JP2005252637A - Multicarrier cdma transmitter and receiver - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transmitter and receiver which determines appropriate weight for subcarrier combination based on an MMSE (minimum mean square errors) rule without increasing the number of reference signals in an MC-CDMA (multicarrier code division multiple access) system, lowers the ratio of the reference signals and can then utilize the MC-CDMA system in an up link. <P>SOLUTION: The communication system of this invention has a transmitter and a receiver of multicarrier modulation and demodulation. The transmitter disperses MC-CDMA spread codes so as to be transmitted by subcarriers at equal intervals and then allocates the same spread code of adjacent data to an adjacent subcarrier. The receiver collects the dispersed signals, and iterates a pilot signal by utilizing the spread signal of the adjacent data at the time of determining a weight for combination to thereby perform orthogonality reduction compensation of codes at the time of reception by a short pilot signal. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a transmitter and a receiver used for wireless communication.

  As one of the methods to realize high-speed and large-capacity communication in a cellular mobile communication environment where multipath exists, multi-carrier method that enables high-speed transmission with multiple subcarriers and multiple access with multiple users with different codes An MC-CDMA (Multicarrier CDMA) system that combines a Code Division Multiple Access (CDMA) system is attracting attention (see Non-Patent Documents 1 and 2). MC-CDMA is one of multi-carrier communications using a plurality of subcarriers. A communication system in which a spread code is arranged on the frequency axis, spread by multiplying with a subcarrier signal, and multiple users are multiple-accessed by the code. It is. By arranging spreading codes in the frequency axis direction, frequency diversity gain can be obtained at the time of despreading, and it can be used repeatedly at a single frequency in a multi-cell environment, so high frequency utilization efficiency can be obtained. (Refer nonpatent literature 3). When MC-CDMA is used in the downlink, which is a channel from the base station to the mobile terminal, the code orthogonality is hardly deteriorated and good characteristics can be obtained. On the other hand, an uplink, which is a line from the mobile terminal to the base station, is received by the base station after receiving a fluctuation (this is called fading) through different propagation paths for each user. Since fading signals are received at the base station at the same time, if the fading state for each subcarrier is different, the orthogonality of the code is greatly lost during signal synthesis between the subcarriers that perform despreading, causing interference between users. There is a problem in that the characteristic is greatly deteriorated. In addition, when asynchronous communication is performed on the uplink, it is known that adjacent symbol signals overlap due to time timing offset between users, resulting in intersymbol interference, and the characteristics are significantly degraded (Non-Patent Document 3). reference).

  Among these, as a countermeasure for time timing offset for each user in the uplink, there is a post guard interval. In normal MC-CDMA, a guard interval for copying the second half of the symbol is set only in the first half of the symbol, and a post guard interval for copying the first half of the symbol is also set in the second half of the symbol. If the symbol timing offset is within the range, demodulation without intersymbol interference is possible. On the other hand, as a method for reducing the interference between users on the uplink, the complex weight of each subcarrier (this is referred to as a weight in the present invention) is determined according to the MMSE (Minimum Mean Square Errors) standard at the time of reception, and the reverse By combining subcarriers for spreading, it is possible to minimize the influence of orthogonality collapse due to fading and increase the signal power (see Non-Patent Documents 4 and 5).

Ohmori, Y. Yamao, and N. Nakajima, “The future generations ofmobile communications based on broadband access technologies,” IEEE Commun.Mag., Vol.38, no.12, pp.134--142, Dec. 2000. H.Atarashi, S. Abeta, and M. Sawahashi, “Broadband packet wireless accessappropriate for high-speed and high-capacity throughput,” Proc. IEEE VTC2001-Spring, pp.566--570, 2001. R.van Nee and R. Prasad, “OFDM for Wireless Multimedia Communications,” ArtechHouse, 2000. Tsumuraand S. Hara, “Design and performance of quasi-synchronous multi-carrier CDMA system,” Proc. VTC2001-Fall, 2002. Toshiya Kobashi, Chihiro Fujita, Munehiro Ura, Kazuaki Tsukakoshi, Yoshitaka Hara, Yasuhide Kamio, “Indoor Experiments on Multicarrier CDMA on Uplink,” IEICE Tech. RCS2001-252, pp.127-134, Jan, 2002 .

