A method for accelerating the precoding and pre-decoding of symbols in ofdm systems

Method for accelerating precoding and predecoding of symbols in orthogonal frequency division multiplexing (OFDM) systems

A METHOD FOR ACCELERATING THE PRECODING AND PRE-DECODING OF SYMBOLS IN OFDM SYSTEMS

The present invention generally relates to an Orthogonal Frequency Division Multiplexing (OFDM) communication system, and specifically relates to the precoding technology implemented by such systems.

This application claims the rights of U.S. Provisional Application No. 61/035,394 filed on March 11, 2008.

The WiMedia standard defines the media access control (MAC) layer and physical (PHY) layer specifications based on OFDM transmission. The WiMedia standard enables short-distance multimedia files to be transmitted at a rate of up to 480Mbps with low power consumption. This standard operates in the frequency band between 3.1 GHz and 10.6 GHz of the ultra-wideband (UWB) spectrum. The maximum data rate of the WiMedia standard rate cannot meet future wireless multimedia applications, such as high-definition television (HDTV) wireless connectivity. People are working hard to increase the data rate to 1Gbps and above.

For this reason, it has been envisaged to use weak channel (or non-channel) coding and higher-order symbol constellation technology in future high data rate wireless systems. For example, if you change<img file="TWI458302B_D0001.tif" he="58" id="i0001" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="41" />Convolutional codes are used together with 16 Quadrature Amplitude Modulation (QAM), and the WiMedia PHY data rate can be increased to 960Mbps. However, due to the characteristics of OFDM transmission, this will reduce channel performance. Specifically, with a weak channel code, this OFDM cannot effectively utilize frequency diversity. Therefore, the channel performance is almost determined by the worst sub-carrier, which has the lowest signal-to-noise ratio (SNR). This limits the number of high data rate applications that can be carried by conventional OFDM wireless systems.

The industry has proposed several precoding techniques to overcome this problem. Generally speaking, all precoding techniques are based on jointly modulating transmission symbols to multiple sub-carriers. This allows the receiver to resume sending symbols even if some of these sub-carriers are in deep fading. In "OFDM or single-carrier block transmissions? (OFDM or single-carrier block transmissions?)" published by Z. Wang, X. Ma and GB Giannakis, IEEE Communications Proceedings Vol. 52, pages 380-394, March 2004 In and published by Z. Wang and GB Giannakis in Third IEEE Signal Processing Workshod on Signal Processing Advances in Wireless communications in Taoyuan, Taiwan, March 20-23, 2001 Examples of precoding technology can be found in "Linearly Precoded or Coded OFDM against Wireless Channel Fades".

Precoding is usually implemented by a precoder circuit coupled to an input of an IFFT OFDM modulator of a transmitter and a predecoder circuit coupled to the output of an FFT OFDM demodulator of a receiver. A well-designed full-power precoder can effectively utilize the frequency diversity provided by the multipath channel. However, implementing a full-power precoder increases the complexity of the transmitter and receiver because it requires more sophisticated decoding and symbol mapping techniques. For example, using dual carrier modulation (DCM) technology as a precoder requires 16QAM symbol constellation to replace QPSK symbol constellation.

In addition, in order to ensure full frequency diversity (ie, second-order diversity) in the high data rate mode, a higher constellation diagram (for example, 256QAM) is required. For example, if the DCM technology is used to jointly modulate the two information symbols s(i) and s(i+50) formed by the QAM16 constellation diagram, the following second-order frequency diversity is achieved:

<maths><img file="TWI458302B_D0002.tif" he="479" id="i0002" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1405" /></maths>

The information symbols s(i) and s(i+50) are formed using a 16QAM constellation diagram. However, the precoding operation extends the symbol constellation to 256QAM, that is, the constellation diagram of the precoding symbols x(i) and x(i+50) is 256QAM.

Due to the high-order precoding symbol constellation and complexity of the precoder and predecoder circuit, it is not feasible to implement and design a receiver and transmitter with this high constellation without affecting the diversity gain and the overall performance of the channel. In addition, the time required for these circuits to process signals actually makes these conventional precoding techniques impossible to implement in high data rate wireless systems.

Therefore, it would be advantageous to provide an effective pre-coding and pre-decoding solution that overcomes the shortcomings of the prior art.

