Updating a filter of an equalizer
View Patent ↗In one aspect, a tuner includes an analog front end to receive a radio frequency (RF) signal and to downconvert the RF signal to a second frequency signal, a digitizer to convert the second frequency signal to a digitized signal, a channel equalizer including a filter to filter the digitized signal, and a first controller to update the filter according to a frequency response of the filter.
1. A tuner comprising:
an analog front end to receive a radio frequency (RF) signal and to downconvert the RF signal to a second frequency signal;
a digitizer to convert the second frequency signal to a digitized signal;
a channel equalizer including a filter to filter the digitized signal;
a first controller to update the filter according to a frequency response of the filter, the first controller comprising:
a frequency response determination logic to receive a plurality of taps of the filter and determine based thereon a magnitude value of the frequency response of the filter;
a comparison logic to compare the magnitude value to a reference value to generate an error signal based on a cost function according to
min
h
J
=
{
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
w
0
*
h
(
m
)
]
2
-
1
}
2
;
and
a tap update logic to update one or more taps of the plurality of taps of the filter based on the error signal.
2. The tuner of claim 1 , further comprising a second controller to update the filter based on a blind adaptive algorithm, wherein the first controller and the second controller are configured to independently update the filter.
3. The tuner of claim 1 , wherein the first controller is to update the one or more taps of the filter to increase a bandwidth of the filter.
4. The tuner of claim 1 , wherein the frequency response determination logic is to determine the magnitude response of the filter for each of a first plurality of frequency points.
5. The tuner of claim 4 , wherein the first plurality of frequency points are within a first portion of a radio channel.
6. The tuner of claim 1 , wherein the tap update logic is to update the one or more taps of the filter based on the error signal according to a least mean squared technique according to
h
(
k
)
n
=
h
(
k
)
n
-
1
-
μ
×
{
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
ω
o
×
h
(
m
)
]
2
-
1
}
×
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
ω
o
×
h
(
m
)
]
×
ⅇ
ⅈ
k
ω
o
.
7. The tuner of claim 1 , further comprising a control logic to selectively enable the first controller based at least in part on a condition of the RF signal.
8. The tuner of claim 7 , wherein the control logic comprises a deviation logic to enable the first controller based at least in part on a deviation of a demodulated signal obtained from the RF signal.
9. The tuner of claim 7 , wherein the control logic comprises a fading logic to enable the first controller based at least in part on a fading of the RF signal being less than a first threshold.
10. The tuner of claim 1 , wherein the first controller is to perform a discrete Fourier transform on the plurality of taps of the filter to determine the frequency response of the filter.
11. The tuner of claim 1 , wherein the first controller is to determine the frequency response of the filter based on a discrete Fourier transform performed in a frequency range corresponding to a selected portion of a channel bandwidth of a radio channel.
12. A system comprising:
an antenna to receive a radio frequency (RF) signal; and
a radio receiver coupled to the antenna to receive and process the RF signal to output an audio signal, the radio receiver including:
an analog front end to receive the RF signal and downconvert the RF signal to a second frequency signal;
an analog-to-digital converter (ADC) to convert the second frequency signal to a digitized signal;
a digital signal processor (DSP) to receive and demodulate the digitized signal and to provide a digital audio output, the DSP having a signal processing path including a filter having controllable coefficients, the coefficients controllable based at least in part on a frequency response of the filter determined based at least in part on the controllable coefficients; and
a first controller to update the controllable coefficients according to the frequency response of the filter, wherein the first controller is to determine an error signal according to
min
h
J
=
{
[
∑
m
=
0
N
-
1
ⅇ
-
imw
0
*
h
(
m
)
]
2
-
1
}
2
,
and to update one or more of the controllable coefficients based on the error signal according to a least mean squared technique according to
h
(
k
)
n
=
h
(
k
)
n
-
1
-
μ
×
{
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
ω
o
×
h
(
m
)
]
2
-
1
}
×
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
ω
o
×
h
(
m
)
]
×
ⅇ
ⅈ
k
ω
o
.
13. The system of claim 12 , further comprising a second controller to update the controllable coefficients based on a blind adaptive algorithm, wherein the first controller and the second controller are configured to independently update the controllable coefficients.
14. The system of claim 12 , further comprising a control logic to selectively enable the first controller based at least in part on a condition of the RF signal, wherein the control logic is to enable the first controller based on one or more of a deviation of a demodulated signal obtained from the RF signal and a fading level of the RF signal.
15. The system of claim 12 , wherein the first controller is to perform a discrete Fourier transform on the controllable coefficients of the filter to determine the frequency response of the filter.
16. A method comprising:
obtaining filter taps of an equalizer filter of a tuner processing an incoming radio frequency (RF) signal, in a frequency response controller of the tuner;
obtaining a magnitude response corresponding to a frequency response of the equalizer filter, based at least in part on a discrete Fourier transform performed on the filter taps;
comparing the magnitude response to a reference value to obtain an error signal based on a cost function according to
min
h
J
=
{
[
∑
m
=
0
N
-
1
ⅇ
-
ⅈ
m
w
0
*
h
(
m
)
]
2
-
1
}
2
;
and
adjusting one or more of the filter taps based on the error signal.
17. The method of claim 16 , further comprising adjusting the one or more filter taps independently of a constant modulus filter tap update mechanism.
18. The method of claim 16 , further comprising:
weighting a first set of filter tap updates obtained using the error signal and a second set of filter tap updates obtained using a constant modulus filter tap update mechanism; and
adjusting the one or more filter taps based on the weighting.
19. The method of claim 16 , further comprising selectively enabling the frequency response controller based on behavior of the incoming RF signal.
20. The method of claim 19 , further comprising:
enabling the frequency response controller if a fading value of the incoming RF signal is less than a first threshold value; and
disabling the frequency response controller if the fading value is greater than a second threshold value.
21. The method of claim 16 , further comprising adjusting the one or more filter taps to maintain a bandwidth of the equalizer filter above a predetermined level.