CA 02484277 2002-O1-30
27879-167D
-1-
TITLE OF THE INVENTION
RECEIVING PROCESS METHOD AND RECEIVING
APPARATUS IN MOBILE COMMUNICATION SYSTEM.
This is a divisional application of Canadian
Application Serial Number 2,369,863.
BACKGROUND OF THE INVENTION
1 . Field of the .Invention
The present invention relates to a mobile
communication system adopting CDMA (DS-CDMA) for
'performing spread spectrum multiple access. More
particularly, the present invention relates to a
receiving process method for canceling interference
due to multipath when high speed data transmission
is performed for down-link transmission from a base
station in cellular communication using DS-CDMA.
In addition, the present invention relates
to a receiving apparatus which can.remove multipath
interference according to the receiving process .
method.
2. Description of the Related Art
Wide band DS-CDMA (W-CDMA) has been
adopted as a wireless access method~in the next
generation mobile communication method IMT-2000
(International Mobile Telecommunication 200Q) .
The. maximum information transmission speed in IMT-
2000 is 144kbps in a mobile environment, 384kbps in
a walking environment and 2Mbps in a quasi-still
environrrient. Thus, it is predicted that multimedia
services in addition to voice services will be
provided in the next generation.mobile communication
system. On the other hand, when considering rapid
popularization of the Internet, multiplicity of
information, enlarging capacity and developments of
the 'next generation Internet in recent years, it
becomes necessary to develop a wireless access_
method for realizing information transmission speed
exceeding 2Mbps in. mobile communication.- Especially
in down-link transmission from the base~station, it
CA 02484277 2002-O1-30
27879-167D
-2-
is considered that high-speed and large capacity
traffic due to download of images, files, video and
the like from databases and web sites will increase.
Therefore, a high speed packet transmission
technology suitable for such data traffic becomes
necessary. Against this backdrop, ultrahigh-speed
packet transmission by extending the W-CDMA wireless
interface is being studied for realizing high-speed
packet transmission exceeding 2Mbps. For example,
application of adaptive modulation/demodulation and
error correction (channel coding) and ARQ (Automatic
Repeat reQuest) based on adaptive wireless link
control (link adaptation) is studied. The adaptive
modulation/demodulation and error correction based
on the link adaptation are methods for switching
modulation level (number of bits in one symbol), SF
(spreading factor), multi-code multiplexing number,
and coding ratio of error correction according to
propagation environment of each user in order to
perform high speed data transmission effectively.
For example, as the propagation envixonment.for a
user becomes better, maximum throughput of the
mobile communication system can be increased by
switching modulation methods of W-CDMA from QPSK to
more effective multilevel modulation, that is, 8PSK,
16QAM, 64QAM modulation. For example, when SF=4,
multi-code number is 3, error correction coding
ratio is 1/2, and 64QAM is used for data modulation,
8.5Mbps ultrahigh-speed data transmission is
theoretically possible by using a W-CDMA wireless
interface of chip rate 3.84Mcps.
As mentioned above, 8.5Mbps ultrahigh-
speed data transmission is theoretically possible by
increasing the modulation level, decreasing SF
(increasing multi-code number), and increasing
coding ratio of error correction. However,
increasing the modulation level leads to increasing
CA 02484277 2002-O1-30
. 27879-167D
-3-
required desired wave signal power to interference
power ratio (SIR) which is necessary for satisfying
the same receiving quality (bit error rate).
In addition, when applying the adaptive
modulation/demodulation and error correction to an
actual mobile communication environment, tolerance
to multipath fading (frequency selective fading)
becomes important. For example, orthogona.lization
between users (between code 'channels) is possible in
the same transmission path in a down-link in the W-
CDMA. However, degradation of transmission quality
occurs due to interference between multipaths under
multipath environment.
Generally in DS-CDMA, this multipath
interference can be suppressed tb 1/SF of received.
signal power on average for each code channel like
general multi-user interference. However, for
performing ultrahigh-speed data transmission of
8.5Mbps by using W-CDMA wireless interface having
chip rate 3.84Mcps, it is necessary to decrease SF
near to 1 and increase the multi-code number for
increasing data rate. In this case, degradation of
received SIR due to multipath interference becomes
very large. As a result, even when other user does
not exist and even when background noise such as
thermal noise is small, an area for realizing high-
speed packet transmission of multilevel.modulation,
low SF and high coding ratio is limited to an area
very close to a base station where there is na
multipath interference so that average throughput of
the system deteriorates.
SUMMARY OF THE INVENTION
A first object of the present invention is
to provide a receiving process method for avoiding
degradation of receiving characteristics due to
multipath interference which is a problem for
CA 02484277 2002-O1-30
27879-167D
-4-
performing high-speed transmission such as down-link
high-speed packet transmission using DS-CDMA in a
mobile communication system.
A second object of the present invention
is to provide a receiving apparatus for removing
multipath interference according to the receiving
process method.
The above object is achieved by a
receiving process method of a receiving apparatus
used in a mobile communication system in which a
sending apparatus sends a plurality o_f code channels
as code channel .groups to which spreading codes are
assigned to a receiving apparatus, and the receiving
apparatus receives the code channels, the receiving
process method comprising the steps of:
when spreading codes used for the code
channel groups are orthogonal code sequences,
generating received spreading signal
sequences of the code channel groups according to
the number of received paths; and
removing received spreading signal
sequences of a received path of own code channel
group of the receiving apparatus which should be
removed from received signals.
According to the receiving process method,
received spreading signal sequences (= multipath
interference replicas) used for canceling
interference of code channel groups of received
paths which become interference under multipath
30. environment are generated. In the receiving
apparatus, interference path occurring between own
code channel groups due to multipath is removed by
using the multipath interference replicas.
