Pulse or digital communications – Spread spectrum – Direct sequence
Reexamination Certificate
1999-08-17
2004-09-07
Deppe, Betsy L. (Department: 2634)
Pulse or digital communications
Spread spectrum
Direct sequence
C375S343000
Reexamination Certificate
active
06788736
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a matched filter for use in an inverse spread process of a received signal using a spreading signal such as a pseudo-noise (PN) signal in a spread spectrum receiver.
BACKGROUND OF THE INVENTION
Conventionally, as the method of an inverse spread process in a spread spectrum receiver, two methods have been used. One is a passive correlation method which uses a matched filter as inverse spreading means and the other is an active correlation method which uses a correlator as inverse spreading means.
First, the operation of the matched filter will be described.
FIG. 6
shows, as an example of the arrangement of the matched filter, a matched filter which uses a spreading signal as a coefficient in an N tap transversal type filter.
In
FIG. 6
, D(m) and R(m) indicate a received signal and a correlation signal at instant m, respectively. Also, in
FIG. 6
, N spreading signals P(n) (n=0, 1, . . . , N−2, N−1) indicate spreading signals of period N. The received signal D(m) is sampled with respect to time with the period equal to chip section length T
c
, where T
c
is a section length (chip section length) of the spreading signal P(n).
Note that, the value in the brackets of the spreading signal P(n) indicates time so that the larger the value, the further back from the present time. Namely, for example, comparing P(n) and P(n+1), P(n+1) indicates a signal which is further back from P(n) by the amount of time equal to the chip section length T
c
. On the other hand, the value in the brackets of the other signals such as the received signal D(m) and correlation signal R(m) indicates time so that the smaller the value, the further back from the present time. Namely, for example, comparing D(m) and D(m+1), D(m) indicates a signal which is further back from D(m+1) by the amount of time equal to the chip section length T
c
.
When the section length (symbol section length) of data to be subjected to a spread process on the sending side is T
s
, the spread ratio N has the relationship of N=T
s
/T
c
with the chip section length T
c
and symbol section length T
s
.
As shown in
FIG. 6
, in a common matched filter, the tap number is equal to the spread ratio N. In the following, for convenience of explaining the operation, the received signal is regarded as a signal within a baseband.
A delay circuit
61
has the structure in which N−1 delay elements
61
(
p
) (p=1, . . . , N−2, N−1) are connected in series. To the delay element
61
(
1
) is inputted the received signal D(m). Each delay element
61
(
p
) outputs signal D(m-p). The delay time of each delay element
61
(
p
) is equal to the chip section length T
c
.
In a multiplier
62
(
n
), the spreading signal P(n) is multiplied with each of the input signal D(m) and the output signal D(m-p) of each delay element
61
(
p
) (i.e., signal D(m-n)), and the result of this multiplication is outputted as a signal. Then, the output signals of all the multipliers
62
(
n
) are added in an adder
63
. As a result, a correlation signal R(m) with respect to symbol section length T
s
, which is the length of one period of the spreading signal P(n), is determined. This is represented by the following equation (1).
R
(
m
)
=
∑
n
=
0
N
-
1
D
(
m
-
n
)
·
P
(
n
)
(
1
)
A common spreading signal only takes the two values of +1 and −1, and for this reason in a common multiplier
62
(
n
), the polarity of the input of the adder
63
is reversed in accordance with the spreading signal P(n). As is clear from the structure of
FIG. 6
, in the matched filter, the spreading signal P(n) is fixed, and the cross-correlation function with the received signal D(m), which changes per chip section length T
c
, is calculated. The absolute value of the correlation signal R(m) becomes maximum at the time when the received signal D(m) and the spreading signal P(n) are in-phase. By the periodicity of the received signal D(m) and the spreading signal P(n), the time of in-phase arrives per symbol section length T
s
. Thus, it is ensured that the inverse spread using a matched filter is carried out with the period equal to the symbol section length T
s
. Accordingly, the operation for making a coincidence of the phases of the received signal D(m) and the spreading signal P(n) is not required. For this reason, the inverse spread method using the matched filter is called the passive correlation method.
The following describes the operation of the correlator.
FIG. 7
shows an example of the structure of the correlator. First, the products of the received signal D(m) and N spreading signals P(n) (n=0, 1, 2, . . . , N−1) are determined in the multiplier
64
, and the products are integrated in an integrator
65
. The received signal D(m) has been sampled with respect to time, and for this reason in this integration operation, cumulative addition of the products of the received signal D(m) and the spreading signals P(n) is carried out. The correlation signal R(m) of the correlator is represented by the following equation (2).
R
(
m
)
=
∑
n
=
0
N
-
1
D
(
m
-
n
)
·
P
(
(
n
+
i
)
mod
N
)
(
2
)
Here, i is an integer in the range of 0≦i≦N−1 indicating the phase of the spreading signal P(n) at the start of integration, and i is set by a controller
67
. “mod” is residual operator.
A spreading signal generator
66
outputs the spreading signal P(n) per chip section length T
c
. For example, when i=5, and N≧7, the spreading signal generator
66
outputs spreading signal P(
4
) at instant m−N+1, spreading signal P(
3
) at instant m−N+2, . . . , spreading signal P(
0
) at instant m−N+5, spreading signal P(N−1) at instant m−N+6, . . . , and spreading signal P(
5
) at instant m.
In this manner, in the above correlator, the integrator
65
performs integration of the product of the received signal D(m) and the spreading signal P(n) for one period (symbol section length T
s
) of the spreading signal from instant m−N+1 to instant m so as to determine the correlation signal R(m) at instant m as the cross-correlation function of the received signal D(m) and the spreading signals P(n).
Note that, the output of the integrator
65
at the instant other than instant m is the value in calculation of the cross-correlation function in progress, and the length of an integration section does not reach the symbol section length T
s
, and therefore is called a partial cross-correlation function. In the case where the spread ratio N is significantly large, the length of integration section may be made shorter than the symbol section length T
s
, and the partial cross-correlation function is used as the correlation signal R(m).
The integrator
65
sets the integrated value which had been accumulated before the start of integration to zero. This operation of making the integrated value to zero is called damping or reset.
In order to carry out the inverse spread process with this correlator, unlike the above matched filter, it is required beforehand to make a coincidence of the phases of the received signal D(m) and the spreading signal P(n). Thus, the phase i of the spreading signal P(n) at the start of integration is controlled by the controller
67
. The method using the correlator is called the active correlation method because the operation of actively matching the phases of the received signal D(m) and the spreading signal P(n) is required. Note that, instead of the integrator
65
of the correlator, a low-pass filter may be used.
The operation of actively matching the phases of the received signal and the spreading signal is the synchronism acquisition. In the matched filter, the synchronism acquisition can be carried out with the period equal to the symbol section length T
s
. On the other hand, in the correlator, the controller
67
determines the integration result (correlation fun
Iizuka Kunihiko
Kawama Shuichi
Birch & Stewart Kolasch & Birch, LLP
Deppe Betsy L.
Sharp Kabushiki Kaisha
LandOfFree
Matched filter does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Matched filter, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Matched filter will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3191709