Method for transmitting data in multiple input multiple output wireless communication system

Abstract

A method for transmitting data in a multiple input multiple output (MIMO) wireless communication system is disclosed. The method for transmitting a signal to a receiver by a transmitter in a multiple input multiple output (MIMO) wireless communication system includes: generating a bit stream having the size of specific bits through channel coding of data; dividing the bit stream into a first bit stream having a first bit size and a second bit stream having a second bit size; allocating the second bit stream having the second bit size to an antenna sequence codeword on the basis of a signal transmission time; and transmitting the first bit stream having the first bit size to the receiver according to an order of antenna pairs indicated by the allocated antenna sequence codeword.

Claims

1. A method for transmitting a signal to a receiver by a transmitter in a multiple input multiple output (MIMO) wireless communication system, comprising: generating a bit stream having a size of specific bits through channel coding of data; dividing the bit stream into a first bit stream having a first bit size and a second bit stream having a second bit size; allocating the second bit stream having the second bit size to an antenna sequence codeword selected from a plurality of antenna sequence codewords based on a signal transmission time, wherein the antenna sequence codeword comprises indexes of a plurality pair of antenna ports not overlapped with each other and indexes of each pair of antenna ports indicates one bit of the second bit stream; and transmitting the first bit stream having the first bit size to the receiver using the plurality of pair of antenna ports according to an order of antenna pairs indicated by the allocated antenna sequence codeword.

2. The method according to claim 1, wherein the antenna sequence codeword is defined by 2 timeslots and 2 antenna indexes.

3. The method according to claim 1, wherein the transmitting the first bit stream having the first bit size includes: transmitting the first bit stream having the first bit size using QO (quasi-orthogonal)-STBC (space time block code), according to the order of antenna pairs indicated by the allocated antenna sequence codeword, on a timeslot basis.

4. The method according to claim 3, wherein the allocated antenna sequence codeword indicates a pair of antennas needed to transmit data during 2 timeslots.

5. The method according to claim 1, wherein the sum of the first bit size and the second bit size is identical to the specific bit size.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principle of the invention. In the drawings:

(2) FIG. 1 is a conceptual diagram illustrating the SM scheme.

(3) FIG. 2 is a conceptual diagram illustrating the STBC-SM scheme.

(4) FIG. 3 is a conceptual diagram illustrating a method for generating an antenna index sequence according to an embodiment of the present invention.

(5) FIG. 4 illustrates the relationship between the number of antennas and the maximum amount of information when the antenna index sequence is used according to an embodiment of the present invention.

(6) FIG. 5 is a block diagram illustrating a transmitter according to an embodiment of the present invention.

(7) FIG. 6 is a block diagram illustrating a receiver according to an embodiment of the present invention.

(8) FIG. 7 is a block diagram illustrating a receiver according to another embodiment of the present invention.

DESCRIPTION OF SPECIFIC EMBODIMENTS

(9) In order to increase the frequency efficiency per cell, a large amount of information must be allocated to each Tx symbol, such that there is a need to transmit the symbol having a high modulation level. However, when the high-modulation-level symbol is transmitted, the minimum distance between the above symbols over the signal constellation is gradually reduced, such that a BER performance may be deteriorated at the same SNR. The closed loop MIMO system in which a base station (BS) can utilize channel information between the transmitter and the receiver increases a valid reception (Rx) SNR using the MIMO beamforming scheme, such that it can directly increase the Achievable Rate (AR).

(10) However, the Rx SNR gain (array gain) of the open loop MIMO system is determined by the number of Rx antennas, such that the Rx SNR gain (array gain) is decided by the number of Rx antennas. As a result, although many antennas of the BS are present, it may be difficult to directly increase the transfer rate in so far as Tx power is not amplified. When the open loop MIMO system transmits the same amount of data, the present invention does not improve the transfer rate due to the increased Rx SNR, but increases the BER performance when the same amount of data is transmitted in the open loop MIMO system, as compared to the legacy open loop MIMO data Tx/Rx methods.

(11) The present invention is characterized in that the GBD-QOSTBC codeword matrix is modified into another within only the limitation condition in which a diversity gain is maintained, from the standpoint of the BER performance of the QO-STBC scheme. Such codeword modification may be identical to that of the method for changing an antenna to be used according to a data transmission time, and a sequence composed of the Tx antenna index is defined, such that the sequence may be used to transmit data. As a result, the total amount of data to be transmitted may be divided into the Tx symbol and the antenna index sequence and then transmitted. The above-mentioned data transmission scheme of the present invention will hereinafter be referred to as STBC-SSC (Space Time Block Coded Spatial Sequence Coding). The codebook according to the embodiment of the present invention will hereinafter be described.

(12) A) Codebook Definition

(13) (1) .sub.QAM,QOSTBC: QO-STBC Symbol Vector Codebook (Comprised of M-PSK and M-QAM Symbols)

(14) The QO-STBC symbol vector codebook is used to construct the (M.sub.TM.sub.T) codeword matrix based on the legacy QO-STBC scheme. In accordance with the proposed scheme, the QO-STBC symbol vector codebook may be used for GBD-QOSTBC symbol transformation (or conversion). The QO-STBC codeword may be acquired by extending the 22 Alamouti codeword (Orthogonal STBC) to the (M.sub.TM.sub.T) matrix based on the ABBA codeword in association with M.sub.T=2.sup.2,2.sup.3, . . . ,2.sup.r. Accordingly, the QO-STBC codeword matrix may have half-orthogonal characteristics. That is, each row vector (or each column vector) constructing the QO-STBC codeword matrix may be perpendicular to M.sub.T/2 different row vectors (column vectors). In order to easily the above-mentioned characteristics, C.sub.4.sup.HC.sub.4 may be represented by the following equation 12 using the QO-STBC codeword shown in Equation 3.