According to the receiving process method
of the present invention, interference path due to
multipath is canceled, received SIR (Signal-to-
interference power ratio) in the receiving apparatus
CA 02484277 2002-O1-30
f , 27879-167D
-5-
can be improved even under multipath environment.
That is, even when high speed data transmission is
performed by multiplexing a plurality of code
channels under multipath environment, degradation of
mean throughput of information transmission.speed
can be.avoided since receiving quality improves. As
a result, an area where a base station can provide
high speed data transmission in required quality can
be enlarged.
From the viewpoint of canceling multipath
interference of other code channel group (orthogonal
channel) of the sending apparatus to which the
receiving apparatus is connected, the receiving
process method may includes' the steps of:
when spreading codes of other code
channels used for control or used for other channels
in the code channel group are orthogonal code
sequences,
generating received spreading signal
sequences of the code channel groups according to
the number of received paths; and
removing received spreading signal
sequences of received paths of other code channels
which should be removed from received signals.
According to the receiving process,
multipath interference occurring between multipaths
of other code channel groups sent from the sending
apparatus to which own code channels are connected
can be canceled. Therefore, received SIR of the
receiving apparatus can be further improved.
From the viewpoint of canceling multipath
interference of other code channel group (non-
orthogonal channel) of the sending apparatus to
which the receiving apparatus is connected in
multipath environment, the receiving process method
may includes the steps of:
when all or a part of the spreading codes
CA 02484277 2002-O1-30
27879-167D
used for the code channel groups are non-orthogonal
code sequences,
generating received spreading signal
'sequences of the code channel groups according to
the number of received paths; and
removing received spreading signal
sequences of other code channels which arewon-
orthogonal in the same received path from received
signals.
According to the receiving process method,
it becomes possible to cancel multipath interference
occurring between multipaths of other code channel
groups (non-orthogonal channel) sent from the
sending. apparatus to which own. code channels are
connected and multipath interference occurring.in
the same path. Therefore, received SIR of the
receiving apparatus can be further improved.
From the viewpoint of canceling
interference due to received code channel group from
an adjacent sending apparatus, the receiving process.
method may includes the steps of:
when the receiving apparatus receives a
code channel group from another sending apparatus
which is not connected to the receiving apparatus,
generating received spreading signal
sequences of the code channel group from another
sending apparatus according to the number of
received paths; and
removing received spreading signal
sequences of the code channel group from received
signals.
In this case, multipath routes are.
different between the code channel group sent from
the other sending apparatus to which the receiving
apparatuses is not connected and other code channel
groups (non-orthogonal channel) sent from the
sending apparatus to which the own code channel
CA 02484277 2002-O1-30
27879-167D
group is connected. Thus, every channel become
interference_
Therefore, according to the receiving
process method mentioned above, since all receiving
spreading signal sequences generated for non-
orthogonal channels and channels of the other
sending apparatus are subtracted, received SIR of
the receiving apparatus can be further improved.
In the receiving process method, the
receiving spreading code sequence may be generated
on the basis of an estimated value of channel
variations and an estimated value of data modulation
obtained for each code channel.
Iw addition, The receiving process method
may include the steps of:
the sending apparatus sending pilot
signals of which the receiving apparatus knows
sending phase and sending amplitude to the receiving
apparatus periodically; and
20, the receiving~apparatus measuring received
phase and received amplitude of the pilot signals,
and obtaining the estimated value of the channel
variations by comparing the sending phase and
sending amplitude with received phase and received
amplitude.
The receiving process method may include
the steps of:.
the receiving apparatus obtaining the
estimated value of channel variations by averaging
the estimated value of channel variations obtained
by using the pilot signals and an estimated value of
channel variations obtained by comparing decision
results of data modulation with receiving phase and
amplitude for data signals.
From the viewpoint of improving generation
accuracy of the multipath interference replica by
updating estimated value-of the channel variations,
CA 02484277 2002-O1-30
27879-167D
_g_
the receiving process method may include the steps
of:
obtaining the estimated value of channel
variations on the basis of the pilot signals, the
data signals and the decision results of the data
modulation;
updating data modulation decision results
by using the estimated value of channel variations;
and
updating the estimated value of. channel
variations on the basis of the updated data
modulation decision results.
In the receiving process method, the same
estimated value may be used as the estimated value
of channel variations for code channels sent form
the same sending apparatus.
As for estimation of data modulation used
for generating the multipath interference replicas,
the receiving process method may includes the steps
of:
performing coherent detection by using the
estimated value of channel variations for received
despread signals of data signals obtained by
despreading received signals from which the received
spreading signal sequences have been subtracted;
wherein, when the receiving apparatus
receives signals by path diversity or by antenna
diversity, the receiving apparatus estimates data
modulation by performing hard decision for signals
on which antenna diversity has been performed.
The receiving process method may include
the steps of:
when the sending apparatus performs data
modulation for sending original information data
sequences which have been error correction coded
beforehand,
the receiving apparatus performing
CA 02484277 2002-O1-30
27879-167D
_g_
coherent detection by using the estimated value of
channel variations for received despread signal of
data signals obtained by despreading received
signals from which the received spreading signal
sequences have been subtracted, performing error
correction decoding on signals after performed
antenna diversity combining when signals were
received by path diversity o~r antenna diversity so
that original information data sequences are
estimated;
the receiving apparatus performing error
correction coding on the original information data
sequences which is estimated; and
the receiving apparatus performing data
modulation by using data sequences which are
obtained by performing error correction coding on
the original information data sequences so that data
modulation is estimated.
From the viewpoint of improving receiving
quality of signals to be demodulated by using
received spreading signal sequences (multipath
interference replicas) having high generation
accuracy, the receiving process method may include
the steps of:
updating the received spreading signal
sequences on the basis of updated estimated values
of channel variations; and
demodulating code channels to be
demodulated by using signals obtained by subtracting
the updated received spreading signal sequences from
received signals.
