JP2009147939A - Method of constituting orthogonal codes to be paired in cdma system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of constituting orthogonal codes to be paired in a CDMA system. <P>SOLUTION: The disclosed method includes the steps of: determining the number n<SB>T</SB>of transmission antennas and a spreading rate M at a transmission side in time or frequency domain; constituting a fundamental matrix (the formula shown in the figure) of orthogonal codes on the basis of the determined n<SB>T</SB>and M in such a way that the orthogonal codes achieve perfect spatial diversity; calculating a spreading code matrix S using constituted fundamental matrices A and B in such a way that columns in the spreading code matrix S can be orthogonal with each other; and dividing each of the columns of the spreading code matrix S into M pieces of column vectors of which the number of dimensions is n<SB>T</SB>×1, constituting a matrix S<SB>u</SB>of which the number of dimensions is n<SB>T</SB>×M so that rows of the constituted matrix can be orthogonal with each other, constituting a u<SP>th</SP>column of the spreading code matrix S on the basis of the matrix S<SB>u</SB>, and using each of the columns in the spreading code matrix S as a spreading code for one user. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

 The present invention relates to a method for forming a pair of orthogonal codes in a CDMA system, and in particular, based on a technique of multiple-input multiple-output / orthogonal frequency division multiplexing (MIMO-OFDM), a pair of orthogonal codes in a downlink of a wireless communication system The invention relates to a method of constructing a code and its application in CDMA technology.

   Multiple-input multiple-output (MIMO) technology has become an important breakthrough in the wireless mobile communication technology field. A wireless communication system that employs multiple transmit and receive antennas is commonly referred to as a multiple-input multiple-output (MIMO) system. In a wireless fading environment, the channel capacity of a MIMO system has a direct proportional relationship with the number of transmission / reception antennas. That is, by increasing the number of antennas, the vector utilization efficiency of the wireless communication system can be doubled. Orthogonal frequency division multiplexing (OFDM) technology is a high-efficiency broadband access technology that can effectively suppress frequency selective fading at a lower cost. An orthogonal frequency division multiplexing system employing a plurality of transmission / reception antennas is a so-called multiple-input multiple-output / orthogonal frequency division multiplexing (MIMO-OFDM) system. This system has many advantages of MIMO technology and OFDM technology, and is said to be one of the main physical layer technologies of future high-speed wireless communication systems.

   Conventionally, studies on MIMO-OFDM technology have been mainly focused on single user systems. However, most of the actual systems are multi-user systems. In order to apply MIMO-OFDM technology to an actual system, it is necessary to consider unique problems such as interference and multiuser detection in a multiuser system. Therefore, a simple and effective multiple access technique suitable for the MIMO-OFDM system has been proposed.

   Conventional multiple access techniques include time division multiple access (TDMA), frequency division multiple access (FDMA), and code division multiple access (CDMA). Compared to the other, the CDMA technology has a high vector efficiency, a large system capacity, a high ability to suppress fading and interference, and can realize efficient and flexible user access. CDMA technologies applicable to normal OFDM systems include multi-carrier direct spreading code division multiple access (MC-DS-CDMA), multi-carrier code division multiple access (MC-CDMA), and orthogonal frequency division multiplexing (OFCDM). Including. In CDMA technology based on MIMO-OFDM, each data information is spread in three dimensions of space (antenna), time and frequency (subcarrier), so here the technology is called space-time-frequency spread (Space). -Time-Frequency Spreading (STFS) CDMA technology. The conventional CDMA technology based on OFDM can be directly applied to a multi-user MIMO-OFDM system without any other processing. However, these methods are originally designed for an OFDM system using one transmission antenna, and are not optimized for the features of MIMO-OFDM.

   In the CDMA system, spreading is performed for each user using a corresponding spreading code. On the other hand, in a multi-user system in space, time, and frequency, one column in the coding matrix may be used as a spread code of space, time, and frequency for one user. For example, the spreading factor is 12 as a 12 × 12 coding matrix. That is, a total of 12 users can be supported. One row of elements may be used as a spread code for one user's space, time, and frequency. In this way, the user can be spread in three dimensions of space, time, and frequency. For example, each user may use 3 antennas in space and 4 time slots in time. Alternatively, communication is performed using three antennas and four subcarriers.

