U.S. patent application number 12/476473 was filed with the patent office on 2009-10-15 for spread spectrum rake receiver.
This patent application is currently assigned to FUJITSU LIMITED. Invention is credited to Tsuyoshi Hasegawa, Masahiko Shimizu.
Application Number | 20090257478 12/476473 |
Document ID | / |
Family ID | 34632418 |
Filed Date | 2009-10-15 |
United States Patent
Application |
20090257478 |
Kind Code |
A1 |
Hasegawa; Tsuyoshi ; et
al. |
October 15, 2009 |
SPREAD SPECTRUM RAKE RECEIVER
Abstract
Even when the number of paths is increased, interfering noises
can be effectively reduced by a rake receiver for use in a spread
spectrum communication system. The rake receiver includes a timing
detecting unit detecting a reception timing t.sub.i (i=1 to N) of
each of N paths when direct spread spectrum signals of the N paths
are received; an inverse spreading timing setting unit setting, as
a timing for inverse spreading, a timing t.sub.i,j,k (k=1 to N,
k.noteq.j) at which an inverse spread value is obtained that has
interference and correlation from the jth (j=1 to N, j.noteq.i)
path included in the inverse spread value of the ith path counted
from the reception timing t.sub.i (i=1 to N) detected by the timing
detecting unit; a plurality of correlators each obtaining an
inverse spread signal of the received signal corresponding to each
timing set by the inverse spreading timing setting unit; and a
signal composing unit composing outputs of the plurality of
correlators.
Inventors: |
Hasegawa; Tsuyoshi;
(Kawasaki-shi, JP) ; Shimizu; Masahiko;
(Kawasaki-shi, JP) |
Correspondence
Address: |
KATTEN MUCHIN ROSENMAN LLP
575 MADISON AVENUE
NEW YORK
NY
10022-2585
US
|
Assignee: |
FUJITSU LIMITED
Kawasaki-shi
JP
|
Family ID: |
34632418 |
Appl. No.: |
12/476473 |
Filed: |
June 2, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11039009 |
Jan 13, 2005 |
7573934 |
|
|
12476473 |
|
|
|
|
PCT/JP02/09414 |
Sep 13, 2002 |
|
|
|
11039009 |
|
|
|
|
Current U.S.
Class: |
375/148 ;
375/E1.032 |
Current CPC
Class: |
H04B 1/7115 20130101;
H04B 1/709 20130101; H04B 1/7107 20130101 |
Class at
Publication: |
375/148 ;
375/E01.032 |
International
Class: |
H04B 1/707 20060101
H04B001/707 |
Claims
1-11. (canceled)
12. A receiver which receives signals through multi-paths,
comprising: a multi-path interference reducing unit to reduce an
interference from another path to an inverse spreading result, at a
receiving timing for a corresponding path; a rake synthesizing unit
to synthesize inverse spreading results at receiving timings for
respective paths among the multi-paths; and a selector to select a
path for which the interference is reduced by the multi-path
interference reducing unit, based on a channel estimated value for
the path, wherein an inverse spreading result at a receiving timing
corresponding to the path selected by the selector is applied to
synthesizing unit after the interference is reduced by the
multi-path interference reducing unit.
13. A receiver which receives signals through multi-paths,
comprising: an inverse spreading unit to obtain a first inverse
spreading result at a first receiving timing for a first path among
multi-paths; and a multi-path interference reducing unit to
synthesize the first inverse spreading result with a second inverse
spreading result, which is obtained at a second receiving timing
preceded by a difference between receiving timings for the first
path and a second path which is later than the first path, from a
receiving timing for any path among the multi-paths except the
second path, wherein the second inverse spreading result
synthesized with the first inverse spreading result is limited to
an inverse spreading result obtained at a timing different from
receiving timings for any paths among the multi-paths.
14. A receiver which receives signals through multi-paths,
comprising: a first multi-path interference reducing unit to
perform multi-path reducing process for an inverse spreading result
obtained by inverse spreading at a timing for a first path among
the multi-paths; a second multi-path interference reducing unit to
perform multi-path reducing process for an inverse spreading result
obtained by inverse spreading at a timing for a second path among
the multi-paths; a level adjusting unit to adjust levels of output
signals from the first multi-path interference reducing unit and
the second multi-path interference reducing unit, respectively; and
a rake synthesizing unit to synthesize the level adjusted output
signals from the first multi-path interference reducing unit and
the second multi-path interference reducing unit.