(15) $\begin{array}{cc}{C}_4^H{C}_4=\left[\begin{array}{cccc}c& 0& d& 0\\ 0& c& 0& d\\ d& 0& c& 0\\ 0& d& 0& c\end{array}\right]& \left[\mathrm{Equation}12\right]\end{array}$

(16) In Equation 12,

(17) $c={.Math.s_1.Math.}^2+{.Math.s_2.Math.}^2+{.Math.{\tilde{s}}_3.Math.}^2+{.Math.{\tilde{s}}_4.Math.}^2$
and d=s.sub.1{tilde over (s)}.sub.3+{tilde over (s)}.sub.3s*.sub.1s.sub.2{tilde over (s)}*.sub.4{tilde over (s)}.sub.4s*.sub.2 may be used. In addition, it can be recognized that the pair of symbols to be joint-ML decoded is composed of (s.sub.1,s.sub.3),(s.sub.2,s.sub.4) as can be seen from Equation 12. If the scope of the present invention is extended for a random value (M.sub.T), each of two pairs composed of M.sub.T/2 symbols needs to be joint-ML decoded.

(18) In conclusion, the .sub.QAM,QOSTBC codebook may be comprised of the codeword vector, the size of which is M.sub.T/21. Elements of the vector may be symbols, for example, M-PSK, M-QAM, etc. The elements of the vector are composed of a total of M.sup.M.sup.T.sup./2 vectors, as represented by the following equation 13.

(19) 0$\left[\mathrm{Equation}13\right]$$\begin{array}{c}{}_{\mathrm{QAM},\mathrm{QOSTBC}}=\left\{\left[\begin{array}{c}{s}_1\left[1\right]\\ \mathrm{.Math.}\\ s_{M_T/2}\left[1\right]\end{array}\right],\left[\begin{array}{c}{s}_1\left[2\right]\\ \mathrm{.Math.}\\ s_{M_T/2}\left[2\right]\end{array}\right],\mathrm{.Math.},\left[\begin{array}{c}{s}_1\left[M^{M_T/2}\right]\\ \mathrm{.Math.}\\ s_{M_T/2}\left[M^{M_T/2}\right]\end{array}\right]\right\}\\ =\left\{s\left[1\right],s\left[2\right],\mathrm{.Math.},s\left[M^{M_T/2}\right]\right\}\end{array}$

(20) In Equation 13,

(21) $s_n\left[k\right]$
may be used, where k{1,2, . . . M.sup.M.sup.T.sup./2},n{1,2, . . . ,M.sub.T/2}.

(22) (2) .sub.G-STBC: GBD-QOSTBC Symbol Vector Codebook

(23) By means of the above equation 4, the codeword vectors of .sub.QAM,QOSTBC defined as the QAM or PSK symbol may be converted into the GBD-QOSTBC symbols as shown in the following equation 14.

(24) $\begin{array}{cc}\begin{array}{cc}\begin{array}{c}{}_{G-\mathrm{STBC}}=\left\{\mathrm{TDs}\left[1\right],\mathrm{TDs}\left[2\right],\mathrm{.Math.},\mathrm{TDs}\left[M^{M_T/2}\right]\right\},\\ \mathrm{where}s\left[k\right]{}_{\mathrm{QAM},\mathrm{QOSTBC}}\\ =\left\{S\left[1\right],S\left[2\right],\mathrm{.Math.},S\left[M^{M_T/2}\right]\right\}\\ =\left\{\left[\begin{array}{c}{S}_1\left[1\right]\\ \mathrm{.Math.}\\ S_{M_T/2}\left[1\right]\end{array}\right],\left[\begin{array}{c}{S}_1\left[2\right]\\ \mathrm{.Math.}\\ S_{M_T/2}\left[2\right]\end{array}\right],\mathrm{.Math.},\left[\begin{array}{c}{S}_1\left[M^{M_T/2}\right]\\ \mathrm{.Math.}\\ S_{M_T/2}\left[M^{M_T/2}\right]\end{array}\right]\right\}\end{array}& \end{array}& \left[\mathrm{Equation}14\right]\end{array}$

(25) Equation 14 may be understood as one-to-one mapping (1:1 mapping) between the codeword vectors as shown in the following equation 15.