   The already proposed coupling Hadamard code can obtain a spatial diversity gain by fully utilizing the features of MIMO-OFDM. When applied to STFS-CDMA, the function of the multi-user system is improved, but the structure of the coupling Hadamard code is limited to that realized by the conventional Walsh Hadamard matrix, so the structure of the codeword And the length is greatly limited.

  An object of the present invention is to provide a method for constructing a pair of orthogonal codes in a CDMA system. By applying this method to an STFS-CDMA system, spatial diversity can be achieved, and the functions of a multi-user MIMO-OFDM system can be effectively enhanced.

を構成するステップｂと、
構成した基本行列ＡとＢを用いて拡散コード行列Ｓの列同士が相互に直交するように拡散コード行列Ｓを計算するステップｃと、
拡散コード行列Ｓの各列のそれぞれを次元数がｎ_Ｔ×１のＭ個の列ベクトルに分割し、構成された行列の行同士が相互に直交するように、次元数がｎ_Ｔ×Ｍの行列Ｓ_ｕを構成し、行列Ｓ_ｕに基づき拡散コード行列Ｓの第ｕ列を構成し、拡散コード行列Ｓにおける各列のそれぞれをシステムにおける１ユーザの拡散コードとして使用するステップとを含むことを特徴とする方法を提供する。 According to one aspect of the present invention, a method for constructing a pair of orthogonal codes in a CDMA system, comprising:
Determining the number n _T of transmitting antennas on the transmitting side and the spreading factor M in the time or frequency domain;
Based on the determined n _T and M, the basic matrix of the orthogonal code is calculated so that the orthogonal code can realize complete spatial diversity.

Comprising step b,
Calculating the spreading code matrix S using the constructed basic matrices A and B so that the columns of the spreading code matrix S are orthogonal to each other;
Each of the columns of the spreading code matrix S is divided into M column vectors having the number of dimensions n _T × 1, and the number of dimensions is n _T × M so that the rows of the constructed matrix are orthogonal to each other. Forming a matrix S _u , forming a u-th column of the spreading code matrix S based on the matrix S _u, and using each of the columns in the spreading code matrix S as a spreading code for one user in the system. A featured method is provided.

 The above and other objects, features, and advantages of the present invention will become more apparent by describing the preferred embodiment of the present invention with reference to the drawings.

  The method of constructing a pair of orthogonal codes according to the present invention can effectively suppress the influence of radio fading. In addition, multi-user interference can be reduced and the functions of all users can be arranged in a balanced manner, and the overall performance of the MIMO-OFDM multi-user system can be enhanced.

  Furthermore, according to the method of the present invention, the utilization rate of frequency resources is greatly improved. Such a pair of orthogonal codes can be directly applied to the conventional CDMA technology. It can also be applied to other CDMA systems such as MC-DS-CDMA MC-CDMA and OFCDM.

 Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the description, details and functions that are not necessary for the present invention are omitted in the description so that the understanding of the present invention is not confused.

 Next, embodiments of the present invention will be described in detail with reference to the drawings.

 FIG. 1 is a schematic diagram of a transmitter and a receiver 100 in the downlink of a multiuser MIMO-OFDM system based on STFS-CDMA.

 As shown in FIG. 1, a user data stream is generated by a data source 110 on the transmission side. The generated data stream is supplied to the encoder and mapper 112. The encoder and mapper 112 performs encoding and modulation on the generated data stream, and outputs the encoded and modulated data stream to the serial / parallel conversion unit 114. The serial / parallel conversion unit 114 converts the serial data stream into a parallel sub data stream. The sub-data stream is supplied to the space / time / frequency spreading unit 116. In the space / time / frequency spreading unit 116, the sub-data stream is spread and superimposed on the space domain, the time domain and the frequency domain by an address code. The spread signal is input to the interleave unit 118 and subjected to interleave processing. The interleaved signal is supplied to an inverse fast Fourier transform (IFFT) unit 120 where an inverse fast Fourier transform is performed so that the frequency domain signal is transformed into the time domain. Next, the signal converted into the time domain is transmitted to the guard interval unit 122 to add a guard interval (GI) to the encoded signal. Next, the radio frequency (RF) unit 124 converts the baseband signal subjected to the above processing into an RF signal. Finally, the RF signals of all users in the system are transmitted via the antenna unit arrays 126a-126g.