15. A method for use in a receiver which receives signals through
multi-paths, comprising: reducing an interference from another path
to an inverse spreading result, at a receiving timing for a
corresponding path; synthesizing inverse spreading results at
receiving timings for respective paths among the multi-paths; and
selecting a path for which the interference is reduced, based on a
channel estimated value for the path, wherein an inverse spreading
result at a receiving timing corresponding to the selected path is
applied to the synthesizing after the interference is reduced.
16. A method for use in a receiver which receives signals through
multi-paths, comprising: obtaining a first inverse spreading result
at a first receiving timing for a first path among multi-paths; and
synthesizing the first inverse spreading result with a second
inverse spreading result, which is obtained at a second receiving
timing preceded by a difference between receiving timings for the
first path and a second path which is later than the first path,
from a receiving timing for any path among the multi-paths except
the second path, wherein the second inverse spreading result
synthesized with the first inverse spreading result is limited to
an inverse spreading result obtained at a timing different from
receiving timings for any paths among the multi-paths.
17. A method for use in a receiver which receives signals through
multi-paths, comprising: performing a first multi-path interference
reducing process for an inverse spreading result obtained by
inverse spreading at a timing for a first path among the
multi-paths; performing a second multi-path interference reducing
process for an inverse spreading result obtained by inverse
spreading at a timing for a second path among the multi-paths;
adjusting levels of output signals from the first multi-path
interference reducing proces and the second multi-path interference
reducing process, respectively; and synthesizing the level adjusted
output signals from the first multi-path interference reducing
process and the second multi-path interference reducing process.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates generally to a spread spectrum
scheme, and more particularly, to a rake receiver that executes, as
a reception diversity scheme in a multi-path environment,
maximal-ratio composition in the time domain of signals arriving at
an antenna with various differences in delay time thereof caused by
multiple reflections of propagation paths of the signals.
[0003] 2. Description of the Related Art
[0004] Spread spectrum or spread spectrum communication scheme is
utilized extensively as a basic technique for mobile communication.
In the direct spread (DS) scheme as the simplest model of the
spread spectrum communication, an information signal is transmitted
to the receiving side after the spectrum of the information signal
is spread by modulating, that is, multiplying the information
signal to be transmitted by a PN signal having the chip width Tc of
1/100 to 1/1000 of the cycle T of the information signal to be
transmitted as a spread signal.
[0005] On the receiving side, the signal component is detected from
the signal buried in noises by inverse spreading. The inverse
spreading basically refers to executing demodulation by multiplying
a received signal by a same PN signal having the same phase as that
of the PN signal in the received signal.
[0006] However, in a multi-path environment for many reflected
waves to be present in addition to a direct wave, it is necessary
to detect a true signal component by composing appropriately
signals received with various differences in delay time.
[0007] As one of such conventional techniques, a rake scheme can be
listed. "Rake" means a rake in English and the rake scheme is a
diversity scheme for executing the maximal-ratio composition by
collecting power dispersed due to the delay dispersion of
transmission paths, into one like a "rake".
[0008] In a conventional rake receiver, a desired signal is
demodulated by finding a plurality of path timings at which
multiple paths arrive using a known signal, informing a demodulator
of these path timings, executing inverse spreading at these timings
in the demodulator and composing signals of the multiple paths.
[0009] FIG. 1 is a block diagram showing generally an example of
the construction of a rake receiver as, for example, a mobile
communication terminal. In the figure, the receiver has an antenna
100, a wireless receiving unit 101, an A/D converting unit 102, a
searcher 103 for detecting a plurality of timings of the multiple
paths and an inverse spreading timing generating and inverse
spreading unit 104 for executing an inverse spreading to the
plurality of paths according to the timings of the plurality of
paths detected by the searcher 103.
[0010] The receiver further has a signal composing unit 105 for
composing signals of the plurality of paths obtained by the inverse
spreading timing generating and inverse spreading unit 104, a
signal processing unit 106 such as a channel codec for receiving an
output of the signal composing unit 105 and outputting received
signals to a display, speaker, etc., and a level measuring unit 107
for measuring the level of the received signals of the plurality of
paths, providing reliability degree information and signal level
information to the signal composing unit 105 and providing to a
transmission unit 108 control information of transmission power to
a base station.
[0011] The transmission unit 108 transmits input from a keyboard or
a microphone, from the antenna 100 through a duplexer 109 in
response to the control information from the level measuring unit
107.
