Imaging methods and apparatus using model-based array signal processing

• US 6,130,641 A
• Filed: 09/04/1998
• Issued: 10/10/2000
• Est. Priority Date: 09/04/1998
• Status: Expired due to Term

• Current Assignee: Simon Fraser University
• Original Assignee: Simon Fraser University
• Inventors: Bird, John, Kraeutner, Paul
• Primary Examiner(s): Gregory, Bernarr E.
• Application Number:
  US09/148,131
• Time in Patent Office: 767 Days
• Field of Search: 367/7-11, 367/87, 367/88, 367/117, 367/131, 342/22, 342/25, 342/59, 342/175-186, 342/190-197
• US Class Current: 342/179
• CPC Class Codes: G01S 15/8902 Side-looking sonar
• CPC Class Codes: G01S 3/8006 Multi-channel systems specially adapted for direction-finding, i.e. having a single aerial system capable of giving simultaneous indications of the directions of different signals
• CPC Class Codes: G01S 7/52003 Techniques for enhancing spatial resolution of targets G01S7/52046 takes precedence

Abstract

An apparatus and method for imaging uses a transducer array which receives a scattered signal from a region of interest. The method analyses the scattered signal on the basis of a model which assumes several independent point scatterers at each range. The method determines the amplitudes and angles of arrival for the several scatterers. The method provides much higher resolution than is possible with conventional beam formed sonar arrays of practical size. The method is capable of resolving features which cannot be resolved by conventional sidescan sonar or interferometric side scan sonar. The method may be used with electromagnetic or acoustic scattering signals.

Claims

1. A method for imaging a region, the method comprising the steps of:

(a) transmitting a pulse toward the region;

(b) at a receiver transducer array comprising a plurality of N elements detecting a signal scattered from the region to yield a multi-channel scattering signal;

(c) obtaining L snapshots, where L is an integer with L≧

1, by sampling the scattering signal at one or more instants after transmitting the pulse to yield a plurality of complex samples;

(d) in a processing device, processing the plurality of complex samples to obtain image data comprising angles of arrival from M principal scattering components, where M is an integer, with 1≦

M≦

N-1, by the steps of;

(i) constructing a sample matrix S_L,M,N from the complex samples;

(ii) from the sample matrix computing a null-space vector w comprising a plurality of entries;

(iii) computing a plurality of roots of a polynomial formed from the entries of the null-space vector;

(iv) from the roots of the polynomial computing angles of arrival for M principal scattering components; and

,

(v) storing the angles of arrival as image data.

2. The method of claim 1 wherein the step of computing the null space vector comprises numerically finding a vector w in a search space according to a best fit criterion.

3. The method of claim 2 wherein the best fit criterion is selected from the group consisting of:

maximum likelihood, sub-space decomposition fitting or least squares fitting.

4. The method of claim 3 comprising computing the null space vector by a least squares estimator by finding a vector w in a search space which minimizes the quantity e given by:

space="preserve" listing-type="equation">ε

=|S.sub.L,M,N w|.sup.2.

5. The method of claim 4 wherein computing the null space vector while minimizing c comprises:

(1) forming a sub-matrix by taking all but the first column of the sample matrix;

(2) computing a Moore-Penrose pseudo-inverse matrix of the sub-matrix using singular value decomposition; and

,

(3) multiplying the pseudo-inverse matrix by the first column of the sample matrix.

6. The method of claim 4 wherein the procedure for obtaining w while minimizing e comprises the steps of:

(a) forming an estimate of an array covariance matrix R_L,M,N by performing the operation;

space="preserve" listing-type="equation">R.sub.L,M,N =S.sub.L,M,N.sup.H S.sub.L,M,N

(b) performing an eigen decomposition on R_L,M,N and selecting an eigen vector associated with a smallest eigen value of R_L,M,N.

7. The method of claim 3 comprising computing the null space vector from the sample matrix using a sub-space decomposition method comprising the steps of:

a. obtaining an estimate of a covariance matrix from the sample matrix;

b. performing an eigen decomposition on the estimated covariance matrix to yield eigen values and eigen vectors; and

,

c. forming the null space vector w from a combination of the eigen vectors.

8. The method of claim 3 wherein processing the complex samples comprises computing amplitudes for scatterers corresponding to the selected roots from the sample matrix and the angles of arrival and storing the amplitudes with the image data.

