Channel estimation technique
View Patent ↗A method includes determining a sequence of first coefficient estimates of a communication channel based on a sequence of pilots arranged according to a known pilot pattern and based on a receive signal, wherein the receive signal is based on the sequence of pilots transmitted over the communication channel. The method further includes determining a sequence of second coefficient estimates of the communication channel based on a decomposition of the first coefficient estimates in a dictionary matrix and a sparse vector of the second coefficient estimates, the dictionary matrix including filter characteristics of at least one known transceiver filter arranged in the communication channel.
1. A method, comprising:
determining a sequence of first coefficient estimates of a communication channel based on a sequence of pilots arranged according to a known pilot pattern and based on a receive signal, wherein the receive signal is based on the sequence of pilots transmitted over the communication channel; and
determining a sequence of second coefficient estimates of the communication channel based on a decomposition of the first coefficient estimates into a dictionary matrix and a sparse vector of the second coefficient estimates, wherein the dictionary matrix comprises filter characteristics of at least one known transceiver filter arranged in the communication channel,
wherein the second coefficient estimates are based on impulse response coefficients of the communication channel excluding the at least one known transceiver filter.
2. The method of claim 1 , wherein the sequence of second coefficient estimates is determined at positions of the pilots.
3. The method of claim 1 , wherein the pilots are encoded together with data symbols according to an orthogonal frequency division multiplexing technique.
4. The method of claim 1 , wherein the decomposition is based on a compressive sensing inference model, in particular a Sparse Bayesian Learning model.
5. The method of claim 1 , wherein the dictionary matrix is determined based on a sequence of time delays generated with a sequence of time delays determined arbitrarily or by other algorithmic means.
6. The method of claim 1 , wherein the dictionary matrix is determined based on a Fourier transform of impulse response coefficients of the at least one known transceiver filter.
7. The method of claim 1 , wherein the at least one known transceiver filter comprises a transmitter radio frequency shape filter and a receiver radio frequency shape filter.
8. The method of claim 7 , wherein the dictionary matrix is determined based on a Discrete Fourier Transform matrix of impulse response coefficients of the transmitter radio frequency shape filter convolved with impulse response coefficients of the receiver radio frequency shape filter.
9. A device, comprising:
a first determination unit configured for determining a sequence of first coefficient estimates of a communication channel based on a sequence of pilots arranged according to a known pilot pattern and based on a receive signal, wherein the receive signal is based on the sequence of pilots transmitted over the communication channel; and
a second determination unit configured for determining a sequence of second coefficient estimates of the communication channel based on a decomposition of the first coefficient estimates in a dictionary matrix and a sparse vector of the second coefficient estimates, wherein the dictionary matrix comprises filter characteristics of at least one known transceiver filter arranged in the communication channel,
wherein the second coefficient estimates are based on impulse response coefficients of the communication channel excluding the at least one known transceiver filter.
10. The device of claim 9 , wherein the first coefficient estimates are based on impulse response coefficients of the communication channel including the at least one known transceiver filter.
11. The device of claim 9 , wherein the decomposition is based on the following compressive sensing inference model:
t=Hα+w
where t denotes the sequence of first coefficient estimates, H denotes the dictionary matrix, α denotes the sparse vector of the second coefficient estimates and w denotes a vector of noise samples.
12. The device of claim 9 wherein the dictionary matrix is determined on the basis of the following relation:
H
j
,
k
=
N
∑
n
=
0
N
-
1
F
p
j
,
n
ϕ
(
nT
S
-
τ
k
(
s
)
)
,
where H j,k denotes a coefficient of the dictionary matrix with respect to pilot subcarrier of index j and time delay of index k, F p j ,n denotes a Fourier matrix of the pilots p j and φ(nT S −τ k (s) )denotes the filter characteristics of the at least one known transceiver filter with respect to sampling time T S and time delay τ k (s) .
13. A device, comprising:
a first determination unit configured for determining a sequence of first coefficient estimates of a communication channel based on a sequence of pilots arranged according to a known pilot pattern and based on a receive signal, wherein the receive signal is based on the sequence of pilots transmitted over the communication channel; and
a second determination unit configured for determining a sequence of second coefficient estimates of the communication channel based on a decomposition of the first coefficient estimates in a dictionary matrix and a sparse vector of the second coefficient estimates, wherein the dictionary matrix comprises filter characteristics of at least one known transceiver filter arranged in the communication channel, wherein the decomposition is based on the following compressive sensing inference model:
t=Hα+w,
where t denotes the sequence of first coefficient estimates, H denotes the dictionary matrix, α denotes the sparse vector of the second coefficient estimates and w denotes a vector of noise samples.
14. The device of claim 13 , wherein the first coefficient estimates are based on impulse response coefficients of the communication channel including the at least one known transceiver filter.
15. The device of claim 13 , wherein the second coefficient estimates are based on impulse response coefficients of the communication channel excluding the at least one known transceiver filter.
16. The device of claim 13 , wherein the dictionary matrix is determined on the basis of the following relation:
H
j
,
k
=
N
∑
n
=
0
N
-
1
F
p
j
,
n
ϕ
(
nT
S
-
τ
k
(
s
)
)
,
where H j,k denotes a coefficient of the dictionary matrix with respect to pilot subcarrier of index j and time delay of index k, F p j ,n denotes a Fourier matrix of the pilots p j and φ(nT S −τ k (s) ) denotes the filter characteristics of the at least one known transceiver filter with respect to sampling time T S and time delay τ k (s) .