JP2002135165A - Communication method and communication device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide communication methods and communication devices which can reduce computation volumes necessary for demodulation and improve accuracy for demodulation. SOLUTION: Data symbols dc,WMF determined in the past, channel impulse responses h(k), and convolution series b(k) of diffusion code series c(k) for each user and channel impulse responses h(k) are held in a data-holding unit 102. At an adder 104, the result of computation of multiplying data symbols dc,WMF, determined in the past by convolution series b(k) of user diffuse code series c(k) by channel impulse response h(k), is subtracted from reception signal e. At a WMF(whitening matched filter) 106, data symbols dc,WMF are derived from multiplication of the computation result by AH.Rn-1 at the adder 104.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a communication device for demodulating a received signal and a communication method in the communication device.

[0002]

2. Description of the Related Art Code division multiple access (CDMA) systems are
This is a multiple access method that enables a plurality of line connections by assigning a code sequence to each user when a plurality of mobile stations communicate simultaneously in the same frequency band.
The direct spread communication system, which is one of the CDMA systems, is a system in which an information signal is multiplied by a spreading code sequence to spread and transmit the signal in a band sufficiently wider than the bandwidth of the original information signal. The information transmission rate per channel is uniquely determined from the number of spreading code sequences, the spreading factor, and the transmission rate of the spectrum-spread signal. Therefore, by changing the number of spreading code sequences and the spreading factor in a certain transmission bandwidth, it is possible to simultaneously support services of various transmission rates from low to high.

[0003] As a receiving technique in the CDMA system, joint detection (Joint Detection) is used.
2.) There is a multi-user detection demodulation method called demodulation. In the joint detection demodulation, a system matrix A in which a sequence obtained by convolving a spreading code sequence of each user with a channel impulse response is arranged as an element is generated. When this system matrix A is used, the data symbol d to be transmitted and the received signal e are expressed by a relation of e = Ad + n, where n is the noise. Therefore, the data symbol d is estimated by the operation related to the system matrix A. be able to. Hereinafter, the relationship among the received signal e, the system matrix A, the data symbol d, and the noise n will be specifically described.

A spreading code of a user k (k = 1, 2,..., K)
Issue series c^(K)And channel impulse response h
^(K)To c^(K)= (C₁ ^(K), C₂ ^(K), ..., c_Q ^(K))^T・ ・ ・ ・ ・ (1) h^(K)= (H₁ ^(K), H₂ ^(K), ..., h_Q ^(K))^T... (2) Spreading code sequence c of user k^(K)And channel
Impulse response h^{( k)}Convolution series b with^(K)
Is the spreading factor Q and the channel impulse response width
W, where N is the number of demodulated symbol sets for each user, b^(K)= (B₁ ^(K), B₂ ^(K), ..., b_{Q + W-1} ^(K))^T= C^{( k)} * H^(K)  ・ ・ ・ ・ ・ (3) (However, “*” indicates convolution
You. ), Determinant,

[0005]

(Equation 1) Becomes The system matrix A is represented by the convolution sequence b ^(k)
Are arranged as elements,

[0006]

(Equation 2) It is expressed as As described above, the data symbol d and the received signal e have a relationship of e = Ad + n. Here, the noise n is omitted and e = Ad, and the set d ^(k) of the data symbol d and the corresponding Set of received signals e ^(k)
Each _{^{d (k) = (d 1}} (k), d 2 (k), ..., d N (k)) T ····· (6) e (k) = (e 1 (k), e ₂ ^(k) ,..., E _{NQ + W−1} ^(k) ) ^T ... (7), the relationship between the data symbol set d ^(k) and the received signal set e ^(k) Using matrix A,

[0007]

(Equation 3) Becomes By performing an operation on the system matrix A based on the equation (8), the data symbol d can be estimated.

[0008]

By the way, when the object to be demodulated is divided into a plurality of sets by the above-described joint detection demodulation and each data symbol d is demodulated in order, conventionally, the data symbol set d ^(k)
Among them, the data symbols obtained in the past are also estimated as unknowns each time.