(26) $\begin{array}{cc}\begin{array}{c}s\left[1\right].fwdarw.S\left[1\right]\\ s\left[2\right].fwdarw.S\left[2\right]\\ \mathrm{.Math.}\\ s\left[M^{M_T/2}\right].fwdarw.S\left[M^{M_T/2}\right]\end{array}& \left[\mathrm{Equation}15\right]\end{array}$

(27) Referring to Equations 14 and 15, since mapping is achieved on a vector basis, the codeword vector of .sub.G-STBC located nearest to the Rx signal vector is decided, and the symbol vector composed of the QAM symbol of .sub.QAM,QOSTBC is detected through the inverse operation. Symbols used in Equation 14 are summarized as follows. S[k]: M.sub.T/21-sized codeword vector S.sub.j[k]: elements of the vector S[k], j{1,2, . . . M.sub.T/2} S.sub.j[k]

(28) (3) .sub.Ant: Antenna Index Sequence Codebook

(29) 2 consecutive timeslots may be denoted by one unit as represented by t=(1,2), (3,4), . . . , (M.sub.T1,M.sub.T) and an antenna index to be used in response to the transmission (Tx) time may be defined. That is, two antenna indexes to be used in 2 consecutive timeslots may be defined. Two antennas may be combined into one pair so that the two antennas may serve as one symbol constructing the antenna index sequence. Therefore, M.sub.T/ 2 antenna pairs may construct one sequence, and the set of different antenna index sequences is denoted by .sub.Ant. The antenna sequence codebook .sub.Ant may be represented by the following equation 16.

(30) $\begin{array}{cc}{}_{\mathrm{Ant}}=\left\{I_j,u_j,\mathrm{where}j\left\{1,2,\mathrm{.Math.}{\mathrm{m2}}^{B_{\mathrm{SSC}}}\right\}\right\}& \left[\mathrm{Equation}16\right]\end{array}$

(31) In Equation 16, B.sub.SSC is the amount of information allocated to the antenna index sequence and then transmitted, and is represented on a bit basis. In addition, I.sub.j is the j-th antenna index sequence, and u.sub.j is a bit sequence corresponding to I.sub.j. In more detail, I.sub.j and u.sub.j may be represented by the following equation 17.

(32) $\begin{array}{cc}{I}_j=\left(l_1,l_2\right),\left(l_3,I_4\right),\mathrm{.Math.},\left(l_{M_T-1},l_{M_T}\right),& \left[\mathrm{Equation}17\right]\\ u_j=\left[u_1,u_2,u_3,\mathrm{.Math.},u_{B_{\mathrm{SSC}}}\right]& \\ l_i{l}_j,i,& \\ j\left\{1,2,\mathrm{.Math.},2^{B_{\mathrm{SSC}}}\right\},& \\ u_b\left\{0,1\right\},& \\ b\left\{1,\mathrm{.Math.},B_{\mathrm{SSC}}\right\}& \\ u_i{u}_j,& \\ i,j\left\{1,2,\mathrm{.Math.},2^{B_{\mathrm{SSC}}}\right\}.& \end{array}$

(33) (4) .sub.H: The Set of Effective Channel Matrices

(34) .sub.H is the set of (M.sub.TM.sub.T)-sized effective channel matrices corresponding to the antenna index sequence defined in the codebook .sub.Ant. This information is owned by only the open loop MIMO system, and may be represented by the following equation 18.

(35) $\begin{array}{cc}{}_H& \left[\mathrm{Equation}18\right]\end{array}$

(36) In Equation 18, B.sub.SSC is the amount of information allocated to the antenna index sequence in the same manner as described above. The matrix may be decided by I.sub.j of .sub.Ant. Assuming that data is transmitted using the j-th antenna sequence and the pair of Tx antennas used in 2 timeslots (t.sub.0, t.sub.0+1) is denoted by (m.sub.1,m.sub.2), four elements of the effective channel corresponding to (m.sub.1,m.sub.2) may be represented by the following expression.

(37) [Expression]
(m.sub.1,t.sub.0)=h(m.sub.1)
(m.sub.1,t.sub.0+1)=h(m.sub.2)
(m.sub.2,t.sub.0)=h(m.sub.1)*
(m.sub.2,t.sub.0+1)=h(m.sub.2)*

(38) In the above expression,

(39) $h=\left[\begin{array}{cccc}{h}_1& h_2& \mathrm{.Math.}& h_{M_T}\end{array}\right]$
may be used, and h(m.sub.2) is the m.sub.2-th element of the vector (h), where a subscript * may denote a conjugate complex number. It is assumed that respective elements may be independently from each other, and may have the same independent and identically distributed Gaussian elements, and different vector channels may be independent of each other.

(40) For example, if it is assumed that the antenna indexes sequentially used in the time slots (1,2), (3,4), (5,6), (7,8) are (1,3), (2,4), (5,7), (6,8), the matrix may be represented by the following equation 19.

(41) $\begin{array}{cc}{j}_{}=\left[\begin{array}{cccccccc}{h}_1& h_3& 0& 0& 0& & & \\ 0& 0& h_2& h_4& \mathrm{.Math.}& & & \\ -h_3^{*}& h_1^{*}& 0& 0& \mathrm{.Math.}& \mathrm{.Math.}& & \\ 0& 0& -h_4^{*}& h_2^{*}& 0& 0& & \\ \mathrm{.Math.}& \mathrm{.Math.}& 0& 0& h_5& h_7^{*}& & \\ & & & & 0& 0& h_6& h_8\\ & & & & -h_7^{*}& h_5^{*}& 0& 0\\ & & & & 0& 0& -h_8^{*}& h_6^{*}\end{array}\right]& \left[\mathrm{Equation}19\right]\end{array}$

(42) B) Antenna Index Sequence Generation

(43) In accordance with the present invention, a maximum space time diversity gain provided from the GBD-QOSTBC scheme is maintained, and at the same time the GBD-QOSTBC codeword matrix formed in a block diagonal matrix shape may be modified. A detailed description thereof will hereinafter be described with reference to the attached drawings.

