Target detection apparatus and method using angle estimation of MPSK-MIMO FMCW radar
View Patent ↗A target detection method according to the present disclosure includes generating a range-velocity map from a radar signal of an MPSK-MIMO FMCW radar, detecting a plurality of target signals including a real target signal and a ghost target signal for a target with respect to a velocity axis, estimating an angle of arrival of the target by applying a Capon beamforming algorithm to a target detection result, configuring a phase sequence by extracting phase values for the plurality of target signals from the range-velocity map and arranging the phase values in descending order of velocity value, and arranging differently a plurality of prediction phase values derived using the angle of arrival according to a preset rule to obtain a plurality of candidate phase sequences, and calculating a correlation coefficient between the phase sequence and the plurality of candidate phase sequences and identifying a real target signal in the range-velocity map.
1 . A target detection method performed by a target detection device based on an MPSK-MIMO FMCW radar, the target detection method comprising:
generating a range-velocity map from a radar signal transmitted from the MPSK-MIMO FMCW radar and then reflected and received;
detecting a plurality of target signals including a real target signal and a ghost target signal for a target with respect to a velocity axis from the range-velocity map;
estimating an angle of arrival of the target by applying a Capon beamforming algorithm to a target detection result on the range-velocity map;
configuring a phase sequence by extracting phase values for the plurality of target signals from the range-velocity map and arranging the phase values in descending order of velocity value;
deriving a plurality of prediction phase values by using the angle of arrival and arranging differently the plurality of prediction phase values according to a preset rule to obtain a plurality of candidate phase sequences; and
calculating a correlation coefficient between the phase sequence and the plurality of candidate phase sequences and identifying a real target signal among a plurality of target signals of different velocities in the range-velocity map by using the candidate phase sequence representing a highest correlation coefficient.
2 . The target detection method of claim 1 , wherein
the detecting of the plurality of target signals includes detecting M target signals with different velocities per target for each velocity axis in N range-velocity maps obtained from M transmission antennas and N reception antennas.
3 . The target detection method of claim 2 , wherein the configuring of the phase sequence includes:
extracting 2N phase values in response to a total of 2N target signals obtained for the N range-velocity maps in a BPSK method in which M=2; and
configuring the phase sequence including a total of 2N phase values by arranging N phase values extracted from a signal with a higher velocity value in order of reception channel among two target signals in each of the N range-velocity maps and then arranging N phase values extracted from other signals with lower velocity values in order of the reception channel.
4 . The target detection method of claim 2 , wherein, in the obtaining of the plurality of candidate phase sequences,
a total of 2N prediction phase values from e j0ω to e i(2N-1)ω are calculated in response to a total of 2N target signals detected from signals of a total of N reception antennas in a BPSK method in which M=2, and
ω is a phase difference between signals and is defined by an equation
ω
=
2
π
d
sin
(
θ
)
λ
where d represents an interval between the N reception antennas, θ represents an angle of arrival, and λ represents a radar wavelength.
5 . The target detection method of claim 4 , wherein
the obtaining of the plurality of candidate phase sequences includes generating a first candidate phase sequence S1 in which a target signal corresponding to a higher velocity value among two target signals in each of range-velocity maps is assumed to be the real target signal and a second candidate phase sequence S2 in which a target signal corresponding to a lower velocity value is assumed to be the real target signal, when M=2 and N=4, as follows
S
1
=
[
e
j
0
e
jw
e
j
2
w
e
j
3
w
e
j
4
w
e
j
5
w
e
j
6
w
e
j
7
w
]
,
and
S
2
=
[
e
j
4
w
e
j
5
w
e
j
6
w
e
j
7
w
e
j
0
e
jw
e
j
2
w
e
j
3
w
]
.
