Crosstalk estimation and power setting based on interpolation in a multi-channel communication system
View Patent ↗An access node of a communication system comprises a plurality of transmitters adapted for communication with at least one receiver over a plurality of channels. The access node is operative to obtain estimated crosstalk coefficients between a joining channel and an active channel, and to set a power level of at least one signal transmitted over the joining channel based on the estimated crosstalk coefficients. The access node obtains the estimated crosstalk coefficients by first obtaining a subset of the estimated crosstalk coefficients and subsequently determining additional ones of the estimated crosstalk coefficients by applying an interpolation process to the estimated crosstalk coefficients in the subset. The access node sets the power level of the signal transmitted over the joining channel in a manner that ensures maintenance of a desired performance characteristic for the active channel.
1. A method for use in a communication system comprising a plurality of transmitters and at least one receiver, the method comprising the steps of:
obtaining estimated crosstalk coefficients between a joining channel and an active channel of the system; and
setting a power level of at least one signal transmitted over the joining channel based on the estimated crosstalk coefficients;
wherein the obtaining step comprises first obtaining a subset of the estimated crosstalk coefficients and subsequently determining additional ones of the estimated crosstalk coefficients by applying an interpolation process to the estimated crosstalk coefficients in the subset; and
wherein the setting step comprises setting the power level of the signal transmitted over the joining channel in a manner that ensures maintenance of a desired performance characteristic for the active channel.
2. The method of claim 1 wherein the system comprises a DSL system in which the active channel and the joining channel comprise respective active and joining lines each comprising M tones.
3. The method of claim 2 wherein the subset of estimated crosstalk coefficients comprises crosstalk coefficients for a subset {0, 2Ω, 4Ω, . . . , (L/2−1)2Ω} of the M tones, where L=M/Ω, the subset thereby providing a vector of estimated crosstalk coefficients given by ĥ 2Ω =(ĥ 0 , ĥ 2Ω , ĥ 4Ω , . . . ĥ (L/2−1)2Ω ).
4. The method of claim 2 wherein the subset of estimated crosstalk coefficients comprises crosstalk coefficients for a subset {0, Ω, Ω, . . . , (L−1)Ω} of the M tones, where L=M/Ω, the subset thereby providing a vector of estimated crosstalk coefficients given by ĥ Ω =(ĥ 0 , ĥ Ω , ĥ 2Ω , . . . ĥ (L−1)Ω ).
5. The method of claim 3 wherein the interpolation process comprises a linear-model interpolation process.
6. The method of claim 5 wherein linear-model interpolation process applied to the subset of the estimated crosstalk coefficients to obtain the additional ones of the estimated crosstalk coefficients is given by:
h int =A 2Ω ĥ 2Ω ,
where h int =(h 0 int , h 1 int , . . . , h M−1 int ) is a vector comprising the estimated crosstalk coefficients for respective ones of the M tones, ĥ 2Ω =(ĥ 0 , ĥ 2Ω , h 4Ω , . . . ĥ (L/2−1)2Ω ) is the vector of estimated crosstalk coefficients in the subset, and A 2Ω is an M×(L/2) matrix.
7. The method of claim 6 wherein the matrix A 2Ω has entries a m,j given by:
a
m
,
j
=
1
L
1
-
ω
Lm
/
2
1
-
ω
2
jΩ
-
m
if
2
j
Ω
-
m
≠
0
mod
M
,
and
a
m
,
j
=
1
if
2
j
Ω
-
m
=
0
mod
M
,
where ω denotes a complex root of unity.
8. The method of claim 2 wherein the step of setting a power level based on the estimated crosstalk coefficients comprises assigning power levels P 0 join , P 1 join , . . . , P M−1 join to respective ones of M tones of the joining line.
9. The method of claim 8 wherein the power levels are assigned to the tones of the joining line such that if a given one of the estimated crosstalk coefficients ĥ j obtained by application of the interpolation process is utilized for precoding of the active line then interference in the j-th tone of the active line from the joining line will exceed a designated level only with a negligibly small probability.
10. The method of claim 2 wherein the desired performance characteristic for the active line comprises SINRs for respective tones of the active line.
