TWI502936B - Method for signal precoding - Google Patents

Description

Signal precoding method

The invention relates to a wireless communication technology, in particular to a signal precoding method of an orthogonal frequency division multiplexing system.

Orthogonal frequency-division multiplexing (OFDM) is one of the most important technologies in the field of wireless communication. The main concept of the technology is to divide the data into several subcarrier channels into a single carrier channel. Carriers are transmitted in parallel to solve the problem of inter-symbol interference (ISI) and de-modulation of transmitted data in high-rate systems. The orthogonal frequency division multiplexing system has many advantages: if the symbol time of each subcarrier is long, it can effectively resist the inter-symbol interference effect caused by multipath delay spread, compared with the single carrier system. The hardware complexity of the equalizer is low; in the slow time varying channel, each channel can adjust the transmission rate according to the signal-to-noise ratio to improve each The capacity of the channel; it can effectively resist narrowband interference; and can realize modulation and demodulation with Fast Fourier Transform (FFT), which greatly reduces system complexity.

Compared with a single carrier system, the orthogonal frequency division multiplexing system still has the disadvantage of being relatively sensitive to the carrier frequency offset (CFO), which causes severe signal distortion. The carrier frequency offset (CFO) is mostly caused by a mismatch between the transmitter and the receiver oscillator. The oscillator responsible for generating the carrier frequency itself may have errors, such that there is inter-carrier interference (ICI) between each subcarrier channel and there is a problem of complex multiplicative distortion (CMD). Inter-carrier interference (ICI) can cause the orthogonality between the sub-carrier channels to be corrupted, so that the signal energy of each sub-carrier channel itself is attenuated and interferes with the signals of other sub-carrier channels. In addition, in the wireless communication environment, the transmitting end and the receiving end may be in a moving state. If there is relative motion between the transmitting end and the receiving end, a Doppler effect is generated to make the transmitting end and the receiving end Carrier frequency offset (CFO) occurs when the carrier frequency is inconsistent. In high-speed mobile environments, the influence of carrier frequency offset (CFO) will be more serious.

In order to mitigate the impact of carrier frequency offset (CFO), many signal precoding techniques have begun to develop. The so-called signal precoding is the orthogonal frequency division multiplexing system that encodes the data signal first. Then, fast Fourier transform (FFT) is used to modulate the signal to be transmitted, and the signal is received at the receiving end for demodulation and decoding. Such technology can effectively improve the relevant performance of communication systems, such as improving spectrum efficiency, improving system diversity performance, suppressing inter-carrier interference (ICI), reducing coding complexity, saving power consumption, and reducing data errors. Rate and so on. Because signal precoding technology has low complexity and easy to implement characteristics, it is widely used in orthogonal frequency division multiplexing systems to solve the problem that the orthogonal frequency division multiplexing system is susceptible to carrier frequency offset (CFO). .

In the current signal precoding method, it is assumed that the channel and the carrier frequency shift effect (CFO effect) remain unchanged when transmitting a plurality of data blocks. However, in a real environment, the CFO effect will change over time as the different data blocks are transmitted, resulting in a time variant of the channel. More specifically, when transmitting a large number of data blocks, the complex multiplicative distortion (CMD) is a time-varying distortion source that changes with the data block, which causes signal distortion and greatly reduces the system fault tolerance. Therefore, the assumptions based on the channel system invariance in the prior art are not in line with the actual situation when the carrier frequency offset exists. In other words, if the channel is not changed with time, the effect of the carrier frequency shift effect (CFO effect) caused by the complex multiplicative distortion (CMD) is ignored, and the signal precoding method developed is developed. It is not effective to overcome complex multiplicative distortion (CMD) and can only overcome inter-carrier interference (ICI).

Therefore, how to provide a signal precoding method in orthogonal frequency division multiplexing system to overcome carrier-to-interference (ICI) and complex multiplicative distortion (CMD) to effectively suppress carrier frequency offset (CFO) The impact of this is an urgent problem to be solved by those skilled in the art.