Specific embodiments of the present invention include a method for implementing fast multiple subcarrier joint modulation (MSJM) precoding. The method includes grouping input information bits into bit blocks; converting the equal bit blocks into bit vectors; and mapping a first group of bits of each bit vector to real-dimensional symbols in a symbol vector ; The second group of bits of each bit vector is mapped to the imaginary dimensional symbol in the symbol vector, wherein the real dimensional mapping and the imaginary dimensional mapping are simultaneously implemented; and the symbol vector is modulated into a data subcarrier.

Certain embodiments of the present invention also include a computer-readable medium having computer-executable codes stored thereon for implementing fast joint multiple-subcarrier modulation (MSJM) precoding. The computer executable code causes the computer to perform the following processes: group the input information bits into bit blocks; convert the equal bit blocks into bit vectors; map the first group of bits of each bit vector to a The real-dimensional symbol in the symbol vector; the second group of bits of each bit vector is mapped to the imaginary-dimensional symbol in the symbol vector, wherein the real-dimensional mapping and the imaginary-dimensional mapping are simultaneously implemented; and the symbol vector is adjusted Become a data sub-carrier.

Certain embodiments of the present invention also include an Orthogonal Frequency Division Multiplexing (OFDM) transmitter for implementing multiple subcarrier joint modulation (MSJM) precoding. The OFDM transmitter includes: a first serial-to-parallel (S/P) converter for converting bit blocks into bit vectors; and a precoder for converting the first set of each bit vector Bits are mapped to real-dimensional symbols in a symbol vector and the second group of bits of each bit vector is mapped to imaginary symbols in the symbol vector, wherein the real-dimensional mapping and the imaginary-dimensional mapping are simultaneously implemented ; And a second S/P converter for grouping symbol vectors and mapping these symbol vectors into data subcarriers.

The scope of patent application at the end of this specification specifically points out and clearly claims the subject matter of the present invention. The above and other features and advantages of the present invention will be apparent from the following detailed description in conjunction with the accompanying drawings.

It is important to note that the embodiments disclosed in the present invention are only examples of the many advantageous uses of the innovative teachings herein. Generally speaking, the statements made in the specification of this application do not necessarily limit any of the various claimed inventions. In addition, some statements may apply to some inventive features but not to others. Generally speaking, unless otherwise indicated, singular elements may be plural, and vice versa in order not to lose broadness. In the figures, the same number refers to the same part throughout several figures.

Fig. 1 shows a non-limiting and exemplary block diagram of an OFDM-based wireless system 100 for describing the principles of the present invention. The system 100 jointly precodes m (m&gt;2) subcarriers to achieve at least second-order frequency diversity. The system 100 may be any type of OFDM-based wireless system operating according to wireless communication standards, such wireless communication standards including but not limited to WiMedia UWB versions 1.0, 1.5, and 2.0, IEEE 802.11n, WiMax, and so on.

The system 100 includes a transmitter 110 and a receiver 120, which communicate using wireless media. The transmitter 110 includes serial-to-parallel (S/P) converters 111 and 113, a precoder 112, and an OFDM modulator 114. The receiver 120 includes an OFDM demodulator 121, a serial-to-parallel (S/P) converter 122, and a pre-decoder 123. The system 100 also includes a transmitting antenna 130 and a receiving antenna 140.

According to the present invention, the input information bits are preferably divided into bit blocks after being encoded and interleaved. Each bit block contains n*g bits, where "n" is the number of available data subcarriers and "g" is the number of bits that will be transmitted per subcarrier. The S/P converter 111 converts each bit block into k bit vectors. For example, the i-th bit vector of the q-th bit block can be expressed as follows:

<maths><img file="TWI458302B_D0003.tif" he="137" id="i0003" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1102" /></maths>

The number "k" is equal to the number of data subcarriers divided by the number of joint precoding subcarriers, that is, n/m. Each bit vector contains m*g bits.

