Electromagnetic (EM) solver using a shooting bouncing ray (SBR) technique

• US 7,880,671 B2
• Filed: 09/18/2008
• Issued: 02/01/2011
• Est. Priority Date: 09/18/2008
• Status: Expired due to Fees

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Lee, Chul J., Villanueva, Axel R.
• Primary Examiner(s): Tarcza; Thomas H
• Assistant Examiner(s): Brainard; Timothy A
• Application Number:
  US12/212,783
• Publication Number:
  US 20100066595A1
• Time in Patent Office: 866 Days
• Field of Search: 342/169
• US Class Current: 342/169
• CPC Class Codes: G01S 7/412 based on a comparison between measured values and known or stored values

Abstract

In one aspect, a system to generate radar signatures for multiple objects in real-time includes a first module including at least one processor to perform a shooting and bouncing (SBR) technique to solve for physical optics and multi-bounce characteristics of the objects. The at least one processor includes a central processing unit to perform dynamic ray tracing and a graphics processing unit (GPU) to perform far field calculations. The GPU includes a hit point database to store entries associated with rays that intersect an object.

Claims

1. A system comprising:

a first module comprising at least one processor to perform a shooting and bouncing (SBR) technique to determine characteristics of multiple objects, the characteristics comprising;

physical optics of electromagnetic (EM) physics; and

multi-bounce interaction of EM physics, wherein the at least one processor comprising;

a central processing unit to perform dynamic ray tracing;

a graphics processing unit (GPU) to perform far field calculations, wherein the GPU comprises a hit point database to store entries associated with rays that intersect an object, and

wherein the system generates radar signatures for the multiple objects in real-time based on the characteristics determined.

2. The system of claim 1 wherein the GPU further comprises threads and each thread operates on a single entry in the hit point database.

3. The system of claim 2 wherein the GPU further comprises:

registers to receive hit point entries to store a state of each hit point, a hit point entry comprising direction, origin and distance;

a shared memory; and

a far field results memory.

4. The system of claim 3 wherein the GPU further comprises:

a constant memory; and

a plurality of multiprocessors comprising stream processors,

wherein a multiprocessor is assigned a block of the threads to process and a stream processor operates on a single thread in the block.

5. The system of claim 4 wherein the GPU is configured, on a per frequency basis, to:

read entries of the hit point database into the registers;

enable the threads to;

process register entries; and

save results into the shared memory on a per thread basis;

sum the results to form far field results; and

save the far-field results in the far field results memory.

6. A method, comprising:

generating radar signatures for multiple objects in real-time based on characteristics comprising;

performing a shooting and bouncing (SBR) technique to determine the characteristics of the multiple objects, the characteristics comprising;

physical optics of electromagnetic (EM) physics; and

multi-bounce interaction of EM physics, wherein performing the SBR technique comprises;

performing dynamic ray tracing using a central processing unit; and

performing far field calculations using a graphics processing unit (GPU) by using a hit point database to store entries associated with rays that intersect an object.

7. The method of claim 6, further comprising using each thread of the GPU to operate on a single entry in the hit point database.

8. The method of claim 7, further comprising using registers to store a state of each hit point, a hit point entry comprising direction, origin and distance.

9. The method of claim 7, further comprising assigning a multiprocessor to process a block of threads.

10. The method of claim 9, further comprising assigning a stream processor in the multiprocessor to operate on a single thread in the block.

11. The method of claim 9, further comprising on a per frequency basis:

reading entries of the hit point database into registers;

enabling the threads to;

process entries of the registers; and

save results into a shared memory on a per thread basis;

summing the results to form far field results; and

saving the far-field results in a far field results memory.

12. An article comprising:

a machine-readable medium that stores executable instructions to generate radar signatures for multiple objects in real-time based on characteristics, the instructions causing a machine to;

perform a shooting and bouncing (SBR) technique to determine the characteristics of the multiple objects, the characteristics comprising;

physical optics of electromagnetic (EM) physics; and

multi-bounce interaction of EM physics,

wherein the instructions causing a machine to perform the SBR technique comprises instructions causing a machine to;

perform dynamic ray tracing using a central processing unit; and

perform far field calculations using a graphics processing unit (GPU) by using a hit point database to store entries associated with rays that intersect an object.

13. The article of claim 12, further comprising instructions causing a machine to use each thread of the GPU to operate on a single entry in the hit point database.

14. The article of claim 12, further comprising instructions causing a machine to use registers to store a state of each hit point, a hit point entry comprising direction, origin and distance.

15. The article of claim 14, further comprising instructions causing a machine to assign a multiprocessor to process a block of threads.

16. The article of claim 15, further comprising instructions causing a machine to assign a stream processor in the multiprocessor to operate on a single thread in the block.

17. The article of claim 15, further comprising instructions causing a machine to on a per frequency basis:

read entries of the hit point database into registers;

enable the threads to;

process entries of the registers; and

save results into a shared memory on a per thread basis;

sum the results to form far field results; and

save the far-field results in a far field results memory.

18. A system to generate radar signatures for multiple objects in real-time comprising:

a first module comprising at least one processor to perform a shooting and bouncing (SBR) technique to solve for physical optics and multi-bounce characteristics of the objects, the at least one processor to perform the SBR technique comprising;

a central processing unit to perform dynamic ray tracing;

a graphics processing unit (GPU) to perform far field calculations;

the GPU comprising a hit point database to store entries associated with rays that intersect an object; and

a second module comprising at least one processor to perform a physical theory (PTD) technique to solve for material edges of the objects; and

a third module comprising at least one processor to perform an incremental length diffraction coefficient (ILDC) technique to solve for material boss/channel,

wherein results from the first, second and third modules are coherently integrated by frequency to generate RCS values of the objects in real-time.

19. The system of claim 18 wherein the GPU further comprises:

threads and each thread operates on a single entry in the hit point database;

registers to receive hit point entries to store a state of each hit point, a hit point entry comprising direction, origin and distance;

a shared memory;

a far field results memory;

a constant memory; and

a plurality of multiprocessors comprising stream processors,

wherein a multiprocessor is assigned a block of the threads to process and a stream processor operates on a single thread in the block.

20. The system of claim 19 wherein the GPU is configured, on a per frequency basis, to:

read entries of the hit point database into the registers;

enable the threads to;

process register entries; and

save results into the shared memory on a per thread basis;

sum the results to form far field results; and

save the far-field results in the far field results memory.