地面辐射源多普勒定位系统工作原理(一):多普勒定位系统原理
正巧遇到一个利用多普勒测量进行定位的相关课题,就以此为契机开始分享之路。
–记于2023.05.02 南京
基本原理
无人机接收地面发射机发送的无线电信号,测定无人机接收多普勒频谱,得到以发射机位置为未知变量的方程,通过三次及以上测量,联立观测方程,利用牛顿迭代法进行迭代解算,能够推算出发射机位置。
假设地面发射机用固定频率 f s {f_s} fs发射连续的无线电波,由于无人机运动,其和地面发射机间存在多普勒效应,多普勒频移 f d {f_d} fd可以表示为: f d = f r − f s = v r − v s c ⋅ x s − x r ∥ x s − x r ∥ f s {f_d} = {f_r} - {f_s} = {{{{\bf{v}}_r} - {{\bf{v}}_s}} \over c} \cdot {{{{\bf{x}}_s} - {{\bf{x}}_r}} \over {\left\| {{{\bf{x}}_s} - {{\bf{x}}_r}} \right\|}}{f_s} fd=fr−fs=cvr−vs⋅∥xs−xr∥xs−xrfs
其中 f r {f_r} fr为无人机接收信号频率, v s = [ v x s v y s v z s ] {{\bf{v}}_s} = \left[ {\begin{array}{ccccccccccccccc}{v_x^s}&{v_y^s}&{v_z^s}\end{array}} \right] vs=[vxsvysvzs], v r = [ v x r v y r v z r ] {{\bf{v}}_r} = \left[ {\begin{array}{ccccccccccccccc}{v_x^r}&{v_y^r}&{v_z^r}\end{array}} \right] vr=[vxrvyrvzr]分别为发射机运动速度和无人机运动速度; x s = [ x s y s z s ] T {{\bf{x}}_s} = {\left[ {\begin{array}{ccccccccccccccc}{{x^s}}&{{y^s}}&{{z^s}}\end{array}} \right]^{\rm{T}}} xs=[xsyszs]T , x r = [ x r y r z r ] T {{\bf{x}}_r} = {\left[ {\begin{array}{ccccccccccccccc}{{x^r}}&{{y^r}}&{{z^r}}\end{array}} \right]^{\rm{T}}} xr=[xryrzr]T 分别为发射机位置和无人机接收机位置。
当地面发射机静止时,多普勒频移可以进一步表示为:
f
d
=
v
r
c
⋅
x
s
−
x
r
∥
x
s
−
x
r
∥
f
s
=
v
r
⋅
e
c
f
s
{f_d} = {{{{\bf{v}}_r}} \over c} \cdot {{{{\bf{x}}_s} - {{\bf{x}}_r}} \over {\left\| {{{\bf{x}}_s} - {{\bf{x}}_r}} \right\|}}{f_s} = {{{{\bf{v}}_r} \cdot {\bf{e}}} \over c}{f_s}
fd=cvr⋅∥xs−xr∥xs−xrfs=cvr⋅efs其中
e
{\bf{e}}
e为发射机指向无人机的方向矢量。当考虑接收机钟偏和噪声时,接收机瞬时多普勒频移观测方程可以表示为:
f
d
=
v
r
c
⋅
x
s
−
x
r
∥
x
s
−
x
r
∥
f
s
+
δ
r
+
w
{f_d} = {{{{\bf{v}}_r}} \over c} \cdot {{{{\bf{x}}_s} - {{\bf{x}}_r}} \over {\left\| {{{\bf{x}}_s} - {{\bf{x}}_r}} \right\|}}{f_s} + {\delta _r} + w
fd=cvr⋅∥xs−xr∥xs−xrfs+δr+w其中
δ
r
{\delta _r}
δr表示接收机时钟频率误差,
w
w
w为热噪声、电离层和对流层延迟率以及其他次要影响造成的偏差。
令
u
=
[
x
s
δ
r
]
T
{\bf{u}} = {\left[ {\begin{array}{ccccccccccccccc}{{{\bf{x}}_s}}&{{\delta _r}}\end{array}} \right]^{\rm{T}}}
u=[xsδr]T为发射机三维坐标和接收机钟偏4个待求未知数。假设发射机初始估计解为
[
x
0
y
0
z
0
δ
r
0
]
T
{\left[ {\begin{array}{ccccccccccccccc}{\begin{array}{ccccccccccccccc}{{x_0}}&{{y_0}}&{{z_0}}\end{array}}&{{\delta _{r0}}}\end{array}} \right]^{\rm{T}}}
[x0y0z0δr0]T,将多普勒观测方程线性化可得:
f
(
u
)
=
f
(
u
0
)
+
∂
f
∂
x
∣
x
=
x
0
(
x
−
x
0
)
+
∂
f
∂
y
∣
y
=
y
0
(
y
−
y
0
)
+
∂
f
∂
z
∣
z
=
z
0
(
z
−
z
0
)
+
(
δ
r
−
δ
r
0
)
+
o
˙
\begin{array}{c}f\left( {\bf{u}} \right) = f\left( {{{\bf{u}}_0}} \right) + {\left. {\frac{{\partial f}}{{\partial x}}} \right|_{x = {x_0}}}\left( {x - {x_0}} \right) + {\left. {\frac{{\partial f}}{{\partial y}}} \right|_{y = {y_0}}}\left( {y - {y_0}} \right) + \\{\left. {\frac{{\partial f}}{{\partial z}}} \right|_{z = {z_0}}}\left( {z - {z_0}} \right) + \left( {{\delta _r} - {\delta _{r0}}} \right) + \dot o\end{array}