6 . The target detection method of claim 5 , wherein the identifying of the real target signal includes:
comparing a first correlation coefficient between the phase sequence and the first candidate phase sequence with a second correlation coefficient between the phase sequence and the second candidate phase sequence; and
identifying a target signal located at a higher velocity value among the two target signals in each of the range-velocity maps as the real target signal when the first correlation coefficient is greater than the second correlation coefficient, and identifying a target signal located at a lower velocity value as the real target signal when the second correlation coefficient is greater than the first correlation coefficient.
7 . A target detection device based on a MPSK-MIMO FMCW radar, the target detection device comprising:
a signal processor configured to generate a range-velocity map from a radar signal transmitted from the MPSK-MIMO FMCW radar and then reflected and received;
a signal detector configured to detect a plurality of target signals including a real target signal and a ghost target signal for a target with respect to a velocity axis from the range-velocity map;
an angle estimator configured to estimate an angle of arrival of the target by applying a Capon beamforming algorithm to a target detection result on the range-velocity map;
a phase sequence generator configuring a phase sequence by extracting phase values for the plurality of target signals from the range-velocity map and configured to arrange the phase values in descending order of velocity value;
a candidate phase sequence generator configured to derive a plurality of prediction phase values by using the angle of arrival and arranging differently the plurality of prediction phase values according to a preset rule to obtain a plurality of candidate phase sequences; and
a signal identifier configured to calculate a correlation coefficient between the phase sequence and the plurality of candidate phase sequences and identify a real target signal among a plurality of target signals of different velocities in the range-velocity map by using the candidate phase sequence representing a highest correlation coefficient.
8 . The target detection device of claim 7 , wherein
the signal detector detects M target signals with different velocities per target for each velocity axis in N range-velocity maps obtained from M transmission antennas and N reception antennas.
9 . The target detection device of claim 8 , wherein
the phase sequence generator extracts 2N phase values in response to a total of 2N target signals obtained for the N range-velocity maps in a BPSK method in which M=2, and configures the phase sequence including a total of 2N phase values by arranging N phase values extracted from a signal with a higher velocity value in order of reception channel among two target signals in each of the N range-velocity maps and then arranging N phase values extracted from other signals with lower velocity values in order of the reception channel.
10 . The target detection device of claim 8 , wherein
the candidate phase sequence generator calculates a total of 2N prediction phase values from e j0ω to e i(2N-1)ω in response to a total of 2N target signals detected from signals of a total of N reception antennas in a BPSK method in which M=2, and
ω is a phase difference between signals and is defined by an equation
ω
=
2
π
d
sin
(
θ
)
λ
where d represents an interval between the N reception antennas, θ represents an angle of arrival, and λ represents a radar wavelength.
11 . The target detection device of claim 10 , wherein
The candidate phase sequence generator generates a first candidate phase sequence S1 in which e i0ω to e i3ω are arranged in a front portion by assuming a target signal corresponding to a higher velocity value among two target signals in each of range-velocity maps as the real target signal and a second candidate phase sequence S2 in which e i4ω to e i7ω are arranged in a front portion by assuming a target signal corresponding to a lower velocity value as the real target signal, when M=2 and N=4, as follows:
S
1
=
[
e
j
0
e
jw
e
j
2
w
e
j
3
w
e
j
4
w
e
j
5
w
e
j
6
w
e
j
7
w
]
,
and
S
2
=
[
e
j
4
w
e
j
5
w
e
j
6
w
e
j
7
w
e
j
0
e
jw
e
j
2
w
e
j
3
w
]
.
12 . The target detection device of claim 11 , wherein
the signal identifier compares a first correlation coefficient between the phase sequence and the first candidate phase sequence with a second correlation coefficient between the phase sequence and the second candidate phase sequence, and identifies a target signal located at a higher velocity value among the two target signals in each of the range-velocity maps as the real target signal when the first correlation coefficient is greater than the second correlation coefficient, and identifies a target signal located at a lower velocity value as the real target signal when the second correlation coefficient is greater than the first correlation coefficient.