Antenna array design and processing to eliminate false detections in a radar system

1. A method of eliminating false detections in a radar system, the method comprising:

arranging an array of antenna elements into two or more sub-arrays with a spacing between adjacent ones of the antenna elements of one of the two or more sub-arrays being different than a spacing between adjacent ones of the antenna elements of at least one other of the two or more sub-arrays;

receiving reflected signals at the two or more sub-arrays resulting from transmitting transmit signals from the antenna elements of the two or more sub-arrays;

processing the reflected signals to distinguish an actual angle from the radar system to an object that contributed to the reflected signals from ambiguous angles at which the false detections of the object are obtained; and

determining a location of the object as a result of the processing.

2. The method according to claim 1 , wherein the arranging the array of the antenna elements includes the spacing between the adjacent ones of the antenna elements of the one of the two or more sub-arrays and the spacing between the adjacent ones of the antenna elements of the at least one other of the two or more sub-arrays are both greater than half a wavelength of the transmit signals.

3. The method according to claim 1 , wherein the arranging the array of the antenna elements includes the one of the two or more sub-arrays including three or more of the antenna elements and the spacing between one pair of the adjacent ones of the three or more of the antenna elements is different than the spacing between another pair of the adjacent ones of the three or more of the antenna elements.

4. The method according to claim 1 , wherein the arranging the array of the antenna elements includes the one of the two or more sub-arrays including three or more of the antenna elements and the spacing between all adjacent pairs of the three or more of the antenna elements is a same.

5. The method according to claim 1 , wherein the processing the reflected signals includes generating clusters of detection points obtained with each of the two or more sub-arrays, each of the detection points having an associated angle and range.

6. The method according to claim 5 , wherein the processing the reflected signals includes determining a detection overlap metric γ and a distribution uniformity metric δ for each of the clusters.

7. The method according to claim 6 , wherein the determining the detection overlap metric γ includes computing:

γ

=

1

N

⁢

∑

i

-

0

N

-

1

⁢

(

μ

i

-

μ

¯

)

2

,

where

N is a number of the two or more sub-arrays, μ i is a mean of the detection points of an i th sub-array among the two or more sub-arrays, and μ i is a mean of all the detection points of all the sub-arrays, and a decreasing value of the detection overlap metric γ corresponds with an increasing overlap among the detection points and a correspondence of the detection points to the object.

8. The method according to claim 6 , wherein the determining the detection overlap metric γ includes computing:

γ

=

Σ

i

,

j

⁢

i

≠

j

⁢

X

i

⋂

X

j

X

i

⋃

X

j

,

or

γ

=

min

i

,

j

⁢

X

i

⋂

X

j

X

i

⋃

X

j

,

where

X i is a bounding box enclosing the detection points associated with an i th sub-array among the two or more sub-arrays and X j is a bounding box enclosing the detection points associated with any other of the two or more sub-arrays, and an increasing value of the detection overlap metric γ indicates corresponds with an increasing overlap among the detection points and a correspondence of the detection points to the object.

9. The method according to claim 6 , wherein the determining the distribution uniformity metric δ includes computing:

δ

=

∑

n

=

0

N

-

1

⁢

P

n

⁢

log

⁡

(

P

n

)

,

where

P

n

=

S

n

∑

n

=

0

N

-

1

⁢

S

n

,

and

S n is a number of the detection points of an n th sub-array among the two or more sub-arrays, and an increasing value of the distribution uniformity metric δ corresponds with an increasing likelihood of the number of detection points of the two or more sub-arrays being similar and a correspondence of the detection points to the object.

10. The method according to claim 6 , wherein the processing the reflected signals includes generating a combined metric η from the detection overlap metric γ and the distribution uniformity metric δ for each of the clusters as:

η

=

C

1

γ

1

+

C

2

⁢

δ

,

or

η

=

C

1

⁢

γ

2

+

C

2

⁢

δ

,

where

C 1 and C 2 are scale factors, and the processing includes using a value of the combined metric η exceeding a threshold as an indication that the detection points correspond with the object.