There are several methods for determining the subcarrier synthesis weight of the MMSE standard. The first is a method in which a channel correlation matrix is obtained by grasping a channel state for each user, and an optimum weight is directly obtained and synthesized by inverse matrix calculation. However, this method has a problem that a channel estimation for each user is required after separating a signal received from an antenna by combining waves from a plurality of users into a signal for each user. On the other hand, as a second method, there is a method of sequentially updating weights by using a reference signal. Specifically, MMSE RLS (Recursive
The Least Squares) algorithm and the LMS (Least Mean Squares) algorithm have been proposed. However, in order to update sequentially, it is necessary to update using a certain number of symbols or more, and the reference signal used for the update must be lengthened. The increase in the reference signal has a problem that the ratio of the data signal in the wireless line is reduced.

  The present invention has been made in view of the above-mentioned circumstances, and pays attention to the latter method of obtaining by sequentially updating, and determines an appropriate weight for subcarrier synthesis based on the MMSE norm without increasing the number of reference signals. An object of the present invention is to provide a transmitter and a receiver that can use the MC-CDMA scheme on the uplink while reducing the ratio of the reference signal.

  In order to achieve the above object, the present invention provides a transmitter for performing multi-carrier modulation on a plurality of symbols constituting data in parallel, and transmitting each of the plurality of symbols subjected to multi-carrier modulation in parallel. Creates multiple subsymbols that are duplicates of a symbol and generates a subsymbol group having a plurality of subsymbols, and a plurality of chips for each of a plurality of subsymbol groups transmitted in parallel A plurality of sub-symbol groups that are transmitted in parallel, and a plurality of sub-symbols that are multiplied by the same chip are modulated by successive sub-carriers. A distributor for associating subcarriers and a plurality of subsymbol groups transmitted in parallel include subcarriers included in the plurality of subsymbol groups. Utilizing a transmitter and having a multi-carrier modulation section for modulating with the number of subcarriers in symbol.

  According to the present invention, in the MMSE combining used to cope with the interference between users at the time of subcarrier combining of the MC-CDMA receiver, the provisional weight obtained by another subcarrier and the inverse matrix of the correlation matrix are utilized, By repeatedly using the same reference signal, good characteristics can be obtained with a short reference signal.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following explanation, a receiver receives a spread signal distributed and arranged in subcarriers, generates a weight for subcarrier combining using a reference signal, and combines the subcarriers based on the weight. The case where it performs is illustrated.

  FIG. 1 shows a multi-carrier CDMA transmitter configuration according to the first embodiment of the present invention. The OFDM transmitter of this embodiment is provided in a mobile device of a mobile communication system, and spreads and transmits a multicarrier signal by multiplying it with a spreading code. A multicarrier signal can be transmitted at high speed even when one symbol length is long by transmitting a plurality of subcarriers in parallel. In the present invention, this multicarrier signal is multiplied by a spreading code. As the spreading method, an MC-CDMA signal obtained by multiplying the spreading code in the frequency axis direction of the symbols constituting the subcarrier is assumed. The MC-CDMA signal is transmitted from the antenna. Here, a method of multiplying the spread code in the frequency axis direction is described, but in addition, a method of multiplying the spread code in the time axis direction may be used in combination.