However, if the data symbols obtained in the past are also estimated as unknowns each time, the size of the system matrix A increases, and the amount of calculation increases. Moreover, the accuracy of the past data symbols estimated there is reduced because the data symbols before the first symbol and after the last symbol in the data symbol set are not considered. This is the data symbol set d ^(k)
The received signal e corresponding to the first and last few symbols of
Although the signals corresponding to the other data symbol sets overlap, the system matrix A and the data symbol set d
This is because the component corresponding to ^(k) is not arranged. Therefore, the error propagates to the data symbol d estimated at the present time.

The present invention has been made to solve the above problems, and has as its object to reduce the amount of computation required for demodulation;
It is another object of the present invention to provide a communication method and a communication device capable of improving the accuracy of demodulation.

[0011]

According to one aspect of the present invention, there is provided a communication method in a communication apparatus for demodulating a received signal. The received signal is demodulated by subtracting the component containing the held past demodulated signal from the held and currently demodulated received signal.

In such a communication method, when a signal to be demodulated includes a signal component which has been demodulated in the past, the component is subtracted and demodulation is performed to thereby reduce the amount of computation at the time of demodulation. Can be reduced, and the accuracy of demodulation can be improved as compared with the case where a past demodulated signal is estimated as an unknown number.

According to a second aspect of the present invention, there is provided a communication method in a communication apparatus for demodulating a received signal as described above. Based on the matrix in which the convolutional sequence of the code sequence and the channel impulse response are arranged as components, the demodulation target is divided into a plurality of sets,
When sequentially demodulating each data symbol, hold the past demodulated data symbols, from the received signal to be demodulated now, to the held past demodulated data symbols spread code sequence and channel impulse response The demodulation of the data symbol is performed by subtracting the signal multiplied by the convolution sequence of.

In this case, demodulation is performed by subtracting a signal obtained by multiplying a demodulated data symbol of the past by a convolution sequence of a spreading code sequence and a channel impulse response from a received signal to be demodulated at present, thereby performing each demodulation. Since the size of the matrix in which the convolutional sequence of the user's spreading code sequence and the channel impulse response is arranged as a component can be reduced, the amount of calculation can be reduced, and the past demodulated data symbols can be converted into unknown numbers. It is possible to improve the accuracy of demodulation as compared with the case of estimating.

According to a third aspect of the present invention, there is provided a communication method in a communication apparatus for demodulating a received signal including a fixed pattern signal, wherein the fixed pattern signal is held and demodulated now. Then, the received signal is demodulated by subtracting the component including the held fixed pattern signal from the received signal.

In this case, since the content of the fixed-pattern signal is known without demodulation, if the received signal includes the fixed-pattern signal, the received signal to be demodulated is to be demodulated. By performing demodulation after subtracting the signal component of the fixed pattern from the above, it is possible to improve the demodulation accuracy as compared with a case where the fixed pattern is estimated as an unknown or not considered.

According to a fourth aspect of the present invention, there is provided a communication method in a communication apparatus for demodulating a received signal including a signal of a fixed pattern as described above. In the multiple access method,
Based on a matrix in which convolutional sequences of spreading code sequences of each user and channel impulse responses are arranged as components,
When demodulating each data symbol, the data symbol is demodulated by subtracting the convolution sequence of the held fixed pattern signal and the channel impulse response from the received signal to be demodulated at present.

In this case, the demodulation accuracy can be improved as compared with the case where the fixed pattern is estimated as an unknown number or is not considered, as in the third aspect of the invention.

Further, the invention described in claims 5 to 8 is:
A communication device suitable for the communication method according to claim 1.

[0020]

Embodiments of the present invention will be described below. First, in a CDMA mobile communication system in a direct sequence communication system, a base station apparatus or a mobile station apparatus performs joint detection based on a matrix in which convolutional sequences of spreading code sequences of respective users and channel impulse responses are arranged as components. A first embodiment for performing the demodulation is described.

