Apparatuses and methodologies for blind equalization in a communication system
A method and apparatus for equalization in a communication system. The method includes receiving an input of a block of symbols, filtering the input in the frequency domain, calculating an error factor in the time domain based on the filtered input, updating tap weights of the equalizer in the frequency domain using circular correlation based on the error factor and the conjugate of the input in the frequency domain, and outputting the filtered input.
1. A method for equalization in a single carrier frequency division multiple access communication system, the method comprising:
receiving an input of a block of symbols, the block symbols being periodic transmitted symbols and being circularly convoluted by a channel;
filtering, using processing circuitry of an equalizer of a receiver, the input in the frequency domain;
calculating an error factor in the time domain based on the filtered input;
updating tap weights of the equalizer in the frequency domain using circular correlation based on a conjugate of an element wise multiplication between the error factor and the conjugate of the input in the frequency domain; and
outputting the filtered input, wherein updating the tap weights includes applying
Ω k+1 =Ω k −μD (Ψ* k εE k )*
where Ω is a tap weight vector in the frequency domain, Ψ is the input in the frequency domain, E is an error factor in the frequency domain, D is a predefined matrix, ε is element wise multiplication, k is a number of blocks, and μ is a predetermined factor.
2. The method of claim 1 , further comprising:
initializing the tap weights of the equalizer using central spike.
3. The method of claim 1 , further comprising:
normalizing each frequency bin as a function of the spectral power associated with input data associated with each frequency bin.
4. The method of claim 1 , wherein calculating the error factor includes applying
e
k
(
n
)
=
z
k
,
R
(
n
)
(
1
-
z
k
,
R
(
n
)
R
2
,
R
)
+
j
z
k
,
I
(
n
)
(
1
-
z
k
,
I
(
n
)
R
2
,
I
)
where R is a constant, k is a number of blocks, n is a number of equalizer input, and z is the equalizer output.
5. The method of claim 1 , wherein calculating the error factor includes applying
e k ( n )=4 z ( n )(| z ( n )| 2 −R )
where R is a constant, k is a number of blocks, n is a number of equalizer input, and z is the equalizer output.
6. The method of claim 1 , wherein calculating the error factor includes applying
e k ( n )=2[ z R ( n )( z R 2 ( n )− R 1,R )+ jz I ( n )( z I 2 ( n )− R 1,I )]
where R is a constant, n is a number of equalizer input, k is a number of blocks, and z is the equalizer output.
7. An apparatus for equalization in a single carrier frequency division multiple access communication system, the apparatus comprising:
processing circuitry configured to
receive an input of a block of symbols, the block symbols being periodic transmitted symbols and being circularly convoluted by a channel,
filter the input in the frequency domain based on tap weights,
calculate an error factor in the time domain based on the filtered input,
update the tap weights in the frequency domain using circular correlation based on a conjugate of an element wise multiplication between the error factor and the conjugate of the input in the frequency domain, and
output the filtered input, wherein updating the tap weights includes applying
Ω k+1 =Ω k −μD (Ψ* k εE k )*
where Ω is a tap weight vector in the frequency domain, Ψ is the input in the frequency domain, E is an error factor in the frequency domain, D is a predefined matrix, ε is element wise multiplication, k is a number of blocks, and μ is a predetermined factor.
8. The apparatus of claim 7 , wherein the processing circuitry is configured to:
initialize the tap weights using central spike.
9. The apparatus of claim 7 , wherein the processing circuitry is configured to:
normalize each frequency bin as a function of the spectral power associated with input data associated with each frequency bin.
10. The apparatus of claim 7 , wherein calculating the error factor includes applying
e
k
(
n
)
=
z
k
,
R
(
n
)
(
1
-
z
k
,
R
(
n
)
R
2
,
R
)
+
j
z
k
,
I
(
n
)
(
1
-
z
k
,
I
(
n
)
R
2
,
I
)
where R is a constant, k is a number of blocks, n is a number of equalizer input, and z is the filtered input.
11. The apparatus of claim 7 , wherein calculating the error factor includes applying
e k ( n )=4 z ( n )(| z ( n )| 2 −R )
where R is a constant, k is a number of blocks, n is a number of equalizer input, and z is the filtered input.
12. The apparatus of claim 7 , wherein calculating the error factor includes applying
e k ( n )=2[ z R ( n )( z R 2 ( n )− R 1,R )+ jz I ( n )( z I 2 ( n )− R 1,I )]
where R is a constant, k is a number of blocks, n is a number of equalizer input, and z is the filtered input.
13. A non-transitory computer readable medium storing computer-readable instructions therein which when executed by a computer cause the computer to perform a method for equalization in a single carrier frequency division multiple access communication system, the method comprising:
receiving an input of a block of symbols, the block symbols being periodic transmitted symbols and being circularly convoluted by a channel;
filtering the input in the frequency domain;
calculating an error factor in the time domain based on the filtered input;
updating tap weights of a filter in the frequency domain using circular correlation based on a conjugate of an element wise multiplication between the error factor and the conjugate of the input in the frequency domain; and
outputting the filtered input, wherein updating the tap weights includes applying
Ω k+1 =Ω k −μD (Ψ* k εE k )*
where Ω is a tap weight vector in the frequency domain, Ψ is the input in the frequency domain, E is an error factor in the frequency domain, D is a predefined matrix, ε is element wise multiplication, k is a number of blocks, and μ is a predetermined factor.