Method and system for sensing objects in a scene using transducer arrays and coherent wideband ultrasound pulses
View Patent ↗Pulses are transmitted into a scene by an array of the transducers, wherein each pulse has wideband ultrasound frequency, wherein the pulses are transmitted simultaneously, and wherein a pattern of the wideband ultrasound frequencies in each pulse is unique with respect to the patterns of each other pulse. The pulses as received when the pulse are reflected by the scene and objects in the scene. Each received pulse is sampled and decomposed using a Fourier transform to produce frequency coefficients, which are stacked to produce a linear system modeling a reflectivity of the scene and the objects. Then, a recovery method is applied to the linear system to recover the reflectivity of the scene and the objects.
1. A method for sensing a scene, comprising the steps of:
transmitting pulses into the scene by an array of transducers, wherein each pulse has a wideband ultrasound frequency, wherein the pulses are transmitted simultaneously, and wherein a pattern of the wideband ultrasound frequencies in each pulse is unique with respect to the patterns of each other pulse;
receiving the pulses by each transducer as received pulses when the pulse are reflected by the scene;
sampling and decomposing each received pulse using a Fourier transform to produce a vector of frequency coefficients;
partitioning a representation of the scene into voxels, wherein a reflectivity of each voxel is to be determined;
stacking the reflectivities to be determined to form a vector of the reflectivities to be determined;
forming a linear system, wherein the linear system models the vector of frequency coefficients given the vector of the reflectivities, wherein the transmitted pulses are represented as a multiplication of a transmission operator by the vector of the reflectivities to be determined; and
applying a sparse recovery method to the linear system to determine the reflectivity of the scene.
2. The method of claim 1 , wherein the array is regular.
3. The method of claim 1 , wherein the array is irregular.
4. The method of claim 1 , wherein the pulses transmitted by each transducer are represented by discrete-time samples at a Nyquist rate.
5. The method of claim 1 , wherein the representation is three dimensional.
6. The method of claim 1 , wherein the voxels are linearly indexed.
7. The method of claim 1 , further comprising:
determining a distance to each voxel having a non-zero reflectivity.
8. The method of claim 1 in which the linear system is represented as r=Ax, where r denotes the vector of reflectivities to be determined, A is the transmission operator, and x represents the reflectivity in the scene.
9. The method of claim 8 , wherein the sparse recovery method optimizes
x
^
=
arg
min
x
x
1
such
that
r
≈
Ax
,
wherein “^” indicates an estimate.
10. The method of claim 8 , wherein the sparse recovery method optimizes
x
^
=
arg
min
x
x
1
+
λ
r
-
Ax
2
2
,
wherein “^” indicates an estimate, and λ is a regularization parameter.
11. The method of claim 1 , wherein the sparse recovery method is a greedy sparse recovery algorithm.
12. The method of claim 8 , further comprising:
moving the array of transducers while transmitting and receiving the pulses, and wherein the linear system is r q =A q x for each position q of the array of transducers.
13. The method of claim 1 , wherein the array of transducers is mounted on a vehicle.
14. The method of claim 1 , wherein the array of transducers is mounted on a robot.
15. The method of claim 1 , wherein the scene is under surveillance by the array of transducers.
16. The method of claim 1 , wherein the array of transducers monitors occupancy of the scene.
17. The method of claim 1 , wherein the representation is two dimensional.
18. The method of claim 1 , wherein the transmission operator incorporates a frequency transform of the transmitted pulses and delays between the transmitted pulses and the received pulses.
19. The method of claim 1 , wherein the transmission operator is in a form of a sensing matrix.
20. A system for sensing a scene, comprising:
an array of transducers configured to transmit pulses into the scene from each transducer, wherein each pulse has a wideband ultrasound frequencies, wherein the pulses are transmitted simultaneously, and wherein a pattern of the wideband ultrasound frequencies in each pulse is unique with respect to the patterns of the other pulses, and wherein each transducer is configured to receive all pulses when the pulse are reflected by an object in the scene;
means for associating each received pulse with a particular transducer according to the unique pattern;
means for sampling and decomposing each received pulse using a Fourier transform to produce a vector of frequency coefficients;
means for partitioning a representation of the scene into voxels, wherein a reflectivity of each voxel is to be determined;
means for stacking the reflectivities to be determined to form a vector of the reflectivities to be determined;
means for forming a linear system, wherein the linear system models the vector of frequency coefficients given the vector of the reflectivities, wherein the transmitted pulses are represented as a multiplication of a transmission operator by the vector of the reflectivities to be determined, and wherein the transmission operator is in a form of a sensing matrix; and
means for recovering the reflectivity of the scene and the objects.