The above object is also achieved by a
receiving apparatus which receives code channel
groups each including code channels from sending
apparatuses, the receiving apparatus including an
interference canceler which comprises a plurality of
stages,
CA 02484277 2002-O1-30
27879-167D
-10-
a first stage of the stages comprising:
a data modulation estimation part and a
channel estimation part for estimating data
modulation and channel variations for each received
code channel which is a subject for interference
canceling;
a multiplier for multiplying an estimated
data modulation signal by an estimated value of
channel variations; and
a received spreading signal sequence
generation part for obtaining a received spreading
signal sequence for each multipath by performing
spreading a received signal by using a corresponding
spreading code;
a stage after the first stage in the
stages comprising:
an other channel multipath interference
canceling part for subtracting received spreading
code sequences of other code channels obtained in
the previous stage from received signals for each
received code channel which is a subject for
interference canceling;
an own channel multipath interference
canceling part for subtracting received spreading
signal sequences of own code channels obtained in
the previous stage corresponding to a path which is
a subject for demodulation;
a part for preparing signals corresponding
to the number of multipaths obtained by subtracting
received spreading signal sequences from received
signals by the other channel multipath interference
canceling part and by the own channel multipath
interference canceling part, and updating estimated
values of data modulation and channel variations by
using the prepared signals;
a received spreading signal sequence
updating part for updating received spreading signal
CA 02484277 2002-O1-30
27879-167D
-11-
sequences on the basis of updated estimated values
of data modulation and channel variations;
a last stage in the stages comprising a
data demodulation part for demodulating data by
using signals obtained by subtracting received
spreading code sequences obtained in the previous
stage from received signals for code channels to be
demodulated.
The receiving apparatus may include:
a data modulation estimation adaptive
switching part for switching between a first data
modulation estimation. part and a second data
modulation estimation part for performiwg estimation
of data modulation in each stage of the interference
canceler;
the first data modulation estimation part
estimating data modulation by performing hard
decision for signals on which antenna diversity has
been performed, when the receiving apparatus
receives signals by path diversity or by antenna
diversity;
the second data modulation estimation part
performing error correction coding on the original
information data sequences which were estimated, and
performing data modulation by using data sequences
which are obtained by performing error correction
coding on.the original information data sequences so
that data modulation is estimated.
In addition, the receiving apparatus may
include:.
. a subtracting part for subtracting
received spreading signal sequences from received
signals after multiplying the received spreading
signal sequences by predetermined interference
removing weight coefficients.
BRIEF DESCRIPTION OF THE DRAWINGS
CA 02484277 2002-O1-30
27879-167D
-12-
Other objects, features and advantages of
the present invention will become more apparent from
the following detailed description when read in
conjunction with the accompanying drawings, in
which:
Fig.l shows a configuration of a mobile
communication system in which a multipath
interference canceling method is applied according
to an embodiment of the present invention;
Fig.2 shows the same configuration as
Fig.l in which the mobile station 10 obtains the
spreading code information of each code channel
other than the own channel by using a way different
from the way shown in Fig. l;
Fig.3A shows a sending format example in a
case in which a pilot channel is code-multiplexed;
Fig.3B shows a sending format example in a
case in which the pilot channel is time-multiplexed;
Fig.4 shows a first configuration example
of an interference canceler of the present
invention;
Fig.S shows a configuration example of an
interference estimator in the interference canceler
shown in Fig.4;
Fig.6 shows a second configuration example
of an interference canceler of the present
invention;
Fig.7 shows a third configuration example
of an interference canceler of the present
invention;
Fig.8 indicates a case where code channe l
signals sent from the base stations (base station120
and base stationZ21) are received by the mobile
station 10;
Figs.9A-9E shows an interference
decreasing effect when the interference canceler of
the present invention is applied;
CA 02484277 2002-O1-30
. ~ 27879-167D
-13-
Fig.lO shows throughput characteristics
obtained by computer simulation for a case that the
interference canceler of the present invention is
applied in a multipath environment;
Fig.ll shows data. used for the computer
simulation shown in Fig.lO.
DETATLED DESCRIPTION OF THE PREFERRED EMBOQIMENTS
In the following, embodiments of the
present invention will be described with reference
to figures:
Fig.l shows a configuration of a mobile
communication system in which a multipath
interference canceling method is applied according
to an embodiment of the present invention.
In Fig. l, the mobile communication system
adopts, for example, CDMA for wireless access. The
mobile communication system includes a receiving
apparatus 10 (which will be called a mobile.station
hereinafter); a sending apparatus) 20 (which will be
called a base stations hereinafter), a sending
apparatus2 21 (which will be called a base station2
hereinafter), and an upper station (example .
wireless circuit control apparatus). Each base
station 20, 21 sends a plurality of code channels.
A part of code channels sent from the same base
station are orthogonalized by using orthogonal
spreading code and other code channels are not
orthogonalized. In a commercial system of W-CDMA
and IS-95, all code channels of down-link are
basically orthogonalized. However, since_special
spreading code is used in synchronization channels
of W-CDMA, a few number of channels which are not
orthogonal to other channels exist. In addition,
since~the number of spreading codes for
orthogonalization is. limited, non-orthogonal code
channel transmission by using the non-orthogonal
CA 02484277 2002-O1-30
~ ~ 27879-167D
-14-
spreading codes is performed when code channels the
number of which exceeds the limited number are
necessary.