The precoder 112 maps each bit vector to a symbol vector. A symbol vector contains m symbols. The symbol vector corresponding to the i-th bit vector of the q-th bit block can be expressed as follows:

<maths><img file="TWI458302B_D0004.tif" he="120" id="i0004" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1120" /></maths>

According to a specific embodiment of the present invention, the precoder 112 independently maps the bits of the bit vector to the real dimension and the imaginary dimension of each symbol in the symbol vectors. Specifically, a symbolic vector<img file="TWI458302B_D0005.tif" he="62" id="i0005" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="65" />It is generated by two independent mapping operations: 1) The one-bit vector<img file="TWI458302B_D0006.tif" he="64" id="i0006" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="53" />M*g bits of y₁Bits are mapped to<img file="TWI458302B_D0007.tif" he="62" id="i0007" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="65" />The real dimensional value of the symbol in the vector; and 2) the<img file="TWI458302B_D0008.tif" he="64" id="i0008" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="55" />The remaining m*gy of the vector₁Bits are mapped to this<img file="TWI458302B_D0009.tif" he="61" id="i0009" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="62" />The imaginary dimension value of the symbol in the vector. These mapping operations significantly reduce the complexity and time required for symbol precoding.

In an embodiment of the present invention, two mapping functions defined as follows can be used to implement a one-bit vector<img file="TWI458302B_D0010.tif" he="64" id="i0010" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="48" />Mapping to a symbolic vector:

1. A mapping function<i>K</i>_<i>I</i> ：

<maths><img file="TWI458302B_D0011.tif" he="195" id="i0011" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1639" /></maths>

2. A mapping function<i>K</i>_<i>Q</i> ：

<maths><img file="TWI458302B_D0012.tif" he="123" id="i0012" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1840" /></maths>

Among them, Re{x} and Im{x} represent the real and imaginary dimensions of the complex number x, respectively. The mapping function<i>K</i>_<i>I</i>and<i>K</i>_<i>Q</i>Each must achieve second-order diversity, that is, for any two different input bit vectors,<i>K</i>_<i>I</i>(or<i>K</i>_<i>Q</i>The corresponding output of) has at least two different symbols.

In a preferred embodiment of the present invention, the precoder 112 uses a look-up table to map the bit vector to the symbol vector. When using a lookup table, the value of the bit vector is the index used to retrieve the value of the symbol. Various embodiments for constructing such look-up tables are described in more detail below.

The symbol vectors are combined and mapped into "n" data subcarriers by the S/P converter 113. Then at the i-th of the q-th OFDM symbol_eThe symbol S is transmitted on each data subcarrier_q,i(i_e)(e=1,...,m). The OFDM modulator 114 performs an IFFT operation to generate a time-domain transmission signal, which is then transmitted via the transmission antenna 130.

A signal received at the receiver 120 is converted into a frequency domain signal by the FFT operation performed by the OFDM demodulator 121. Then, the S/P converter 122 outputs m symbol vectors<img file="TWI458302B_D0013.tif" he="61" id="i0013" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="611" />,in<i>r</i>_<i>q</i>_,_<i>i</i> (<i>i</i>_<i>e</i>) Is the i-th of the q-th OFDM symbol_eThe received signal of a data sub-carrier, and "T" represents the matrix transposition operation. The predecoder 123 uses the following formula to generate information bits:

<maths><img file="TWI458302B_D0014.tif" he="127" id="i0014" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="577" /></maths>

in<i>H</i>_<i>i</i> =<i>diag</i> {<i>h</i> (<i>i</i>_<i>e</i>)} is the diagonal matrix of m times m, where the e-th diagonal element is equal to h(i_e)(e=1,...,m), where h(i_e)(e=1,...,m) means the i-th_eChannel parameters of data subcarriers,<img file="TWI458302B_D0015.tif" he="53" id="i0015" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="47" />Is the received symbol vector.<img file="TWI458302B_D0016.tif" he="63" id="i0016" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="629" />Is the transmitted symbol vector, and<img file="TWI458302B_D0017.tif" he="66" id="i0017" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="97" />Is the additive white Gaussian noise (AWGN) vector.