In the communication system using MC-CDMA, the transmitter first outputs a transmission data symbol sequence to the serial-parallel converter 2 and outputs M parallel data symbol sequences to be transmitted in parallel to the data modulator 4. The data modulator 4 modulates each of these parallel data symbol sequences by a predetermined modulation method, and outputs each of these parallel data symbol sequences to the multiplexer 6 as M parallel modulation symbol sequences. As a modulation method performed by this data modulator, for example, there is a quadrature phase shift keying (QPSK). In addition, two-phase phase modulation (BPSK: Binary
Phase Shift Keying) and QAM (Quadrature Amplitude Modulation) can also be used. The multiplexer 6 generates a reference signal whose phase and amplitude generated by the reference signal symbol generator 22 are also known by the receiving base station, and multiplexes each of the reference signals to a plurality of parallel symbol sequences at a constant time symbol interval. A known signal is generated on the receiving side, and each parallel symbol sequence is output to the duplicator 8. Next, each duplicator 8 duplicates these parallel modulation symbol sequences into signals of spreading gain SF, and outputs the duplicated signals to the multiplier 10 respectively. At this time, a group of parallel sub-symbols obtained by replicating the same data for SF is called a spread frame.

  Thereafter, the spreading sequence generator 24 generates spreading codes on the frequency axis and outputs them to the multiplier 10. Multiplier 10 multiplies the inputted spreading code with each inputted parallel data symbol sequence, and outputs it to divider 11 for each spread frame. The spreading codes to be multiplied at this time are the same between spreading frames, and the same chip of the same spreading code sequence is output when the output from the spreading code generator 24 in FIG. 1 is connected. Become.

  The distributor 11 disperses and arranges the input parallel modulation symbols so that spread signals composed of the same data are equally spaced, and outputs them to the IFFT circuit 12 (Inverse Fast Fourier Transform). Each of these parallel modulation symbol sequences arranged in parallel constitutes each subcarrier component of the multicarrier signal. In the present invention, each subcarrier component is called a subsymbol. The flow of parallel modulation symbol sequences distributed at this time can be expressed by mathematical formulas.

As shown. As a result, spread signals from the same chip in different spread frames are arranged in adjacent subcarrier components.

  The IFFT circuit 12 regards the in-phase component of the input parallel modulation symbol sequence as the real part of the complex number and the orthogonal component as the imaginary part of the complex number, converts the parallel complex data into an orthogonal multicarrier CDMA signal by inverse Fourier transform, To the parallel-serial converter 14 as a sample signal. The parallel / serial converter 14 performs parallel / serial conversion on the input parallel signal to convert it into a serial sample symbol sequence, which is output to the multiplexer 18 and the guard interval generator 16. This serial sample symbol sequence is an MC-CDMA signal.

  The guard interval generator 16 generates a part of the second half and a part of the first half of the input symbol to add them to the original sample signal, and the second half of the symbol is before the symbol and the first half of the symbol is the symbol. The data is output to the multiplexer 18 to be added later. The multiplexer multiplexes 18 input serial sample symbol sequences and guard interval signals, converts these sample signals to radio frequencies, and outputs them to the antenna 20. The antenna transmits the input signal to space. Thereafter, the transmitted signal reaches the receiver.

FIG. 2 is a block diagram showing the configuration of the MC-CDMA receiver according to this embodiment. As shown in the figure, the receiver includes an antenna 50. A signal received via the antenna is converted to a baseband frequency and then output to the guard interval removal unit 52. The guard interval removing unit 52 removes the guard interval added before and after the symbol from the input signal, and outputs it to the serial-parallel converter 54 as a serial sample sequence. The serial-parallel converter 54 performs serial-parallel conversion on the input serial sample sequence for each symbol to generate Nc parallel sample signals, and an FFT circuit 56 (Fast Fourier Transform Circuit: Fast Fourier Transform Circuit).
Output to (Transform).

  The FFT circuit 56 outputs Nc subcarrier components in parallel by performing Fourier transform on the input Nc parallel sample signals, and separates them into parallel modulation symbols. The FFT circuit 56 outputs the parallel modulation symbol sequence to the distributor 57. The distributor 57 collects the sub-symbols belonging to the spread frame distributed on the transmission side into M sets of SFs and outputs them to the multiplier 58. The spread code generator 66 generates the same spread code as that used on the transmission side and inputs it to the multiplier 58 to perform despreading. Multipliers 58 each output signals belonging to the same spread frame to MMSE combiner 60.