(44) FIG. 3 is a conceptual diagram illustrating a method for generating an antenna index sequence according to an embodiment of the present invention.

(45) Referring to FIG. 3, two antennas used in two consecutive timeslots may be used as one symbol constructing the sequence, and the GBD-QOSTBC codeword matrix may be modified in units of a column vector as shown in FIG. 3. In FIG. 3, the left GBD-QOSTBC codeword matrix may be denoted by C.sub.2k, and the right modified GBD-QOSTBC codeword matrix may be denoted by .

(46) In this case, although is not configured in a block diagonal matrix shape, matrix may still be a diagonal matrix. In addition, C.sub.2k.sup.HC.sub.2k and do not correspond to the same diagonal matrix, and the same principles may also be applied to the effective channel matrix corresponding to . Therefore, user equipments (UEs) may be identified by different kinds of information.

(47) A maximum number of sequences capable of being generated by modification of the M.sub.TM.sub.T GBD-QOSTBC matrix may be the number of cases achieved when every 2 Tx antennas from among a total of M.sub.T Tx antennas are selected a total of M.sub.T/2 times without duplication, and may be represented by the following equation 20.
N.sub.Max=.sub.M.sub.TC.sub.2.sub.M.sub.T.sub.-2C.sub.2 . . . .sub.4C.sub.2[Equation 20]

(48) In this way, a maximum amount of information capable of being allocated to the antenna sequence index may be represented by the following equation 21.

(49) $\begin{array}{cc}\begin{array}{c}{B}_{\mathrm{SSC}}=\frac{1}{M_T}{\log }_2\left(\frac{M_T!}{2^{M_T/2}}\right)\left(\mathrm{bits}\text{/}\mathrm{channel}\mathrm{use}\right)\\ =\frac{1}{M_T}{\log }_2\left(M_T!\right)-\frac{1}{2}\\ {\log }_2\left(M_T\right)+\underset{\underset{\left(\right)}{}}{\frac{1}{M_T\ln 2}+\frac{1}{2M_T}{\log }_2\left(M_T\right)}-\\ \underset{\underset{\left(\right)}{}}{\left(\frac{1}{\ln 2}+\frac{1}{2}\right)}{\log }_2\left(M_T\right)-2\end{array}& \left[\mathrm{Equation}21\right]\end{array}$

(50) In Equation 12, if M.sub.T is at a very high value, ().fwdarw.0 may be decided, () is a constant (or an invariable number) that always be less than 2. Therefore, when many Tx antennas are present, B.sub.SSC may be represented by Equation 21. In addition, as the number of antennas increases as shown in the graph of FIG. 4, a maximum amount of information may increase in proportion to the logarithmic function as shown in Equation 21.

(51) C) Transmission/Reception (Tx/Rx) Procedures of the STBC-SSC Based Signal

(52) First of all, it is assumed that the transmitter and the receiver may have the codebooks (1) to (3), and the receiver further includes the codebook (4) through channel estimation.

(53) FIG. 5 is a block diagram illustrating a transmitter according to an embodiment of the present invention.

(54) Referring to FIG. 5, data to be transmitted may be channel-encoded, such that the encoded bit stream (i.e., the codeword) may occur. For convenience of description, it is assumed that one codeword is used and a total number of bits is denoted by N=B.sub.QOSTBC+B.sub.SSC. In this case, B.sub.QOSTBC bits from among a total of N bits may be transmitted using the QO-STBC scheme, and B.sub.SSC bits may be allocated to the antenna sequence and then transmitted. A reference for dividing the encoded bit stream corresponding to the length of N into two portions must be recognized by both the transmitter and the receiver in order to implement the encoding and decoding processes.

(55) 2.sup.B.sup.SSC bit streams, each of which has the length of B.sub.SSC, may be allocated to a total of 2.sup.B.sup.SSC antenna sequences. Information regarding the bit stream allocated to each antenna sequence has already been defined in the codebook .sub.Ant. In addition, as can be seen from the following equation 22, specific information indicating which one of antenna index sequences will be used by the bit stream to be transmitted may be decided according to a data transmission (Tx) time.

(56) 0$\begin{array}{cc}{u}_j=\left[u_1,u_2,u_3,\mathrm{.Math.},u_{B_{\mathrm{SSC}}}\right].fwdarw.I_j& \left[\mathrm{Equation}22\right]\end{array}$

(57) If the antenna index sequence codeword is decided, the order of antenna pairs to be used in the timeslot pair (t.sub.1,t.sub.2)=(1, 2), . . . ,(M.sub.T1,M.sub.T) may be used. The GBD-QOSTBC symbols S.sub.1,S.sub.2, . . . ,S.sub.M.sub.T may be transmitted according to the decided order.

(58) The operations of the receiver configured to reconstruct the transmitted data will hereinafter be described. FIG. 6 is a block diagram illustrating a receiver according to an embodiment of the present invention. In FIG. 6,

(59) $y={\left[\begin{array}{ccccc}{y}_1& y_2^{*}& y_3& \mathrm{.Math.}& y_{M_T}^{*}\end{array}\right]}^T$
may denote the Rx signal vector received during the M.sub.T timeslot.