As shown in Fig. T, other code channel
groups (~2, ~3) sent from the base station120 can be
considered to be common channels such as common
control channels or communication channels to other
carrier. Code channels of other base station (base
station221 in this case) can be orthogonalized in
the same-path. However, the code channel from the
base station221 can not be orthogonalized to the
code channel sent from the base station120 since
they are received by the mobile station 10
asynchronously. In addition, in this example, the
base station120 and the base station221 send
spreading.code information to the mobile station 10
as down-link control information for the mobile
station 10 to obtain spreading.code information of
each code channel other than its own channel.
Fig.2 shows the same configuration as
Fig.l in which the mobile station 10 obtains the
spreading code information of each code channe l
other than the own channel by using a way different
from the way shown in Fig.l. In the example of
Fig.2, the base stationz2l sends its own spreading
code information to the base station120 via an upper
station, and the base station120 sends spreading
code information of the both base stations (base
station120 and base station221) to the mobile
station 10 as down-link control information.
As mentioned above, although methods in
which the base station120 sends the spreading code
information to the mobile station 10 as the control
information have been described, there is a method
in which the mobile station 10 recognizes spreading
codes without. information from the base station120.
For example, the mobile station 10 prepares
CA 02484277 2002-O1-30
27879-167D
-15-
estimated spreading code candidate beforehand,
obtains correlation between received signal and the
spreading code candidate. As a result, the mobile
station 10 can recognize spreading code used in
other base station by obtaining spreading code
having large correlation.
A sending format of packet data sent from
the base stations 20 is as shown in Figs.3A and 3B,
wherein packet data is a sending unit of data, and.
is obtained by dividing sending data into each
constant amount of data.
Fig.3A shows a sending format example in a
case in which a pilot channel is code-multiplexed.
In this case, one packet includes Ns slots, and the
pilot signals for estimating channels are code-
multiplexed to data channels as a code channel which
is spread by a spreading. code (~l). On the other
hand, Fig.3B shows a sending format example in a
case in which the pilot channel is time-multiplexed.
In this case, the pilot signals are inserted into
data signals periodically (for each slot) (2~)- In
addition., one packet (=Ns slot) includes K code
channels (#1~-#K) in both cases.
In the following, the first case in which
the pilot signal is code-multiplexed will be taken
for explaining embodiments of the present invention.
The pilot channel will be called a common pilot
channel (CPICH) since the pilot channel can. be also
used for channel estimation of other transmission
data channels.
A receiving apparatus (which will be
called an interference canceler hereinafter) to
which the receiving process method of the present
invention is applied is configured as shown in Fig.4
for example. In this example, it is assumed that
the interference canceler is used for a down-link by
which the base station sends data and the mobile
CA 02484277 2002-O1-30
27879-167D
-16-
station receives the data.
In Fig.4, the interference canceler (first
configuration example) includes a plurality of
stages of interference estimators 100 and 110, delay
lines 121-124, subtractors 150-153, multipliers 140,
141, and interference estimator signal output parts
130, 160. In this example, Ib,~'~ represents a
received spreading signal sequence of lth path
(1<1<Lj of bth antenna branch (1<b<B) in pth stage
(1<p<P) of the interference estimator. The received
spreading signal sequence will be called a multipath.
interference replica hereinafter. In the first
stage interference estimator 100, signals received
by the receiving antennas #1 and #2 are directly
input. In and after the second stage interference
estimator 110, received signals from which all other
multipath interference replicas Ib,~'~ generated in
the previous stage have been subtracted are input.
The interference estimators 100, 110 estimate
channels (channel estimation). The channel
estimation value, is updated for each stage (for each
interference estimator stage) by using.common pilot
channel, or, in. addition to this, by using decision
data modulation and data symbol. Decision of data
modulation by using this is also updated. The
multipath interference replica is updated by using
the channel estimation value for each stage.
Therefore, as channel .estimation accuracy and data
decision accuracy improve, generation accuracy of
the multipath interference replica improves.,
Next, the configuration and the operation
of the interference estimator 100, 110 will be
described with reference to Fig. S. In the following,
the first stage interference estimator 100 will be
described as an example.
CA 02484277 2002-O1-30
27879-167D
-17-
This interference estimator 100 includes
Rake/antenna diversity combining parts 200, 210, a
parallel/serial converter (=P/S converter), an error
correction decoder 230, hard decision parts 240, 241,
an error correction coder 250, a data modulator 260,
a serial/parallel converter (=S/P converter) 270,
channel estimators 300, 310, multipath interference
replica generators 320, 330, spreading parts 340,
341, computing units 350-353, 360-363, an antenna
signal input part 400, and a multipath-interference
replica signal output part 410. In addition, a
multiplier 280 which performs complex conjugate
operation between signals of the Rake/antenna
diversity combining parts 200, 210 and signals from
the data modulator 260 is provided in the
interference estimator. In addition, the
Rake/antenna diversity combining part 200 includes
despreading parts 201, 202, multipliers 203, 204 and
an adder 205. The channel estimator 300 includes
despreading parts 301. 302, and channel estimators
303, 304. The multipath interference replica
generator 320 includes spreading parts 321, 322, and
multipliers 323, 324.
Canceling of multipath interference due to
multipath between the own channel code groups. is
performed in the following way in the above-
mentioned configuration.
An input signal to the despreading part
which performs despreading of the lth path of the
bth antenna in the pth stage in the interference
estimator in the multipath interference canceler is
a signal in which all other multipath interference
replicas are subtracted from the received signal.
A narrow band modulation signal wave form
dk(t) and spreading signal modulation wave form ck(t)
of kth code channel is represented by the following
equations 1 and 2.
CA 02484277 2002-O1-30
27879-167D
-18-
dk(t) =~,8k(1)'eXP~.l~k(1)~' ua(t -1Te) ( 1 )
~k(t)=~exP~.l~k(l)} u~(t-iT~) (2)
In the equations (1) and (2), Td indicates
a symbol interval, an Tc indicates a chip interval.