f(u)=f(u0)+∂x∂f
x=x0(x−x0)+∂y∂f
y=y0(y−y0)+∂z∂f
z=z0(z−z0)+(δr−δr0)+o˙其中
o
˙
\dot o
o˙为高阶小量。
当无人机接收机获得多次多普勒测量时,联立观测方程得到联立方程组以及状态更新矩阵:
G
[
x
−
x
0
y
−
y
0
z
−
z
0
δ
r
−
δ
r
0
]
+
ς
=
G
⋅
Δ
u
k
+
ς
{\bf{G}}\left[ {\begin{array}{ccccccccccccccc}{x - {x_0}}\\{y - {y_0}}\\{z - {z_0}}\\{{\delta _r} - {\delta _{r0}}}\end{array}} \right] + \varsigma = {\bf{G}} \cdot \Delta {{\bf{u}}_k} + \varsigma
G
x−x0y−y0z−z0δr−δr0
+ς=G⋅Δuk+ς
Δ
f
=
[
f
(
1
)
(
u
)
−
f
(
1
)
(
u
k
)
f
(
2
)
(
u
)
−
f
(
2
)
(
u
k
)
⋮
f
(
n
)
(
u
)
−
f
(
n
)
(
u
k
)
]
\Delta {\bf{f}} = \left[ {\begin{array}{ccccccccccccccc}{{f^{\left( 1 \right)}}\left( {\bf{u}} \right) - {f^{\left( 1 \right)}}\left( {{{\bf{u}}_k}} \right)}\\{{f^{\left( 2 \right)}}\left( {\bf{u}} \right) - {f^{\left( 2 \right)}}\left( {{{\bf{u}}_k}} \right)}\\ \vdots \\{{f^{\left( n \right)}}\left( {\bf{u}} \right) - {f^{\left( n \right)}}\left( {{{\bf{u}}_k}} \right)}\end{array}} \right]
Δf=
f(1)(u)−f(1)(uk)f(2)(u)−f(2)(uk)⋮f(n)(u)−f(n)(uk)
G
=
[
∂
f
(
1
)
∂
x
∂
f
(
1
)
∂
y
∂
f
(
1
)
∂
z
1
∂
f
(
2
)
∂
x
∂
f
(
2
)
∂
y
∂
f
(
2
)
∂
z
1
⋮
⋮
⋮
⋮
∂
f
(
n
)
∂
x
∂
f
(
n
)
∂
y
∂
f
(
n
)
∂
z
1
]
{\bf{G}} = \left[ {\begin{array}{ccccccccccccccc}{\frac{{\partial {f^{\left( 1 \right)}}}}{{\partial x}}}&{\frac{{\partial {f^{\left( 1 \right)}}}}{{\partial y}}}&{\frac{{\partial {f^{\left( 1 \right)}}}}{{\partial z}}}&1\\{\frac{{\partial {f^{\left( 2 \right)}}}}{{\partial x}}}&{\frac{{\partial {f^{\left( 2 \right)}}}}{{\partial y}}}&{\frac{{\partial {f^{\left( 2 \right)}}}}{{\partial z}}}&1\\ \vdots & \vdots & \vdots & \vdots \\{\frac{{\partial {f^{\left( n \right)}}}}{{\partial x}}}&{\frac{{\partial {f^{\left( n \right)}}}}{{\partial y}}}&{\frac{{\partial {f^{\left( n \right)}}}}{{\partial z}}}&1\end{array}} \right]
G=
∂x∂f(1)∂x∂f(2)⋮∂x∂f(n)∂y∂f(1)∂y∂f(2)⋮∂y∂f(n)∂z∂f(1)∂z∂f(2)⋮∂z∂f(n)11⋮1
其中k为迭代次数,
u
k
{{\bf{u}}_k}
uk为第k次迭代解算 的估计值,
ς
\varsigma
ς为残差向量,
Δ
f
\Delta f
Δf为多普勒观测量真实值与预测值的差值向量。根据最小二乘原理解得:
Δ
u
k
=
(
G
T
G
)
−
1
G
T
⋅
Δ
f
\Delta {{\bf{u}}_k} = {\left( {{{\bf{G}}^{\rm{T}}}{\bf{G}}} \right)^{ - 1}}{{\bf{G}}^{\rm{T}}} \cdot \Delta {\bf{f}}
Δuk=(GTG)−1GT⋅Δf将解算出的
Δ
u
k
\Delta {{\bf{u}}_k}
Δuk对上一步估计解
u
k
{{\bf{u}}_k}
uk进行更正,则更新后的估计解为
u
k
+
1
=
u
k
+
Δ
u
k
{{\bf{u}}_{k + 1}} = {{\bf{u}}_k} + \Delta {{\bf{u}}_k}
uk+1=uk+Δuk将其代入下一次迭代运算,直到
Δ
u
k
\Delta {{\bf{u}}_k}
Δuk的二范数小于某个设定的阈值时推出迭代。此时的
u
k
{{\bf{u}}_k}
uk即为最终的发射机3维位置坐标和接收机钟偏估计解。
结语
本篇文章整理了基于牛顿迭代法的地面辐射源多普勒定位系统工作原理,后续将整理如何根据发射信号测量多普勒。
如理解有误,恳请大家批评指正!!!