According to the present invention, the pre-decoder 123 independently pre-decodes the real and imaginary dimensions of the symbols. To this end, the predecoder 123 implements two demapping functions, which can be expressed as follows:

<maths><img file="TWI458302B_D0018.tif" he="111" id="i0018" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1661" /></maths>

<maths><img file="TWI458302B_D0019.tif" he="121" id="i0019" img-content="drawing" img-format="tif" inline="yes" orientation="portrait" wi="1726" /></maths>

in<img file="TWI458302B_D0020.tif" he="56" id="i0020" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="362" />Yes<i>H</i>_<i>i</i>Conjugate complex number of, and<i>diag</i> {|<i>h</i> (<i>i</i>_<i>e</i> )|²} Is the diagonal matrix of m by m, where the e-th diagonal element is equal to |<i>h</i> (<i>i</i>_<i>e</i> )|² ,<i>e</i> =1,...<i>m</i>. These demapping functions are executed in parallel, and each demapping function only jointly pre-decodes m*g/2 bits of the symbol. Therefore, the complexity of the pre-decoding job is 2^(mg/2)Order, which is O(2^(mg/2 )。

According to a preferred embodiment of the present invention, the pre-decoding can be simplified. In this embodiment, the receiver 120 (from m sub-carriers) selects at least m-1 sub-carriers with the highest signal-to-noise ratio. The symbol indexes of the respective selected subcarriers are pre-decoded separately. Because the pre-coding technique has second-order diversity, the m-1 indexes of the pre-decoding are sufficient to restore the bit vector. This significantly reduces the complexity of the pre-decoding operation. It should be noted that this simplified pre-decoding independently restores the real and imaginary dimensional values of a symbol.

It should be noted that due to the structure of digital modulation technology (such as QAM), independent precoding (and pre-decoding) of real and virtual dimensions can be implemented. Specifically, the symbol of a square/rectangular QAM constellation can be described as two symbols (one real dimension and one imaginary dimension) from two pulse amplitude modulation (PAM) constellations. For example, if a symbol c comes from 64QAM, the real dimension Re(c) and the imaginary dimension Im(c) can be described as P_RConstellation diagram and P_IConstellation. Such P_RAnd P_IThe constellation diagrams are each an 8PAM constellation diagram.

The following is a non-limiting example describing the precoding technology disclosed in the present invention. In the following example, the number of joint precoding subcarriers "m" is three (3), the number of available data subcarriers "n" is 102 data, and the number of bits to be transmitted per subcarrier "g" is four ( 4). The input information bits are grouped into bit blocks, and each block contains 102*4=408 bits. Then, 34 (102/3=34) bit vectors are generated, and each vector contains 12 (4*3=12) bits. The bit vector is:<img file="TWI458302B_D0021.tif" he="75" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="62" />=[<i>b</i>_{<i>q</i> ,<i>i</i>} (0),<i>b</i>_{<i>q</i> ,<i>i</i>} (1),...,<i>b</i>_{<i>q</i> ,<i>i</i>}(11)],i=0,1,...,33.

Since the input bit vector<img file="TWI458302B_D0022.tif" he="80" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="75" />There are 12 bits, so the number of code words required for this precoding is 2¹² =64². Therefore, for the symbol<i>s</i>_<i>m</i> (<i>k</i>) The minimum constellation size to achieve second-order diversity is 64. In this example, 64QAM modulation is selected. The precoder 112 converts each bit vector<img file="TWI458302B_D0023.tif" he="77" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="65" />Map to a symbolic vector<img file="TWI458302B_D0024.tif" he="62" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="67" />=[<i>s</i>_<i>q</i> (<i>i</i> ),<i>s</i>_<i>q</i> (<i>i</i> +34),<i>s</i>_<i>q</i> (<i>i</i>+68)], where three symbols are from a 64-QAM constellation diagram, and<i>s</i>_<i>q</i> (<i>i</i> )、<i>s</i>_<i>q</i> (<i>i</i>+34) and<i>s</i>_<i>q</i> (<i>i</i>+68) are respectively transmitted on the i-th, (i+34)-th and (i+68)-th data subcarriers of the qth OFDM symbol.

The precoding can be implemented using two lookup tables, one table for real dimension values and another table for imaginary dimension values. The two lookup tables can be the same. In this example, each table contains 2⁶Columns and 3 rows. Figure 2 is an example of a lookup table 210, in which three rows correspond to<i>s</i>_<i>q</i> (<i>i</i> )、<i>s</i>_<i>q</i> (<i>i</i>+34) and<i>s</i>_<i>q</i> (<i>i</i>+68) are the real values Re(c1), Re(c2) and Re(c3) of these symbols. Specifically, the symbol index of Re(c1) is bit [<i>b</i>_{<i>q</i> ,<i>i</i>} (0),<i>b</i>_{<i>q</i> ,<i>i</i>} (1),<i>b</i>_{<i>q</i> ,<i>i</i>}(2)] and use the following formula to calculate:<maths><img alt="" file="TWI458302B_D0025.tif" he="178" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="621" /></maths>