In addition, Ud(t)=1(0) for 0<t<Td (otherwise) and
Uc(t)=1(0) for 0<t<Tc (otherw.ise), and N=Td/Tc is SF.
~k(i)E~q~l2+~cl4;q=0,1,2,3 indicates QPSK spreading
modulation by spreading codes in wh~.ch code channels
are orthogonalized by using orthogonal spreading
codes. gk(i) and ~k(i) indicate data modulation
amplitude and phase respectively. gk(i) and ~k(i)
are represented by the following equations
respectively according to the data modulation method.
1. OPSK modulation
8k (i) _
~k(i)e~q~/2+~c/4;q=0,1,2,3} (4)
2. 8PSK modulation
gk(i)=~ (5)
~k (t) E {qn l4; q = 0,1,...,7} ( 6 )
3. 16QAM modulation
Sk (1) = xk2 '~' Ykz ( 9 )
~k(1)=~ l Yk (10)
xk
wherein
zk E ~ (2qs + 1) 215 ; qx = -2,-1,0,1 ( '7 )
CA 02484277 2002-O1-30
27879-167D
-19-
yk a (2qy + 1) 215 ; qy = -2,-1,0,1 ( 8 )
4. 64QAM modulation
8k (1) = xkz + yk2 ( 1 3 )
~k (1) = tan-I yk ( 14 )
xk
wherein
xk E (2qs + 1) ~ ; qs = -4,-3,...3 ( 11 )
yk E (2qY +1) ~ ; qy = -4,-3,..,3 ( 12 )
In the same way, a narrow band modulation
signal wave form d~pich(t) and a spreading modulation
signal wave form c~p;~h (t) of the common pilot channel
can be represented as follows.
d~~~h(t) _ ~exp~jtc l4~ u~P;~b(t-iT pr~n) ( 15 )
Ccpich(t) - ~exp[J~lcpich(1)} u~(t-iT ) ( 16 )
r
wherein T~p;~n indicates a symbol interval , and
u~p;~n(t)=1 (0) for 0<t<T~pich (otherwise) . Since
spreading by orthogonal spreading code is performed
also for the common pilot channel, code channels in
the same path are orthogonalized. The sending
signals are transmitted in L multipath channels and
are received by B receiving antennas. A received
signal rb(t) at the bth antenna can be represented
by
L K
rb(t)-~~b.l(t) ~dk(t 'il)~ Ck(t Zl)+dcP~ch(t '~()~ Ccpich(t 'C!) +Y1(t)
t=1 k=I
(17), wherein ~b,lindicates complex fading envelope
of the lth path of the bth antenna, zlindicates
CA 02484277 2002-O1-30
27879-167D
-20-
propagation delay of the lth path: n(t) indicates
additive Gaussian noise component of one-sided power
spectrum densely NO/2. A despreading part output of
the interference estimator corresponding to the mth
symbol of the nth slot for the the lth path.of the
bth antenna of the kth code channel in the lth stage
is indicated by the following equation.
_ 1 r~"+cm+pr,+.r, ,
Zk.b.l (n° m) - ~.d .IIT,,oI+mT,+~r, rb (r~ Ck (t ZI )dt ( 1 8 )
In the same way, despreading output of the common
pilot channel is represented by the following
equation.
(n~m) - 1 ~r,"+cm+ord~,+r, r (t~ ~~ (t-,~ )dt ( 19 )
cpich,b.l T T~a+mT d~h+t, b cpich l
cpich
wherein Tslot indicates a slot interval. A channel
estimation value used for Rake combining in rth time
of repeated channel estimation (1<r<R) of the pth
~ ( p:r)
stage (1<p<P) is represented by fib., (n) . A first
" a.u
stage channel estimation value ~b,,(n) is obtained by
the following equation by using the common pilot
channel.
"u.u
~u
2 0 ~ b., (n) _ ~ Zc~ch.b.l (n' m)' dcprcr~ (n, m) ( 2 0 )
Ncpich i_-1
wherein Ncpich is a symbol number of the common
pilot channel included in one slot. That is, by
multiplying the received complex signal by complex
conjugate of sending complex signal, complex envelop
change of channel is obtained. By using the channel
estimation value and by multiplying BL multipath
"c~.u
components by complex conjugate of ~b.,(n), coherent
~ (p=l,r=t)
Rake combining output dk (n,m) in the mth symbol of
CA 02484277 2002-O1-30
27879-167D
-21-
the nth slot of the kth code channel is obtained by
the following equation.
"(t.t) a t "f.1)~
dk (n'm)-~~zxlb.l(n'm)~~b.l (n) (21)
m i=t
When tentative data decision is performed
after Rake combining is performed, hard decision for
)
data sequence dt (n, m) is performed, and tentative
decision data symbol sequence
_ (t.t) _ (~=',r=t)
d k O~ m) = gk°='.r=') ~n~ m). exp j ~,r (n, m) ( 2 2 ) i s
reproduced. On the other hand, the tentative data
decision is performed after error correction
" ('.u .
decoding, branch metric is calculated for dk (n, m),
the branch metric of K code channels is
parallel/serial-converted, error correction decoding
is performed, so that binary information data
n (P=l.r=1)
sequence b (i) is obtained. Other methods can be
used for error correction decoding. Error
correction coding is performed for the decoded
information data sequence, and the decoded
information data sequence is assigned to K code
channels by serial/parallel conversion. After that,
data modulation is performed so that tentative
decision data symbof sequence
_ u.') _ ('.n
d,r (n, m)=gk'v)(n,m)~exp j~x (n, m) (23) is reproduced.
Then, by multiplying data symbol zk'b.r(n,m) of the
despreading part output by complex conjugate of
_ u.')
dx (n, m) (reverse modulation) so that data modulation
CA 02484277 2002-O1-30
' ' 27879-167D
-22-
component is removed, the data symbol can be used a.s
pseudo-pilot symbol. Thus, channel estimation is
performed again by using the KNd pseudo-pilot
symbols in addition to the common pilot channel,
Rake combining is performed and the tentative data
decision value is updated. The channel estimation
o..+o
value ~b,, (n) obtained after repeating this process
r times is.represented by
"a.r+n . 1 "i~..+~> 1 1 x N, _ a..r
b.r (n) -1 + w ~ b'~ (n) + 1 + w x ~'~ - o..~ ~, ~ zklb.r (n' m) ~ d k (n~
g k (n' m)2 k=I m=I
k=1 m=l
(24) .
In the above description, although channel
estimation using the common pilot.channel is
performed by averaging pilot channels of one slot
interval, this can be also done by other methods
which are generally known. For example, the first
term of the equation (24) indicates a channel
estimation value using the common pilot channel, the
second term indicates a channel estimation value in
which decision feedback data symbol is regarded as
the pseudo-pilot symbol. The channel estimation
value by decision feedback data is obtained by using
weighted mean value according to amplitude. In
addition, w in the equation (24) is a weight
coefficient in averaging the channel estimation
value by the common pilot channel and the channel
estimation value by the decision feedback data
symbol. Optimal estimation accuracy can be obtained
by using a small value for w when data decision
error is large and by using a large value when data
decision error is small. w=0 indicates a case where
the data symbol is not used for channel estimation.
Since the symbol number used for channel estimation
increases by adding channel estimation by the
CA 02484277 2002-O1-30
27879-167D
-23-
decision feedback data, effects of averaging noise
and interference increase. Therefore, channel
estimation accuracy improves. (However, since
decision error is included in the decision feedback
data symbol, the decision error may affect the
channel estimation accuracy.) In the interference
estimator; by using thus obtained channel estimation
tt,R+t)
value fib,, (n) and the tentative decision data symbol
(1,R)
sequence dk (n, m), the multipath interference
replica of the lth path of the bth antenna can b.e
obtained by a following equation.
w (t) ~ L a (1.R+t) K _ (t.R) n ~ w a
Ib.~(t-'Cr)=~~b.r (t) ~dk (t-~e)~ ~k(t-'Cr)+d~~~~(r-'Ct)' ~~;~h(t-W )
r=t k=t
(25)
By using the multipath interference,
replica, a despreading part input signal of the
interference estimator of the lth path of the bth
antenna in the second stage can be represented by
the following equation.
L n(1) w
r~;~(t)=rb(t)-tx~le.j(t-~>) (26)
j=t
Jxt
wherein a is an interference canceling weight
coefficient and 0<c~. When error included the
generated interference replica is large, this effect
can be alleviated by using a small a. However,
. since too small a decreases the effect of removing
interference, the effect of multipath interference
removal can be increased by setting optimal a
according to generation accuracy of the interference
replica. For example, when there are many
multipaths, since accuracy of channel estimation
degrades, there is a case in which using smaller a
is more effective. In addition, as the number of
CA 02484277 2002-O1-30
27879-167D
-2 4-
stages becomes large, since the accuracy of the
multipath interference replica improves, receiving
characteristics can be improved by using~larger a.
In the same way, a despreading input
signal of the interference estimator of the pth
stage (p>2) can be obtained by the following
equation by using the multipath interferenc a
replicas generated in previous stage.
c ~(P-i)
r~;)(t)=rb(t)-(x~le.l (t-zi) (2'7)
;m
~*~
In each stage; for s signal in which the multipath
interference replicas are subtracted, channel
estimation, tentative decision of data modulation
are performed like in the first stage so that the
multipath interference replica is updated. Then, in
the final stage (p=P), data sequence after Rake
(P.R)
combining dx (n, m) is error-correction-decoded (when
error correction coding was performed), and binary
~(P)
information data sequence b (i) is demodulated.
As mentioned, since a signal in which all
( P-~)
other multipath interference replicas Ib.r generated
in previous stage have been subtracted from the
received signal is input into each interference
estimator in and after second stage, the multipath
interference replica is updated for each stage.
Therefore, the multipath. interference replica having
high accuracy can be used for removing multipath
interference of the own channel. Thus, interference
canceling of, high receiving quality can be realized.
In.~the above embodiment, although a case
in which the common pilot channel is code-
multiplexed has been described, the interference
canceler of the present invention can be easily
CA 02484277 2002-O1-30
~ ~ 27879-167D
-25-
applied to a sending format shown in Fig.3B (a case
where common pilot channels are time-multiplexed).
Next, an example for removing multipath
interference due to code channel group other than
the own channel code group with reference to figures.
Compared with the multipath interference
canceler shown in Fig.4, the multipath interference
canceler shown in Fig.6 includes interference
estimators 510, 540 of other orthogonal code. channel,
and interference estimators 520, 550 of other non-
orthogonal code channel. In this embodiment,
although the multipath interference canceler
includes the interference estimators 520, 550 of
other non-orthogonal code channel, it is possible to
use processing parts of other sending apparatus
instead of the interference estimators 520,.550 of
other non-orthogonal code channel.
In the same way as the above-mentioned
interference estimator, each interference estimator
performs channel estimation and decision of data
modulation so that the multipath interference
replica is output. In Fig.6, the multipath
interference replica of the own channel is indicated
nCP)
by Ib.r, the multipath interference replica of other
~ (P)
orthogonal channel is indicated by Ob.~, and the
multipath interference replica of the other non-
w (P)
orthogonal channel is indicated by U.b.i. In the
multipath interference canceling method shown in
Fig.6, estimation of the multipath interference
replica for each code channel group is performed in
parallel for each stage. That is, in the first
stage, the multipath. interference replica is
generated by using the received signal itself. In
the second stage, the multipath interference replica
CA 02484277 2002-O1-30
27879-167D
-2 6-
is estimated more accurately on the basis of a
signal in which the multipath interference replica
estimation value of the first stage is subtracted
from the received signal.