9. The method of claim 2 wherein the step of processing the plurality of complex samples comprises:

(a) selecting those roots of the polynomial which have a magnitude in the range of 1-Δ

₁ to 1+Δ

₂, where Δ

₁ and Δ

₂ are predetermined threshold values; and

(b) storing as image data only the angles of arrival corresponding to the selected roots.

10. The method of claim 9 wherein Δ

₁ =Δ

₂.

11. The method of claim 9 wherein processing the complex samples comprises computing amplitudes for the M scatterers from the sample matrix and the angles of arrival and storing the amplitudes with the image data.

12. The method of claim 11 comprising comparing the amplitudes to a threshold and storing a binary value indicating whether the amplitudes are greater than or less than the threshold with the image data.

13. The method of claim 1 wherein the step of processing the plurality of complex samples comprises:

(a) selecting those roots of the polynomial which have a magnitude in the range of 1-Δ

₁ to 1+Δ

₂, where Δ

₁ and Δ

₂ are predetermined threshold values; and

(b) storing as image data only the angles of arrival corresponding to the selected roots.

14. The method of claim 13 wherein Δ

₁ =Δ

₂.

15. The method of claim 13 wherein processing the complex samples comprises computing amplitudes for scatterers corresponding to the selected roots from the sample matrix and the angles of arrival and storing the amplitudes with the image data.

16. The method of claim 13 comprising displaying the image data on a display device and subsequently accepting user specified values for Δ

₁, Δ

₂, or both Δ

₁ and Δ

₂, generating new image data corresponding to the user specified values and displaying the new image data on the display device.

17. The method of claim 1 wherein processing the complex samples comprises computing amplitudes for the M scatterers from the sample matrix and the angles of arrival and storing the amplitudes with the image data.

18. The method of claim 17 comprising comparing the amplitudes to a threshold and storing a binary value indicating whether the amplitudes are greater than or less than the threshold with the image data.

19. The method of claim 1 comprising displaying the image data on a display device.

20. The method of claim 1 wherein the sample matrix comprises a first plurality of complex samples, all of the first plurality of complex samples detected at a first time.

21. The method of claim 20 wherein the sample matrix consists of the first plurality of complex samples and is of the form:

##EQU15## where s_n (l) is a complex sample taken at the n^th element of the transducer array, l is an index representing the first time, and n is an integer in the range 1≦

n≦

N.

22. The method of claim 20 wherein the sample matrix comprises a second plurality of complex samples detected by the transducer array at a second time different from the first time.

23. The method of claim 22 wherein the sample matrix is of the form:

##EQU16## where s_n (l) is a complex sample taken at the n^th element of the transducer array, l is an index representing the time at which the sample is taken, and n is an integer in the range 1≦

n≦

N.

24. The method of claim 1 wherein the polynomial has N-1 roots where N is the number of elements in the transducer array.

25. The method of claim 1 comprising the step of reorienting the transducer array after step (c) and then repeating steps (a) (b) and (c) to produce another set of complex samples.

26. The method of claim 1 wherein the step of transmitting the pulse comprises coding the pulse so as to increase a time bandwidth product of the transmitted pulse and the method includes decoding the received scattered signals so as to increase a signal-to-noise ratio of the multi-channel scattering signal.

27. The method of claim 1 wherein the step of processing the plurality of complex samples includes the steps of:

(a) forming Q beams with offset phase centers by taking linear combinations of the complex samples, where Q is an integer and Q≦

N, and,

(b) forming the sample matrix from outputs of the beams. ##EQU17##

28. The method of claim 27 wherein the sample matrix has the form:

where s_q (1) is the complex output from the g^th beam at an 1^th time interval, with 1 an integer in the range 1≦

1≦

L, L is an integer such that L≧

1, q is an integer in the range 1≦

q≦

Q, and M is an integer with 1≦

M≦

Q-1.

29. The method of claim 1 comprising repeating the steps (b) through (d) to produce image data corresponding to groups of principal scatterers at different distances from the receive transducer array.

30. The method of claim 1 wherein N≧

3.

