Estimating Noise Variances by Using Multiple Observers

Unknown variances of the noises that excite a time-invariant, linear dynamic system and/or in the observation of its output can be estimated by use of multiple observers. One observer is needed, in general, for each unknown variance. Each observer is time invariant and has different gains from the others. It is shown that each unknown variance is a linear combination of the variances of the residuals of the observers. The required estimates of the noise variances are obtained by using the measured variances of the residuals. The method presented in this paper is illustrated by an application to determining noise parameters in a ring laser gyro.     

Instrumental variable adaptive array processing

An instrumental variable (IV) approach is presented for estimating the weights of an adaptive antenna array. Theoretical analysis of the IV method shows that the antenna gain weights are independent of finitely correlated noise, so that unbiased estimation of signal arrival angles is possible. Only matrix inversions are required to compute the weight estimates. In this sense, the IV method provides performance comparable with eigenvector techniques but with lower computational burden. Both minimal and overdetermined IV estimators are derived. The overdetermined estimators give the same theoretical array weights as minimal estimators, but yield more accurate weight estimates in real data situations. Simulation results are presented to compare these IV methods with one another and with conventional matrix inversion weight estimators. In these examples it is seen that IV methods are able to resolve closely spaced interference sources when conventional matrix inversion techniques cannot. It is also shown that overdetermined methods are capable of providing weight estimates with lower variances than those of minimal methods     

Estimating Range Derivatives Using Phase Differences

The problem of estimating radar range derivatives using phase differences is considered. A minimum-variance, unbiased estimator is obtained. The resultant variances of the estimates so obtained are shown to be significantly lower than those of a least-squares estimator. Finally, a recursive form for the estimator is derived.     

Stability of Parameter Estimates for a Gaussian Process

Maximum-likelihood estimates for the levels of the mean value function and the covariance function of a Gaussian random process are investigated. The stability of these estimates is examined as the actual covariance function of the process deviates from the form assumed in the estimators. It is found that the time-bandwidth product for stationary processes represents an upper bound on the number of estimator terms that can be safely used when estimating with uncertainty about the process covariance function. This result is consistent with other interpretations of the time-bandwidth product and tempers the conclusion that, in principle, an infinite number of estimator terms can be used to obtain a perfect estimate of the covariance level. In practice, the estimate of the level can never be perfect, and the accuracy of the estimate depends on the observation interval. Finally, conditions are established to ensure asymptotic stability of the estimates and physical interpretations are presented.     

Radar Measurement Noise Variance Estimation with Several Targets of Opportunity

A number of methods exist to track a target's uncertain motion through space using inherently inaccurate sensor measurements. A powerful method of adaptive estimation is the interacting multiple model (IMM) estimator. In order to carry out state estimation from the noisy measurements of a sensor, however, the filter should have knowledge of the statistical characteristics of the noise associated with that sensor. The statistical characteristics (accuracies) of real sensors, however, are not always available, in particular for legacy sensors. A method is presented of determining the measurement noise variances of a sensor, assumed to be constant, by using multiple IMM estimators while tracking targets whose motion is not known---targets of opportunity. Combining techniques outlined in [2] and [6], the likelihood functions are obtained for a number of IMM estimators, each with different assumptions on the measurement noise variances. Then a search is carried out over a varying grid of IMMs to bracket the variances of the sensor measurement noises. The end result consists of estimates of the measurement noise variances of the sensor in question.     

Stability of Gyro with Harmonic Nonlinearity in Spinning Vehicle

A stability analysis of a single-axis rate gyroscope mounted in a space vehicle which is spinning with uncertain angular velocity ?z about the spin axis of the gyro is presented. The complete nonlinear equation of motion, which includes the fundamental and second harmonic nonlinear terms, arising due to ?z, is considered. For time-varying ?z(t), using the circle criterion, it is shown that the gimbal motion is globally asymptotically stable if the Nyquist plot of the linear transfer function of the gyro lies in the interior of a certain disk. For the case of uncertain constant ?z, using the Lyapunov approach, conditions for global asymptotic stability (GAS) and asymptotic stability are derived. Stable regions in parameter space of the gyro and state space are obtained. Analytical relations for the selection of gyro parameters are derived.     