[0012] FIG. 2 is a block diagram of the detailed construction of
the inverse spreading timing generating and inverse spreading unit
104 of FIG. 1, that is, a signal demodulating unit. In the figure,
the signal demodulating unit comprises a spread code generator 110,
a plurality of delay control units 111-1 to 111-n and a plurality
of correlators 112-1 to 112-n corresponding thereto.
[0013] The spread code generator 110 generates a code for inverse
spreading. The plurality of delay control units 111-1 to 111-n
control respectively delay operations of the plurality of
correlators 112-1 to 112-n corresponding respectively to timings t1
to 1N of the multiple paths detected by the searcher 103. Each of
the correlators 112-1 to 112-n executes inverse spreading on the
received signals from the A/D converting unit 102 according to the
inverse spread timings controlled by the corresponding delay
control units 111-1 to 111-n.
[0014] Thereby, the correlators 112-1 to 112-n respectively provide
inverse spread signals 1 to N to the signal composing unit 105 and
the signal composing unit 105 composes these signals and outputs a
demodulated signal.
[0015] Such an inverse spread signal includes a channel estimation
signal corresponding to a propagation coefficient of each of the
multiple paths.
[0016] As described above, for example, in FIG. 2, inverse
spreading is executed using the timings themselves of each path of
the multiple paths. When inverse spreading is executed at a timing,
signals corresponding to paths other than the path of this timing
are all interference. Especially, in the case where an orthogonal
spread code is used for a plurality of channels in a downlink from
a base station in the CDMA scheme, a problem exists that the
reception property is degraded due to the multi-path
interference.
[0017] Considering the above point, the inventor has previously
proposed a rake receiver capable of suppressing multi-path
interference when the spread spectrum scheme is used in a
multi-path environment in Japanese Patent No. 2001-332510.
[0018] Here, the schematic construction of such a rake receiver as
proposed previously will be described. FIG. 3 is a block diagram of
the principle construction of a rake receiver constituting a spread
spectrum communication system in a multi-path environment,
previously proposed.
[0019] In FIG. 3, path timing detecting 1 correspond to, for
example, the path searcher 103 of FIG. 1 and FIG. 2 detect timings
of, for example, N paths.
[0020] Inverse spreading timing setting 2 set the detected timings
of the paths as timings for inverse spreading, that is, timings for
demodulating spread encoding signal by multiplying an inverse
spread code. Concurrently, settings are made to all combinations of
two (2) paths such that, taking the center at a timing of one (1)
path of arbitrary two (2) paths, two (2) timings at positions
symmetrical to the timing of the other path on the time axis by the
delayed time of the timings of the two (2) paths are timings of the
inverse spreading.
[0021] A plurality of correlators 3-1 to 3-n respectively obtains
an inverse spreading signal of a signal resulted from, for example,
A/D conversion of a signal sent from the transmitting side in
response to each timing having been set. Signal composing 4 compose
outputs of the plurality of correlators 3-1 to 3-n and output a
demodulated signal.
[0022] As described above, in the present invention proposed in the
previous application, interference component contained in a desired
signal is reduced using a multi-path interference correlative
signal (MICS) reproduced using only information of selected two (2)
paths.
[0023] However, as described above, in the present invention of the
previous application, a drawback is recognized that the effect of
the reduction of the interference component becomes smaller as the
number of the paths increases because the information of only
selected two (2) paths is utilized when the interference component
is reproduced. That is, information that must be contained in paths
other than the noted two (2) paths can not be utilized.
SUMMARY OF THE PRESENT INVENTION
[0024] It is therefore the object of the present invention to
provide a spread spectrum rake receiver capable of overcoming such
disadvantage of the present invention of the previous
application.
[0025] In order to achieve the above object, according to a first
aspect of the present invention there is provided a rake receiver
for use in a spread spectrum communication system, comprising
timing detecting operable to detect a reception timing t.sub.i (i=1
to N) of each of N paths when direct spread spectrum signals of the
N paths are received; inverse spreading timing setting operable to
set, as a timing for inverse spreading, a timing t.sub.i,j,k (k=1
to N, k.noteq.j) at which an inverse spread value is obtained that
has interference and correlation from the jth (j=1 to N, j.noteq.i)
path included in the inverse spread value of the ith path counted
from the reception timing t.sub.i (i=1 to N) detected by the timing
detecting; a plurality of correlators each operable to obtain an
inverse spread signal of the received signal corresponding to each
timing set by the inverse spreading timing setting; and signal
composing operable to compose outputs of the plurality of
correlators.
