanic Engineering, 2006, 31(4):964-974.
[3] CAFFERY J J, STUBER G L. Overview of radio location in CDMA cellular systems[J]. IEEE Communications Magazine, 1998, 16(4):38-45.
[4] GEZICI S, ZHI T, GIANNAKIS G B, et al. Localization via ultra-wideband radios:A look at positioning aspects for future sensor networks[J]. IEEE Signal Processing Magazine, 2005, 22(1):70-84.
[5] 刘洋, 杨乐, 郭福成, 等. 基于定位误差修正的运动目标TDOA/FDOA无源定位方法[J]. 航空学报, 2015, 36(5):1617-1626. LIU Y, YANG L, GUO F C, et al. Moving targets TDOA/FDOA passive localization algorithm based on localization error refinement[J]. Acta Aeronautica et Astronautica Sinica, 2015, 36(5):1617-1626(in Chinese).
[6] STEIN S. Algorithms for ambiguity function processing[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1981, 29(3):588-599.
[7] TOLIMIERI R, WINOGRAD S. Computing the ambiguity surface[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1985, 33(5):1239-1245.
[8] AUSLANDER L, TOLIMIERI R. Computing decimated finite cross-ambiguity functions[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1988, 36(3):359-364.
[9] OZDEMIR A K, ARIKAN O. Fast computation of the ambiguity function and the Wigner distribution on arbitrary line segments[J]. IEEE Transactions on Signal Processing, 2001, 49(2):381-393.
[10] TAO R, ZHANG W Q, CHEN E Q. Two-stage method for joint time delay and Doppler shift estimation[J]. IET Radar, Sonar and Navigation, 2008, 2(1):71-77.
[11] SHIN D C, NIKIAS C L. Complex ambiguity functions using nonstationary higher order cumulant estimates[J]. IEEE Transactions on Signal Processing, 1995, 43(11):2649-2664.
[12] NIU X X, CHING P C, CHAN Y T. Wavelet based approach for joint time delay and Doppler stretch measurements[J]. IEEE Transactions on Aerospace & Electronic Systems, 1999, 35(3):1111-1119.
[13] BELANGER S P. Multipath TDOA and FDOA estimation using the EM algorithm[C]//ICASSP IEEE Computer Society. Piscataway, NJ:IEEE Press, 1993:168-171.
[14] BEICHL I. The importance of importance sampling[J]. Computing in Science & Engineering, 1999, 1(2):71-73.
[15] WANG H, KAY S, SAHA S. An importance sampling maximum likelihood direction of arrival estimator[J]. IEEE Transactions on Signal Processing, 2008, 56(10):5082-5092.
[16] WANG H, KAY S. Maximum likelihood angle-Doppler estimator using importance sampling[J]. IEEE Transactions on Aerospace and Electronic Systems, 2010, 46(2):610-622.
[17] MASMOUDI A, BELLILI F, AFFES S, et al. A maximum likelihood time delay estimator using importance sampling[C]//2011 IEEE Global Telecommunications Conference (GLOBECOM 2011). Piscataway, NJ:IEEE Press, 2011:1-6.
[18] MASMOUDI A, BELLILI F, AFFES S, et al. A maximum likelihood time delay estimator in a multipath environment using importance sampling[J]. IEEE Transactions on Signal Processing, 2013, 61(1):182-193.
[19] PINCUS M. A closed form solution of certain programming problems[J]. Operations Research, 1968, 16(3):690-694.
[20] KAY S M. Comments on "Frequency estimation by linear prediction"[J]. IEEE Transactions on Acoustics Speech & Signal Processing, 1979, 27(2):198-199.] CAFFERY J J, STUBER G L. Overview of radio loca-tion in CDMA cellular systems[J]. IEEE Communica-tions Magazine, 1998, 16(4): 38-45.
[4] GEZICI S, ZHI T, GIANNAKIS G B, et al. Localization via ultra-wideband radios: A look at positioning aspects for future sensor networks[J]. IEEE Signal Processing Magazine, 2005, 22(1): 70-84.
[5] LIU Y, YANG L, GUO F, JIANG W. Moving targets TDOA/FDOA passive localization algorithm based on localization error refinement[J]. Acta Aeronautica et As-tronautica Sinica, 2015, 36(5): 1617-1626.
刘洋, 杨乐, 郭福成, 等. 基于定位误差修正的运动目标TDOA/FDOA无源定位方法[J]. 航空学报, 2015, 36(5): 1617-1626.
[6] STEIN S. Algorithms for ambiguity function pro-cessing[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1981, 29(3): 588-599.
[7] TOLIMIERI R, WINOGRAD S. Computing the ambi-guity surface[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1985, 33(5): 1239-1245.
[8] AUSLANDER L, TOLIMIERI R. Computing decimated finite cross-ambiguity functions[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1988, 36(3): 359-364.
[9] OZDEMIR A K, ARIKAN O. Fast computation of the ambiguity function and the Wigner distribution on arbi-trary line segments[J]. IEEE Transactions on Signal Processing, 2001, 49(2): 381-393.
[10] TAO R, ZHANG W Q, CHEN E Q. Two-stage method for joint time delay and Doppler shift estimation[J]. IET Radar, Sonar and Navigation, 2008, 2(1): 71-77.
[11] SHIN D C, NIKIAS C L. Complex ambiguity functions using nonstationary higher order cumulant estimates[J]. IEEE Transactions on Signal Processing, 1995, 43(11): 2649-2664.
[12] NIU X X, CHING P C, CHAN Y T. Wavelet based approach for joint time delay and Doppler stretch measurements[J]. IEEE Transactions on Aerospace & Electronic Systems, 1999, 35 (3): 1111-1119.
[13] BELANGER S P. Multipath TDOA and FDOA estima-tion using the EM algorithm[C]// ICASSP IEEE Com-puter Society, 1993:168-171.
[14] BEICHL I. The importance of importance sampling[J]. Computing in Science & Engineering, 1999, 1(2):71-73.
[15] WANG H, KAY S, SAHA S. An Importance Sampling maximum likelihood direction of arrival estimator[J]. IEEE Transactions on Signal Processing, 2008, 56(10): 5082-5092.
[16] WANG H, KAY S. Maximum likelihood angle-Doppler estimator using importance sampling[J]. IEEE Transac-tions on Aerospace and Electronic Systems, 2010, 46(2): 610-622.
[17] MASMOUDI A, BELLILI F, AFFES S, et al. A Maxi-mum Likelihood Time Delay Estimator Using Im-portance Sampling[C]// Global Telecommunications Conference (GLOBECOM 2011), 2011 IEEE. IEEE, 2011: 1-6.
[18] MASMOUDI A, BELLILI F, AFFES S, et al. A maxi-mum likelihood time delay estimator in a multipath en-vironment using importance sampling[J]. IEEE Trans-actions on Signal Processing, 2013, 61(1): 182-193.
[19] PINCUS M. Pincus M. A closed form solution of certain programming problems[J]. Operations Research, 1968, 16(3): 690-694.
[20] KAY S M. Comments on "Frequency estimation by linear prediction"[J]. IEEE Transactions on Acoustics Speech & Signal Processing, 1979, 27(2): 198-199.
/
| 〈 |
|
〉 |
CJA