In view of the above-mentioned shortcomings of the prior art, it is an object of the present invention to provide a The signal precoding method, in addition to effectively suppressing the inter-carrier interference (ICI), can simultaneously estimate the complex multiplicative distortion (CMD) to help the receiving end suppress the complex multiplicative distortion effect, thereby effectively suppressing the carrier frequency offset. The effect of shifting (CFO) improves the performance of the system's fault tolerance.

To achieve the foregoing or other objects, the present invention provides a signal precoding method for applying to a transmitting end of an orthogonal frequency division multiplexing system, the method comprising: according to a chain precoding algorithm, Transmitting, by each of the plurality of signal symbols, a plurality of signal symbols into a plurality of precoding symbols; and processing the plurality of precoding symbols by using an orthogonal frequency division multiplexing algorithm to generate a plurality of transmission symbols And transmitting the plurality of transmission symbols through the antenna, and passing the multi-path channel, the plurality of transmission symbols are received by a receiving end.

The orthogonal frequency division multiplexing algorithm converts the plurality of pre-coded symbols from the frequency domain to the time domain by adding a trans-discrete Fourier transform, and adds the cyclic prefix to generate the plurality of transmission symbols. .

The plurality of signal symbols of each of the frames of the transmission signal form L signal blocks, and the index of the L signal blocks is 0 to L-1 indexes, and the L signal regions Each of the blocks contains N-2 signal symbols.

The signal block of the 0th index includes N-2 training symbols.

Each of the signal blocks of the first to L-1 indexes includes M pilot symbols and N-M-2 data symbols.

The M pilot symbols are located at the forefront of the N-M-2 data symbols.

The interlocking precoding algorithm sequentially uses a precoding matrix modified by a correlation of ( N -1) × ( N -2) and a degenerate Hadamard code matrix for the first to the Lth. - 1 index signal block is calculated to obtain the plurality of pre-coded symbols.

The degenerate Hada code matrix is a row in which a row of 1 is deleted and normalized in a Hada code matrix of size N × N.

Each of the rows of the degenerate Hadamard code matrix has a property of zero sum and orthogonality.

The chained precoding algorithm sequentially uses a precoding matrix modified by a correlation of size N × ( N - 2) and a degenerate Hadam code matrix, and the 0th to L-1 indexes are used. The signal block is calculated to obtain the plurality of pre-coded symbols.

Compared with the prior art, the signal precoding method of the present invention can not only effectively suppress inter-carrier interference (ICI), but also simultaneously estimate complex multiplicative distortion (CMD) to suppress complex multiplicative distortion (CMD). The influence of carrier frequency offset (CFO) prevents the orthogonality between subcarriers from being destroyed, which in turn improves the system fault tolerance and increases the overall performance of the system. In addition, since the signal precoding method of the present invention only needs an adder to implement degenerate Hada code encoding, the present invention can simplify the overall architecture of the system and reduce the complexity of the system design, and has the effect of cost saving.

10‧‧‧Orthogonal Frequency Division Multiplex System

101‧‧‧Transport

102‧‧‧Correct modification before the encoder

103‧‧‧Degraded Hadacode pre-encoder

104‧‧‧Orthogonal Frequency Division Multiplex Modulator

105‧‧‧ channel

106‧‧‧Orthogonal Frequency Division Multiplexer Demodulator

107‧‧‧ channel estimation

108‧‧‧Minimum mean square error frequency equalizer

109‧‧‧Degraded Hadacode decoder

110‧‧‧ Viterbi decoder

111‧‧‧ Receiver

41, 42, 43‧ ‧ formalized estimation error variability curve

S01~S03‧‧‧Steps

1 is a frame structure of an embodiment of the present invention; FIG. 2 is a flowchart of a signal precoding method according to the present invention; and FIG. 3 is an orthogonal frequency division multiplexing system using signal precoding of the present invention. The system architecture diagram; and FIG. 4 is a simulation result of the normalized estimation error variance graph of the time-variable complex multiplicative distortion of the signal precoding method of the present invention in different environments.