[0026] The rake receiver may further comprise, between the
correlators and the signal composing, a circuit operable to compose
a multi-path interference signal (mics (i,j,k)) of a path k
(k.noteq.j) from the following Eq. (1),
MICS ( i , j ) = r i , j k = j r k ' mics ( i , j , k ) , Eq . ( 1
) ##EQU00001##
to reproduce interference .alpha..sub.jI.sub.i,j and subtract the
interference MICS(i,j) from the ith path.
[0027] When the multi-path interference signal (mics (i,j,k)) is
composed, coefficients r.sub.i,j and r'.sub.k of Eq. (1) may be
obtained from the following Eq. (2) and Eq. (3),
r.sub.i,j=.alpha..sub.jI.sup.2/{(I/N).sub.j+1} Eq. (2),
r'.sub.k=.alpha.*.sub.k/{.SIGMA.
.sub.I.noteq.k|.alpha..sub.I|.sup.2I.sup.2+n.sup.2} Eq. (3),
and the multi-path interference signal (mics(i,j,k)) may be
composed using the maximal ratio composition.
[0028] When the multi-path interference signal (mics(i,j,k)) is
composed, noises of the multi-path interference signal
(mics(i,j,k)) may be approximated to be constant and the
coefficients r.sub.i,j and r'.sub.k of Eq. (1) may be obtained from
the following Eq. (4) and Eq. (5),
r.sub.i,j=.alpha..sub.j(I/N).sub.j/{(.SIGMA.k.noteq.j|.alpha..sub.k|.sup-
.2) ((I/N).sub.j+1)} Eq. (4),
r'.sub.k=.alpha.*.sub.k Eq. (5),
and the multi-path interference signal (mics (i, j, k)) may be
composed using the maximal ratio composition.
[0029] The circuit operable to subtract the interference MICS(i,j)
may include a circuit operable to select a plurality of paths
having high power, and the circuit operable to subtract the
interference MICS(i,j) may subject the selected paths to processes
for composing the multi-path interference signal (mics(i,j,k)) and
subtracting the interference MICS(i,j).
[0030] The circuit operable to subtract MICS(i,j) from the ith path
may select a plurality of paths i having high power and may be
provided in the quantity corresponding to the number of the
selected paths.
[0031] The inverse spreading timing setting may detect coincidence
between the timing t.sub.i,j,k for the inverse spreading and a
reception timing t.sub.i and, may not subject the paths between
which the coincidence has been detected, to the processes for
composing the multi-path interference signal (mics(i,j,k)) and
subtracting the interference MICS(i,j).
[0032] The rake receiver may further comprise a level compensating
circuit disposed between the circuit operable to subtract the
interference MICS(i,j) and the signal composing , the level
compensating circuit acting to compensate the levels of signals
after reduction of the interference in the circuit operable to
reduce the interference MICS(i,j) to keep the level of the noises
constant.
[0033] The rake receiver may further comprise, at the preceding
stage of the correlators, a circuit operable to compose a
multi-path interference signal (mics(i,j,k) of a path k (k.noteq.j)
from Eq. (1)
MICS ( i , j ) = r i , j k = j r k ' mics ( i , j , k ) , Eq . ( 1
) ##EQU00002##
to reproduce interference .alpha..sub.jI.sub.i,j, and subtract the
interference MICS(i,j) from the ith path.
[0034] The circuit operable to subtract the interference MICS(i,j)
may include a circuit operable to select a plurality of paths
having high power, thereby subjecting the selected paths to the
processes for composing the multi-path interference signal
(mics(i,j,k)) and subtracting the interference MICS(i,j).
[0035] The circuit operable to subtract the interference MICS(i,j)
from the ith path may select a plurality of paths i having high
power and may be provided in the quantity corresponding to the
number of the selected paths.
[0036] The above and other features of the present invention will
become more apparent from the description of the embodiments of the
invention when taken in conjunction with the accompanying drawings
which follow.