  Each MMSE combiner 60 multiplies the input parallel modulation symbol sequences by complex weights and combines them to obtain M parallel data symbol sequences from which interference components are adaptively removed. The operation of the MMSE combiner 60 at this time in the present invention will be described. The configuration of the MMSE synthesizer 60 is shown in FIG. Here, each of the SF parallel data symbol sequences input to the MMSE combiner 60 is multiplied by the weight obtained by the weight generator to synthesize a spread frame. The despread output signal of the mth spread frame at this time is shown as follows.

Where the weight matrix is

The received signal matrix is

It becomes.
However,

Represents.

  Here, FIG. 4 shows an example of a packet configuration used in the present invention. This figure shows the symbol arrangement of MC-CDMA generated by FIG. 1, with the vertical axis representing sub-symbols and the horizontal axis representing symbols. Here, sub-symbols in which the data symbol portion is painted in the same pattern represent signals of the same spread frame, indicating that the same data is spread and distributed. In this packet configuration, a reference signal symbol is added to the head of the packet. Different reference signal symbols are assigned in advance for each transmitting terminal. A data signal is transmitted following this reference signal symbol. In addition, in order to reduce the influence of the time timing offset for each user, each symbol has a pre-guard interval (pre-GI) in the first half of the symbol before and after the symbol, and a second half of the symbol in the first half of the symbol. Is inserted a post guard interval (post-GI).

  The MMSE combiner 60 shown in FIG. 3 first outputs the input parallel modulation symbol sequence to the storage device 602. The storage device 602 accumulates the input parallel modulation symbol sequence and outputs the reference signal symbol portion to the first weight generator 608 and the data symbol portion to the multiplier 604 so that the parallel modulation symbol sequence can be taken out at any time. . The first weight generator 608 determines composite weights for these signals using a weight generation algorithm. Here, the composite weight is obtained by an RLS (Recursive Least Squares) algorithm, which is a high-speed and practical amount of calculation. The RLS algorithm is an algorithm that sequentially updates the composite weight using a plurality of reference signal symbols, and uses the composite weight value and the inverse matrix of the correlation matrix used in the previous update. This update formula is shown below. The error signal of the mth spread frame ith update is

At this time, the weight update formula is

It becomes. The inverse matrix is sequentially obtained by the following update formula.

Here, I represents a unit matrix.

  In the present invention, an appropriate weight is obtained using a short reference signal symbol by repeatedly using the same reference signal symbol based on the RLS algorithm. First, in the first stage, each of the M first weight generators 608 in parallel uses the reference signal symbol to update the composite weights sequentially by the RLS algorithm, and reverses the temporary composite weight and the temporary correlation matrix. Find the matrix. These values, when the number of pilot signals is small, are not updated accurately and are not the inverse of the optimal weight and correlation matrix. Good characteristics can be obtained even if they are used as weights for data demodulation as they are. I can't.

  Here, due to the effect of the distributor 11 provided in the transmitter, chips having the same spreading code of adjacent spreading frames are transmitted from the adjacent subcarriers, respectively. Since adjacent subcarriers are close to the fading state, the weight determined by MMSE is expected to take a close value in the adjacent spread frame. Therefore, each first weight generator outputs the obtained temporary combined weight and the inverse matrix of the temporary correlation matrix to the first weight generator 608 of the MMSE combiner 60 of the spread frame existing in parallel. To do. On the other hand, the first weight generator 608 of FIG. 3 includes an inverse matrix of a temporary weight of another spread frame and a temporary correlation matrix generated by the first weight generator 608 of the MMSE combiner 60 existing in parallel. A predetermined interpolation operation is performed on these, and a new inverse matrix of weight and correlation matrix is obtained. In the present invention, the multistage update is performed by using the weight and the inverse matrix value of the correlation matrix generated from the separate weight generator obtained by the interpolation operation as the initial value of the weight and the inverse matrix of the correlation matrix in the spread frame. I do. Here, using the initial value generated in the previous stage, the same reference signal symbol portion as that used earlier is read again from the storage device 602, and the weight is updated by the RLS algorithm as in the previous stage.