Particle flters for probability hypothesis density flter with the presence of unknown measurement noise covariance

In Bayesian multi-target fltering,knowledge of measurement noise variance is very important.Signifcant mismatches in noise parameters will result in biased estimates.In this paper,a new particle flter for a probability hypothesis density（PHD）flter handling unknown measurement noise variances is proposed.The approach is based on marginalizing the unknown parameters out of the posterior distribution by using variational Bayesian（VB）methods.Moreover,the sequential Monte Carlo method is used to approximate the posterior intensity considering non-linear and non-Gaussian conditions.Unlike other particle flters for this challenging class of PHD flters,the proposed method can adaptively learn the unknown and time-varying noise variances while fltering.Simulation results show that the proposed method improves estimation accuracy in terms of both the number of targets and their states.     

Position Finding with Prior Knowledge of Covariance Parameters

Butterly [1] presents a Bayesian approach as an alternative to the classical methods for solving the position-finding problem. Butterly assumes that bearing errors are independent and normally distributed with known variances. In the paper, the assumption of known variances is relaxed and it is shown that uncertainty about these variances can be incorporated into the model while also retaining the computational advantages of the Butterly formulation. It is also shown that the Bayes estimate and the classical maximum likelihood estimate will agree in certain cases.     

Analysis of an adaptive algorithm for unbiased multipath time delay estimation

The multipath equalization time delay estimator (METDE) provides an adaptive approach for estimating the difference in arrival times of a signal received at spatially separated sensors as well as the multipath channel characteristics. However, the parameter estimates of the METDE are biased in the presence of noise. The METDE algorithm is improved for unbiased parameter estimation via minimizing a modified cost function. Convergence behaviors and variances of the system variables of the amended method are derived. Computer simulations are included to corroborate the theoretical analysis and to evaluate the adaptive multipath delay estimation performance of the proposed algorithm.     

Self-Tuning Multisensor Weighted Measurement Fusion Kalman Filter

For the multisensor systems with unknown noise variances, based on the solution of the matrix equations for the correlation function, the on-line estimators of the noise variance matrices are obtained, whose consistency is proved using the ergodicity of sampled correlation function. Further, two self-tuning weighted measurement fusion Kalman filters are presented for the multisensor systems with identical and different measurement matrices, respectively. Based on the stability of the dynamic error system, a new convergence analysis tool is presented for a self-tuning fuser, which is called the dynamic error system analysis (DESA) method. A new concept of convergence in a realization is presented, which is weaker than the convergence with probability one. It is rigorously proved that the proposed self-tuning Kalman fusers converge to the steady-state optimal Kalman fusers in a realization or with probability one, so that they have asymptotic global optimality. A simulation example for a target tracking system with 3 sensors shows their effectiveness.     

Analysis of Min-Norm and MUSIC with arbitrary array geometry

A nonasymptotic performance comparison is presented between the Min-Norm and MUSIC algorithms for estimating the directions of arrival of narrowband plane waves impinging on an array of sensors. The analysis is based on a finite amount of sensor data. The analysis makes the assumption of high signal-to-noise ratio (SNR), and it applies to arrays of arbitrary geometry. It is shown that Min-Norm can be expressed as a certain data-dependent weighted MUSIC algorithm, and that this relationship allows a unified performance comparison. It is also shown that the variances of the estimated directions-of-arrival from the MUSIC algorithm are always smaller than those of the Min-Norm algorithm at high SNR when both algorithms employ a numerical search procedure to obtain the estimates     

A Hybrid Navigation Concept Using a Spinning Satellite?Borne Interferometer and Self?Contained Equipment

This paper analyzes a hybrid navigation concept that uses signals from a radio interferometer mounted on a spinning geostationary satellite, preliminary position estimates from self-contained equipment, and stored a priori information on the past performance of this equipment. The craft-borne processor, optimum in the maximum a posteriori (MAP) sense, is designed to estimate position coordinates using only the incoming radio signals, although improved estimates result if the other two items are available. An error analysis starts with the derivation of an estimation error covariance matrix, whose elements depend on additive receiver noise and the physical parameters of the system. A minimum 1? estimation error in position is obtained by trading off these parameters. The effects of other major error sources, such as tropospheric phase fluctuations, multipath, and craft altitude uncertainty, are added to the estimation error to give a total 1? position error on the order of 3.7 to 5.6 km.     