BRIEF DESCRIPTION OF THE DRAWINGS
[0037] FIGS. 1 is a block diagram showing generally an example of
the construction of a rake receiver as a mobile communication
terminal;
[0038] FIG. 2 is a block diagram of the detailed construction of
the inverse spreading timing creating and inverse spreading unit
104 of FIG. 1, that is, a signal demodulating unit;
[0039] FIG. 3 is a block diagram of the principle construction of a
rake receiver constituting a spread spectrum communication system
in a multi-path environment, previously proposed;
[0040] FIG. 4 shows timings of signals of two (2) paths received by
a CDMA (Code Division Multiple Access) mobile terminal;
[0041] FIG. 5 shows N path signals, and a plurality of Multi-path
Interference Correlative Timings (MICTs) that can be utilized for
reduction of interference of a path j contained in a path i;
[0042] FIG. 6 shows an example of the construction of a CDMA
receiver applied with the present invention;
[0043] FIG. 7 shows an example of the construction of a composing
unit 27 of FIG. 6;
[0044] FIG. 8 shows another exemplary embodiment of the present
invention;
[0045] FIG. 9 shows the detailed construction of a multi-path
interference exchange reduction circuit 28;
[0046] FIG. 10 shows the details of MICS units 280-1 to 280-N
represented by a MICS unit 280-i;
[0047] FIG. 11 shows an example of the construction of an MRC
128;
[0048] FIG. 12 shows another example of the construction of the
MICS unit 280-i;
[0049] FIG. 13 shows a construction that is provided with a
selector circuit 31 at the preceding stage of a MIXR circuit 28, as
an exemplary embodiment;
[0050] FIG. 14 shows yet another example of the construction of the
MICS unit 280-i:
[0051] FIG. 15 shows a construction that is provided with a level
correcting unit 32 between the MIXR unit 28 and a rake composing
unit 27 that is the following stage of the level correcting unit
32;
[0052] FIG. 16 shows yet another exemplary embodiment of the
present invention; and
[0053] FIG. 17 shows the construction of the MICS unit 280-i
constituting the MIXR unit 28 in the construction of FIG. 16.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0054] Here, prior to the description of an exemplary embodiments
of the present invention, the principle of the previously applied
invention by the present inventor described above will be further
described for the full understanding of the present invention.
[0055] When signals of multiple paths are inversely spread at a
timing, signals of paths that occur interference are determined by
mutual correlation value of an inverse spreading signal, the
attenuation coefficient of a propagation path, etc. The correlation
value of an inverse spreading signal is a constant determined by
the delay between the timing of a signal arrived through a path and
the timing of inverse spreading.
[0056] FIG. 4 shows timings of signals of two (2) paths received by
a CDMA (Code Division Multiple Access) mobile terminal. In the
figure, .cndot..cndot.YZABCD.cndot..cndot. are labels indicating
signal timings of each path and "A" is assumed to be the correct
inverse spreading timing. The channels of a path 1 and a path 2 are
denoted respectively by .alpha..sub.1 and .alpha..sub.2. Inverse
spreading timings are respectively denoted by t.sub.1 and t.sub.2
and signals inversely spread at these timings are respectively
denoted by x.sub.1 and x.sub.2.
[0057] Here, defining a special timing
t.sub.0=t.sub.1-(t.sub.2-t.sub.1) and denoting a signal inversely
spread at the timing t.sub.0 by x.sub.0, x.sub.1 and x.sub.0 can be
represented as follows.
x.sub.1=.alpha..sub.1S+.alpha..sub.2I.sub.z+n.sub.1
x.sub.0=.alpha..sub.1I.sub.z+.alpha..sub.2I.sup.y+n.sub.0
[0058] .alpha..sub.1S is a desired signal obtained by inversely
spreading the path 1 from A and .alpha..sub.2I.sub.Z is
interference obtained by inversely spreading the path 2 from Z.
Furthermore, .alpha..sub.2I.sub.Y is a signal obtained by inversely
spreading the path 2 from Y, and n.sub.1 and n.sub.0 are
respectively noises thereof.
[0059] The above signal x.sub.0 is a signal obtained by inversely
spreading at a timing that can not be obtained, of a received
signal S, and .alpha..sub.1I.sub.Z is contained therein. That is,
it can be seen that .alpha..sub.1I.sub.Z has correlation with an
interference component .alpha..sub.2I.sub.Z of x.sub.1. In this
sense, a signal like x.sub.0 is referred to a "Multi-path
Interference Correlative Signal (MICS)" of the path 1 to the path
2, and a timing like t.sub.0 is referred to as a "Multi-path
Interference Correlative Timing (MICT)" of the path 1 to the path
2.
[0060] The interference component of x.sub.1 can be reduced by
multiplying an appropriate coefficient r from x.sub.1 to x.sub.0
because x.sub.0 has correlation with the interference component of
x.sub.1.
