Method for signal processing in user equipment of CDMA mobile communication system
View Patent ↗The invention discloses to a signal processing method for user equipment in a CDMA mobile communication system. The signal processing method is a combination of time diversity and joint detection in order to raise diversity gain of a receiver. The method comprises: sample a received signal on time domain, estimate spatial channel impulse response value, obtain matrixes, constitute a system matrix, perform joint detection algorithm and obtain an estimated value of original sending data. When the method is used in a mobile user equipment, the uplink and downlink performance unmatched problem can be solved, using same frequency multiplexing can be reached and the joint detection algorithm is a simpler one. The method also can be used in a base station.
1. A signal processing method in a code division multiple access mobile communication system, comprising:
sampling a signal at a plurality of instances in a time domain to thereby obtain a plurality of sampled data; and
performing a joint detection process based upon a plurality of the sampled data to thereby obtain an estimated value of the signal, wherein performing a joint detection process for said plurality of sampled data comprises:
estimating spatial channel impulse response for each sampled data e i of the plurality of sampled data e i , i=1 . . . M,
getting M spatial channel impulse response estimated values h i , wherein M references the number of sample times;
getting M matrixes A i from the obtained each spatial channel impulse response estimated value h i ; and
estimating the original sending data estimated value according to the sampled data e i and the matrixes A i .
2. The method according to claim 1 , wherein the spatial channel impulse response estimated value h i is obtained by first using formula
e
m
i
d
=
∑
i
m
i
⊗
h
i
+
n
m
,
and then formula h i =G −1 e mid ; wherein {circle around (×)} represents convolution, wherein e mid is the training sequence midamble received by user equipment, G is a coefficient matrix generated by the training sequence midamble of user equipment, m i is a training sequence midamble sent by the i th user equipment, and n m represents noise interference at training sequence midamble.
3. The method according to claim 1 , wherein the original sending data estimated value d sum is estimated by using formula:
d
s
u
m
=
∑
i
=
1
M
d
i
=
∑
i
=
1
M
(
A
i
*
T
A
i
)
-
1
A
i
*
T
e
i
wherein *T represents conjunction transpose and diag is a diagonal matrix symbol.
4. The method according to claim 1 , wherein the original sending data estimated value d est is estimated by using formulas:
d
e
s
t
=
(
∑
i
=
1
M
A
i
*
T
A
i
)
-
1
A
s
u
m
*
T
e
s
u
m
,
wherein A sum =(A 1 T , A 2 T . . . A M T ) T , e sum =(e 1 T , e 2 T . . . e M T ) T and *T represents conjunction transpose.
5. The method according to claim 1 , wherein when M is larger, the original sending data estimated value d est is estimated by using formula:
d
e
s
t
=
(
diag
(
∑
i
=
1
M
i
*
T
A
i
)
)
-
1
A
s
u
m
*
T
e
s
u
m
,
wherein A sum =(A 1 T , A 2 T . . . A M T ) T , e sum =(e 1 T , e 2 T . . . e M T ) T and *T represents conjunction transpose.