CRLB and ML for parametric estimate: New results
Introduction
The Cramér–Rao Lower Bound (CRLB) plays an important role in the estimation theory because it theoretically predicts the best achievable second-order estimation error performance [1]. Traditionally, the main source of uncertainty in the parameter estimation and the derivation of the corresponding CRLB is the measurement noise. In many engineering problems (e.g. target tracking), due to the imperfections in target detection, there is an additional source of uncertainty in estimation, caused by the unknown measurement origin (target detections or false alarm occurrences). In other words, the assumption of detection probability and false alarm probability is unrealistic in these applications; this consideration applies also when the Maximum Likelihood Estimator (MLE) is determined and the CRLB of the same is evaluated.
The effect of uncertain measurements has been first observed in [2], [3], with respect to a deterministic target motion: it is shown that the Fisher Information Matrix (FIM) is scaled by a constant factor less than unity, the so-called Information Reduction Factor (IRF). In mathematical terms, these results are synthesized via the equation (see also Eq. (32) of [2] and Eq. (52) of [3]):The bound resulting from Eq. (1.1) is effective in modeling the measurement origin uncertainty as the product of two distinct factors: the FIM relative to the ideal case (, ) and the IRF accounting for missed detections and false alarms.
The CRLB reported in [4] is a generalization allowing for non-linear measurements and non-deterministic dynamics; also in this extended context, the measurement origin uncertainty can be expressed by means of a suitable IRF.
In [5], the theoretical CRLB for and (i.e., in absence of false alarms) has been calculated via the enumeration of all possible sequences of detections and missed detections, given a certain scan number. The computational complexity of this solution grows exponentially with time, but the bound has been verified to be the exact one, by means of numerical simulations, both in linear and non-linear cases. An approximation of the theoretical bound for practical applications, where the number of sensor scans is large, has also been proposed.
The comparison of the two bounds (IRF and enumeration method) has been reported in [6], showing, both theoretically and with numerical examples, that the CRLB computed via the IRF is overly optimistic. The enumeration based CRLB is the true bound for and case, since all possible sequences of detections and missed detections are averaged.
In the present paper, we extend the enumeration method for CRLB computation, to the parameter estimation problem with and . A case study, which compares the MLE with this new CRLB, is presented in the context of a ballistic target tracking with a radar. The rationale of the work rests on the features of ML estimators, the most suitable for parametric problems, and on the yardstick for the second-order error performance that the CRLB analysis provides.
The paper is organized as follows:
Section 2 presents the mathematical formulation of the enumeration method, illustrating all possible sequences of detections and missed detections for true and false targets. The theoretical CRLB for non-linear parameter estimation is calculated with and .
Section 3 presents a case study. It describes the application of the enumeration method, with and , to the estimation of the impact point of a ballistic target, e.g. an artillery shell.
Section 4 illustrates the Monte Carlo simulation and compares its results to the CRLB.
Conclusions are reported in Section 5.
Section snippets
Background
This section extends the enumeration method illustrated in [5] to all the possible sequences of detections and missed detections in presence of true and false targets. The method proposed in the following relies on the assumption of constant measurement accuracy.
Assume we have a radar that acquires measurements on the target position. Let us consider a discrete-time deterministic target motion. The measurement equation when a target is detected is given bywhere k is the
A case study
In this section we apply the above mentioned theory to the tracking of a ballistic target; in particular, we assume that a radar is acquiring few measurements during the ballistic target flight: the problem is to estimate the impact point of the target [7].
To calculate the accuracy of the impact point, a simplified model of a ballistic target has been considered. The algorithm that has been conceived is based on processing the acquired measurements with a batch approach; it accounts for
Monte Carlo simulation results
A Monte Carlo simulation has been run to study the performance of the MLE given by Eq. (3.2) and validate expression (2.14) we found for the CRLB.
The simulation refers to the case study described in Section 3 and has been organized as follows:
- 1.
ideal trajectory generation out of the Monte Carlo loop;
- 2.
generation of a random variable in the [0,1] range to assess the detection or missed detection of the kth target measurement (out of n) in the specific Monte Carlo run;
- 3.
in case of target detection,
Conclusions
This paper has extended the classical theory of parametric estimation of MLE to the case in which the measurements are acquired with and also in presence of potential false alarms (i.e. ). The analysis was built around the assumption of having just one false alarm. A suitable expression of the CRLB has been found and applied to the estimation study of a ballistic target flight. A Monte Carlo simulation has successfully validated the new theoretical CRLB enumeration method. Our future
References (8)
Target tracking with bearings—only measurements
Signal Processing
(October 1999)- S. Kay, Fundamentals of statistical signal processing. estimation theory, in: Signal Processing Theory, Prentice-Hall,...
- et al.
Track formation with bearings and frequency measurements in clutter
IEEE Trans. Aerosp. Electron. Syst.
(1990) - et al.
Low observable target motion analysis using amplitude information
IEEE Trans. Aerosp. Electron. Syst.
(1996)