In the present embodiment, the base station apparatus or the mobile station apparatus divides demodulation targets into a plurality of sets, and demodulates each data symbol in order. Here, assuming that the number N of demodulated symbol sets for each user is 5, a data symbol set (hereinafter, referred to as a “demodulated symbol set”) d ^(k) ,
When a corresponding received signal set e ^(k) is formed,
The demodulated symbol set d ^(k) , received signal set e ^(k) and system matrix A shown in Expression (8) are

[0022]

(Equation 4) The relationship shown in FIG. The base station apparatus or mobile station apparatus, the data symbols d ₃ only to the extracted, demodulated symbol set d for each extract ^(k) and the received signal set e ^(k located at the center among the demodulated symbols set d ^{(k) )} , Demodulation is performed successively while sliding.

In this case, the base station device or the mobile station device
May be held in the data holding unit having a built-in data symbols d _1, d ₂ obtained in the past, for the data symbols d _1, d ₂ obtained in these past, represented by the formula (4), for each user Multiply the convolution sequence b ^(k) of the spreading code sequence c ^(k) and the channel impulse response h ^(k) . An operation result obtained by multiplying the data symbols d ₁ and d ₂ by the convolution sequence b ^(k) indicates components in the received signal e corresponding to the data symbols d ₁ and d ₂ . The base station device or the mobile station device subtracts this calculation result from the received signal e. At this time, equation (9) becomes

[0024]

(Equation 5) And can be transformed. According to this equation (10), the system matrix A has the demodulated symbol set number reduced to the size of N = 3 and the size of each matrix of the demodulated symbol set d ^{(k )} and the received signal set e ^(k). Is also reduced.

The base station apparatus or mobile station apparatus is given by the following equation (10).
Using need only extract data symbols d ₃ of the first position of the demodulated symbol set d ^(k). Then, the base station apparatus or the mobile station apparatus slides the demodulated symbol set d ^(k) and the received signal set e ^(k) , and again relocates the data symbol d ^(k) located at the beginning of the demodulated symbol set d ^(k).
₃ (data symbol d ₄ before slide) is extracted. By repeating such an operation, a demodulated symbol set d
Each data symbol in ^(k) is demodulated.

[0026] In this way, demodulated before the data symbols from the data symbols d ₃ past, by utilizing the fact is known, the system matrix A, the demodulated symbol set d ^(k) and the received signal set e ^(k) Since the size of each matrix can be reduced, the amount of calculation for matrix calculation can be reduced. Further, the data symbols before the first symbol in the demodulated symbol set d ^(k) are also held in the data holding unit when demodulating the data symbol, and are used for demodulating the data symbols in the demodulated symbol set d ^(k) . Therefore, the demodulation accuracy can be improved as compared with the case where the data symbols obtained in the past are estimated as unknowns.

It is to be noted that the number of demodulated symbol sets is other than N = 5.
However, in general, the corresponding past demodulated data
The spreading code sequence c for each user^(K)And channel in
Pulse response h^(K)Convolution series b with^(K)To
By multiplying and subtracting from the received signal e, the system row
Column A, demodulated symbol set d^(K)And the received signal set e
^(K)Since the size of each matrix is reduced,
The amount of computation can be reduced, and at the same time
Data symbol as an unknown,
The accuracy of demodulation can be improved.

Although it depends on the estimation accuracy, the present invention is not limited to the method of extracting one data symbol from the demodulated symbol set d ^(k), but it is also possible to extract a plurality of data symbols at any position. . Further, the order of demodulating each data symbol in order may be from the beginning of the received signal or after.

Also, any demodulation may be required as long as it involves an operation of subtracting a component including a past demodulated signal from a received signal to be demodulated at present, and is not limited to the CDMA demodulation. Further, when the spreading factor Q is set to 1,
Although a single-user detection demodulation method is used, the present embodiment is also applicable to this single-user detection demodulation method.

FIGS. 1 to 3 show examples of the configuration of a receiver as a base station apparatus or a mobile station apparatus when the system matrix is reduced. In the following, the noise n is represented by n = (n ₁ , n ₂ ,..., N _{NQ + W−1} ) ^T (11) R _n and R _d indicate R _n = E (n · n ^T ) (12) R _d = E (d · d ^T ) (13), respectively, and A ^H Denotes the conjugate matrix of A.