11. A radar system in a vehicle, comprising:

an array of antenna elements arranged as two or more sub-arrays with a spacing between adjacent ones of the antenna elements of one of the two or more sub-arrays being different than a spacing between adjacent ones of the antenna elements of at least one other of the two or more sub-arrays; and

a processor configured to receive reflected signals at the two or more sub-arrays resulting from transmitting transmit signals from the antenna elements of the two or more sub-arrays, to process the reflected signals to distinguish an actual angle from the radar system to an object that contributed to the reflected signals from ambiguous angles at which the false detections of the object are obtained, and to determine a location of the object as a result of the processing.

12. The radar system according to claim 11 , wherein the spacing between the adjacent ones of the antenna elements of the one of the two or more sub-arrays and the spacing between the adjacent ones of the antenna elements of the at least one other of the two or more sub-arrays are both greater than half a wavelength of the transmit signals.

13. The radar system according to claim 11 , wherein the one of the two or more sub-arrays includes three or more of the antenna elements and the spacing between one pair of the adjacent ones of the three or more of the antenna elements is different than the spacing between another pair of the adjacent ones of the three or more of the antenna elements, or the spacing between all adjacent pairs of the three or more of the antenna elements is a same.

14. The radar system according to claim 11 , wherein the processor is further configured to generate clusters of detection points obtained with each of the two or more sub-arrays, each of the detection points having an associated angle and range.

15. The radar system according to claim 14 , wherein the processor is configured to determine a detection overlap metric γ and a distribution uniformity metric δ for each of the clusters.

16. The radar system according to claim 15 , wherein the processor is configured to determine the detection overlap metric γ by computing:

γ

=

1

N

⁢

∑

i

-

0

N

-

1

⁢

(

μ

i

-

μ

¯

)

2

,

where

N is a number of the two or more sub-arrays, μ i is a mean of the detection points of an i th sub-array among the two or more sub-arrays, and μ i is a mean of all the detection points of all the sub-arrays, and a decreasing value of the detection overlap metric γ corresponds with an increasing overlap among the detection points and a correspondence of the detection points to the object.

17. The radar system according to claim 15 , wherein the processor is configured to determine the detection overlap metric γ by computing:

γ

=

Σ

i

,

j

⁢

i

≠

j

⁢

X

i

⋂

X

j

X

i

⋃

X

j

,

or

γ

=

min

i

,

j

⁢

X

i

⋂

X

j

X

i

⋃

X

j

,

where

X i is a bounding box enclosing the detection points associated with an i th sub-array among the two or more sub-arrays and X j is a bounding box enclosing the detection points associated with any other of the two or more sub-arrays, and an increasing value of the detection overlap metric γ corresponds with an increasing overlap among the detection points and a correspondence of the detection points to the object.

18. The radar system according to claim 15 , wherein the processor is configured to determine the distribution uniformity metric δ by computing:

δ

=

∑

n

=

0

N

-

1

⁢

P

n

⁢

log

⁡

(

P

n

)

,

where

P

n

=

S

n

∑

n

=

0

N

-

1

⁢

S

n

,

and

S n is a number of the detection points of an n th sub-array among the two or more sub-arrays, and an increasing value of the distribution uniformity metric δ corresponds with an increasing likelihood of the number of detection points of the two or more sub-arrays being similar and a correspondence of the detection points to the object.

19. The radar system according to claim 15 , wherein the processor is configured to generate a combined metric η from the detection overlap metric γ and the distribution uniformity metric δ for each of the clusters as:

η

=

C

1

γ

1

+

C

2

⁢

δ

,

or

η

=

C

1

⁢

γ

2

+

C

2

⁢

δ

,

where

C 1 and C 2 are scale factors, and the processor is further configured to use a value of the combined metric η exceeding a threshold as an indication that the detection points correspond with the object.

20. The radar system according to claim 11 , wherein the processor is further configured to control an operation of the vehicle based on the location of the object.