Method for noise energy estimation in TDMA systems
View Patent ↗The present invention proposes a method for estimating a signal variance of a received signal (y) transmitted via a transmission channel, said method comprising the steps of: estimating the energy (RSSI) of the received signal (y), estimating the energy (Emax) of channel impulse response (h) taps, and estimating said signal variance (VAR) of said received signal by subtracting said estimated energy of channel impulse response taps (Emax) from said estimated energy (RSSI) of said received signal, such that said signal variance is obtained as VAR=RSSI−Emax. The present invention also proposes a correspondingly adapted device.
1. A method for estimating a signal variance (VAR) received signal transmitted via a transmission channel, said method comprising the steps of:
estimating an energy (RSSI) of the received signal,
estimating an energy of (Emax) of channel impulse response taps,
wherein
estimating said signal variance of said received signal by subtracting the estimated energy of channel impulse response taps from the estimated energy of said received signal,
such that said signal variance is obtained as:
VAR=RSSI−Emax,
wherein
said estimated energy of the received signal is represented by a sum of all vector components of the received signal according to a relation of
RSSI
=
(
∑
i
(
Re
{
y
i
}
2
+
Im
{
y
i
}
2
)
)
/
N
=
E
(
y
H
y
)
/
N
with
Re{y i } denoting a real component of the i-th vector component of the received signal y in vector representation,
Im{y i } denoting an imaginary component of the i-th vector component of the received signal y in vector representation,
y denotes the vector of the received signal, and
y H denotes a conjugate complex transposition of the vector of the received signal,
and wherein
said estimated energy of said channel impulse response taps is represented by a sum of all vector components of the channel impulse response according to a relation of
Emax
=
(
∑
i
(
Re
{
h
i
}
2
+
Im
{
h
i
}
2
)
)
/
N
=
E
(
h
H
X
H
hX
)
/
N
with
Re{h i } denoting a real component of the i-th vector component of the channel impulse response h in vector representation,
Im{h i } denoting an imaginary component of the i-th vector component of the channel impulse response h in vector representation,
h denotes the vector of the channel impulse response,
h H denotes a conjugate complex transposition of the vector of the channel impulse response, and
X represents a training sequence in vector representation, the training sequence being such that a product of the training sequence with its conjugate complex transposition X H makes “1” and
wherein N (N is a natural number) represents the length of a symbol sequence of the received signal y.
2. A method according to claim 1 , wherein
said step of estimating said signal variance of said received signal further comprises
a step of weighting said estimated energy of channel impulse response taps by a respective first and second predetermined factor, and wherein
the estimated signal variance is obtained by subtracting
said estimated energy of channel impulse response taps weighted with the first predetermined factor, a, from
said estimated energy of said received signal,
such that said signal variance is obtained as: VAR=RSSI−a*Emax, and said method further comprises a selecting step for selecting an actual estimated variance (PCT-VAR) as the maximum value of said estimated signal variance and said estimated energy of channel impulse response taps weighted with the second predetermined factor, b, such that the actual estimated variance is obtained as:
ACT 13 VAR=MAX ( VAR, b*Emax ).
3. A method according to claim 2 , wherein the first and second predetermined factors are user specific constants.
4. A device for estimating a signal variance (VAR) of a received signal transmitted via a transmission channel, said device comprising:
first estimation means adapted to estimate an energy (RSSI) of the received signal,
second estimation means adapted to estimate an energy (Emax) of channel impulse response taps, and
third estimation means adapted to estimate said signal variance of said received signal and comprising a subtracting unit adapted for subtracting
the estimated energy of channel impulse response taps from
the estimated energy of said received signal, such that said signal variance is obtained as:
VAR=RSSI−Emax
wherein
said estimated energy of the received signal is represented by a sum of all vector components of the received signal according to a relation of
RSSI
=
(
∑
i
(
Re
{
y
i
}
2
+
Im
{
y
i
}
2
)
)
/
N
=
E
(
y
H
y
)
/
N
with
Re{y i } denoting a real component of the i-th vector component of the received signal y in vector representation,
Im{y i } denoting an imaginary component of the i-th vector component of the received signal y in vector representation,
y denotes the vector of the received signal, and y H denotes a conjugate complex transposition of the vector of the received signal, and wherein
said estimated energy of said channel impulse response taps is represented by a sum of all vector components of the channel impulse response according to a relation of
Emax
=
(
∑
i
(
Re
{
h
i
}
2
+
Im
{
h
i
}
2
)
)
/
N
=
E
(
h
H
X
H
hX
)
/
N
with
Re{h i } denoting a real component of the i-th vector component of the channel impulse response h in vector representation,
Im{h i } denoting an imaginary component of the i-th vector component of the channel impulse response in vector representation,
h denotes the vector of the channel impulse response,
h H denotes a conjugate complex transposition of the vector of the channel impulse response, and
X represents a training sequence in vector representation, the training sequence being such that a product of the training sequence with its conjugate complex transposition X H makes “1,” and
wherein N (N is a natural number) represents the length of a sequence of the received signal y.
5. A device according to claim 4 , wherein
said third estimation means adapted to estimate said signal variance of said received signal further comprises
a weighting unit adapted to weight said estimated energy of channel impulse response taps by a respective first and second predetermined factor, and
said subtracting unit outputs the estimated signal variance as a result of subtracting
said estimated energy of channel impulse response taps weighted with the first predetermined factor, a, from
said estimated energy of said received signal, such that said signal variance is obtained as
VAR=RSSI−a*Emax,
and said device further comprises
a selecting means adapted to select an actual estimated variance (ACT_VAR) as the maximum value of said estimated signal variance and said estimated energy of b, channel impulse response taps weighted with the second predetermined factor, b, such that the actual estimated variance is obtained as
ACT — VAR=MAX ( VAR, b*Emax ).
6. A device according to claim 5 , wherein the first an second predetermined factors are user specific constants.