The present invention is applicable to a wireless communication system implementing orthogonal frequency division multiplexing, such as IEEE 802.11n standard, IEEE 802.16 series standard, 3G, 4G, long term evolution (LTE) or the like. While the features of the present invention can be incorporated into an integrated circuit (IC) or configured in a circuit that includes multiple interconnected components.

Please refer to FIG. 1 for a frame structure according to an embodiment of the present invention. In an Orthogonal Frequency-Division Multiplexing (OFDM) system, a transmitted signal transmitted by a transmitting end is composed of a plurality of frames. Each frame structure is as shown in Fig. 1, and the index of the block is indexed from 0th to L-1, and there are a total of L signal blocks, and each signal block Both have N-2 symbols, and their mathematical formulas can be expressed as follows:

The signal block of the 0th index can be used as a leading training block, which contains N-2 training symbols, and the mathematical formula is d ⁽⁰ )=[ t _n ; n Z _{n -2} ], where t _{n is a} plurality of known training symbols. The role of the leadership training block is the channel estimation of the receiver of the auxiliary orthogonal frequency division multiplexing system, enabling it to implement frequency domain equalization (FDE). The signal blocks of the first to L-1 indexes can be used as data blocks, and each index data block contains M pilot symbols and NM-2 data symbols. Data symbols, and the M pilot symbols are a plurality of known and identical pilot symbols, and the M pilot symbols are located at the forefront of the NM-2 data symbols. In other words, each signal block of the first to L-1 indexes starts with M pilot symbols, and then NM-2 data symbols to form N-2 symbols, so the following mathematics can be used. The formula says:

Please refer to FIG. 2 and FIG. 3, wherein FIG. 2 is a flowchart of the signal precoding method of the present invention, and the flow of the signal precoding method of the present invention is used for the transmission of the orthogonal frequency division multiplexing system 10 shown in FIG. In the terminal 101, a plurality of transmission symbols are transmitted to a receiving end 111. The step of the signal precoding method of the present invention includes: step S01, according to a chain precoding algorithm, encoding a plurality of signal symbols of each frame in a transmission signal into a plurality of precoding symbols; step S02 Processing the plurality of pre-coded symbols using an orthogonal frequency division multiplexing algorithm to generate a plurality of transmission symbols; and S03 transmits the plurality of transmission symbols through the antenna, and the plurality of transmission symbols are received by a receiving end via a multipath channel.

In step S01, the chain precoding algorithm sequentially uses a precoding matrix modified by a correlation of ( N -1) × ( N - 2) and a degenerate Hadamard code matrix, and the data area is used. The block (ie, the signal block of the first to L-1 indexes) is calculated to obtain the plurality of pre-coded symbols. The orthogonal frequency division multiplexing algorithm converts the plurality of pre-coded symbols from a frequency domain to a time domain by an inverse discrete Fourier transform. The plurality of transfer symbols are generated after the cyclic prefix is generated. Since the present invention performs the chained precoding algorithm before performing the orthogonal frequency division multiplexing algorithm, and the chained precoding algorithm also performs the correlation modification precoding matrix and the degenerate Hada code in sequence. The calculation of the matrix, and the calculation of the degenerate Hada code matrix, can estimate the complex multiplicative distortion (CMD) by receiving the leading pilot symbols in the front end of each signal block, and receiving When the terminal performs decoding, the complex multiplicative distortion is compensated to enable the transmission signal to be correctly restored.