(60) Referring to FIG. 6, the receiver may search for the codeword indexes of the codebooks (.sub.QAM,QOSTBC, .sub.G-STBC, .sub.H, .sub.Ant) through decoding. The overall decoding order is as follows. The concept of FIG. 6 is based on the Joint ML (maximum likelihood) scheme capable of simultaneously searching for the antenna sequence (.sub.Ant) and the symbol (.sub.G-STBC). A method for searching for the .sub.Ant symbol will first be described, and a method for searching for the .sub.G-STBC symbol will be described later.

(61) (a) Joint-ML decoding

(62) The receiver may perform Joint-ML decoding using the effective channel codebook (.sub.H) in which the GBD-QOSTBC codebook (.sub.G-STBC) and the antenna index sequence are reflected. From among the effective channel matrices defined in the codebook .sub.H={,, . . . ,}, if a conjugate transpose matrix of the j-th matrix is multiplied by the Rx signal vector, the multiplication result is represented by the following equation 23.

(63) $\begin{array}{cc}{\tilde{y}}_j=j_H^{}\left[\begin{array}{c}{y}_1\\ y_2^{*}\\ y_3\\ \mathrm{.Math.}\\ y_8^{*}\end{array}\right],\mathrm{where}j\left\{1,2,\mathrm{.Math.},2^{B_{\mathrm{SSC}}}\right\}& \left[\mathrm{Equation}23\right]\end{array}$

(64) First elements corresponding to the odd indexes and second elements corresponding to the even indexes of the vector ({tilde over (y)}.sub.j) of Equation 23 are distinguished from one another so that the first and second elements are divided into two vectors ({tilde over (y)}.sub.j.sup.Even,{tilde over (y)}.sub.j.sup.Odd). {tilde over (y)}.sub.j.sup.Even,{tilde over (y)}.sub.j.sup.Odd may be defined as shown in the following equation 24.

(65) $\begin{array}{cc}{\tilde{y}}_j^{\mathrm{Even}}=\left[\begin{array}{c}{\tilde{y}}_{j,2}\\ {\tilde{y}}_{j,4}\\ \mathrm{.Math.}\\ \mathrm{.Math.}\\ {\tilde{y}}_{j,M_T}\end{array}\right],{\tilde{y}}_j^{\mathrm{Odd}}=\left[\begin{array}{c}{\tilde{y}}_{j,1}\\ {\tilde{y}}_{j,3}\\ \mathrm{.Math.}\\ \mathrm{.Math.}\\ {\tilde{y}}_{j,M_T-1}\end{array}\right]& \left[\mathrm{Equation}24\right]\end{array}$

(66) Assuming that data is transmitted according to the j-th antenna index sequence, the GBD-QOSTBC codeword located nearest to the Rx signal vector is searched for in .sub.G-STBC. In this case, although it is assumed that the distance measurement method for searching for the nearest codeword vector is l.sub.2-norm, it may be possible to use other measurement methods as necessary. In association with two symbol vectors ({tilde over (y)}.sub.j.sup.Even,{tilde over (y)}.sub.j.sup.Odd), the k-th codeword vector of the codebook .sub.G-STBC can be represented by the following equation 25, and the l.sub.2-norm value may be represented by the following equation 26.

(67) $\begin{array}{cc}{d}_{\mathrm{Even}}^{\left[k,j\right]}=\underset{k}{\min }{.Math.{\tilde{y}}_j^{\mathrm{Even}}-\left[\begin{array}{cccc}{a}_j& & & \\ & b_j& & \\ & & c_j& \\ & & & d_j\end{array}\right]S\left[k\right].Math.}_2& \left[\mathrm{Equation}25\right]\end{array}$

(68) $\begin{array}{cc}{d}_{\mathrm{Odd}}^{\left[k,j\right]}=\underset{k}{\min }{.Math.{\tilde{y}}_j^{\mathrm{Odd}}-\left[\begin{array}{cccc}{a}_j& & & \\ & b_j& & \\ & & c_j& \\ & & & d_j\end{array}\right]S\left[k\right].Math.}_2& \left[\mathrm{Equation}26\right]\end{array}$

(69) In Equations 25 and 26, S[k].sub.G-STBC may be used, and a.sub.j,b.sub.j,c.sub.j,d.sub.j may be derived from the following equation 27.

(70) $\begin{array}{cc}{j}_H^{}{j}_{}=\left[\begin{array}{cccc}{a}_j{I}_2& & & \\ & b_j{I}_2& & \\ & & c_j{I}_2& \\ & & & d_j{I}_2\end{array}\right],\mathrm{where}{I}_2=\left[\begin{array}{cc}1& 0\\ 0& 1\end{array}\right]& \left[\mathrm{Equation}27\right]\end{array}$

(71) In addition, in association with the j-th codeword matrix of .sub.H, Equation 25 and Equation 26 can be calculated in consideration of all codeword vectors (s[k],k{1,2, . . . ,M.sup.M.sup.T.sup./2}) of .sub.G-STBC. In association with all codeword matrices of .sub.H, the above-mentioned processes are repeatedly performed as denoted by (,j{1,2, . . . ,2.sup.B.sup.SSC}), d.sub.Odd.sup.[k,j] and d.sub.Even.sup.[k,j] values may be searched for in all values (k, j). Thereafter, the antenna sequence index and the GBD-QOSTBC codeword index may be decided by the following equation 28 according to the following reference based on the calculated value.