[0061] However, here, it should be noted that another interference
component I.sub.Y contained in x.sub.0 is increased when the
coefficient r is determined such that I.sub.Z contained in x.sub.1
is completely cancelled. Therefore, the total magnitude of the
interference may be increased instead of being decreased.
Therefore, the appropriate coefficient r needs to be a coefficient
that is determined such that the total power of the interference
becomes minimum remaining the original interference I.sub.Z.
[0062] In the previously applied invention described above, the
information of only two (2) paths selected when interference
components are reproduced is used. Therefore, information contained
in paths other than the noted two (2) paths can not be utilized.
Therefore, the reduction effect of the interference components is
small.
[0063] Therefore, the present invention solves such a drawback of
the previously applied invention and improves the interference
reproduction accuracy by using all the paths except interference
sources in order to reproduce interference components. The
principle of the present invention will be described as
follows.
[0064] FIG. 5 shows N path signals, and a plurality of Multi-path
Interference Correlative Timings (MICTs) that can be utilized for
reduction of interference of a path j contained in a path i. A
signal inversely spread at a timing of the path i is represented as
follows.
x i = .alpha. i S + j = i .alpha. j I i , j ##EQU00003##
[0065] Here, S is a desired signal, I.sub.i,j is an interference
component by the path j contained in x.sub.i. In FIG. 5, a timing
t.sub.i,j,i is a multi-path interference correlative timing (MICT)
used in the previous application. The timing t.sub.i,j,i is a
timing shifted from the timing t.sub.i by the time difference
.DELTA. t between the path i and the path j. A signal having a
correlation with I.sub.i,j can be obtained by inversely spreading
at this timing t.sub.i,j,i.
[0066] Here, noting paths other than the paths i and j, similarly
to the timing t.sub.i,j,i, a signal having a correlation with
I.sub.i,j can be obtained by inversely spreading at a timing
t.sub.i,j,k (k is a value from 1 to N except j. ) shifted from each
of the paths by .DELTA. t. The signals obtained by inversely
spreading at the timings t.sub.i,j,k and t.sub.i,j,k are
represented as follows.
t i , j , k = t i - t j + t k ##EQU00004## m i , j , k = I .alpha.
I I i , j , k + n i , j , k = .alpha. k I i , j + I = k .alpha. I I
i , j , k , I + n i , j , k ##EQU00004.2##
where I.sub.i,j,k,I=I (t.sub.i-t.sub.j+t.sub.k-t.sub.I) and,
especially I.sub.i,j,k,k=I.sub.i,j.
[0067] Next, an embodiment of the present invention will be
described based on the above principle of the present
invention.
[0068] FIG. 6 shows an example of the construction of a CDMA
receiver applied with the present invention. A CDMA signal received
by an antenna 20 is converted into a base-band signal by a
down-converter 21.
[0069] The base-band signal is inputted into an A/D converter 23
through an AGC amplifier 22. Here, the base-band signal is
converted into a digital signal and is inputted into inverse
spreading circuit units 24-1 to 24-n corresponding to the number of
paths n and into a path searching unit 25.
[0070] In the path searching unit 25, a timing of each of the paths
of the multi-paths is obtained from the received signal. Based on
these path timings, inverse spreading timings t.sub.i,j,k,n are
created by a timing generating circuit 26 according to the
following equation.
t.sub.i,j,k,n=t.sub.i-(t.sub.j-t.sub.k) n
[0071] The inverse spreading timings t.sub.i,j,k,n are sent
respectively to the corresponding inverse spreading circuit units
24-1 to 24-n and inverse spreading processes are executed at the
respective timings. Inverse spread outputs obtained in the inverse
spreading circuit units 24-1 to 24-n are composed in the composing
unit 27 and an inverse spread signal is obtained.
[0072] FIG. 7 shows an example of the construction of a composing
unit 27 of FIG. 6. Here, the construction of an MMSE receiver is
shown as an embodiment. Therefore, the composing unit 27 has an
MMSE coefficient generating unit 270.
[0073] The MMSE coefficient generating unit 270 obtains composition
coefficients that maximizes the S/N of received signals. These
coefficients are multiplied as the coefficients to multipliers
271-1 to 271-n respectively corresponding to each of fingers. The
MMSE coefficient generating unit 270 further comprises an adder 272
that adds the outputs of these multipliers 271-1 to 271-n. Thereby,
an output that maximizes the S/N of the received signals can be
obtained from the adder 272.