In detail, in step S01, before the orthogonal frequency division multiplexing step 104 performs the orthogonal frequency division multiplexing step at the transmitting end 101 of the orthogonal frequency division multiplexing system 10, the correlation must be modified first. The pre-encoder 102 and the degraded Hada code pre-encoder 103 perform a chain pre-coding algorithm to encode a plurality of signal symbols of each frame in a transmitted signal into a plurality of pre-encoded symbols. . And this encoding is processed for each data block d ^{( l )} in each frame. d ^{( l )} is a linear precoding technique of N × ( N - 2 ) in - T _g - T _d /2 t -1 T <T _d / 2 transmission interval to generate a plurality of precoded symbols, i.e., , c ^{( l )} = Bd ^{( l )} , where T = T _d + T _g , T _d represents the required subinterval length for transmitting useful symbols, and T _g represents the transfer cycle prefix protector The required subinterval length of the cyclic-prefix guard symbols. For the convenience of calculation, N must be limited to N 4, and M +1 must be limited to the power of 2 of the integer 2, such as 2 ¹ , 2 ² , 2 ^{3 ,} etc. In detail, M +1 2, 4, 8... to simplify the calculation, get M 1,3,7..., that is, M may be a Mersenne number, but the invention is not limited thereto. And the matrix N × ( N -2) is generated by the correlation pre-encoder encoder 102 and the degenerate Hada code pre-encoder 103 combined by a concatenated precoder. And c ^{( l )} is a plurality of pre-coded symbols calculated as d ^{( l )} and matrix N × ( N -2) as described above. In calculation, because of l Z _L - Z ₁ , all data symbols Both are assumed to be independent and the zero mean of the same distribution, and at m Expected value in the case of Z _{n -2} - Z _M .

In step S02, i.e., in the orthogonal frequency division multiplexing modulator 104, the plurality of precoding symbols are processed using an orthogonal frequency division multiplexing step. First, the trans-discrete Fourier transform is used to transform the plurality of pre-encoded symbols from the frequency domain to the time domain, and the relevant mathematical expression is expressed as . After the plurality of pre-encoded symbols are modulated in the frequency domain to the time domain, the step of adding the cyclic prefix as the guard symbol is performed. After this orthogonal frequency division multiplexing step, a plurality of transmission symbols are generated. In one embodiment, the plurality of symbols transmitted signal waveforms to transmit a particular embodiment, the waveform of a transmission signal can be expressed as s ^{(l) (t).}

In step S03, the transmitting end 101 transmits the transmitted signal waveform s ^{( l )} ( t ) through the antenna to be received by the receiving end 111 via the channel 105. The channel 105 can be specifically a multi-path channel. This channel 105 is assumed to be quasi-static and varies with each frame. In addition, it is assumed that the length of the channel impulse response is less than the cyclic prefix protection symbol. Therefore, when the receiving end 111 removes the cyclic prefix, there is no problem of intersymbol interference in the adjacent block.

The algorithm for modifying the pre-encoder 102 and the degraded Hada code pre-encoder 103 will be described in detail below. The correlation modification pre-encoder 102 performs an operation on a precoding matrix modified by a correlation of ( N -1) × ( N - 2), which is expressed as , the normalization trip And m Z _M time; And m When Z _{N -2} - Z _M , where g = [ g ₀ , g ₁ ] ^t , | g ₀ | ² + | g ₁ | ² =1. In this operation, the M pilot symbols remain unchanged. However, the NM-2 data symbols are calculated by the correlation precoding of g , and the data symbols are given inter-carrier interference when receiving ( ICI) the ability to self-eliminate and allow data detection by the Viterbi decoder 110.

The calculation of the degraded Hada code pre-encoder 103 is performed by modifying the correlation before the operation of the encoder 102, and performing the calculation by a degenerate Hadamard code matrix. The so-called reduced Hadamard matrix is obtained by deleting the row of 1 in the Hada code matrix of size N × N and normalizing the remaining rows. The reason for this calculation is that all Each row is orthogonal and its sum is zero. Take the mathematical formula as an example. ,among them and , n , m is a binary extension and is at n ₁ , m ₁ Z ₂ . Since the degenerate Hada code matrix is composed of matrices of equal size, the multiplication can be effectively implemented as a series of additions, so that it can be easily implemented corresponding to the encoder and the decoder without requiring complicated multiplication operations. The mathematical formula of the degenerate Hada code matrix can be expressed as And must limit and satisfy the sum of rows to zero ( And the nature of the orthogonal row ( V ^t V =I _{N -1} ) to enable the receiving end 111 to time-variable complex-time multiplicative CMD (TCMD) to eliminate inter-carrier interference (ICI) The estimation is performed under the influence and can assist the Viterbi decoder 110 to compensate for the time-varying complex multiplicative distortion (TCMD) when decoding.

