Next Article in Journal
Adapting an Ant Colony Metaphor for Multi-Robot Chemical Plume Tracing
Previous Article in Journal
On the Selection of Transmission Range in Underwater Acoustic Sensor Networks
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Rao and Wald Tests for Nonhomogeneous Scenarios

1
The State Key Laboratory of Acoustics, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China
2
Elettronica S.P.A., Via Tiburtina Valeria, Rome 00131, Italy
*
Author to whom correspondence should be addressed.
Sensors 2012, 12(4), 4730-4736; https://doi.org/10.3390/s120404730
Submission received: 12 March 2012 / Revised: 28 March 2012 / Accepted: 29 March 2012 / Published: 12 April 2012
(This article belongs to the Section Physical Sensors)

Abstract

: In this paper, we focus on the design of adaptive receivers for nonhomogeneous scenarios. More precisely, at the design stage we assume a mismatch between the covariance matrix of the noise in the cell under test and that of secondary data. Under the above assumption, we show that the Wald test is the adaptive matched filter, while the Rao test coincides with the receiver obtained by using the Rao test design criterion in homogeneous environment, hence providing a theoretical explanation of the enhanced selectivity of this receiver.

1. Introduction

In recent years adaptive detection of targets embedded in Gaussian disturbance with unknown spectral properties has received an increased attention in the signal processing community. In the seminal paper by Kelly [1], the generalized likelihood ratio test (GLRT) is used to design an adaptive decision scheme capable of detecting a signal known up to a scaling factor in presence of Gaussian disturbance with unknown covariance matrix. Moreover, it is assumed that a set of secondary data free of signal components, but sharing the same spectral properties of the noise in the cell under test (CUT), is available (homogeneous environment). In [2] the authors derive the so-called adaptive matched filter (AMF), whose design relies on the so-called two-step GLRT-based design procedure. In fact, its design is split into two steps: first a non-adaptive GLRT for known covariance matrix is derived; then, the fully-adaptive version of the GLRT is obtained by replacing the unknown matrix with a proper estimate. Other design criteria have been investigated as an alternative to the GLRT over the years. In particular, in [3] a novel derivation of the AMF is proposed by resorting to the Wald test design criterion for the homogeneous environment. In [4], the Rao test is used to derive a detector (referred to in the following as H-RAO) that exhibits enhanced rejection capabilities of mismatched signals [57].

However, the homogeneous environment is an assumption that might not be met in realistic situations: see, for example, ([8], and references therein). A slightly general model assumes that the covariance matrix of the CUT and that of secondary data coincide only up to a scale factor. This scenario is referred to as partially-homogeneous environment and has been firstly proposed in [9], where the authors apply the GLRT to derive a fully-adaptive detector, referred to as the adaptive coherence estimator (ACE) but also as adaptive normalized matched filter [10]. Interestingly, in [11] it is proven that Rao and Wald tests for partially-homogeneous environment coincide with the ACE.

Another model of interest assumes that CUT and secondary data share the same covariance matrix of the thermal noise plus clutter, but, in addition, the CUT contains a noise-like interferer [12]. Such a covariance mismatch can lead to receivers that are less inclined to reveal signals that produce steering vectors different from the nominal one (selective receivers) [13]. In [12] the authors show that the GLRT for this problem is the ACE, providing, as a byproduct, a theoretical explanation of the good rejection capabilities of such a receiver. The above model is modified in [14], where it is assumed that the noise-like interferer is orthogonal to the nominal steering vector in the whitened observation space. The goal is to induce the rejection of only those signals which are sufficiently far from the nominal steering vector. As a matter of fact, the GLRT for this constrained problem is Kelly's detector, which offers a good tradeoff between detection performance of mainlobe targets and rejection capabilities of sidelobe interferers [14].

In this work, we use the Wald test and the Rao test design criteria to solve the latter detection problem. In particular, we show that these design criteria yield the decision statistics that have been obtained for the homogeneous environment, i.e., the Wald test is the AMF, while the Rao test coincides with the H-RAO. Moreover, the latter coincidence provides, as a byproduct, an alternative explanation of the excellent selectivity exhibited by the H-RAO in the homogeneous scenario; in fact, taking into account the noise-like interferer at the design stage can make it possible to reject the signal-plus-noise hypothesis as the degree of mismatch between the received signal and the postulated one increases. The interested reader is referred to ([5] and references therein) for the performance assessment of the above detectors in comparison to existing receivers in open literature.

The remainder of the paper is organized as follows: the next section is devoted to the problem formulation while Section III addresses detector designs. Finally, Section IV contains some concluding remarks and potential directions for future works.