 On the receiving side, the antenna arrays 128a to 128h receive signals from the transmitting side. The received signal is sent to the RF demodulator 130. The RF demodulator 130 converts the received RF signal into a baseband signal, and supplies the converted baseband signal to the guard interval unit 132 again. The baseband signal from which the guard interval has been removed is input to the fast Fourier transform unit 134. The fast Fourier transform unit 134 performs fast Fourier transform on the baseband signal, converts the time domain signal to a frequency domain signal, and supplies the converted baseband signal to the deinterleave unit 136. Then, the deinterleave unit 136 performs deinterleave processing on the baseband signal. The deinterleaved data stream is sent to a specific user detector 138. The detector 138 detects a parallel data stream of a specific user in the received data stream, and supplies the detected parallel data stream to the parallel / serial conversion unit 140. The parallel / serial conversion unit 140 converts the user's parallel data stream into a serial data stream, and supplies the converted serial data stream to the decoder and demapping unit 142. The decoder and demapping unit 142 performs decoding and demapping processing on the serial data stream to restore the original signal. Further, the restored original signal is supplied to the data receiving unit 144.

ただし、Ｓ＝［ｓ_１ ｓ_２ ・・・ｓ_Ｕ］は拡散コード行列、Ｕは利用可能なユーザ数、Ｓ_ｕはユーザｕの長さＰの拡散コードを表す。従って、拡散ブロックが収容できる最大のユーザ数もＰである。Ｘ＝［ｘ_１ ｘ_２ ・・・ｘ_Ｕ］^Ｔは利用可能なユーザのデータシンボルベクトル、ｘ_ｕはｕ番目のユーザのシンボル（［・］^Ｔは転置演算子を表す）、

はノイズベクトル、ｎ_Ｔは送信アンテナの数、

（ｋ＝１、２、・・・、Ｐ／ｎ_Ｔ）はｋ番目の時間・周波数スロットにおける次元がｎ_Ｒ×１であるノイズベクトル、ｎ_Ｒは受信アンテナの数を表し、

であって、ｙ_ｋはｋ番目の時間周波数スロットにおける次元がｎ_Ｒ×１の受信信号ベクトルを表す。Ｈは時間周波数ブロックにおける効果が同じなベースバンドチャネル行列であり、次の数式（２）で表すことができる。

ただし、Ｈ_ｋはｋ番目の時間周波数スロットにおけるチャネル行列を表し、ｋ＝１、２、・・・、Ｐ／ｎ_Ｔである。なお、ここで拡散コードの長さはＰ＝ｎ_ＴＭに限定されており、Ｍは時間領域と周波数領域の拡散率である。すなわち、Ｐはｎ_Ｔの倍数となる。ここで、拡散コードの長さはｎ_ＴとＭの二つの値によるものである。つまり、所要の拡散の長さとアンテナ数が時間領域または周波数領域での拡散率を決める。拡散コードの長さをＰに限定することは、従来の直交コード及び従来技術でのカップリング直交コードの長さにかけられた２の累乗の制限を解除するためである。 In the STFS-CDMA system, the input / output relationship of the spreading block can be expressed by the following equation (1).