In an embodiment of the present invention, the chain precoding algorithm of the present invention is not only capable of calculating the data block (ie, the signal blocks of the first to L-1 indexes), and may also be 0. The leadership training blocks of the index signal blocks are included in the calculation. In other words, the chain precoding algorithm can sequentially use a precoding matrix of a size modified by N × ( N - 2) and a degenerate Hadam code matrix, and the 0th to L-1 indexes are The signal block is calculated. In this way, it can be avoided that the signal block of the 0th index must adopt the other codec mode to increase the system complexity. Therefore, the signal blocks of the 0th and 1st to 1st index can be encoded and decoded by the chain precoding algorithm of the present invention at the same time, or can be separately coded for the signal block of the 0th index. The decoding method, the signal block for the first to L-1 indexes, and the chain precoding algorithm of the present invention are used to calculate the plurality of precoding symbols. The following description continues with the case where the chain type precoding algorithm of the present invention is applied to the signal blocks of the first to the L-1 indexes, but the present invention is not limited thereto.

The demodulated received signal r ^{( l ) is} obtained by transmitting signal waveform s ^{( l )} ( t ) through channel 105 and demodulating by orthogonal frequency division multiplexing demodulator 106. The so-called demodulation in the orthogonal frequency division multiplexing system 10 is performed by trans-discrete Fourier transform (IDFT) using a trans fast Fourier transform (IFFT) circuit in the transmitting end 101, and this is reversed at the receiving end 111. operating. Briefly, the receiving end 111 operates in a reverse direction with respect to the transmitting end 101. If the system is not full duplex, the transmitting end 101 and the receiving end 111 will share some of the components. Therefore, in the orthogonal frequency division multiplexing demodulator 106, the transmitted signal waveform s ^{( l )} ( t ) is calculated by a discrete Fourier transform to obtain the transmitted signal waveform s ^{( l )} ( t) after receiving the demodulated signal r ^(l), the r ^(l) can be represented by the following equation, wherein =E- γ ( ε ;0)I _N :

ε is a normalized carrier frequency offset (CFO). In this embodiment, ε is limited to | ε |<0.5. When ε ≠ 0, carrier-to-interference (CFO)-induced inter-carrier interference (ICI) disturbs the original transmitted signal. φ ^{( l )} ( ε ) can be expressed as a time-varying complex multiplicative distortion (TCMD), and γ ( ε ; 0) is expressed as a non-time-varying complex multiplicative distortion (constant CMD). Minimum Average (least squares) estimation of the channel 107 to provide an approximate maximum likelihood (maximum-likelihood) α to the estimated minimum mean square error (minimizing the mean-square estimation error , MMSE) equalizer frequency 108, a frequency Domain equalization (FDE) can be operated by the minimum mean square error (MMSE) standard and using the equal channel matrix diag ( α ) to obtain an equalized , . The specific mathematical formula can be expressed as follows:

among them Included circularly symmetric complex Gaussian (CSCG) noise samples with zero mean and variance . due to Similar to ,among them , u _{n , m} = ψ (< m - n 〉 _N ) and n ≠ m , u _{n , n} =0, γ ( ε ; k) δ ( ε ) ψ (k), and . therefore Can be further expressed as Where δ ^{( l )} ( ε ) = φ ^{( l )} ( ε ) δ ( ε ):

In order to estimate the complex multiplicative distortion (TCMD) for time-varying, Input to the decoder 109 of the degraded Hada code for further calculation, and Expressed as . Since V ^t V =I _{N -1} , To produce And this It can be used as an input value for the Viterbi decoder 110. The mathematical operation steps of the time-varying complex multiplicative distortion (TCMD) estimation are described in detail below.