(72) $\begin{array}{cc}\left(k^{*},j^{*}\right)=\underset{k,j}{\min }\left(d_{\mathrm{Even}}^{\left[k,j\right]}+d_{\mathrm{odd}}^{\left[k,j\right]}\right)& \left[\mathrm{Equation}28\right]\end{array}$

(73) In Equation 28, k* is a codeword vector index of the codebook .sub.G-STBC, and j* is a codeword matrix index of the codebook .sub.Ant.

(74) (b) As can be seen from Equation 29, the symbol vector .sub.QAM,QOSTBC may be searched for through the inverse operation of Equation 14.

(75) $\begin{array}{cc}\hat{s}=s\left[k^{*}\right],\mathrm{where}s\left[k^{*}\right]{}_{\mathrm{QAM},\mathrm{QOSTBC}}& \left[\mathrm{Equation}29\right]\end{array}$

(76) (c) Not only through the decision of the Tx symbol such as QAM/PSK or the like, but also through the demodulation of the decided symbol, binary data can be extracted.

(77) (d) The bit stream allocated to the decided j*-th antenna sequence may be extracted from the codebook .sub.Ant.

(78) (e) The encoded bit stream formed by adding the binary data extracted from the above methods (c) and (d) to one bit stream may be constructed, and the resultant encoded bit stream is decoded so that an original signal (binary data) is reconstructed.

(79) An example of the STBC-SSC signal transmission method will hereinafter be given on the basis of the above-mentioned description. First of all, 8 bits may be transmitted according to the QO-STBC scheme (i.e., 8 bits may be respectively transmitted through 8 antennas). It is assumed that 2-bit data is transmitted using SSC, such that the following definition may be achieved. 88 GBD-QOSTBC, M.sub.T=8 B.sub.STBC=8 bits and B.sub.SSC=2 bits Constellation: BPSK (M=2)

(80) In addition, the internal parameters (T,D) according to the system environment may be defined as follows. D=diag{e.sup.j.sup.0,e.sup.j.sup.1,e.sup.j.sup.2,e.sup.j.sup.3}, If 8 antennas are used and BPSK is then used, phases needed for acquiring a maximum space time diversity of GBD-QOSTBC are denoted by

(81) $\begin{array}{c}{}_0=0,\\ {}_1=\frac{}{4}\\ {}_2=\frac{}{2},\\ {}_3=\frac{3}{4}.\end{array}$ It is assumed that Hadamard matrix is denoted by

(82) 0$T=\left[\begin{array}{cccc}1& -1& 1& -1\\ 1& 1& -1& -1\\ -1& 1& 1& -1\\ 1& 1& 1& 1\end{array}\right].$

(83) In order to transmit a total of 8 bits for each Tx data stream, one BPSK symbol for each antenna may be transmitted during the time slot (M.sub.T=8). Each element of the codebooks (.sub.QAM,QOSTBC, .sub.G-STBC) nay be denoted by a (41) vector. Each element of one vector may be set to one of 2 BPSK signals. Therefore, each codebook (.sub.QAM,QOSTBC,.sub.G-STBC) may have a total number (M.sup.M.sup.T.sup./2=2.sup.4=16) of the (41) codeword vectors, such that the total of (M.sup.M.sup.T.sup./2=2.sup.4=16) (41) codeword vectors may serve as the element. .sub.QAM,QOSTBC and .sub.G-STBC may be represented as follows.
.sub.QAM,QOSTBC={s[1],s[2], . . . ,s[16]}, s.sub.n[k]{1,+1}
Parameter range: k{1,2, . . . ,16}, n{1,2, . . . 4}
.sub.G-STBC=TD.sub.QAM,QOSTBC={S[1],S[2], . . ,S[16]}

(84) The amount of information applied to the Tx antenna index sequence is a total of 2 bits, and the Tx antenna index sequence and the binary signal thereto may be defined as follows.
.sub.Ant={I.sub.1,I.sub.2,I.sub.3,I.sub.4,u.sub.1,u.sub.2,u.sub.3,u.sub.4}
I.sub.1=(1,2), (3,4), (5,6), (7,8), I.sub.2=(1,3), (2,4), (5,7), (6,8)
I.sub.3=(1,4), (2,5), (3,8), (6,7), I.sub.4=(1,5), (2,6), (3,7), (4,8)
u.sub.1=[0,0],u.sub.2=[0,1],u.sub.3=[1,0],u.sub.4=[1,1]

(85) The effective channel matrix codebook .sub.H={,,,} corresponding to .sub.Ant={I.sub.1,I.sub.2,I.sub.3,I.sub.4,u.sub.1,u.sub.2,u.sub.3,u.sub.4} is as follows.

