Method for suppressing electromagnetic interference in an amplitude modulated radio signal, in particular received in an electric or hybrid vehicle
View Patent ↗A method for attenuating electromagnetic interference in an AM radio signal received by a radio receiver, so as to produce a filtered radio signal Z n defined by Z n =W n T Y n at the time n, where Y n is a vector the components of which correspond to the radio signal received by a first antenna and to a second signal received by a second antenna connected to the source of interference, respectively, and W n is a vector the components of which correspond to the complex coefficients of an impulse response filter, with: Zn= w 1,n ·y 1,n + w 2,n ·y 2,n , by the introduction of a correlation between the real and imaginary parts of the complex coefficients. The complex coefficients being expressed in polar coordinates, so that w 1 , n = g 1 , n e j 2 πθ 1 , n and w 2 , n = g 2 , n e j 2 πθ 2 , n , and of the implementation of a predetermined iterative algorithm configured to determine the g 1,n , g 2,n , θ 1,n and θ 2,n able to minimize the following cost function: J CPA = E { z z n - 1 2 } = E { e j θ z n - 1 } .
1. A method for attenuating electromagnetic interference in an amplitude-modulated radio signal received by a radio receiver, said electromagnetic interference being produced by a source of interference, so as to produce a filtered radio signal Z n defined by Z n =W n T Y n at the time n, where Y n is a vector the components of which correspond to the radio signal received by a first antenna of the radio receiver, said received radio signal containing electromagnetic interference, and a second signal received by a second antenna connected to the source of interference, respectively, said second signal corresponding to the electromagnetic interference, in complex baseband, and W n is a vector the components of which correspond to the characteristic complex coefficients of an impulse response filter, said complex coefficients being intended to be applied to the received radio signal expressed in complex baseband and to the second radio signal expressed in complex baseband, respectively, in order to form, after combination, the filtered radio signal Z n in which the electromagnetic interference is attenuated, the method comprising determining complex coefficients of the vector W n such that:
Z n =W n T Y n = w 1,n ·y 1,n + w 2,n ·y 2,n ,
by the introduction of a correlation between the real and imaginary parts of said complex coefficients, said complex coefficients being expressed in polar coordinates, so that
w
1
,
n
=
g
1
,
n
e
j
2
πθ
1
,
n
and
w
2
,
n
=
g
2
,
n
e
j
2
πθ
2
,
n
,
and of the implementation of a predetermined iterative algorithm configured to determine the g 1,n , g 2,n , θ 1,n and θ 2,n able to minimize the following cost function:
J
CPA
=
E
{
z
n
z
n
-
1
2
}
=
E
{
e
j
θ
z
n
-
1
}
so as to obtain the vector W n of characteristic complex coefficients of the impulse response filter to be applied to the received radio signal Y n in order to attenuate the electromagnetic interference thereof.
2. The method as claimed in claim 1 , wherein said iterative adaptation algorithm is a constant-phase adaptation algorithm.
3. The method as claimed in claim 2 , comprising the implementation of the iterative adaptation algorithm consisting in determining the complex gains G n and phase Θ n able to minimize over time the cost function characterized by the following instantaneous gradient:
∇
J
CPA
=
∇
z
_
n
×
f
(
z
n
)
+
∇
z
n
×
f
(
z
_
n
)
with
f
(
∝
)
=
1
2
∝
3
(
∝
2
-
∝
2
)
,
and
∇
z
n
=
∂
z
n
∂
Wn
=
[
∂
z
n
∂
g
n
∂
z
n
∂
θ
n
]
=
[
exp
(
-
j
Θ
n
)
Y
n
-
j
G
n
exp
(
-
j
Θ
n
)
Y
n
]
,
where G n is a vector composed of the gains of the complex coefficients at the time n and Θ n is the vector composed of the phases of the complex coefficients.
4. The method as claimed in claim 1 , comprising the implementation of the iterative adaptation algorithm consisting in determining the complex gains G n and phase Θ n able to minimize over time the cost function characterized by the following instantaneous gradient:
∇
J
CPA
=
∇
z
_
n
×
f
(
z
n
)
+
∇
z
n
×
f
(
z
_
n
)
with
f
(
∝
)
=
1
2
∝
3
(
∝
2
-
∝
2
)
,
and
∇
z
n
=
∂
z
n
∂
Wn
=
[
∂
z
n
∂
g
n
∂
z
n
∂
θ
n
]
=
[
exp
(
-
j
Θ
n
)
Y
n
-
j
G
n
exp
(
-
j
Θ
n
)
Y
n
]
,
where G n is a vector composed of the gains of the complex coefficients at the time n and Θ n is the vector composed of the phases of the complex coefficients.
5. The method as claimed in claim 4 , wherein the respective variations in G n , Θ n over time are computed by the following formulae:
G
n
+
1
=
G
n
-
μ
g
1
z
n
3
×
Re
[
(
z
n
2
-
z
n
2
)
exp
(
j
Θ
n
)
Y
n
_
]
Θ
n
+
1
=
Θ
n
-
μ
θ
1
z
n
3
×
Im
[
(
z
n
2
-
z
n
2
)
G
n
exp
(
j
Θ
n
)
Y
n
_
]
where G n is a vector composed of the gains of the complex coefficients at the time n, Θ n is the vector composed of the phases of the complex coefficients, and μ g and μ g are constants chosen depending on the rapidity and precision desired for the convergence of the iterative adaptation algorithm implemented.
6. A radio receiver comprising a microcontroller configured to implement the method as claimed in claim 1 .
7. A motor vehicle comprising a radio receiver as claimed in claim 6 .
8. The motor vehicle as claimed in claim 7 , said vehicle comprising an electric motor, wherein the first antenna is a receiving antenna of the radio receiver and the second antenna is an antenna connected to said electric motor of the motor vehicle, said electric motor being the source of interference.