FIG. 1 shows an example of the configuration of a receiver using a whitening matched filter (WMF). The receiver 100 shown in FIG. 1 includes a data holding unit 102, an adder 104, and a WMF 106.

The data holding unit 102 stores data symbols dc _{, WMF} , channel impulse responses h ^(k) , and spreading code sequences c ^{(k) of} each user, which have been obtained in the past.
Convolution sequence of the channel impulse response ^{h (k) b (k)} is held with.

[0033] The adder 104 is provided with a data synth obtained in the past.
Bol d_{c, WMF}To each user's spreading code sequence c^(K)When
Channel impulse response h^(K)Convolution with
Column b ^(K)Is subtracted from the received signal e.
Good. Thereby, the matrix on the left side of Expression (10) is obtained.
be able to. The WMF 106 calculates the operation result of the adder 104.
For the fruit, A^H・ R_n ^-1Multiplied by the data symbol d
_{c, WMF}Is derived.

FIG. 2 shows ZF-BLE (Zero Forcing? B).
This is a configuration example of a receiver using locking Linear Equalizer). The receiver 200 shown in FIG.
2. It comprises an adder 204 and a ZF-BLE 206.

The data holding unit 202 stores data symbols _dc , _ZF-BLE , channel impulse response h ^(k) , and spreading code sequence c of each user obtained in the past.
Convolution sequence ^(k) and the channel impulse response ^{h (k) b (k)} is held.

The adder 204 is provided with a data synth obtained in the past.
Bol d_{c, ZF-BLE}To each user's spreading code sequence c
^(K)And channel impulse response h^(K)Tatami with
Containing series b^(K)Is multiplied by the received signal e
Subtract from it. ZF-BLE 206 is a function of adder 204.
For the calculation result, (A^H・ R_n ^-1・ A)^-1・ A^H・ R
_n ^-1Multiplied by the data symbol d_{c, ZF-BLE}Led
Put out. By using ZF-BLE206, FIG.
Is demodulated more than when the WMF 106 shown in FIG.
The accuracy of data symbols can be improved.

FIG. 3 shows an MMSE-BLE (Minimum Mean)
This is a configuration example of a receiver using Square Error (Blocking LinearEqualizer). The receiver 300 shown in FIG.
Data holding unit 302, adder 304, MMSE-BLE
306.

The data holding unit 302 stores data obtained in the past.
Data symbol d_{c, MMSE-BL E}, Channel imp
Lus response h^(K), And the spreading code system of each user
Column c ^(K)And channel impulse response h^(K)When
Convolution series b^(K)Is held.

The adder 304 multiplies the data symbol _{dc, MMSE-BLE} obtained in the past by the convolution sequence b ^(k) of the spreading code sequence c ^(k) of each user and the channel impulse response h ^(k) . The result of the operation is represented by the received signal e
Subtract from MMSE-BLE306 is ZF-BL
E308, Maximum likelihood sequence estimation type equalizer (Whiner estimator)
310 is provided. ZF-BLE 308 is an adder 3
04 with respect to the calculation result of (A ^H · R _n ⁻¹ · A) ⁻¹ ·
A ^H · R _n ^-1 . Maximum likelihood sequence estimation type equalizer 310
Calculates W ₀ = (I + R _d · A ^H · R
^{_{n -1 · A) - 1 ·}} A) multiplied by ^-1, the data symbols d
_{c, MMSE-BLE} is output. MMSE-BLE3
06, the ZF-BLE2 shown in FIG.
It is possible to further improve the accuracy of the data symbol to be demodulated as compared with the case where 06 is used.

Next, CDMA in the direct spread communication system
Mobile communication system, the base station apparatus or mobile station apparatus performs joint detection demodulation based on a matrix in which convolutional sequences of spreading code sequences and channel impulse responses of each user are arranged as components. A second embodiment in which a received signal includes a signal component of a fixed pattern in addition to a signal component of a data symbol will be described.