The symbol index of Re(c2) is bit [<i>b</i>_{<i>q</i> ,<i>i</i>} (3),<i>b</i>_{<i>q</i> ,<i>i</i>} (4),<i>b</i>_{<i>q</i> ,<i>i</i>}(5)] and use the following formula to calculate:<maths><img alt="" file="TWI458302B_D0026.tif" he="178" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="605" /></maths>

According to an embodiment, any 8th order Latin 2-hypercube (or an 8 by 8 The Latin square matrix) determines the mapping value of Re(c3). The Latin b-hypercube of order a is a b-dimensional array in which each column is a permutation of symbols 0, 1,..., a-1. The Latin matrix is an "I by I" square matrix, each unit of which contains "I" different symbols (from 0 to (1-1)), and the symbol appears only once in any column or row. Specifically, c₃The dth element of the row is the (x+1,y+1)th element of an 8-by-8 Latin square matrix, where<i>d</i> -1=<i>x</i> *8+<i>y</i> ,0<img file="TWI458302B_D0027.tif" he="46" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="35" /><i>x</i> ,<i>y</i>&lt;8. As an example, the following Latin matrix can be used to map the value of Re(c3):<maths><img alt="" file="TWI458302B_D0028.tif" he="418" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="608" /></maths>

This matrix allows the use of an eight-element parity check code to express the value of Re(c3), that is, Re(c3)=mod(c2+c1,8). The Re(c3) values shown in Table 210 are calculated using the modulo-8 operation. This precoding allows to achieve 2nd order diversity because any two different columns of the look-up table contain at least two different symbols.

It should be noted that any type of Latin matrix (special or non-special) can be used to determine the value of these symbols. It should be further noted that using different precoding techniques may result in different Latin matrices and different symbol constellation map markers (ie, mapping between symbol indexes and points on the symbol constellation map), which may have different performances. This allows the selection of constellation map markers and a Latin square matrix to optimize system performance.

In another embodiment, the Re(c3) values can be determined according to the binary operation defined as follows: [<i>p</i>_{<i>q</i> ,<i>i</i>} (0),<i>p</i>_{<i>q</i> ,<i>i</i>} (1),<i>p</i>_{<i>q</i> ,<i>i</i>} (2)]=[<i>b</i>_{<i>q</i> ,<i>i</i>} (0),<i>b</i>_{<i>q</i> ,<i>i</i>} (1),...,<i>b</i>_{<i>q</i> ,<i>i</i>} (5)]*<i>G</i> ,

Where G is equal to:<maths><img alt="" file="TWI458302B_D0029.tif" he="456" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="385" /></maths>

And the value of Re(c3) line can be calculated as follows:<maths><img alt="" file="TWI458302B_D0030.tif" he="182" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="655" /></maths>

The final Latin matrix is:<maths><img alt="" file="TWI458302B_D0031.tif" he="619" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="822" /></maths>

It should be noted that the lookup table used to map the virtual dimension values is constructed using the same technique described herein. It should be further noted that different techniques can be used to construct different "real dimension" and "virtual dimension" lookup tables.

FIG. 3 shows the simulation results verifying the performance of the precoding technique implemented according to an embodiment of the present invention. In this simulation, 102 data subcarriers are used. This precoding ("MSJM precoding") is based on the assumption that all data subcarrier channels are iid-end fading channels, using a Gray PAM marker and Latin matrix M1 for both real and imaginary dimensions as shown above. Combined realization. As illustrated in Figure 3, the MSJM precoding gain (indicated by the curve 310) is higher than the gain of the conventional DCM precoding with 16QAM (as indicated by the curve 320 instructions) is better. Using this simplified precoding/predecoding technique results in lower gain performance than the MSJM precoding/predecoding (as indicated by curve 330), but predecoding is less complicated.