(86) $\begin{array}{c}{1}_{}=\left[\begin{array}{cccccccc}{h}_1& h_2& & & & & & \\ -h_2^{*}& h_1^{*}& & & & & & \\ & & h_3& h_4& & & & \\ & & -h_4^{*}& h_3^{*}& & & & \\ & & & & h_5& h_6& & \\ & & & & -h_6^{*}& h_5^{*}& & \\ & & & & & & h_7& h_8\\ & & & & & & -h_8^{*}& h_7^{*}\end{array}\right],\\ 2_{}=\left[\begin{array}{cccccccc}{h}_1& h_3& 0& 0& 0& & & 0\\ 0& 0& h_2& h_4& \mathrm{.Math.}& & & \\ -h_3^{*}& h_1^{*}& 0& 0& \mathrm{.Math.}& \mathrm{.Math.}& & \\ 0& 0& -h_4^{*}& h_2^{*}& 0& 0& \mathrm{.Math.}& 0\\ \mathrm{.Math.}& \mathrm{.Math.}& 0& 0& h_5& h_7^{*}& 0& 0\\ & & & & 0& 0& h_6& h_8\\ & & & & -h_7^{*}& h_5^{*}& 0& 0\\ 0& 0& \mathrm{.Math.}& 0& 0& 0& -h_8^{*}& h_6^{*}\end{array}\right]\\ 3_{}=\left[\begin{array}{cccccccc}{h}_1& h_4& & & & & & \\ & & h_2& h_5& & & & \\ & & & & h_3& h_8& & \\ -h_4^{*}& h_1^{*}& & & & & & \\ & & -h_5^{*}& h_2^{*}& & & & \\ & & & & & & h_6& h_7\\ & & & & & & -h_7^{*}& h_6^{*}\\ & & & & -h_8^{*}& h_3^{*}& & \end{array}\right],\\ 4_{}=\left[\begin{array}{cccccccc}{h}_1& h_5& & & & & & \\ & & h_2& h_6& & & & \\ & & & & h_3& h_7& & \\ & & & & & & h_4& h_8\\ -h_5^{*}& h_1^{*}& & & & & & \\ & & -h_6^{*}& h_2^{*}& & & & \\ & & & & -h_7^{*}& h_3^{*}& & \\ & & & & & & -h_8^{*}& h_4^{*}\end{array}\right]\end{array}$

(87) Assuming that the k-the symbol vector S[k].sub.G-STBC is transmitted using the sequence (I.sub.2) and the reception noise is omitted, Equation 23 may also be represented by the following equation 30.
{tilde over (y)}.sub.j=S[k],[Equation 30]
where j{1,2,3,4}, k{1,2, . . . ,16}

(88) Specifically, for use in {tilde over (y)}.sub.2=S[k] may have the following equation 31.

(89) $\begin{array}{cc}& \left[\mathrm{Equation}31\right]\\ 2_H^{}{2}_{}=[\begin{array}{cccc}\left(h_1^{}{h}_1^{*}+h_3{h}_3^{*}\right)I_2& & & \\ & \left(h_2{h}_2^{*}+h_4{h}_4^{*}\right)I_2& & \\ & & \left(h_5{h}_5^{*}+h_7{h}_7^{*}\right)I_2& \\ & & & \left(h_6{h}_6^{*}+h_8{h}_8^{*}\right)I_2\end{array}]& \end{array}$

(90) In Equations 25 and 26, the parameters (a.sub.2,b.sub.2,c.sub.2,d.sub.2) corresponding to (j=2) may be a.sub.2=h.sub.1h*.sub.1+h.sub.3h*.sub.3, b.sub.2=h.sub.2h*.sub.2+h.sub.4h*.sub.4, c.sub.2=h.sub.5h*.sub.5+h.sub.7h*.sub.7, d.sub.2=h.sub.6h*.sub.6+h.sub.8h*.sub.8, respectively. Even when (where j=1,3,4) is used, the above parameters may be derived in the same manner as described above.

(91) The matrices (,,) corresponding to the product of the effective channel () of the sequence (I.sub.2) and a transpose matrix of a channel matrix corresponding to another sequence are not denoted by diagonal matrices. Although the odd and even elements of the resultant value {tilde over (y)}.sub.j obtained when the transpose matrix of the j-th element of the codebook (.sub.H) is multiplied by the Rx signal are decoded in different ways, if no noise occurs, a desired Tx symbol vector and the effective channel matrix corresponding to the used sequence can always be searched for. Therefore, {tilde over (y)}.sub.j is divided into {tilde over (y)}.sub.j.sup.Even,{tilde over (y)}.sub.j.sup.Odd as shown in the following equation 32, and {tilde over (y)}.sub.j.sup.Even,{tilde over (y)}.sub.j.sup.Odd may be decoded in different ways.

(92) $\begin{array}{cc}{\tilde{y}}_j^{\mathrm{Even}}=\left[\begin{array}{c}{\tilde{y}}_{j,2}\\ \mathrm{.Math.}\\ {\tilde{y}}_{j,8}\end{array}\right],{\tilde{y}}_j^{\mathrm{Odd}}=\left[\begin{array}{c}{\tilde{y}}_{j,1}\\ \mathrm{.Math.}\\ {\tilde{y}}_{j,7}\end{array}\right],\mathrm{where},j\left\{1,2,3,4\right\}& \left[\mathrm{Equation}32\right]\end{array}$

(93) As can be seen from Equations 25 to 27, d.sub.Even.sup.[k,j] and d.sub.Odd.sup.[k,j] may be searched for, and the Tx symbol vector and the Tx antenna index sequence may be determined using the following equation 33.