Here, S = [s ₁ s ₂ ... S _U ] is a spreading code matrix, U is the number of available users, and _Su is a spreading code of the length P of the user u. Therefore, the maximum number of users that can be accommodated by the spreading block is also P. X = [x ₁ x ₂ ... X _U ] ^T is an available user data symbol vector, x _u is the u th user symbol ([•] ^T represents a transpose operator),

Is the noise vector, n _T is the number of transmit antennas,

(K = 1, 2,..., P / n _T ) is a noise vector whose dimension in the k th time / frequency slot is n _R × 1, n _R represents the number of receiving antennas,

Where y _k represents a received signal vector of dimension n _R × 1 in the k th time frequency slot. H is a baseband channel matrix having the same effect in the time-frequency block, and can be expressed by the following equation (2).

Here, H _k represents a channel matrix in the k-th time frequency slot, and k = 1, 2,..., P / n _T. Here, the length of the spreading code is limited to P = n _T M, where M is the spreading factor in the time domain and the frequency domain. In other words, P is a multiple of _{n T.} Here, the length of the spreading code is due to two values of n _T and M. That is, the required spreading length and the number of antennas determine the spreading factor in the time domain or frequency domain. The reason for limiting the length of the spreading code to P is to remove the power-of-two limitation imposed on the length of the conventional orthogonal code and the coupling orthogonal code in the prior art.

 FIG. 2 shows an example 200 of a chip mapping relationship in the space area, time area, and frequency domain of the address code. In the STFS-CDMA system, each data code is not only spread in the time domain and frequency domain (subcarrier), but also in the spatial domain (antenna).

上記の数式において、数字はアドレスコードの時間周波数スロットの指標を示す。行列のｎ行目のチップがｎ番目のアンテナにマッピングされる。それとともに、ｉ列目のチップがｉ（ｉ＝１、２、・・・、Ｐ／ｎ_Ｔ）番目の時間周波数スロットに割り当てられる。 Chip mapping process STFS-CDMA, in order to ensure the realization of space diversity, by associating each subcode spreading codes S _u is divided into P / n _T sub code of the user u 1 transmit antenna Dividing each antenna sub-code sequence orthogonal to each other, and assigning each chip in the sub-code to the time-frequency slot corresponding to the antenna, the diversity in the time domain and the frequency domain is realized as much as possible. And further improving the function of the system. A set of time frequency slots used in all antennas of one spreading code is called a time frequency block. At this time, the time frequency block includes P / n _T time frequency slots. Here, the number of transmission antennas is 4 and the length of the spreading code is 24, that is, n _T = 4 and P = 24 will be described as an example, but the present invention is not limited to this. Other antenna numbers and spreading code lengths may be used. When n _T = 4 and P = 24, it is obtained by P / n _T that 6 chips are included in one time frequency block. These chips can be distributed over two subcarriers and three time slots. The present invention is not limited to this, and other chip distribution methods may be used as long as the product is 6. Therefore, a vector of codewords having a length of 24 can be expressed by the following equation (3).

In the above formula, the number indicates the index of the time frequency slot of the address code. The chip in the nth row of the matrix is mapped to the nth antenna. At the same time, the i-th chip is assigned to the i (i = 1, 2,..., P / n _T ) -th time frequency slot.

 FIG. 2 shows the situation of four transmit antennas (1, 2, 3, 4). The chip distribution of subcodes 210, 212, 214, and 216 assigned to each transmission antenna is shown. According to Equation (3) above, the chip distribution of the subcode 210 assigned to the transmission antenna 1 is 1, 5, 9, 13, 17, and 21, and the chip distribution of the subcode 212 assigned to the transmission antenna 2 is 2, 6, 10, 14, 18 and 22, the chip distribution of the subcode 214 assigned to the transmission antenna 3 is 3, 7, 11, 15, 19, 23 and the subcode 216 assigned to the transmission antenna 4. It can be seen that the chip distribution is 4, 8, 12, 16, 20, and 24 (see FIG. 2).

 The bi-orthogonality of pairs of orthogonal codes according to the present invention is such that different available user address codes are orthogonal. This is also a basic requirement of the conventional CDMA system. On the other hand, codewords transferred to different antennas of each address code are also reflected so as to be orthogonal. In FIG. 2, the subcode 210 is orthogonal to the subcodes 212, 214, and 216.