2. Problem Formulation

Assume that a linear array formed by Na ∈ IN antennas senses the cell under test and that each antenna collects Nt ∈ IN samples. Denote by z ∈ ℂ N×1 the N-dimensional vector, with N = NaNt, containing returns from the CUT. We want to test whether or not z contains useful target echoes. As customary, we assume that a set of K ∈ IN secondary data, zk ∈ ℂN×1, k = 1,…, K, KN, namely data free of signal components is available.

The detection problem at hand can be formulated as follows:

{ H 0 : { z = n , z k = n k , k = 1 , , K H 1 : { z = α υ + n , z k = n k , k = 1 , , K
where
  • υ ∈ ℂN×1 is the nominal steering vector;

  • nk ∈ ℂN×1, k = 1,…, K, are independent and identically distributed complex normal random vectors with zero mean and unknown, positive definite covariance matrix M ∈ ℂN×N, i.e., nk Sensors 12 04730i1N(0, M), k = 1,…, K, see ([5] and references therein);

  • n Sensors 12 04730i1N(0, M + qq), where q ∈ℂN×1 is due to a noise-like interferer and denotes conjugate transpose. Moreover, we assume that q is orthogonal to the nominal steering vector in the whitened observation space, namely

    q M 1 υ = 0

  • α ∈ ℂ an unknown (deterministic) factor which accounts for both target and channel effects.

Notice that the probability density function (pdf) of z and Z ≡ [z1,…, zK] under H1 (and assuming that Equation (2) holds true) is given by

f ( z , Z ; θ , H 1 ) = exp { tr [ M 1 ( S + ( z α υ ) ( z α υ ) ) ] } π N ( K + 1 ) det ( M ) K + 1 ( 1 + q M 1 q ) × exp { | z M 1 q | 2 1 + q M 1 q }
In Equation (3) S = k = 1 K z k z k , and θ ( 2 + 2 N + N 2 ) × 1 is the parameter vector, i.e.,
θ = [ α r α i q r T q i T f ( M ) T ] T = [ θ A T θ B T ] T
where
  • αr, αi ∈ ℝ denote the real and imaginary part of α, respectively;

  • T denotes transpose;

  • θ A = [ α r α i ] T 2 × 1 and θ B = [ q r T q i T f ( M ) T ] T ( 2 N + N 2 ) × 1; and observe that θA contains the parameters of interest while θB contains the nuisance parameters;

  • f(M) ∈ ℝN2×1 is a vector that contains in univocal way the real and the imaginary parts of the elements of M.

As a preliminary step towards the derivation of the receivers, denote by J(θ) the Fisher information matrix, which can be written as follows [15]

J ( θ ) 1 = [ J A A ( θ ) J A B ( θ ) J B A ( θ ) J B B ( θ ) ] 1 = [ C A A ( θ ) C A B ( θ ) C B A ( θ ) C B B ( θ ) ]
where
C A A ( θ ) = [ J A A ( θ ) J A B ( θ ) J B B 1 ( θ ) J B A ( θ ) ] 1

3. Detector Designs

In this section we solve the binary hypothesis testing problem (1) by applying the Wald and the Rao test design criteria. Notice that problem (1) amounts to testing H0: α = 0 against the alternative that α is unrestricted.

3.1. Wald Test Design

The Wald test is the following decision rule [15]

θ ^ A , 1 T [ C A A ( θ ^ 1 ) ] 1 θ ^ A , 1 H 1 > < H 0 η 1
where θ ^ 1 = [ θ ^ A , 1 T θ ^ B , 1 T ] T with θˆA,1 and θˆB,1 the maximum likelihood estimates of θA and θB under H1, respectively, and η1 is the threshold to be set to achieve a predetermined probability of false alarm (Pfa). In order to evaluate Equation (7), observe that [13]
[ C A A ( θ ) ] 1 = J A A ( θ ) = 2 I 2 υ M 1 υ
where I2 denotes the 2-dimensional identity matrix, and that
θ ^ A , 1 = [ 2 { α ^ } 2 { α ^ } ] = [ 2 { υ S 1 Z υ S 1 υ } 2 { υ S 1 Z υ S 1 υ } ]
where
α ^ = υ S 1 z υ S 1 υ
is the maximum likelihood estimate of α, while ℜ{·} and ℑ{·} denote the real and the imaginary parts of the argument, respectively. Now, it remains to replace M−1 in Equation (8) with its maximum likelihood estimate under H1 given by [14]
M ^ 1 = ( K + 1 ) [ S 1 1 γ S 1 1 ( z α ^ υ β υ ) ( z α ^ υ β υ ) S 1 1 1 + γ ( z α ^ υ β υ ) S 1 1 ( z α ^ υ β υ ) ]
where
S 1 = S + ( z α ^ υ ) ( z α ^ υ )
β = υ S 1 1 ( z α ^ υ ) υ S 1 1 υ
γ = 1 ( K + 1 ) ( z α ^ υ β υ ) S 1 1 ( z α ^ υ β υ ) K ( z α ^ υ β υ ) S 1 1 ( z α ^ υ β υ )