FIG. 4 shows a non-limiting flowchart 400, which describes a method for performing a fast multi-subcarrier joint modulation (MSJM) precoding according to an embodiment of the present invention. In S410, the input information bits are grouped into data blocks, each of which contains n*g bits. The parameter "n" is the number of available sub-carriers, and "g" is the number of bits that will be transmitted per sub-carrier. In S420, the bit block is converted into a bit vector. The number of bit vectors is equal to the number of data subcarriers divided by the number of joint precoding subcarriers, that is, n/m. In S430, each bit vector is mapped into a symbol vector, which contains m symbols. Specifically, this method also converts the y of the one-bit vector₁Bits are mapped to the real dimensions of the precoding symbols and the y of the bit vector₂The bits are mapped to the virtual dimensions of the precoding symbols. The y₁Bits are different from the y₂Bits and y₁Add y₂The number of bits is equal to m*g. In a preferred embodiment of the present invention, two lookup tables are used to implement the mapping: one for having 2^y1Real dimensions of columns and m rows and another one for having 2^y2A virtual dimension of columns and m rows. Although the operation of the method in this article is discussed for generating a "virtual dimension" lookup table, this is only implemented for exemplary purposes. The method under discussion uses the same steps described below to generate a "real-dimensional" lookup table.

In order to construct the virtual dimension lookup table, first generate a Q_I^m-1A table with columns and m rows. Parameter Q_IIs the minimum PAM constellation diagram size and is determined as follows:<maths><img alt="" file="TWI458302B_D0032.tif" he="139" id="" img-content="drawing" img-format="tif" inline="no" orientation="portrait" wi="275" /></maths>

Use one of the techniques described in detail above to set the value of the first "m-1" rows. Then, a Q_IThe order Latin (m-1)-hypercube is inserted (in a column) to the last (m) row. Specifically, the dth element of the last row is a Q_IOrder Latin (m-1)-Hypercube's (<i>x</i>_<i>m</i>_-1 +1,<i>x</i>_<i>m</i>_-2 +1,...,<i>x</i>₁+1) elements, of which<img file="TWI458302B_D0033.tif" he="119" id="i0033" img-content="character" img-format="tif" inline="no" orientation="portrait" wi="698" />. The last step of constructing the lookup table involves starting from the Q_I^m-1Multiplying table selection 2^y2Different columns.

In S440, the symbol vectors are modulated onto "n" data subcarriers and transmitted. Those skilled in the art will understand that the fast MSJM precoding can increase the data transmission rate while having improved gain performance and minimum symbol constellation extension, as well as fast precoding and predecoding.

The above detailed description has set forth some of the many forms that the present invention may take. It is hoped that the above detailed description is understood to illustrate selected forms that the present invention can take and not to be a limitation of the definition of the present invention. Only these claims (including all analogs) are intended to define the scope of the present invention.

Most preferably, the principles of the present invention are implemented as a combination of hardware, firmware, and software. In addition, the software is preferably implemented as an application program embodied on a program storage unit or a computer-readable medium. The application can be uploaded to and executed by a machine including any suitable structure. Preferably, the machine is implemented on a computer platform with hardware such as one or more central processing units ("CPU"), a memory, and input/output interfaces. The computer platform may also include an operating system and microinstruction code. The various processes and functions described herein can be part of the microinstruction code or part of the application program or any combination thereof, which can be executed by a CPU, regardless of whether the computer or processor is explicitly displayed. In addition, various other peripheral units can be connected to the computer platform, such as an additional data storage unit and a printing unit.

<p>100. . . OFDM-based wireless system</p><p>110. . . launcher</p><p>111. . . Serial to parallel converter</p><p>112. . . Precoder</p><p>113. . . Serial to parallel converter</p><p>114. . . OFDM modulator</p><p>120. . . receiver</p><p>121. . . OFDM demodulator</p><p>122. . . Serial to parallel converter</p><p>123. . . Predecoder</p><p>130. . . Transmit antenna</p><p>140. . . Receive antenna</p><p>310. . . MSJM precoding gain</p><p>320. . . The gain of precoding with the conventional DCM of 16QAM</p><p>330. . . Simplified precoding/predecoding gain</p>

Fig. 1 is a block diagram of a conventional OFDM communication system for describing the principle of the present invention;

Figure 2 is an exemplary lookup table showing precoding of symbol vectors;

Figure 3 is a graph showing the simulation results of fast multiple subcarrier joint modulation (MSJM) precoding; and

Fig. 4 is a flowchart describing a method for implementing fast MSJM precoding according to an embodiment of the present invention.