 Next, a process for generating a pair of orthogonal codes will be described with reference to FIG.

を生成する。ここで、ｎ_Ｔは送信アンテナの数、Ｍは時間または周波数における拡散率を表し、そして

である。直交行列

は、ある程度で対となる直交コードの特性を決めている。最後に生成される拡散コード系列の長さは、この二つの行列のサイズによって決められるので、この二つの行列のサイズを制限しなければ、生成される拡散コード系列の長さは、

という制限条件の影響しか受けない。一方、この二つの行列は、対となる直交コードを構成する基本行列であり、持っている性質は、直交コードも持っている。

という条件は、直交コードが完全な空間ダイバシティの実現を可能にすることを確保することによって、高い性能利得をもたらすものである。コード系列の直交性を保証するために、行列の各要素の絶対値をそれぞれ１／Ｍ、ｎ_Ｔ ^−１とする。続いて、ステップＳ４１４において、

の関数関係によりＭ^２×Ｍ次元の行列

を構成することができる。以下述べるように関数「Ｔｒａｎｓ（・）」を取得してもよい。 FIG. 3 shows a flowchart for constructing a pair of orthogonal codes according to an embodiment of the present invention. As shown in FIG. 3, first, in step S412, an orthogonal matrix is used.

Is generated. Where n _T represents the number of transmit antennas, M represents the spreading factor in time or frequency, and

It is. Orthogonal matrix

Determines the characteristics of a pair of orthogonal codes to some extent. Since the length of the spreading code sequence generated at the end is determined by the sizes of the two matrices, if the size of the two matrices is not limited, the length of the spreading code sequence generated is

It is only affected by the restriction condition. On the other hand, these two matrices are basic matrices constituting a pair of orthogonal codes, and have the orthogonal code as a property they have.

This condition results in a high performance gain by ensuring that the orthogonal code enables full spatial diversity. In order to guarantee the orthogonality of the code sequence, the absolute value of each element of the matrix is set to 1 / M and n _T ⁻¹ , respectively. Subsequently, in step S414,

M ² × M dimensional matrix due to the functional relationship of

Can be configured. The function “Trans (•)” may be acquired as described below.

を構成することができる。行列

が次の数式（４）で与えられる。

新しい行列

が次の数式（５）で構成されてもよい。

これにより関数「Ｔｒａｎｓ（・）」を取得することができる。 Therefore, _mn (n = 1, 2,..., N) is the n-th column of the matrix M, and the matrix

Can be configured. line; queue; procession; parade

Is given by the following equation (4).

New matrix

May be configured by the following mathematical formula (5).

As a result, the function “Trans (·)” can be acquired.

を構成することができる。次の数式（６）で示すように、

（ｎ＝１、２、・・・、Ｎ）は、行列Ｅの｛（ｎ−１）Ｍ＋１｝行目の（ｅ）_{（ｎ−１）Ｍ＋１}〜_ｎＭ行目の（ｅ）_ｎＭから構成される行列である。

同じように、行列Ｆのｎ行目の（ｆ）_ｎも構成することができる。新しい（ＮＭ）×（ＮＭ）行列が次の数式（７）で構成してもよい。

次に、ステップＳ４１６において、ステップＳ４１４で取得した行列

によって行列

の１列目からｎ_ＴＭ列目を抽出して、拡散コード行列

を算出する。ただし、演算子

の演算方式は上記の数式（７）のように示す。 (E) _m is the m-th column (m = 1, 2,..., MN) of the matrix E, and the matrix

Can be configured. As shown in the following formula (6),

(N = 1, 2,..., N) is composed of (e) in the {(n−1) M + 1} row of the matrix E, (e) _{nM in the (n−1) M + 1} to _n M rows. Is a matrix.

Similarly, the (f) _{n in} the nth row of the matrix F can also be configured. A new (NM) × (NM) matrix may be constituted by the following equation (7).

Next, in step S416, the matrix acquired in step S414

Matrix by

To extract the n _T M columns from the first column of the spreading code matrix

Is calculated. However, the operator

The calculation method is expressed as in the above equation (7).

 The above-described transformation can guarantee the following two characteristics of the paired orthogonal codes.

 (1) The matrix columns are orthogonal to each other. That is, the code sequences of different users are orthogonal.

(2) If each column of the matrix is configured by a small matrix of n _T × M, the rows of the small matrix are also orthogonal to each other. That is, when the code sequences are spread to the respective transmission antennas, the antenna code sequences are also orthogonal to each other.

 Such double orthogonality not only suppresses multiple access interference of a multi-user system, but also fully utilizes multi-antenna resources and realizes complete antenna diversity.

、ただし、ｕ＝１、２、・・・、ｎ_ＴＭ。ここで、ｐａｒは１つの列ベクトルを、上から下まで、Ｍ個の小さな列ベクトルに分割することを示す。 Next, in step S418, a first user to be spread is set. That is, u = 1. In step S420, each column of the spreading code matrix S is divided into M column vectors having n _T × 1 dimensions. That means

However, u = 1, 2,..., N _TM . Here, par indicates that one column vector is divided into M small column vectors from top to bottom.

にしてもよい。ただし、

はＮ×１の列ベクトルである。次の数式（８）を示すように、

個の列ベクトルに等分することを示す。

また、

にしてもよい。ただし、ｍ_１、ｍ_２、・・・、ｍ_ＮはＭ×１の列ベクトルである。Ｍのベクトル数式を次の数式（９）のように設定してもよい。

さらに、ベクトル行列

にしてもよい。ここで、ｖ_１、ｖ_２、・・・、ｖ_ＮはＮ×１の列ベクトルである。次の数式（１０）によって計算する。

次に、ステップＳ４２２において、上記の数式（１０）を利用して、

により次元数がｎ_Ｔ×Ｍの行列Ｓ_ｕを構成することができる。 For this reason,

It may be. However,

Is an N × 1 column vector. As shown in the following formula (8),

It shows that it is equally divided into a number of column vectors.

Also,

It may be. Here, m ₁ , m ₂ ,..., M _N are M × 1 column vectors. The vector formula of M may be set as the following formula (9).

In addition, the vector matrix

It may be. Here, v ₁ , v ₂ ,..., V _N are N × 1 column vectors. It calculates with the following numerical formula (10).

Next, in step S422, using the above equation (10),

Thus, the matrix S _u having the number of dimensions n _T × M can be formed.

 Through the above processing, the generated spreading code matrix S has the characteristics of a pair of orthogonal codes. Each column in the generated spreading code matrix S may be used as one spreading code by one user in the system.

を構成する。ここで、Ｓ’は次元数がｎ_ＴＭ×ｎ_ＴＭである対となる直交行列であり、Ｓ’の各列は長さがｎ_ＴＭである対となる直交コードである。このように、拡散しようとするすべてのユーザに拡散コードを構成することができる。 In order to simplify the signal processing on the receiving side, the order of the code words in each column in the spreading code matrix S may be adjusted. For this reason, in step S424, all columns of the matrix S _u are _laid out horizontally as one (n _T M) × 1 column vector s ′ _u by s ′ _u = vec (S _u ). Thereby, a spreading code for the user u is obtained. Next, in step S426, 1 is added to the number on the user side. That is, a spread code configuration for the next user is prepared. In step S428, it is determined whether or not a spreading code has been calculated for all users to be spread. If the spreading codes have not been calculated for all users, the process returns to step S420 and the steps from step S420 to S426 are repeated. If it is determined in step S428 that the spreading codes have been calculated for all users to be spread, the process proceeds to step S430, and a matrix of (n _T M) × (n _T M) dimensions is obtained.

Configure. Here, S ′ is a paired orthogonal matrix whose number of dimensions is n _T M × n _T M, and each column of S ′ is a pair of orthogonal codes whose length is n _T M. In this way, a spreading code can be configured for all users who want to spread.

。なお、

という条件は必ず必要ではない。２以上に設定するのは、マルチアンテナシステムを説明するためにすぎない。しかし、本発明は、シングルアンテナシステムにも適用することができる。ただ、空間的な直交性がないだけである。この場合、従来のコードワードのことと同じである。 A pair of orthogonal codes according to the present invention has a generated address code length n _T M, and

. In addition,

This condition is not always necessary. Setting it to 2 or more is only for explaining the multi-antenna system. However, the present invention can also be applied to a single antenna system. However, there is no spatial orthogonality. In this case, it is the same as the conventional code word.

 In a conventional CDMA system, a Walsh Hadamard code is used, and the length of the code varies within a power of two. That is, the length of the code is 2, 4, 8, 16, or the like. The codeword transfer rate also changes within the range of the code transfer rate that is a power of two. This wastes a lot of band resources in the actual system.

Configuration method of spreading codes according to this embodiment, the transfer rate can not be set to an arbitrary natural number, it can be relaxed limit to an integral multiple of n _T. And _nT is good also as arbitrary natural numbers. Therefore, it is possible to generate integer multiples of the address code of n _T. For example, when n _T = 3 and M = 4, the length of the paired address code is 12. Therefore, the configuration method of the orthogonal code to be paired can eliminate the restriction on the length of the double orthogonal code and increase the frequency utilization efficiency. In addition, no loss of function occurs.

なお、対となる直交コードが拡散率の制限の条件をさらに緩和させるので、任意の素数以外の次元における対となる直交コードを構成することができる。 Further, an excellent feature of the paired orthogonal codes is that the encoding matrix is generated from a general orthogonal matrix such as a matrix of discrete Fourier transform. As a limiting condition, the absolute value of each element is n _T ⁻¹ . For example, when n _T = 2 and M = 3, using a discrete Fourier transform code, a pair of orthogonal codes having a dimension number of 6 × 6 is generated as in the following matrix S ′.

Since the paired orthogonal code further relaxes the condition for limiting the spreading factor, a paired orthogonal code in a dimension other than an arbitrary prime number can be configured.

FIG. 4 shows a flowchart of a detection algorithm in a multiuser MIMO-OFDM system based on STFS-CDMA. Maximum number of users STFS-CDMA system can accommodate differs by a proportional relationship between the number of transmit antennas _{n T} and the receiving antenna number _{n R.} Next, it explains in detail.

（送信アンテナ数が受信アンテナ数より小さいまたは等しい）
送信アンテナ数が受信アンテナ数より小さいまたは等しい場合には、ＳＴＦＳ−ＣＤＭＡの方案では、長さがｎ_ＴＭの対となる直交コードを使うとき、最大ｎ_ＴＭユーザを収容することができる。コードワードがＭ個の時間・周波数スロットを使用する。 1.

(The number of transmit antennas is less than or equal to the number of receive antennas)
If the number of transmit antennas is less than or equal to the number of receive antennas, the STFS-CDMA scheme can accommodate a maximum of n _T M users when using orthogonal codes that are n _T M pairs in length. A codeword uses M time / frequency slots.

で表す）を推定することによってチャネル情報を取得する。ステップＳ５１４において、取得したチャネル情報によって判定統計を計算する。 2. n _T > n _R (the number of transmitting antennas is larger than the number of receiving antennas)
Usually, since the maximum number of users needs to be equal to or less than the number of elements of the vector y of the received signal, in the STFS-CDMA system, the maximum n _R M users can be determined using orthogonal codes whose length is n _T M. Can be supported. In the determination, the user may know only his / her address and may not know the address code of another available user (this is also the actual system situation). In this case, the steps of the detection algorithm are as shown in FIG. First, in step S512, the channel matrix H (

Channel information is obtained by estimating In step S514, determination statistics are calculated based on the acquired channel information.

ただし、

はムーア・ペンローズの逆行列、

はＨ_ｋに対する推定、ｋ＝１、２、・・・、Ｐ／ｎ_Ｔ、

はＸ_ｕの判定統計を表す。 Here, the data symbol may be detected by an algorithm of orthogonal reconstruction combining (ORC). It is shown as the following formula (11).