Since code channels sent from. the same
sending station are received by the mobile station
after receiving the same channel variation,
accuracy of channel estimation can be improved and
receiving processing amount can be decreased by
10 performing the above-mentioned channel estimation by
providing commonality of channel estimator and by
using more common pilot channels and data signals.
In the second configuration example of the
interference canceler shown in Fig.6, the
interference estimator in and after the second stage
receives a signal obtained by subtracting all other
multipath interference replicas generated by the
previous stage from the received signal. In this ,
embodiment, as for interference replicas of code
channels which are non-orthogonal, multipath
interference replica of the same path is also
subtracted from the received signal so that
receiving quality improves.
On the other hand, in the configuration
example 3 of the interference canceler shown in
Fig.7, estimation of the multipath interference .
replica for each~code channel group is performed in
series. The configuration of this interference
canceler is the same as that of Fig.6. Interference
estimators 700, 730 of the own code channel,.
interference estimators 710, 7.40 of other orthogonal
code channel, and interference estimators 720, 750
of other non-orthogonal code channel are added. In
this embodiment, although the multipath interference
canceler includes the interference estimators 720,.
~750-~of other non-orthogonal code channel, it is
possible to use parts of other sending
CA 02484277 2002-O1-30
' ' 27879-167D
-27-
apparatus instead of the interference estimators 720,
750 of other non-orthogonal code channel.
In the example of Fig.7, processes are
performed in the order of interference estimation of
own channel, interference estimation of other
orthogonal code channel and interference estimation
of other non-orthogonal code channel. Therefore, in
the first stage, interference estimation can be
performed for succeeding channel by.a signal
obtained by subtracting interference replica of
processed channel from the received signal. Thus,
the performance is better than that of the
configuration example 2 shown in Fig.6. After the
second stage, as for an interference estimator,
multipath interference replicas for channels which
was processed earlier are subtracted from the
received signal, in addition, multipath interference
replicas obtained in the previous stage for the own
channel and channels which will be processes after
the own channel are subtracted from the received
signal. It can be expected that receiving quality
of the configuration example 3 of Fig.7 is bette r
than the configuration example 2. However, since it
can be considered that process delay for signal
processing generally becomes large for the example 3,
the configuration example 2 or the configuration
example 3 is selected according to the circumstances.
In addition, it is possible to adopt a combined
configuration, for example, a configuration
including the configuration example 3 for the first
stage and the configuration example 2 in and after
the second stage.
Next, multipath interference decreasing
effect when the interference canceler (Figs.4, 6, 7)
of the present invention is applied, and improvement .
effect of receiving quality obtained by interferenca
decreasing effect will be described with reference
CA 02484277 2002-O1-30
27879-167D
-28-
to Figs.8 and 9.
Fig.8 indicates a case where.code channel
signals sent from the base stations (base station120
and base stationz2l) are received by the mobile
station 10 by using two paths on the basis of Figs.l
and 2, in which the received paths from the base
station120 are indicated as path 1 and path 2 and
the received paths from the base station221 are
indicated as path 3 and path 4. Therefore, received
signal at the mobile station 10 is one in which
following code channels are multiplexed.
~l received signal of path 1 of the own code
channel group (orthogonal channels) to be
demodulated
2~ received signal of path 2 of the own code
channel group (orthogonal channels) to be
demodulated
~3 received signal of path 1 of other code channel
group (orthogonal channels) of the base station120
(own cell)
~ received signal of path 2 of other code channel
group (orthogonal channels) of the base station120
(own cell)
~5 received signal of path 1 of other code channel
group (non-orthogonal channels) of the base stations
20 (own cell)
~ received signal of path 2 of other code channel
group (non-orthogonal channels) of the base stations
20 (own cell)
~7 received signal of path 3 of code channel group
of the base stationZ21 (other cell)
~ received signal of path 4 of code channel group
of the base stationz2l (other cell)
Accordingly, the mobile station 10
receives the signals of ~1-~ in the same bandwidth
as received spread signals corresponding to the
examples of Figs.l and 2. The mobile station 10
CA 02484277 2002-O1-30
27879-167D
-29-
performs despreading by using spreading code of code
channel desired to be demodulated so that only the
code channel is converted into narrow band signal
and is demodulated. Figs.9A shows a case where the
interference canceling method of the present
invention is not applied. In this case, since the
path 1 of the own code channel to be demodulated is
converted into a narrow band signal and is
demodulated, signal component of the~same path (~1
in Fig.8)~o.f the other orthogonal code channels of
the base station 1 does not remain in the despreaded
signal, however, other received signals (2~-~)
remain 'as interference. Therefore, the receiving
quality becomes one according to SIR shown in Fig.9A.
When the interference canceler shown in
Fig.4 is applied for performing interference
canceling, interference of other multipath (~2 of
Fig.8) of the own code. channel is removed as shown
in Fig.9B. Thus, the SIR of despread signal becomes
larger than that of Fig.9A in which the interference
canceling is not performed so that the receiving
quality improves. In addition, when interference
canceling is performed on the basis of the
interference canceler shown Fig. S. or Fig.6,
interference estimation is performed for the other
orthogonal channels and the ocher non-orthogonal
channels other than the own code channels.