11. The method of claim 3 wherein the step of setting a power level based on the estimated crosstalk coefficients comprises the steps of:
determining a critical value ε j cr of systematic error for tones j ε {Ω, 3Ω, . . . , Ω+(L/2−1)2Ω};
for (2t−1)Ω<j<(2t+1)Ω, t≧1, assigning ε j cr =(ε (2t−1)Ω cr +ε (2t+1)Ω cr )/2;
computing h int =A Ω ĥ Ω , where ĥ Ω =(ĥ 0 , ĥ Ω , ĥ 2Ω , . . . ĥ (L−1)Ω );
for j ε {0, Ω, 2Ω, . . . , (L−1)Ω} assigning ε j cr =0;
for each 0≦j≦M−1 computing e j such that J({circumflex over (ε)} j cr , Σ j,j , R j )≧1−P int.err ; and
assigning
P
j
join
≥
(
P
j
act
S
j
cr
-
V
j
)
/
e
j
2
;
where
J
(
a
,
W
,
R
)
=
∫
C
(
R
)
p
a
,
W
(
x
)
ⅆ
x
;
where C(R) is a circle of radius R centered at the origin;
where P a,W (x) is a probability density function of a random variable x˜CN(a,W); and
where V j is the additive background noise in the j-th tone of the active line, P j act and P j join are the powers used in the j-th tones of the active and joining lines, respectively, and S j cr is a critical SINR value of in the j-th tone of the active line.
12. The method of claim 1 further comprising the step of adjusting precoder coefficients applied to signals sent by the transmitters to respective receivers over respective channels of the system based on the estimated crosstalk coefficients.
13. The method of claim 1 wherein the step of estimating crosstalk coefficients comprises processing at least one of error samples and SNR measurements fed back from one or more receivers of the system.
14. The method of claim 2 wherein particular ones of the M tones to be utilized for determining the estimated crosstalk coefficients for use in the interpolation process are determined at least in part based on knowledge of current channel conditions and error correcting codes of one or more active lines.
15. A non-transitory computer-readable storage medium having embodied therein executable program code that when executed by a processor of an access node of the system causes the access node to perform the steps of the method of claim 1 .
16. An apparatus comprising:
an access node comprising a plurality of transmitters adapted for communication with at least one receiver over a plurality of channels;
the access node being operative to obtain estimated crosstalk coefficients between a joining one of the channels and an active one of the channels, and to set a power level of at least one signal transmitted over the joining channel based on the estimated crosstalk coefficients;
wherein the access node obtains the estimated crosstalk coefficients by first obtaining a subset of the estimated crosstalk coefficients and subsequently determining additional ones of the estimated crosstalk coefficients by applying an interpolation process to the estimated crosstalk coefficients in the subset; and
wherein the access node sets the power level of the signal transmitted over the joining channel in a manner that ensures maintenance of a desired performance characteristic for the active channel.
17. The method of claim 16 wherein the subset of estimated crosstalk coefficients comprises crosstalk coefficients for a subset {0, 2Ω, 4Ω, . . . , (L/2−1)2Ω} of M tones of the joining channel, where L=M/Ω, the subset thereby providing a vector of estimated crosstalk coefficients given by ĥ 2Ω =(ĥ 0 , ĥ 2Ω , ĥ 4Ω , . . . ĥ (L/2−1)2Ω ).
18. The apparatus of claim 16 wherein the access node comprises a processor coupled to a memory with the processor being configured to control said estimation of crosstalk coefficients and set setting of the power level of the signal transmitted over the joining channel in accordance with executable program code stored in said memory.
19. The apparatus of claim 16 wherein the access node comprises at least a portion of at least one central office of a DSL system.
20. A communication system comprising:
a plurality of transmitters; and
a plurality of receivers;
wherein the transmitters are adapted for communication with respective ones of the receivers over respective channels;
the system being operative to obtain estimated crosstalk coefficients between a joining channel and an active channel, and to set a power level of at least one signal transmitted over the joining channel based on the estimated crosstalk coefficients;
wherein the crosstalk coefficients are obtained by first obtaining a subset of the estimated crosstalk coefficients and subsequently determining additional ones of the estimated crosstalk coefficients by applying an interpolation process to the estimated crosstalk coefficients in the subset; and
wherein the power level of the signal transmitted over the joining channel is set in a manner that ensures maintenance of a desired performance characteristic for the active channel.
21. The method of claim 20 wherein the subset of estimated crosstalk coefficients comprises crosstalk coefficients for a subset {0, 2Ω, 4Ω, . . . , (L/2−1)2Ω} of M tones of the joining channel, where L=M/Ω, the subset thereby providing a vector of estimated crosstalk coefficients given by ĥ 2Ω =(ĥ 0 , ĥ 2Ω , ĥ 4Ω , . . . ĥ (L/2−1)2Ω ).