  Various interpolation methods such as averaging between adjacent diffusion frames and spline interpolation can be used for the interpolation formula at this time. As an example, the initial value derivation formula for the p stage and m spread frames when averaging between adjacent spread frames is shown below.

Using these obtained initial values, the first weight generator 608 calls the reference signal symbol from the storage device 602 again, and updates the weight by the RLS method using the same reference signal symbol as the previous stage. The first weight generator 608 can generate an appropriate weight efficiently with a short reference signal symbol by repeating the derivation of these initial values and the weight update by the RLS method a prescribed number of times. The weights determined by these procedures are each output from the first weight generator 608 and output to the multiplier 604 for each sub-symbol. The multiplier multiplies the sub-symbol of the data symbol input from the storage device 602 and the weight determined by the first weight generator 608 and outputs the result to the adder 606. Adder 606 adds the input sub-symbols and outputs the result to data demodulator 62.

  FIG. 5 shows an operation image of the first weight generator 608 used in the present invention. Here, attention is paid to the processing for the (m−1) th to (m + 1) th spreading frames, but since the same operation is performed for each spreading frame as a whole, the RLS algorithm is operated in parallel for all spreading frames. Become. This figure shows that the initial value of the next stage is obtained by averaging the weights obtained in adjacent spread frames and the inverse matrix of the correlation matrix. However, as a method for deriving the initial value, it is possible to use other complementation methods besides averaging. Also, since the end spread frame requires a temporary weight and an inverse matrix of the correlation matrix only on one side, it can be handled by setting the initial value to the value temporarily determined on the adjacent side. Interpolation can also be used. The following example shows the case of averaging.

  First, in the first stage, a reference signal symbol known at the reception side sent at the head of each spread frame is extracted from the storage device 602, and the initial value of the weight and the initial value of the inverse of the correlation matrix are predetermined values. Then, the weight obtained by updating the inverse matrix of the weight and the correlation matrix by the RLS algorithm is set as the temporary weight, and the inverse matrix of the correlation matrix is set as the inverse matrix of the temporary correlation matrix. Thus, inverse matrices of the temporary weight and the temporary correlation matrix are obtained for the spread frames # 1 to #M, respectively.

  Next, in the second stage, the inverse of the temporary weight and the temporary correlation matrix previously obtained in each spread frame is averaged between adjacent spread frames, and the initial value of the inverse matrix of the weight and the correlation matrix in the second stage is obtained. . An equation for averaging for generating the initial value is shown in Equation 9. The first weight generator 608 again reads the same reference signal symbol stored from the storage device 602 and updates the inverse of the weight and the correlation matrix by the RLS algorithm in the same manner as in the first stage. The inverse matrix of the weight and the correlation matrix obtained in step 1 is set as a temporary weight and the inverse matrix of the temporary correlation matrix.

  Thereafter, the same operation as in the second stage is repeated until the P-th stage, and the finally obtained weight and the inverse matrix of the correlation matrix are used as weights for synthesizing between the sub-symbols. However, the value of P at this time is assumed to be predetermined.

  In this way, by averaging the temporary weights and inverse matrixes of the temporary correlation matrix obtained in the adjacent spread frame to obtain the initial value, the influence of the same noise caused by repeatedly using the same reference signal symbol is eliminated. It is possible to suppress other users' signals. It becomes possible to realize the MMSE combiner 60 capable of receiving a desired signal by using a short reference signal symbol. These weight generation procedures are summarized as a flowchart in FIG.

