US 11,994,603 B1 · Method of sparse array oriented approach for DOA estimation of grating lobe target filtering

Abstract

A method of sparse array oriented approach for DOA estimation of grating lobe target filtering includes sub-array division processing of received echoes in multiple channels; digital beamforming is performed on each sub-array obtained after division to realize DOA estimation; the echo power of the beam pointing at each angle is calculated and the peak point is detected; peak threshold discrimination processing is performed based on the determined peak threshold; if the current sub-array satisfies the identified peak threshold, the corresponding power spectrum extreme is calculated to obtain the final DOA estimation. The present invention also relates to a corresponding device, processor and computer-readable storage medium thereof. With the use of this sparse array oriented method, device, processor and computer-readable storage medium for implementing filtered DOA estimation of a gate lobe target, the interference of the gate target is effectively avoided by dividing the subarray and binarizing the angular power spectrum.

A method of sparse array oriented approach for Direction Of Arrival (DOA) estimation of grating lobe target filtering for a radio signal received by an array antenna with a sparse layout, comprising following steps:

(1) receiving echoes of the radio signal in multiple channels by the array antenna and dividing, by a processor, the received echoes into a plurality of sub-arrays;
(2) performing, by the processor, digital beam formation on each sub-array obtained after the division to realize DOA estimation;
(3) calculating, by the processor, an echo power of a beam pointing at each angle and determining a peak point from the calculated echo power;
(4) determining a peak threshold discrimination based on the determined peak threshold;
(5) if a current sub-array satisfies the determined peak threshold, then calculating a corresponding power spectrum extreme to obtain a final DOA estimation;
wherein said step (1) specifically comprises:
(1.1) forming, as a unit of row, the sub-array division in an azimuthal dimension;
(1.2) modeling the received echoes of each array element and calculating the echoes of each channel using the following equations:
S
r
=
[
S
r
1
⁢
S
r
2
⁢
…
⁢
S
rN
]
where the Srn denotes each receiving channel echo, (n=1, 2 . . . N), and the N is the number of receiving channels;
(1.3) processing the received echoes in each channel for wave path-difference cancellation and performing phase compensation by using a steering vector for the digital beam formation, where a two-dimensional guiding vector is expressed using the following equation:
α(θ, φ)=[1, e−j2πϕ1(θ, φ), . . . , e−j2πϕN(θ, φ)]
where ϕn(θ, φ) denotes a phase difference between the nth channel and a reference channel, (n=1, 2 . . . N), the phase difference also includes azimuthal phase difference ϕnazim and pitch phase difference ϕnelev, as follows:
ϕn(θ, φ)=ϕnelev+ϕnazim
ϕ
n
azim
(
θ
,
φ
)
=
Δ
⁢
d
n
x
⁢
cos
⁢
φ
⁢
sin
⁢
θ
λ
ϕ
n
elev
(
φ
)
=
Δ
⁢
d
n
y
⁢
sin
⁢
φ
λ
where Δdnx denotes an azimuthal distance between the nth channel and the reference channel, Δdny denotes a pitching direction distance between the nth channel and the reference channel, λ denotes the wavelength, θ denotes an azimuth angle, and φ denotes the pitch angle;
(1.4) completing the subarray division process;
wherein said step (3) specifically comprises:
(3.1) multiplying the steering vector by the each channel echoes to obtain the echo power of the digital beam pointing at each angle, as calculated using the following equation:
pi(θ)=αH(θ)SrSrHα(θ)
where i denotes the ith subarray, pi(θ) denotes the power spectrum of the ith sub-array, αH(θ) denotes the steering vector at an azimuthal angle of θ, Sr denotes the echo vector, and SrH is the conjugate transpose of Sr;
(3.2) normalizing and logarithmizing the power spectrum of each sub-array according to the following equation to calculate the angular power spectrum maximum and thus determine the peak threshold thr:
p
i
norm
(
θ
)
=
10
⁢
log
10
⁢
p
i
(
θ
)
max
[
p
i
(
θ
)
]
where pinorm(θ) denotes power spectrum after normalization of the ith sub-array;
and
once the normalized power spectrum is complete, determining the peak threshold thr according to ½ of the power maximum;
(3.3) for peak point detection of the normalized spectrogram pinorm(θ) for each sub-array, retaining the power pinorm(θpeak) at all peak points θpeak;
wherein said step (4) specifically comprises:
setting the angular power greater than or equal to said peak threshold thr to 1 and the angular power less than said peak threshold thr to 0, specifically:
p
i
binary
(
θ
)
=
{
1
p
i
norm
(
θ
peak
)
≥
thr
0
p
i
norm
(
θ
peak
)
<
thr
where pibinary(θ)denotes the binary spectrogram of the ith sub-array.

134 more elements

Method of sparse array oriented approach for DOA estimation of grating lobe target filtering

Life of this patent

Method of sparse array oriented approach for DOA estimation of grating lobe target filtering

Life of this patent

Abstract

Claims6