It is tedious but not difficult to show that

[ C A A ( θ ^ 1 ) ] 1 = 2 I 2 υ M ^ 1 υ = 2 I 2 ( K + 1 ) υ S 1 1 υ = 2 I 2 ( K + 1 ) υ S 1 1 υ

Finally, by substituting Equations (9) and (15) into Equation (7) we obtain

| z S 1 v | 2 v S 1 v H 1 > < H 0 η 1

Notice that the left-hand side of Equation (16) is the decision statistic of the AMF [2], which can be also obtained by using the Wald test design criterion assuming q = 0 (homogeneous environment).

3.2. Rao Test Design

The Rao test for the problem (1) can be written as follows [15]

ln f ( z , Z ; θ , H 1 ) θ A | θ = θ ^ 0 T [ C A A ( θ ^ 0 ) ] ln f ( z , Z ; θ , H 1 ) θ A | θ = θ ^ 0 H 1 > < H 0 η 2
where
  • In f(z, Z; θ, H1) is the natural logarithm of the pdf of z and Z under the H1 hypothesis (see Equation (3));

  • CAA(θˆ0) = [JAA(θˆ0)]1, where JAA(θ) is given by Equation (8);

  • θ ^ 0 = [ 00 θ ^ B , 0 T ] T, with θˆB,0 denoting in turn the maximum likelihood estimate of θB under H0;

  • η2 is the threshold to be set in order to ensure the preassigned Pfa.

It is straightforward to show that

ln f ( z , Z ; θ , H 1 ) θ A = [ v M 1 ( z α v ) + ( z α v ) M 1 v j v M 1 ( z α v ) + j ( z α v ) M 1 v ]
= [ 2 { v M 1 ( z α v ) } 2 { v M 1 ( z α v ) } ]
where j denotes the imaginary unit. The maximum likelihood estimate of M under the H0 hypothesis is given by [14]
M ^ = 1 K + 1 [ S 1 + γ ( z ξ v ) ( z ξ v ) ]
where
S 1 = S + z z , ξ = v S 1 1 z v S 1 1 v
γ = 1 ( K + 1 ) ( z ξ v ) S 1 1 ( z ξ v ) K ( z ξ v ) S 1 1 ( z ξ v )

It follows that

ln f ( z , Z ; θ , H 1 ) θ A | θ = θ 0 ^ = [ 2 { v M ^ 1 z } 2 { v M ^ 1 z } ]
= ( K + 1 ) [ 2 { v S 1 z 1 + z S 1 z } 2 { v S 1 z 1 + z S 1 z } ]
and
[ J A A ( θ ^ 0 ) ] 1 = 2 I 2 1 v M ^ 1 v = 2 I 2 1 ( K + 1 ) ( v S 1 v | v S 1 z | 2 1 + z S 1 z )

Gathering the above results, the Rao test can be recast as

| v S 1 z | 2 v S 1 v ( 1 + z S 1 z ) [ 1 1 + z S 1 z | v S 1 z | 2 v S 1 v ] H 1 > < H 0 η 2
or, equivalently, as
| v ( S + z z ) 1 z | 2 v ( S + z z ) 1 v H 1 > < H 0 η 2

which is the H-RAO [4].

4. Conclusions

This work addresses the adaptive detection of point-like targets in nonhomogeneous scenarios. In particular, at the design stage it is assumed that the CUT contains a noise-like interferer in addition to thermal noise, clutter, and to the possible signal of interest; a set of secondary data, free of signal components, is available: such data share a common covariance matrix that is equal to that of thermal noise plus clutter in the cell under test. Observe that the covariance mismatch between the CUT and the secondary data can lead to receivers that are less inclined to reveal a signal with the actual steering vector different from the nominal one (selective receivers). In addition, we assume that the noise-like interferer is orthogonal to the nominal steering vector in the whitened observation space. Under the above assumptions, we show that the Wald test coincides with the AMF, while the RAO test is the H-RAO. The latter result provides an alternative explanation of the good selectivity properties exhibited by the H-RAO [5]. Moreover, observe that the condition (2) makes the H-RAO less selective than the DN-AMF, derived in [13] using the Rao test design criterion without any constraint on the noise-like interferer.