Therefore, the effect of interference decreasing
becomes as shown in Figs.9C-9E. That is, Fig.9C
shows a.case where. interference canceling is
performed for the own code channel and the other
code channel group (orthogonal channel) of the base
station120. In this case, interference signals of
~3 and ~ of Fig.8 are removed. In addition, Fig.9D
shows a case where interference canceling is
performed for the own code channel + other code
channel group (orthogonal channel) of the base
CA 02484277 2002-O1-30
' ' 27879-167D
-30-
station120 + other code channel group (non-
orthogonal channel) of the base station120. In this
case, the interference signals of 05 and ~ shown in
Fig.8 are removed. In addition, Fig.9E shows a case
where interference canceling is performed for the
own code channel + other code channel group
(orthogonal channel) of the base station120 + other
code channel group (non-orthogonal channel) of the
base station120 + code channel group of the base
station221. In this case, the interference signals
from two paths of the code channel of the base
station221 are removed, further improvement of the
received quality_becomes possible as shown in Fig.9E.
Accordingly, by applying the interference
canceler of Fig.S or Fig.6 to mobile communication
in which multipath fading occurs, received quality
is improved further compared with interference
canceler of Fig.4 since SIR of despreaded signal
becomes large.
Next, a result of computer simulation
showing effects of this interference canceling
method will be described. Fig.lO shows the
throughput characteristics, and Fig.ll shows data
used for the computer simulation.
As shown in Fig.ll, the data for the
computer simulation is as follows.
Chip rate (1) . 3.84Mcps, Symbol rate . 240ksps,
Information bit rate . 8.42Mbps, spreading ratio .
16, number .of multicodes . 12, spreading code .
orthogonal code sequences, Gold sequences,
Modulation method . &4QAM for data modulation, OPSK
for first spreading, Channel coding/decoding .
convolutional coding (rate=1/2, constraint length=9)
/ soft decision, Viterbi. decoding, Antenna
diversity . 2branch, Channel model . L-path Rayleigh,
Doppler frequency fp=80Hz.
As shown in Fig.ll, the information bit
CA 02484277 2002-O1-30
27879-167D
-31-
rate 8:42Mbps is realized by using 64QAM data
modulation in 3.84Mcps chip rate, spreading ratio 16,
l2 multicodes, convolutional coding ratio 1/2.
Fig.lO shows the result of the computer
simulation performed by using the data shown in
Fig.ll.
In Fig.lO, the vertical axis indicates
throughput (bits/sec), the lateral axis indicates
average receiving Eb/No(dB) in which Eb/No is signal
power to noise power density ratio per 1 information
bit.
In this computer simulation, propagation
models of L=1 - 4 paths were evaluated in which
interference removing weight coefficients for L=2, 3,
4 are 0.9, 0.7, 0.7 respectively. In addition, the
number P of stages of the interference canceler is '4
and repeated number R of channel estimation is 1.
In addition, throughput characteristics in which the
multipath interference canceler is not applied are
evaluated for checking effects of the case in which
the multipath interference canceler is applied. In
the figure, the cases in which the multipath
interference canceler is applied are indicated by X
(L=1) , O (L=2) , O (L=3) , D (L=4) , and the cases in
which the multipath interference canceler is not
applied are indicated by 1 (L=2) , ~ (L=3) , ~ (L=4) .
As shown in the result of computer
simulation, high throughput of 8.4Mbps is achieved
in a high Eb/No area in L=l path environment (X in
Fig.lO). However, the throughput deteriorates below
2Mbps without the multipath interference canceler in
L=2 path environment (~ in Fig.lO). On the other
hand, by applying the interference canceler (4
stages) of the present invention, high throughput of
8Mbps can be obtained even in the L=2 path
environment (O in Fig.lO). It is understood that
the throughput in the multipath environment can be
CA 02484277 2002-O1-30
27879-167D
-32-
largely improved according to the present invention.
As described above, according to the
present invention, the multipath interference
replicas are generated on the basis of accurate
channel estimation values using decision feedback
data after or before error correction (channe l
coding) decoding of common pilot channels and
communication channels in code channels which are
spread by orthogonal code sequence and non-
orthogonal code sequence, and the multipath
interference replicas are removed from the revived
signals (multipath interference cancel). Thus, it
becomes possible to largely improve received quality
(bit error rate, throughput and the like) in
multipath environment. As a result, since received
signal power required for the same received quality
can be largely decreased, an area where high speed
data transmission is available covered by a base
station can be enlarged compared. with a conventional
technology in which high speed data transmission is
limited to an area very close to the base station
where there is no multipath interference.
The channel estimation can be performed
without the decision feedback data.
In the above-mentioned examples, the
function of the interference canceler corresponds to
the other channel multipath interference canceling
part, the own channel multipath interference
canceling part and the subtracting part. The
function of the interference estimator in the
interference canceler corresponds to the data
modulation estimation part, the channel estimation
part, multiplier, received spreading signal sequence
generation part, the data modulation estimation
updating part, channel estimation updating part; the
received spreading code sequence updating part and
the data demodulation part.
CA 02484277 2002-O1-30
27879-167D
-33-
In addition, the function of interference
estimator of the interference canceler corresponds
to the.first to third channel variation estimation
part, the data modulation decision updating part,
the channel variation estimation updating part, the
same estimation value application part, the coherent
detection part, the first and second data modulation
estimation part, the original, information data
sequence estimation part and data modulation
estimation adaptive switching part.
As mentioned above, according to the
present invention, in cellular communication using
DS-CDMA, especially when ultrahigh-speed data
communication equal .to or higher than the chip rate
is performed, received quality can be largely
improved in multipath environment by generating
multipath interference which largely degrade
received quality and by subtracting the multipath
interference from the received signals.
In addition, according to the present
invention, a receiving apparatus which can cancel
multipath interference even when performing
ultrahigh-speed data transmission in multipath
environment can be provided.
The present invention is not limited to
the specifically disclosed embodiments, and
variations and modifications may be made without
departing from the scope of the invention.
35