However,

Is the inverse of Moore Penrose,

Is an estimate for H _k , k = 1, 2,..., P / n _T ,

Represents a decision statistic of _{X u.}

ただし、

はノイズの二乗誤差である。 Note that the data code may be acquired by a minimum mean square error synthesis (MMSEC) algorithm. It is shown as the following formula (12).

However,

Is the square error of noise.

 Next, in step S516, user data is restored by performing a hard decision on the decision statistics. In addition, when soft information is requested | required by the determination algorithm, you may transmit soft information to a decoder similarly.

 A conventional discrete Fourier transform where the paired orthogonal code is shown in green by functionally comparing the paired orthogonal code composed of a discrete Fourier transform code or transformation matrix to a conventional discrete Fourier transform code I found everything better than the code. Moreover, the length of the diffusion has a greater flexibility, and the unconstrained nature of the pair of orthogonal codes is sufficiently reflected according to the present invention. Further, when the spreading length increases in direct proportion to the number of users, the function of the paired orthogonal code does not change.

 The method for constructing a pair of orthogonal codes has been described above using a discrete Fourier transform code or a transformation matrix as an example. The present invention is not limited to this. For example, a pair of orthogonal codes may be configured by other methods such as a HouseHolder matrix.

   In the above, this invention was demonstrated together with the optimal Example. It should be understood by those skilled in the art that various changes, substitutions, and additions can be made without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should not be understood as limited to the above-described embodiments, but should be limited by the appended claims.

1 shows a schematic diagram of a transmitter and a receiver in the downlink of a multi-user MIMO-OFDM system that employs space-time-frequency-spread CDMA (STFS-CDMA). FIG. The chip mapping relationship in the space domain, time domain, and frequency domain of the address code is shown. 3 shows a flow chart for constructing a pair of orthogonal codes according to an embodiment of the present invention. 2 shows a flowchart of a detection algorithm in a multi-user MIMO-OFDM system employing STFS-CDMA.

Claims (8)

A method for constructing a pair of orthogonal codes in a CDMA system, comprising:
Determining the number n _T of transmitting antennas on the transmitting side and the spreading factor M in the time or frequency domain;
Based on the determined n _T and M, the basic matrix of the orthogonal code is calculated so that the orthogonal code can realize complete spatial diversity.

Comprising step b,
Calculating the spreading code matrix S using the constructed basic matrices A and B so that the columns of the spreading code matrix S are orthogonal to each other;
Each column of the spreading code matrix S is divided into M column vectors having a dimension number of n _T × 1, and the number of dimensions is n _T × M so that the rows of the configured matrix are orthogonal to each other. Forming a matrix S _u , forming a u-th column of the spreading code matrix S based on the matrix S _u, and using each of the columns in the spreading code matrix S as a spreading code for one user in the system. A method characterized by. The step of calculating the spreading code matrix S includes a matrix having dimension M ² × M

And the matrix

To extract the n _T M columns from the first column of the spreading code matrix

The method of claim 1, further comprising the step of calculating. The step d extracts one column in the spreading code matrix S, divides the column into n _T column vectors each having a length M from the top to the bottom, and transposes the column vector to form the matrix S _u. The matrix S _u is configured so as to be each row of and a column vector S ′ _u of (n _T M) × 1 is obtained as all columns of the matrix S _u , thereby obtaining a spreading code of the user u. The method according to claim 1. Configure in step b

Each is an orthogonal matrix and said

The method according to claim 1, characterized in that the absolute value of each element in is 1 / M and n _T ^-1 respectively. The method of claim 1, wherein a code length P of the spreading code is n _TM . The method according to claim 5, further comprising the step of dividing the spreading code S _u of the user u into P / n _T subcodes, each subcode corresponding to one transmit antenna.   The method of claim 6, further comprising assigning each chip in the subcode to a corresponding time / frequency slot in the antenna. The method according to claim 1 generated spreading code is characterized in that it is an integral multiple of n _T.

Images