As a final comment, it would be of interest to investigate under which conditions these design criteria are invariant with respect to different classes of detection problems. A preliminary step towards this direction is given in [16].

Acknowledgments

This work was supported by the National Natural Science Foundations of China under Grant No. 61172166.

References

  1. Kelly, E.J. An adaptive detection algorithm. IEEE Trans. Aerosp. Electron. Syst. 1986, 22, 115–127. [Google Scholar]
  2. Robey, F.C.; Fuhrman, D.L.; Kelly, E.J.; Nitzberg, R. A CFAR adaptive matched filter detector. IEEE Trans. Aerosp. Electron. Syst. 1992, 29, 208–216. [Google Scholar]
  3. de Maio, A. A new derivation of the adaptive matched filter. IEEE Signal Process. Lett. 2004, 11, 792–793. [Google Scholar]
  4. de Maio, A. Rao test for adaptive detection in gaussian interference with unknown covariance matrix. IEEE Trans. Signal Process. 2007, 55, 3577–3584. [Google Scholar]
  5. Bandiera, F.; Orlando, D.; Ricci, G. Advanced Radar Detection Schemes under Mismatched Signal Models; Synthesis Lectures on Signal Process; No. 8, Morgan & Claypool Publishers: San Rafael, CA, USA, 2009. [Google Scholar]
  6. Hao, C.; Liu, B.; Cai, L. Performance analysis of a two-stage Rao detector. Signal Process 2011, 91, 2141–2146. [Google Scholar]
  7. Hao, C.; Liu, B.; Yan, S.; Cai, L. Parametric adaptive radar detector with enhanced mismatched signals rejection capabilities. EURASIP J. Adv. Signal Process 2010, 2010, 375136:1–375136:11. [Google Scholar]
  8. Capraro, G.T.; Farina, A.; Griffiths, H.; Wicks, M.C. Knowledge-based radar signal and data processing (a tutorial review). IEEE Signal Process. Mag. 2006, 23, 18–29. [Google Scholar]
  9. Kraut, S.; Scharf, L.L. The CFAR adaptive subspace detector is a scale-invariant GLRT. IEEE Trans. Signal Process. 1999, 47, 2538–2541. [Google Scholar]
  10. Conte, E.; Lops, M.; Ricci, G. Asymptotically optimum radar detection in compound gaussian noise. IEEE Trans. Aerosp. Electron. Syst. 1995, 31, 617–625. [Google Scholar]
  11. de Maio, A.; Iommelli, S. Coincidence of the Rao test, Wald test, and GLRT in partially homogeneous environment. IEEE Signal Process. Lett. 2008, 15, 385–388. [Google Scholar]
  12. Besson, O. Detection in the presence of surprise or undernulled interference. IEEE Signal Process. Lett. 2007, 14, 352–354. [Google Scholar]
  13. Orlando, D.; Ricci, G. A Rao test with enhanced selectivity properties in homogeneous scenarios. IEEE Trans. Signal Process. 2010, 58, 5385–5390. [Google Scholar]
  14. Besson, O.; Orlando, D. Adaptive detection in nonhomogeneous environments using the generalized eigenrelation. IEEE Signal Process. Lett. 2007, 14, 731–734. [Google Scholar]
  15. Kay, S.M. Fundamentals of Statistical Signal Processing, Detection Theory; Prentice Hall: Upper Saddle River, NJ, USA, 1998; Volume 2, pp. 375–380. [Google Scholar]
  16. de Maio, A.; Kay, S.M.; Farina, A. On the invariance, coincidence, and statistical equivalence of the GLRT, Rao test, and Wald test. IEEE Trans. Signal Process. 2010, 58, 1967–1979. [Google Scholar]

Share and Cite

MDPI and ACS Style

Hao, C.; Orlando, D.; Hou, C. Rao and Wald Tests for Nonhomogeneous Scenarios. Sensors 2012, 12, 4730-4736. https://doi.org/10.3390/s120404730

AMA Style

Hao C, Orlando D, Hou C. Rao and Wald Tests for Nonhomogeneous Scenarios. Sensors. 2012; 12(4):4730-4736. https://doi.org/10.3390/s120404730

Chicago/Turabian Style

Hao, Chengpeng, Danilo Orlando, and Chaohuan Hou. 2012. "Rao and Wald Tests for Nonhomogeneous Scenarios" Sensors 12, no. 4: 4730-4736. https://doi.org/10.3390/s120404730

Article Metrics

Back to TopTop