  The MMSE synthesizer 60 performs sub-symbol synthesis on the input parallel modulation symbol sequence using the synthesis weights obtained by these operations, and outputs the data symbol sequence to the data demodulator 62 in FIG. The data demodulator 62 demodulates each input parallel modulation symbol sequence and outputs the parallel data symbol sequence to the parallel / serial converter 64. The parallel-serial converter 64 converts the input parallel data symbol sequence into a serial data symbol sequence, and generates a data sequence.

  As described above, according to the first embodiment, when a transmitter places a spread frame in a sub-symbol corresponding to a subcarrier, the adjacent sub-symbol with high correlation is multiplied by the spread code of the same chip in the adjacent spread frame. Arrange the sub-symbols to be arranged. Further, when sub-symbol synthesis is performed on the signal sent from the transmitter based on the MMSE-based RLS algorithm, the receiver averages between the adjacent spread frames with respect to the inverse matrix of the weight and the correlation matrix that are obtained. By applying the interpolation formula, the initial value in the next update is generated. By updating the weight again with the initial value, a good sub-symbol synthesis weight is generated by repeatedly using a short reference signal symbol. With this method, it is possible to obtain good transmission characteristics while reducing the proportion of known symbols for the reference signal provided in the packet, and it is possible to reduce the occupied time by the reference signal and improve the frequency utilization efficiency To do.

  By the operation of the first embodiment, the sub-symbol can be synthesized with the semi-optimal weight. This receiver determines the weight with a short reference signal symbol by using the fact that the correlation between adjacent subcarriers is high with respect to the fading state between subcarriers. As the amount of delay increases, the influence of frequency selective fading that the correlation between subcarriers becomes smaller increases, and the improvement in characteristics decreases.

  Accordingly, a configuration is added in which the receiver can update the sub-symbol synthesis weight again by using the entire packet as a reference signal using the data sequence that is temporarily demodulated by the operation of the first embodiment, thereby enabling better weight generation. You can also. The basic configuration is the same as that used in the first embodiment, but in this embodiment, the data symbols that have been temporarily demodulated and determined using the sub-symbol synthesis weights determined by the receiver of the first embodiment. Is regarded as a reference signal to further improve the reception characteristics.

  FIG. 7 shows an MMSE synthesizer 60 of the receiver according to the second embodiment. In the second embodiment, the weight generation has a two-stage configuration, and includes a first weight generator 608 and a second weight generator 612. Here, the first weight generator 608 is the same as the first weight generator 608 of the first embodiment. Here, for the generation of the weight of each sub-symbol, first, as in the first embodiment, the first weight generator 608 obtains the weight of the RLS algorithm of the MMSE standard and the inverse matrix of the correlation matrix in a multistage configuration. In the second embodiment, the obtained weights are output to the multiplier 604, which multiplies each sub-symbol component of the data symbol input from the storage device 602 and the spreading code input from the first weight generator 608, The data is output to the adder 606. Adder 606 adds the input signals and outputs the result to provisional demodulator 610. Temporary demodulator 610 performs provisional determination of the data portion of the spread frame in parallel and outputs the result to second weight generator 612.

  Next, the receiver outputs the reference signal symbol portion and the data symbol portion of the received signal at each antenna accumulated in the storage device to the second weight generator 612. The second weight generator 612 updates the weight of the RLS algorithm using the entire packet using the provisionally determined data series and the known reference signal symbol input from the temporary demodulator 610 as the reference signal of the spread frame.

  An image of this operation is shown in FIG. Here, the weight of the provisionally determined data is updated by the RLS algorithm from the back to the front of the packet. However, since the first half reference signal portion can be used as a reference signal that does not include a data determination error, this data is used. Thus, an example of weight update from the second half of the packet is adopted. However, the direction of weight update is not limited to this method. In addition, the initial value of the inverse matrix of the weight and the correlation matrix can be used as the initial value as a temporary determination value, or may be another value.

  As a result of this operation, a sub-symbol synthesis weight by the RLS algorithm and an inverse matrix of the correlation matrix are obtained using the entire frame. The obtained weight is output to the multiplier 614. Multiplier 614 multiplies each sub-symbol component of the data symbol input from storage device 602 and the spreading code input from second weight generator 608, and outputs the result to adder 616. Adder 616 adds the input signals and outputs the result to provisional demodulator 618. Temporary demodulator 618 performs provisional determination of the data portion of the spread frame in parallel and outputs the result to second weight generator 612. In this way, by inputting the provisional determination result again to the second weight generator 612, the receiver can repeat updating the weight again, and can generate a better weight. The number of repetitions here can be chosen arbitrarily. The weights determined by these procedures are each output from the weight generator 612 and output to the multiplier 614 for each sub-symbol. The multiplier multiplies the sub-symbol of the data symbol input from the storage device 602 and the weight determined by the second weight generator 612 and outputs the result to the adder 616. Adder 616 adds the input sub-symbols and outputs the result to data demodulator 62. FIG. 9 shows a flowchart when the provisional determination result is used for weight generation.

  As described above, according to the second embodiment, even in a frequency selective fading environment where fading correlation between sub-symbols is low, a short reference signal symbol is repeatedly used and a temporarily determined data symbol is used as a reference signal symbol. And updating the weight again, it is possible to obtain a better characteristic than the first embodiment while reducing the ratio of the known symbol for the reference signal provided in the frame.

Modified example

  The above-described embodiment shows one aspect of the present invention, and can be arbitrarily changed within the scope of the technical idea of the present invention. Accordingly, various modifications will be described below.

(1) In each of the embodiments described above, it is assumed that the transmitter is used for a mobile communication terminal and the receiver is used for a mobile communication base station. However, in this embodiment, the terminal and the base station can be interchanged. In particular, when changing the phase and amplitude of a signal output from between transmission antennas by a diversity transmission or a transmission adaptive array antenna from a base station having an array configuration having a plurality of antenna elements in the base station, the base station Similar to communication to the station, different fading is applied to each user, so that the orthogonality between MC-CDMA users is lost. Under such circumstances, the effect of using the transmitter and receiver used in the present invention is enhanced.

(2) Furthermore, since the mobile station and the base station have the transmitter and receiver of the present invention, the present invention can be used for both the uplink and the downlink.

(3) Further, in (1), a modification example in which a base station is used as a transmitter and a terminal is used as a receiver is shown. However, the present invention can also be used when directly communicating between terminals. At this time, the transmitter and receiver of the present invention are switched according to the transmission and reception timings.

It is the figure which showed the transmitter structure. It is the figure which showed the receiver structure. It is the figure which showed the MMSE synthesizer structure. It is the figure which showed the packet structure. It is the figure which showed the weight update image. It is the figure which showed the MMSE synthetic | combination part operation | movement flowchart. It is the figure which showed the MMSE synthesizer structure with a temporary determination part. It is the figure which showed the weight update image by the RLS algorithm using a temporary determination symbol. It is the figure which showed the MMSE synthesizer operation | movement flowchart with a temporary determination part.

Explanation of symbols

2 serial-parallel converter 4 data modulator 6 multiplexer 8 replicator 10 multiplier 11 divider 12 IFFT circuit 14 parallel-serial converter 16 guard interval generator 18 multiplexer 20 antenna 22 reference symbol generator 24 spreading code generator 50 Antenna 52 Guard interval remover 54 Serial-parallel converter 56 FFT circuit 57 Distributor 58 Multiplier 60 MMSE combiner 62 Data demodulator 64 Parallel-serial converter 66 Spreading code generator 602 Storage device 604 Multiplier 606 Adder 608 First Weight generator 610 Temporary demodulator 612 Second weight generator 614 Multiplier 616 Adder 618 Temporary demodulator

Claims (6)

A transmitter that performs multi-carrier modulation in parallel on a plurality of symbols constituting data and transmits the data,
For each of a plurality of symbols that are multi-carrier modulated in parallel, a plurality of sub-symbols that are duplicates of the symbols, and a replica that generates a sub-symbol group having a plurality of sub-symbols;
A spreading code generator for multiplying each of a plurality of subsymbol groups transmitted in parallel by the same spreading sequence composed of a plurality of chips;
For a plurality of sub-symbol groups transmitted in parallel, a distributor that associates sub-symbols with sub-carriers so that a plurality of sub-symbols multiplied by the same chip are modulated by successive sub-carriers,
A transmitter comprising: a multicarrier modulation unit that modulates a plurality of subsymbol groups transmitted in parallel using subcarriers corresponding to the number of subsymbols included in the plurality of subsymbol groups. The transmitter of claim 1, wherein
A transmitter characterized in that a plurality of symbols constituting data includes a plurality of reference signal symbols. A communication system using multi-carrier modulation / demodulation,
A transmitter that multi-carrier modulates and transmits a plurality of symbols constituting data;
A receiver that receives and demodulates a plurality of multi-carrier modulated symbols from the transmitter;
The transmitter is
For each of a plurality of symbols that are multi-carrier modulated in parallel, a duplicator that generates a plurality of sub-symbols that are duplicates of the symbol and creates a sub-symbol group having a plurality of sub-symbols;
For each of a plurality of sub-symbol groups transmitted in parallel, a spreading code generator for multiplying the same spreading sequence composed of a plurality of chips;
For a plurality of sub-symbol groups transmitted in parallel, a distributor that associates sub-symbols with sub-carriers so that a plurality of sub-symbols multiplied by the same chip are modulated by successive sub-carriers,
A multi-carrier modulation section that modulates a plurality of sub-symbol groups transmitted in parallel using as many sub-carriers as the number of sub-symbols included in the plurality of sub-symbol groups;
The receiver
A multicarrier demodulator that multicarrier demodulates a received signal received from the transmitter;
For a plurality of sub-symbol groups demodulated by multicarrier, a multiplier that multiplies the spreading sequence for each sub-symbol group;
A plurality of combiners for combining a plurality of subsymbols constituting the subsymbol group for each of the plurality of subsymbol groups multiplied by the spreading sequence;
A data demodulator that demodulates a signal based on outputs from the plurality of combiners and outputs a plurality of symbols;
Each of the plurality of combiners combines each of a plurality of sub-symbols to be combined by multiplying a weight value,
The weight value multiplied by each of the plurality of sub-symbols constituting the sub-symbol group is
The reference symbols included in the plurality of symbols are transmitted from the transmitter for weight value calculation, and the reference symbols are used to calculate weight values for the plurality of subsymbols constituting another subsymbol group. A communication system that is calculated from the calculated intermediate weight value. A communication system according to claim 3,
Each of the multiple synthesizers of the receiver includes a storage device for storing the input signal,
The weight value multiplied by each of the plurality of sub-symbols constituting the sub-symbol group is
To calculate weight values for a plurality of sub-symbols constituting a reference symbol and other sub-symbol groups, which are included in a plurality of symbols and transmitted from the transmitter for weight value calculation and output from the storage device Calculated from the intermediate weight value calculated using the reference symbol output from the storage device,
A communication system, wherein the calculation is repeated a plurality of times by repeatedly outputting an input signal stored from a storage device. A communication system according to claim 3,
A communication system, wherein a weight value multiplied by each of a plurality of sub-symbols constituting a sub-symbol group is further calculated using a symbol demodulated previously by the receiver. The communication system according to claim 4, wherein
A communication system, wherein a weight value multiplied by each of a plurality of sub-symbols constituting a sub-symbol group is further calculated using a symbol demodulated previously by the receiver.