(94) $\begin{array}{cc}\left(k^{*},j^{*}\right)=\arg \underset{k,j}{\min }\left(d_{\mathrm{Even}}^{\left[k,j\right]}+d_{\mathrm{odd}}^{\left[k,j\right]}\right)& \left[\mathrm{Equation}33\right]\end{array}$

(95) A detailed description of the receiver having lower calculation complexity than the receiver of FIG. 6 will hereinafter be given.

(96) FIG. 7 is a block diagram illustrating a receiver according to another embodiment of the present invention.

(97) The receiver shown in FIG. 6 simultaneously searches for all QO-STBC symbols and all antenna index sequences. If many antennas are present and the modulation order is very high, the complexity may excessively increase. As a result, the low complexity Rx scheme for sequentially searching for antenna indexes two by two may be considered and used.

(98) As can be seen from Equation 25, Equation 26, and Equation 31, although the effective channel matrix () is not denoted by the block diagonal matrix, may be diagonal matrices. Accordingly, one antenna index pair may be decided on the basis of two timeslots. In FIG. 7, the Rx symbol vector may be represented by the following equation 34 according to the symbol reception (Rx) time.
t=(1,2),(3,4), . . . ,(M.sub.T1,M.sub.T), y(1,2),y(3,4), . . . , y(M.sub.T1,M.sub.T)

(99) First, the codebook .sub.G-STBC may be used to search for the pair of antenna indexes. The QAM symbol defined in the codebook .sub.QAM,QOSTBC of Equation 13 may be decoded as long as M.sub.T/2 GBD-QOSTBC symbols, and the inverse matrix of the matrix (TD) of Equation 14 is multiplied by the decoded result, such that it may be search for the QAM symbol using the multiplication result. Therefore, after the antenna index sequence is determined on the basis of .sub.G-STBC according to the Rx scheme of the present invention, it may be search for .sub.QAM,QOSTBC. For convenience of description, four antenna index sequences of Equation 35 may be considered and used.

(100) $\begin{array}{cc}\begin{array}{c}t=\left(1,2\right),\left(3,4\right),\mathrm{.Math.},\left(M_T-1,M_T\right)\\ I_1\left(00\right)=\left(1,2\right),\left(3,4\right),\left(5,6\right),\left(7,8\right)\\ I_2\left(01\right)=\left(1,3\right),\left(2,4\right),\left(5,7\right),\left(6,8\right)\\ I_3\left(10\right)=\left(1,4\right),\left(2,5\right),\left(3,8\right),\left(6,7\right)\\ I_4\left(11\right)=\left(1,5\right),\left(2,6\right),\left(3,7\right),\left(4,8\right)\end{array}& \left[\mathrm{Equation}35\right]\end{array}$

(101) The antenna index pairs capable of being used at the condition (t=1,2) may be limited to (1,2), (1,3), (1,4), (1,5). Therefore, only four combinations may be considered and used at t=1,2. The same results may be acquired not only at t=3,4 but also at the subsequent timeslots located after t=3,4. Assuming that the k-th symbol vector of .sub.G-STBC is transmitted at t=1,2 and the antenna indexes (1,3) are then selected, two GBD-QOSTBC symbols may be (h.sub.1h*.sub.1+h.sub.3h*.sub.3)(S.sub.1[k],S.sub.2[k]), respectively. If there is no influence of interference, the GBD-QOSTBC symbol can always be searched for.

(102) As can be seen from Equation 14, since only S.sub.1[k] and S.sub.2[k] can be transmitted at t=1,2,S.sub.j[k],j3 is not considered at t=1,2. The same principle as in the above case may be used even at t=3,4, so that only S.sub.j[k],j=3,4 of the codebook (.sub.G-STBC) may be used in the above searching process. As a result, although the GBD-QOSTBC symbol and the antenna index are simultaneously searched for, this searching process is performed on the basis of 2 Alamouti blocks, instead of using the method for searching for all sequences, such that the above-mentioned method may have lower calculation complexity than the joint-ML scheme.

(103) In summary, the legacy GBD-QOSTBC structure is modified in a manner that the sequence composed of the Tx antenna indexes may be constructed and the binary data sequence may be allocated to the resultant antenna sequence, such that some parts of data to be loaded on the Tx symbols (i.e., QAM and PSK modulation symbols) and then transmitted can be allocated to the antenna sequence. Accordingly, reduction of a minimum length between symbols can be prevented when the modulation order increases, so that an improved BER performance superior to that of the GBD-QOSTBC scheme can be provided to the SNR region of a predetermined level or higher.

(104) Whereas the legacy STBC-SM scheme must unavoidably accept the diversity performance deterioration instead of obtaining the SM gain when the data stream is transmitted, the proposed scheme of the present invention loads data on the Tx antenna sequence and then transmits the resultant Tx antenna sequence, without loss of a maximum amount of the diversity gain that is capable of being obtained by the QO-STBC scheme, such that a gain similar to that of the SM-MIMO scheme can be obtained.

(105) As is apparent from the above description, the embodiments of the present invention can more effectively transmit data in a MIMO wireless communication system.

(106) It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the inventions. Thus, it is intended that the present invention covers the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents.