Compressed sensing characterization system and method

• US 7,916,052 B2
• Filed: 07/13/2009
• Issued: 03/29/2011
• Est. Priority Date: 07/13/2009
• Status: Active Grant

• Current Assignee: L-3 Communications Integrated Systems LP (L3Harris Technologies, Inc.)
• Original Assignee: L-3 Communications Integrated Systems LP (L3Harris Technologies, Inc.)
• Inventors: Fudge, Gerald Lothair, Yeh, Alex Chen-Chu
• Primary Examiner(s): JEAN PIERRE, PEGUY
• Application Number:
  US12/501,866
• Publication Number:
  US 20110007795A1
• Time in Patent Office: 624 Days
• Field of Search: 341/122, 341/155, 341/144, 341/100, 341/101
• US Class Current: 341/122
• CPC Class Codes: H03M 1/109 for dc performance, i.e. static testing H03M1/1085 takes precedence
• CPC Class Codes: H03M 1/1245 Details of sampling arrangements or methods

Abstract

A system for characterizing a compressed sensing apparatus is described which broadly includes a random vector generator for generating an input vector, a waveform generator in communication with the random vector generator for converting the input vector into an analog signal, a compressed sensing apparatus in communication with the digital-to-analog converter for determining a digital bit stream from the analog signal, a serial-to-parallel converter for converting the digital bit stream into an input vector; and an electronic processor in communication with the random number generator and the serial-to-parallel converter configured for determining a compressed sensing matrix.

Claims

1. A system for characterizing a compressed sensing apparatus, the system comprising:

a random vector generator for generating an input vector;

a waveform generator in communication with the random vector generator for converting the input vector into an analog signal;

a compressed sensing apparatus in communication with the waveform generator for determining a first digital bit stream from the analog signal;

a serial-to-parallel converter for converting the first digital bit stream into a second vector; and

an electronic processor in communication with the random number generator and the serial-to-parallel converter configured for determining a compressed sensing matrix.

2. The system for characterizing a compressed sensing apparatus of claim 1, further comprising:

an electronic memory in communication with the electronic processor for storing a plurality of input vectors and a plurality of output vectors.

3. The system for characterizing a compressed sensing apparatus of claim 2, wherein the processor is configured to compute the compressed sensing matrix by multiplying the plurality of output vectors by the pseudo-inverse of the plurality of input vectors.

4. The system for characterizing a compressed sensing apparatus of claim 1, wherein the waveform generator further comprises:

a parallel-to-serial converter in communication with the random vector generator for converting the input vector into a second digital bit stream; and

a digital-to-analog converter in communication with the parallel-to-serial converter for converting the second digital bit stream into an analog signal.

5. The system for characterizing a compressed sensing apparatus of claim 1, wherein the electronic processor is selected from the group consisting of an FPGA, an ASIC, and a CPU.

6. The system for characterizing a compressed sensing apparatus of claim 1, wherein the random vector generator is operable to output an N-element Gaussian random discrete-time vector.

7. The system for characterizing a compressed sensing apparatus of claim 1, wherein the electronic processor calculates the compressed sensing matrix using a least-square pseudo inverse fit between the input vector and the output vector.

8. A method of characterizing a compressed sensing apparatus, the method comprising:

a data gathering step comprising;

generating an input vector of random numbers;

capturing the input vector;

converting the input vector into an analog signal;

reconstructing the analog signal into a digital bit stream using a compressed sensing apparatus;

converting the digital bit stream into an output vector; and

capturing the output vector;

creating an input matrix from the captured input vector;

creating an output matrix from the captured output vector; and

determining the compressed sensing matrix based on the input vector and the output vector.

9. The method of claim 8, wherein the step of determining the compressed sensing matrix further comprises calculating the least square-pseudo inverse between the plurality of input vectors and the plurality of output vectors.

10. The method of claim 8, wherein the input vector is an N-element Gaussian random discrete-time vector.

11. The method of claim 8, wherein the step of converting the input vector into an analog signal further comprises:

converting the input vector into a series of numbers; and

providing the series of numbers to a digital-to-analog converter for creating the analog signal.

12. The method of claim 8, wherein the step of determining the compressed sensing matrix is performed using a device chosen from the group consisting of an FPGA, an ASIC, and a CPU.

13. The method of claim 8, wherein:

the data gathering step is performed a number of times such that a plurality of input vectors are stored and a plurality of output vectors are stored; and

the input matrix is created from the plurality of captured input vectors and the output matrix is created from the plurality of captured output vectors such that each column of the input matrix corresponds to each column of the output matrix.

14. A system for characterizing a compressed sensing apparatus comprising:

a random vector generator for generating an input vector wherein the elements of the first vector comprise Gaussian random noise;

a parallel-to-serial converter in communication with the random vector generator for converting the input vector into a first digital bit stream;

a digital-to-analog converter in communication with the parallel-to-serial converter for converting the first digital bit stream into an analog signal;

a compressed sensing apparatus in communication with the digital-to-analog converter for determining a second digital bit stream from the analog signal;

a serial-to-parallel converter for converting the second digital bit stream into a second vector;

an electronic processor in communication with the random number generator and the serial-to-parallel converter; and

an electronic memory in communication with the electronic processor for storing a plurality of input vectors and a plurality of output vectors

wherein the electronic processor is configured for determining a compressed sensing matrix based on the plurality of input vectors and the plurality of output vectors by multiplying the plurality of output vectors by the pseudo-inverse of the plurality of input vectors.

15. The system for characterizing a compressed sensing apparatus of claim 14, wherein the electronic processor is selected from the group consisting of an FPGA, an ASIC, and a CPU.