In one embodiment, it is assumed that the time-varying complex multiplicative distortion (TCMD) is estimated in the case where the channel flat fading and channel estimation are ideal. In this embodiment, the Can further Expressed as:

among them And it has independent and identical distribution, zero mean and variance Circular symmetric complex Gaussian (CSCG) noise samples. because = p and m Z _M , Has M symbols, so it can be represented To estimate β , so Can be further expressed as:

among them . Because the sum of each line in V is zero ( ), then you can Further simplified to:

Where Λ = ( β + jρηδ ^{( l )} ( ε ) / ( Nα )) = β (1 + jδ ( ε ) / ( Nγ ( ε ; 0))). For the estimation of the complex multiplicative distortion (TCMD) of time-varying, Is distinguished by the subvector log ₂ ( M +1) and is represented as ,among them The mathematical formula can be expressed as follows:

Where formula (2) a 2 ^k × 2 ^k real square matrix under. Using equations (1) and (2), you can get:

among them Since the Q ^{( k )} system is obliquely symmetric, all the sums of the real numbers in Q ^{( k )} are zero ( ), from the formula (3) means:

In equation (4), after encoding and decoding using the degenerate Hadamard code matrix, inter-carrier interference (ICI) has been completely removed, so that it is possible to avoid inter-carrier inter-carrier variation when estimating the complex multiplicative distortion TCMD. Interference (ICI) is affected. When N >> ε , γ ( ε ; 0 ) / ( 1 / Nδ ( ε )) = cot ( π ε / N ) is close to infinity, at this point, Λ β , Equation (4) can be further expressed as Equation (5):

among them Since z ^{( l )} contains independent and identically distributed circular symmetric complex Gaussian (CSCG) noise samples, the approximation model of equation (5) can obtain complex multiplicative distortion (TCMD) for time-varying β The approximate maximum likelihood is estimated and formulated into And the estimated value of TCMD Can be expressed as . And that The Viterbi decoder 110 can be provided for TCMD compensation.

Please also refer to Fig. 4 for the simulation result of the normalized estimation error variance curve of the time-varying complex multiplicative distortion of the signal precoding method of the present invention in different environments. Assume that the number of subcarriers N is 256, the number of cyclic prefix guards N _g is 32, and the pilot symbol M is seven, and is the child of Quadrature phase-shift keying (QPSK). The carrier modulation mode is used to plot the graph. The normalized estimated error variance curves 41, 42, and 43 in Fig. 4 have four different normalized carrier frequency offsets ε, respectively . In the present embodiment, by the above mathematical analysis, The normalized estimated error variation in the additive white Gaussian noise (AWGN) channel environment can be specifically presented in the normalized estimated error variation curve 41 in FIG. The normalized estimated error variability curves 42 and 43 are computer-generated simulation results. The normalized estimated error variability curves 42 and 43 are simulated in a multipath Rayleigh fading channel environment. Next, and the normalized estimated error variability curve 42 is the delay spread of the channel T _RMS = (3.2 / N ) T _d , channel impulse response (CIR) length V = 32 The simulation parameter; the normalized estimated error variability curve 43 is the delay spread of the channel T _RMS = (1.6 / N ) T _d , channel impulse response (CIR) length V = 16 analog parameters. Observing the normalized estimation error variation curve 41, 42, 43 can be seen, the computerized simulation of the multipath Rayleigh fading channel environment, the normalized estimation error variation curve 42, 43, its trend and mining The above mathematical formula and the normalized estimated error variance curve 41 in the additive white Gaussian noise (AWGN) channel environment are approximately identical. Therefore, it can be proved that the signal precoding method of the present invention can be used in a practically more complicated environment, and can also be combined with the aforementioned time-varying complex multiplication in the case where the channel flat fading and channel estimation are ideal. The result of the estimation of sexual distortion (TCMD). However, the simulation result is an embodiment of the present invention, and the present invention is not limited thereto.