[0074] With the construction of FIG. 6 applied with the present
invention, effective timings for inverse spreading can be easily
obtained and, therefore, a preferable effect can be obtained with a
few inverse spread fingers.
[0075] FIG. 8 shows another exemplary embodiment of the present
invention. A multi-path interference exchange reduction (MIXR)
circuit 28 is provided between the inverse spreading circuit units
24-1 to 24-n and a rake composing unit 27 compared to the exemplary
embodiment of the construction of FIG. 6. Interference of each
finger is reduced by this multi-path interference exchange
reduction (MIXR) circuit 28.
[0076] FIG. 9 shows the detailed construction of the multi-path
interference exchange reduction circuit 28. Interference is reduced
by reproducing the interference caused by the path j contained in
the path i in the MICS units 280-1 to 280-N, adding all outputs of
MICS units 280-1 to 280-N in the adder 281 and subtracting the
result of the adding from a signal of the path i.
[0077] Each of the above MICS units 280-1 to 280-N reproduces
interference entering from the path j to the path i
(j.noteq.i).
[0078] FIG. 10 shows the details of the MICS units 280-1 to 280-N
represented by a MICS unit 280-i. Based on timing information
t.sub.i of a rake path obtained by the path searching unit 25,
multi-path interference correlative timings (MICT) t.sub.i,j,k are
obtained in the timing generating circuit 26i according to the
following equation.
t.sub.i,j,k=t.sub.i-t.sub.j+t.sub.k (j.noteq.i)
[0079] A signal mics(i,j,k) inversely spread by the respectively
corresponding inverse spreading circuits 104-il to 104iN at this
obtained timing t.sub.i,j,k is obtained and these signals are
composed and outputted by an MRC unit 128.
[0080] FIG. 11 shows an example of the construction of the MRC 128.
In the MRC 128, the signal mics(i,j,k) is multiplied in a
multiplier 128-2i by an appropriate coefficient r'.sub.k obtained
by a coefficient generating unit 128-1 based on a channel
estimation value .alpha. .sub.i (i=1 to N) obtained from a channel
estimating unit 29 and the noise power n.sup.2 obtained from a
level measuring unit 30, and the products are added by an adder
128-3. Furthermore, the output of the adder 128-3 is multiplied by
a coefficient r.sub.i,j in an adder 128-4 and the MICS(i,j) is
obtained. Therefore, the MICS(i,j) is represented by the following
equation.
MICS ( i , j ) = r i , j k = j r k ' mics ( i , j , k )
##EQU00005##
[0081] Here, the coefficient r'.sub.k and r.sub.i,j are obtained as
follows.
r'.sub.k=.alpha..sub.k*/{.SIGMA..sub.I.noteq.k|.alpha..sub.I|.sup.2I.sup-
.2+n.sup.2}
r.sub.i,j=.alpha..sub.jI.sup.2/{(I/N).sub.j+1}
where (I/N) .sub.j is the ratio of interference to be reproduced
and the power of interference other than that and is obtained as
follows.
( I / N ) j = k = j .alpha. k 2 I 2 / { I .noteq. k .alpha. I 2 I 2
+ n 2 } ##EQU00006##
[0082] In the previously applied invention, the coefficient r.sub.i
is obtained as follows using only the path i as represented in Eq.
3 when the MICS(i,j) is obtained.
MICS(i, j)=r.sub.imics (i, j, i)
r.sub.i=.alpha..sub.i*.alpha..sub.jI.sup.2/{.SIGMA.|.alpha..sub.I|.sup.2-
I.sup.2+n.sup.2}
[0083] In contrast, in the present invention, the accuracy of the
MICS(i, j) can be improved and the reduction effect of interference
can be improved by composing using paths other than the path i as
already shown.
[0084] Here, in the process of the MRC unit 128 of FIG. 11, the
coefficient r'.sub.k to multiply mics(i,j,k) is obtained as follows
approximating the noises to be constant when the MICS(i,j) is
obtained. r'.sub.k=.alpha.*.sub.k r.sub.i,j=.alpha..sub.j
(I/N).sub.j/{(.SIGMA..sub.k.noteq.j|.alpha..sub.k|.sup.2) ( (I/N)
.sub.j+1)}
[0085] Thereby, the size of the circuitry and the amount to be
processed can be reduced.