In this embodiment, the base station device or mobile
The station device is a data holding unit that contains a series of fixed patterns.
And the fixed pattern sequence and channel
Calculate the convolution sequence with the impulse response. Solid
W-1 series m from the beginning of the fixed pattern^(K)To m^(K)= (M₁ ^(K), M₂ ^(K), ..., m_W-1 ^(K))^T····················· (14) These fixed pattern sequences m^(K)And channels
Impulse response h ^(K)Convolution series p^(K)= (P₁ ^(K), P₂ ^(K), ..., p_W-1 ^(K))^T・ ・ ・ ・ ・ (15)

[0042]

(Equation 6) Becomes The convolution sequence p ^(k) indicates a component corresponding to the fixed pattern sequence m ^(k) in the received signal e.

Next, the base station or the mobile station apparatus transmits the fixed pattern sequence m ^(k) and the channel impulse response h
^{(K) is} the convolution sequence of ^{p (k)} subtracted from the received signal e. Here, if the fixed pattern sequence m ^(k) is included after the received signal demodulated at that time, Expression (8) expresses the convolutional sequence p ^(k) from the received signal set e ^{(k).
(K )}

[0044]

(Equation 7) And can be transformed.

On the other hand, when the fixed pattern sequence m ^(k) is included before the received signal demodulated at that time,
Equation (8) is obtained from the received signal set e ^(k) by using the convolution sequence p
^{(K )}

[0046]

(Equation 8) And can be transformed.

According to these equations (17) and (18),
Fixed pattern sequence m related to the data symbol to be demodulated
^When removing the effect of ^(k) , the fixed pattern sequence m
Focusing on the fact that ^(k) is known, the fixed pattern sequence m ^(k) is held in advance, and the channel impulse response h
^(K) convolution series and p ^(k) by subtracting from the received signal e, it is possible to improve the accuracy of the demodulation. Further, since the fixed pattern sequence m ^(k) is a known sequence, it can be subtracted from the received signal even if it appears after that time, as shown in Expression (17).

Note that the demodulation method shown in the second embodiment is not limited to the case of joint detection demodulation in which an object to be demodulated is divided into a plurality of sets and each data symbol is demodulated in order. The present invention is also applicable when demodulating continuous data symbols. Further, as in the first embodiment described above, when the system matrix is reduced using the data symbols obtained in the past, or when the system matrix is not reduced by estimating the data symbols obtained in the past as unknowns, Applicable.

The demodulation may be any demodulation involving an operation of subtracting a signal component of a fixed pattern from a received signal to be demodulated at present, and is not limited to the CDMA demodulation. Further, when the spreading factor Q is set to 1, the single-user detection demodulation method is used.
This single-user detection demodulation method is also applicable.

The configuration of the receiver as a base station apparatus or a mobile station apparatus in the second embodiment is the same as that of the first embodiment shown in FIG.
3 is similar to the configuration example shown in FIG. However, the data holding unit further fixed pattern sequence ^{m (k),} the channel impulse response ^{h (k),} and, these fixed pattern sequence ^{m (k)} and the channel impulse response ^{h (k)} convolution series of ^{p (k)} And the adder subtracts the convolution sequence p ^(k) from the received signal e.

As described above, in the base station apparatus or the mobile station apparatus in the CDMA mobile communication system in the direct spread communication system, a matrix in which the convolutional sequence of the spreading code sequence of each user and the channel impulse response are arranged as components. In the case where the demodulation target is divided into a plurality of sets and the data symbols are demodulated one after another, the data symbols obtained in the past are held. By subtracting the signal multiplied by the convolution sequence with the response from the received signal, the amount of calculation in joint detection demodulation can be reduced compared to the case where the data symbols obtained in the past are calculated as unknowns, and the estimation accuracy is also improved. Can also be improved.

When a fixed pattern signal is transmitted, the data symbol demodulated at that time is
Regardless of whether the fixed pattern is before or after the time series, the estimation accuracy of demodulation can be improved by subtracting the convolution sequence of the fixed pattern and the channel impulse response from the received signal.

[0053]

As described above, according to the present invention, demodulation is performed by holding a past demodulated signal and subtracting a component including the past demodulated signal from a currently received signal to be demodulated. This makes it possible to reduce the amount of calculation at the time of demodulation, and to improve the accuracy of demodulation as compared with the case where a past demodulated signal is estimated as an unknown number.