The above steps are performed on the receiving end 111 for demodulation, decoding, etc., Since each signal block in the data block of the transmitted signal, the M pilot symbols are located at the forefront of each signal block, and before the correlation modification, the encoder 102 and the degraded Hada code are placed. In the calculation of the encoder 103, the M pilot symbols remain unchanged, and the remaining NM-2 data symbols have inter-carrier interference at the time of reception after using the degenerate Hadamard code matrix. ICI) The ability to self-eliminate, as evidenced by the derivation of the aforementioned decoding mathematical formula. Therefore, in the signal precoding method of the present invention, the combination of the specific positions of the M pilot symbols and the degenerate Hada code matrix can be performed on the degraded Hadamard code matrix after the transmission signal passes through the multipath channel due to inter-carrier interference (ICI). After being encoded and decoded, it can be completely removed, so that the effects of inter-carrier interference (ICI) can be considered, and complex multiplicative distortion (CMD) can be directly estimated to provide a subsequent Viterbi decoder 110. It can compensate for complex multiplicative distortion (CMD), and thus effectively improve the performance of the system fault tolerance. In addition, since the degenerate Hada code matrix only needs to be added, the signal precoding method of the present invention does not need a multiplier, which can reduce the system complexity, achieve the purpose of easy implementation, and has the effect of cost saving.

The above is only the preferred embodiment of the present invention, and is not intended to limit the scope of the technical content of the present invention. The technical contents of the present invention are broadly defined in the following claims. Any technical entity or method performed by any other person that is identical to, or equivalent to, the ones defined in the scope of the claims below will be considered to be included in the scope of the invention.

S01~S03‧‧‧Steps

Claims (10)

A signal precoding method is applied to one of the orthogonal frequency division multiplexing systems, the method comprising the following steps: according to a chain precoding algorithm, a plurality of frames in a transmission signal The signal symbol is encoded into a plurality of precoding symbols, wherein the chain precoding algorithm sequentially performs a correlation using a correlation modified precoding matrix and a degenerate Hadam code matrix to obtain the plurality of precodings. Encoding symbol; processing the plurality of precoding symbols using an orthogonal frequency division multiplexing algorithm to generate a plurality of transmission symbols; and transmitting the plurality of transmission symbols through the antenna to transmit via a multipath channel The plurality of transmission symbols are received by a receiving end. The signal precoding method according to claim 1, wherein the orthogonal frequency division multiplexing algorithm converts the plurality of precoding symbols from a frequency domain to a time domain by using a trans-discrete Fourier transform. And adding the cyclic prefix to generate the plurality of transmission symbols. The signal precoding method according to claim 1, wherein the plurality of signal symbols of each frame in the transmission signal form L signal blocks, and the index of the L signal blocks is 0 to L-1 indexes, and each of the L signal blocks includes N-2 signal symbols. The signal precoding method according to claim 3, wherein the signal block of the 0th index includes N-2 training symbols. The signal precoding method as described in claim 3, wherein Each of the signal blocks of the first to L-1 indexes includes M pilot symbols and N-M-2 data symbols. The signal precoding method according to claim 5, wherein the M pilot symbols are located at the forefront of the N-M-2 data symbols. The signal precoding method according to claim 6, wherein the chain precoding algorithm sequentially uses a precoding matrix modified by a correlation of ( N -1) × ( N - 2) And the degenerate Hada code matrix, the signal blocks of the first to L-1 indexes are calculated to obtain the plurality of pre-encoded symbols. The signal precoding method according to claim 7, wherein the degenerate Hada code matrix is to delete the row of 1 in a Hada code matrix of size N × N and normalize the remaining pedestrians. . The signal precoding method of claim 8, wherein each row of the degenerate Hadamard code matrix has a property of zero sum and orthogonality. The signal precoding method according to claim 6, wherein the chain precoding algorithm sequentially uses a precoding matrix modified by a correlation of size N × ( N - 2) and the degenerate The Hadard code matrix calculates the signal blocks of the 0th to L-1th indexes to obtain the plurality of precoding symbols.