31. An imaging system comprising:

(a) a transmit transducer;

(b) a pulse generator connected to the transmit transducer;

(c) a receive transducer array;

(d) a receiver connected to the receive transducer array, the receiver comprising;

(i) a multi-channel amplifier for amplifying scattered signals detected by each element of the receive transducer array; and

,

(ii) an analog to digital conversion circuit connected to the multi-channel amplifier to sample the amplified scattered signals to produce a series of complex samples;

(e) a processor connected to receive the series of complex samples from the analog to digital conversion circuit, the processor performing the following operations for each of a plurality of ranges;

(i) select a group of complex samples corresponding to a range;

(ii) construct a sample matrix S_L,M,N from the group of complex samples;

(iii) compute from the sample matrix a null space vector w comprising a plurality of entries;

(iv) compute the roots of a polynomial formed from the entries of the null-space vector;

(v) compute angles of arrival for one or more scatterers from the roots of the polynomial;

(vi) compute amplitudes for the one or more scatterers from the sample matrix and the angles of arrival; and

,

(vii) store the range, angles of arrival and amplitudes as image data; and

,

(f) a display device driver connected to receive the image data and to display the image data on a display device.

32. A processing unit for processing scattering signal data received at a multi-channel transducer to yield image data, the processing unit comprising:

(i) an input for receiving scattering signal data, the scattering signal data representing signals scattered by one or more objects and comprising a plurality of sets of complex samples, the sets detected sequentially at a transducer array, each set comprising a complex sample obtained from each of a plurality of elements in the array,

(ii) means for selecting a group of complex samples corresponding to a distance range;

(iii) means for constructing a sample matrix S_K,M,N from the group of complex samples;

(iv) means for computing a null-space vector w comprising a plurality of entries from the sample matrix;

(v) means for computing roots of a polynomial formed from the entries of the null-space vector;

(vi) means for computing angles of arrival for two or more scatterers from the roots of the polynomial;

(vii) means for computing amplitudes for the two or more scatterers from the sample matrix and the angles of arrival;

(viii) means for storing the range, angles of arrival and amplitudes as image data and,

ix output means for delivering an image signal representing the image data, the image signal, when displayed on a display, graphically depicting the objects.

33. A computer readable medium containing computer software, the computer software, when run on a computer providing a processing means for:

(a) constructing a sample matrix S_K,M,N from the complex samples;

(b) obtaining from the sample matrix a null-space vector w;

(c) computing the roots of a polynomial formed from the entries of the null-space vector;

(d) computing angles of arrival for two or more scatterers from the roots of the polynomial;

(e) computing amplitudes for the two or more scatterers from the sample matrix and the angles of arrival; and

,

(f) storing the angles of arrival and amplitudes as image data.

34. A method for imaging a region, the method comprising the steps of:

(a) transmitting a pulse toward the region;

(b) at a receiver transducer array comprising a plurality of N elements, with N≧

2 detecting a signal scattered from the region to yield a multi-channel scattering signal;

(c) obtaining L snapshots, where L is an integer with L≧

1, by sampling the scattering signal at one or more instants after transmitting the pulse to yield a plurality of complex samples; and

,

(d) in a computer, processing the plurality of complex samples to obtain image data comprising angles of arrival from M principal scattering components, where M is an integer, with 2≦

M≦

N-1.

35. The method of claim 34 wherein N≧

2.

36. The method of claim 34 wherein N≧

3.

37. The method of claim 34 wherein processing the plurality of complex samples comprises:

(i) constructing a sample matrix S_L,M,N from the complex samples;

(ii) from the sample matrix generating a null-space vector w comprising a plurality of entries;

(iii) computing values for a plurality of roots of a polynomial formed from the entries of the null-space vector;

(iv) from the roots of the polynomial computing angles of arrival for M principal scattering components; and

,

(v) storing the angles of arrival as image data.

38. The method of claim 36 wherein generating the null-space vector w comprises numerically finding a vector w in a search space according to a best fit criterion.

39. The method of claim 34 wherein processing the complex samples comprises computing an amplitude associated with each of the M principal scattering components and storing the amplitudes with the image data.

40. The method of claim 34 wherein transmitting the pulse comprises coding the pulse so as to increase a time bandwidth product of the transmitted pulse and the method comprises decoding the received scattered signals so as to increase a signal-to-noise ratio of the multi-channel scattering signal.

41. The method of claim 34 comprising changing the orientation of the receiver transducer array after obtaining the L snapshots and then repeating steps (a), (b) and (c) to obtain another plurality of complex samples.