According to the present invention, a fixed pattern is held as an unknown by holding a signal of the fixed pattern and subtracting the signal component of the fixed pattern from a received signal to be demodulated at present to perform demodulation. Or, the accuracy of demodulation can be improved as compared with the case where no consideration is given.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration example of a receiver when a system matrix is reduced, and is a diagram illustrating an example when a WMF is used.

FIG. 2 is a configuration example of a receiver when a system matrix is reduced, and is a diagram illustrating an example when ZF-BLE is used.

FIG. 3 is a diagram illustrating an example of a configuration of a receiver when a system matrix is reduced, and illustrates an example when MMSE-BLE is used.

[Explanation of symbols]

 100, 200, 300 Receiver 102, 202, 302 Data holding unit 104, 204, 304 Adder 106 WMF 206, 308 ZF-BLE 306 MMSE-BLE 310 Maximum likelihood sequence estimation type equalizer

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Toshiyuki Nikata 2-1-1-1, Nagatacho, Chiyoda-ku, Tokyo Inside NTT DoCoMo, Inc. F-term (reference) 2-11-11, NTT DoCoMo, Inc. 5K022 EE02 EE33

Claims (8)

[Claims] 1. A communication method in a communication device for demodulating a received signal, comprising: holding a past demodulated signal; and extracting a component including the held past demodulated signal from a received signal to be demodulated now. A communication method characterized in that a received signal is demodulated by subtraction. 2. The communication method according to claim 1, wherein in the code division multiple access system in the direct spreading communication system, a matrix in which convolutional sequences of a spreading code sequence of each user and a channel impulse response are arranged as components is used. In the case where the demodulation object is divided into a plurality of sets and each data symbol is demodulated sequentially, the past demodulated data symbols are held, and the received past demodulated data symbols are demodulated from the received signal to be demodulated now. A data symbol demodulated by subtracting a signal obtained by multiplying a convolutional sequence of a spreading code sequence and a channel impulse response by the data symbol of (i). 3. A communication method in a communication device for demodulating a received signal including a signal of a fixed pattern, wherein the signal of the fixed pattern is held, and the signal of the held fixed pattern is converted from a received signal to be demodulated at present. And demodulating the received signal by subtracting a component including the following. 4. The communication method according to claim 3, wherein, in the code division multiple access system in the direct spreading communication system, a matrix in which convolutional sequences of spreading code sequences of respective users and channel impulse responses are arranged as components is used. And demodulating each data symbol by subtracting a convolutional sequence of the held fixed pattern signal and a channel impulse response from a received signal to be currently demodulated, and demodulating the data symbol. Method. 5. A communication device for demodulating a received signal, comprising: holding means for holding a past demodulated signal; and a component including the held past demodulated signal from a received signal to be demodulated. And a demodulating means for demodulating the received signal by subtracting the demodulated signal. 6. The communication device according to claim 5, wherein in the code division multiple access system in the direct spreading communication system, a matrix in which a convolution sequence of a spreading code sequence of each user and a channel impulse response is arranged as a component is used. When the demodulation object is divided into a plurality of sets and each data symbol is demodulated sequentially, the holding unit holds the past demodulated data symbols, and the demodulation unit receives the currently received demodulated signal. From
A communication device, wherein a received signal is demodulated by subtracting a signal obtained by multiplying the held past demodulated data symbol by a convolution sequence of a spread code sequence and a channel impulse response. 7. A communication device for demodulating a received signal including a fixed pattern signal, comprising: holding means for holding the fixed pattern signal; And a demodulating means for demodulating a received signal by subtracting a component including the signal. 8. The communication apparatus according to claim 7, wherein in the code division multiple access system in the direct spreading communication system, a matrix in which convolutional sequences of spreading code sequences of respective users and channel impulse responses are arranged as components is used. When demodulating each data symbol, the demodulation means, from a received signal to be currently demodulated,
A communication apparatus for demodulating a data symbol by subtracting a convolution sequence of the held fixed pattern signal and a channel impulse response.