Adaptive Radar Detection with Asymptotically Regulated False-Alarm Rate

An adaptive detection procedure is described by which the detection threshold is so adjusted as to provide an asymptotic false-alarm probability PFA that is approximately invariant with changes in radar clutter return amplitude probability density functions (pdf's) in a broad class. The class includes Rayleigh, chi, Weibull, and lognormal pdf's. The receiver noise is also taken into account. The clutter-plus-noise pdf is approximated by a truncated generalized Laguerre series, the coefficients of which are estimated from the radar returns using "cell averaging" techniques. This estimation is assumed to be perfect. The results obtained indicate that the "bias" error, defined as the normalized difference between the design PFA and the asymptotic PFA corresponding to the computed threshold, lies within a fraction of an order of magnitude for 10-3?PFA ? 10-6. For PFA ?10-6 the bias error is more than an order of magnitude. These results are for the case when a single independent radar return is processed at a time. The bias error decreases as the number of postdetection integrations of independent returns increases.     

The Exact Cramer-Rao Bound for Gaussian Autoregressive Processes

An explicit expression is derived for the Cramer-Rao bound(CRB)on unbiased estimates of the parameters of autoregressiveprocesses, given a finite number of measurements. The expressionconverges to the well-known asymptotic form of the CRB when thenumber of measurements tends to infinity. The behavior of thebound is illustrated by some numerical examples.     

The exact Cramer-Rao bound for Gaussian autoregressive processes

An explicit expression is derived for the Cramer-Rao bound (CRB) on unbiased estimates of the parameters of autoregressive (AR) processes, given a finite number of measurements. The expression converges to the well-known asymptotic form of the CRB when the number of measurements tends to infinity. The behavior of the bound is illustrated by some numerical examples     

地形辅助制导系统

本文叙述一种地形辅助制导系统,用来修正惯性系统或其它导航系统的误差,以提高飞行器的制导精度。这种辅助系统含有一个卡尔曼滤波器,它将飞行器所飞过的实际地形信息与预存的地图进行比较和处理,从而估计出导航系统的误差。对于一般的丘陵地形,辅助制导后的精度可达到数十米的范围。它适用于飞机、巡航导弹、弹道导弹和其它武器运载系统,而且所需的设备简单,易于实现。     

Estimation techniques for GMSK using linear detectors in satellite communications

This paper describes data-aided signal level and noise variance estimators for Gaussian minimum shift keying (GMSK) when the observations are limited to the output of a filter matched to the first pulse-amplitude modulation (PAM) pulse in the equivalent PAM representation. The estimators are based on the maximum likelihood (ML) principle and assume burst-mode transmission with known timing and a block of L₀ known bits. While it is well known that ML estimators are asymptotically unbiased and efficient, the analysis quantifies the rate at which the estimators approach these asymptotic properties. It is shown that the carrier phase, amplitude, and noise variance estimators are unbiased and can achieve their corresponding Cramer-Rao bounds with modest combinations of signal-to-noise ratio and observation length. The estimates are used to estimate the signal-to-noise ratio. It is shown that the mean squared error performance of the ratio increases with signal-to-noise ratio while the mean squared error performance of the ratio in decibels decreases with signal-to-noise ratio. Simulation results are provided to confirm the accuracy of the analytic results.     

Detection Performance of Logarithmic Receivers Employing Video Integrators

The detection performance of logarithmic receivers in Rayleigh and non-Gaussian clutter is investigated. In Rayleigh clutter the performance is determined for steady, Swerling case 1, and Swerling case 2 targets. The detection loss of logarithmic receivers is generally less than the ? log n loss conjectured by Green, but consistent with the 1.08-dB asymptotic loss established by Hansen. The Swerling case 2 loss, important in frequency- agility applications, canbe severe for a small number of integrated pulses and high Pd, and apparently approaches the 1.08-dB asymptotic loss as a lower bound. Graphs of GramCharlier series cumulants are provided to allow determination of logarithmic-receiver performance. Curves are presented to allow the detection performance of logarithmic receivers in log-normal and Weibull clutter to be determineds.     

多阶段含延缓纠正可靠性增长的MS-NHPP模型及其Bayes分析

针对复杂航空机电系统的多阶段含延缓纠正可靠性增长试验,提出了多阶段非齐次泊松过程(MS-NHPP)模型及其Bayes分析方法。该模型将多个阶段试验看做多个非齐次泊松过程(NHPP),它们具有相同的时间起点和不同的强度函数参数,并满足序化约束关系。对该模型采用Bayes方法进行分析,以Dirichlet分布作为验前分布,并在验后计算中采用基于Metropolis-Hastings准则的马尔可夫链蒙特卡罗(MCMC)方法获得各阶段可靠性参数的验后期望、方差与置信限。实例分析表明,该方法可以有效融合多阶段可靠性增长试验数据以及专家经验和历史信息,尤其适用于小样本情况,参数估计值满足序化约束关系,可靠性增长评估结果更为可信。