[0086] FIG. 12 shows another example of the construction of the
MICS unit 280-i. Compared to FIG. 10, this construction has a
selector unit 129 as a characteristic thereof. The selector unit
129 determines the magnitude of a path based on a channel
estimation value from the channel estimating unit 29 and obtains
mics(i,j,k) only for large paths. Thereby, the size of the
circuitry and the amount to be processed can be reduced without
degrading considerably the performance.
[0087] In the example of the construction of FIG. 12, two with k=s1
and s2 are selected from the inverse spreading circuits 104-ia and
ib.
[0088] The MIXR process of the MIXR circuit 28 of FIG. 8 exerts a
high effect when the MIXR process is applied to large paths. FIG.
13 shows a construction that is provided with a selector circuit 31
at the preceding stage of a MIXR circuit 28, as an exemplary
embodiment. The selector unit 31 controls such that the magnitude
of the path is determined based on the channel estimation value
from the channel estimating unit 29 and the MIXR processes are
executed only on large paths.
[0089] Paths that are not targets of the MIXR process are lead
directly to the rake circuit 27 to undergo a rake process without
undergoing any other process before undergoing the rake processes.
The example of the construction of FIG. 13 is a construction for
MIXR-processing two (2) paths and not executing any processes to
other (N-2) processes.
[0090] FIG. 14 shows yet another example of the construction of the
MICS unit 280-i. While MIXR-processing, the inverse spreading
timing t.sub.i,j,k of the mics(i,j,k) may be different from the
timing t.sub.i of the desired signal. For example, in the case
where each of t.sub.i is lining spaced equally from each other,
t.sub.i,j,k may coincide with any one (1) of t.sub.i. Then, the
case where the timing t.sub.i,j,k coincides with t.sub.i is
detected by the timing generating circuit 26i and the signal of the
case of the coincidence is blocked by switches 130-1 to 130-N.
Thereby, the degradation of the characteristic can be
prevented.
[0091] The timing generating unit 26i in the exemplary embodiment
of FIG. 14 creates the timing t.sub.i,j,k to correspond to the
above operation and executes comparison with t.sub.i. Then, when
coincidence or approximation almost equal to coincidence is found,
the signal is prevented from being inputted into the MRC unit 128
by controlling accordingly a switch 130-i of an outputting unit of
the mics(i,j,k) corresponding to the found timing.
[0092] Thereby, the mics(i,j,k) of the timing coinciding with a
signal can be masked.
[0093] Here, in FIG. 8, (1) the noise level of each finger is
constant in any finger in the rake composing unit 27 at the
following stage of the MIXR unit 28. Furthermore, (2) that the
amplitude of the data delivered from the rake composing unit 27 to
a correcting unit not shown indicates the likelihood of the signal,
is a precondition for the process.
[0094] In many cases, in the rake composing unit 27, in order to
create appropriate signals for the correcting unit as above, the
noise level of each finger is constant in any finger in the rake
composing unit 27 at the following stage of the MIXR unit 28.
However, when a MIXR process has been executed to each rake finger
respectively, the noise level of each finger containing
interference is reduced and dispersion occurs in noise power that
was at an almost same level in any finger. Consequently, effect of
the rake composition and error correction may not be exerted.
[0095] Therefore, it is preferable to provide a level compensating
unit 32 between the MIXR unit 28 and the rake composing unit 27 as
a construction shown in FIG. 15. Thereby, the noise power can be
made same as that before the MIXR process by amplifying
appropriately the signal after the MIXR process.
[0096] FIG. 16 shows yet another exemplary embodiment of the
present invention. Compared to the exemplary embodiment shown in
FIG. 8, the embodiment is characterized in that the positions of
the MIXR unit 28 and the inverse spreading circuit unit 24 are
exchanged.
[0097] In the construction of FIG. 16, the MICS unit 280-i
constituting the MIXR unit 28 has a construction shown in FIG. 17.
Compared to FIG. 10, the inverse spreading circuit 104-i is
replaced by the delay circuit 105-i.
[0098] As described above, the circuit construction can be
simplified by arranging the MIXR unit 28 before the inverse
spreading circuit unit 24.
[0099] As set forth hereinabove on the exemplary embodiments, even
when the number of paths is increased, interfering noises can be
effectively reduced by applying the present invention.
[0100] Thereby, a rake receiver can be provided, that executes the
effective maximal ratio composition in the time domain of signals
arriving at an antenna with various differences in delay time
thereof caused by multiple reflections of propagation paths of the
signals in a multi-path environment.
* * * * *