TWI423629B - A method of reducing peak-to-average power ratio for an ofdm systems - Google Patents

Description

Method for reducing peak-to-average power ratio of orthogonal frequency division multiplexing system

The present invention relates to a method for reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system, and more particularly to a method for reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system without sideband information.

Orthogonal Frequency Division Multiplexing (OFDM) is a very popular technology in the field of wireless communication in recent years. Because the orthogonal frequency division multiplexing technology has high data transmission rate, good resistance to channel attenuation and easy to be used. It is implemented on hardware and has been adopted by many wireless communication systems, such as IEEE 802.11 a/g/n Wireless Local Area Networks (WLAN), IEEE 802.16 Worldwide Interoperability for Microwave. Access, WiMAX), Digital Video Broadcasting (DVB), and Digital Audio Broadcasting (DAB).

However, in the orthogonal frequency division multiplexing system, since the subcarrier superposition is likely to generate an excessive peak-to-average power ratio (PAPR), the signal passes through the power amplifier before transmission. The signal whose peak-to-average power ratio is too high easily reaches the nonlinear amplification region of the power amplifier and is nonlinearly amplified, resulting in in-band distortion and out-of-band Radiation of the signal. The bit error rate (BER) of the receiving end is greatly increased.

Although the signal can be forced to operate in the linear amplification region of the power amplifier by reducing the signal power proportionally, this approach reduces power efficiency. In order to reduce the peak-to-average power ratio without reducing the power efficiency, many conventional methods have been proposed, such as: Clipping, Coding, Tone Reservation (TR), Selective Mapping ( Selected Mapping (SLM), Partial Transmit Sequences (PTS), and Active Constellation Extension (ACE). Among them, the selective mapping method can provide a good peak-to-average power ratio and keep the original signal undistorted. However, this method must additionally transmit Side Information and its computational complexity is high. Furthermore, transmitting the sideband information will cause the system's data transmission volume to decrease, and if the receiving end decodes the sideband information, the bit error rate will increase significantly. To resolve this issue, you can add error correction code protection to the sideband information. However, this solution will seriously reduce the data transmission rate and reduce the bandwidth usage efficiency.

In order to solve the problem of transmission sideband information of the selective mapping method, different candidate signals (Candidate Signals) can be represented by different subcarrier powers, for example, disclosed in IEEE Trans. Wireless Commun., vol. 8, no. 7, pp. 3320-3325, Jul. 2009, another conventional method of "Selected mapping without side information for PAPR reduction in OFDM", which utilizes direct amplification of power at several locations of subcarriers to represent different candidate signals. However, this method reduces the power of certain subcarriers and increases the bit error rate of the system.

Another conventional method for solving the problem of high computational complexity of the selective mapping method [disclosed in IEEE Trans. Signal Process., vol. 58, no. 5, pp. 2916-2921, May 2010, "Novel low-complexity" SLM schemes for PAPR reduction in OFDM systems] reveals a low-complexity selective mapping method that uses three features of Inverse Fast Fourier Transform (IFFT) to reduce complexity, as described below:

Feature one:

Time domain vector Equal to the N point circular convolution (Circular Convolution), as shown in the following formula (a):

Wherein, the transmission signal X is a vector of N×1, and the vectors of U disjoint N×1 are added, , Is an N × 1 vector, u = 0, 1, ..., U-1, The kth element in can be expressed as [ k ]={ X [ k ], k ; 0 , otherwise , ={ u + v ‧ U | v =0,1,..., N/U -1}, for a vector on the time domain after IFFT; An annular convolution operation for point N; The mth element in the system is as shown in the following formula (b):

Feature 2:

Equal to subvector And coefficient The result of the multiplication is as shown in the following formula (c):

among them, a subvector of N/U×1; =exp( j 2π uq/U ), q=1, 2,..., U-1.

Feature three:

When the domain signal is cyclically shifted (Cyclic Shift), it will be equivalent to the phase rotation of the subcarrier of the original frequency domain signal showing the progressive rotation angle effect, and will not change the amplitude of the original signal, such as the following formulas (d) and (e) ) shown:

So when X = [ X [0], X [1],..., X [ N -1]] and U=4, four equally spaced signals are obtained. , , and , as shown in the following formulas (f), (g), (h) and (i):

Through the formula (a), the time domain signal can be expressed as the following formula (j):

and , , and It can be known from formula (b), which respectively represents the following formulas (k), (l), (m) and (n):

here, , , and They are respectively vectors with a code length of N. Each vector can be subdivided into four sub-vectors with a code length of N/4. The value of the starting point of each sub-vector belongs to {±1, ± j } and the values of other points Both are 0. Therefore, x [ n ] and The computational complexity of the [ n ] circular convolution operation can be greatly reduced. Known by Fast Fourier Transform (FFT) operation , respectively, as expressed in the following formulas (o), (p), (q), and (r):

From the above formulas (o), (p), (q) and (r), it can be seen that the signals are equally spaced in the frequency domain, and the time domain signal can be equivalently circularly convoluted with a sequence, such as the following formula. (s) shows:

Therefore, all the separated signals can be obtained by only one inverse fast Fourier transform, which greatly reduces the number of inverse fast Fourier transforms. In addition, by feature 2 [ n ] represents the following formula (t):

and [ n ] After doing a cyclic shift of N/4, the following formula (u) can be obtained:

Known by the formula (s) It can be composed of four 1×N/4 matrices a _{u 1} , a _{u 2} , a _{u 3} and a _{u 4} , as shown in the following formula (t):

among them, and The value is as shown in the following formula (u):

From (t) and (u), we can see In time, it is only necessary to find the value of the first N/4, and the value of the latter 3N/4 can be obtained by multiplying the value of the previous N/4 by a complex value.

Obtain four signals on the time domain , , and After that, if different candidate signals are to be generated, the first signal will be Fixed, second to fourth signal , and The cyclic shifts are respectively limited to 0 to N/4-1 to obtain a plurality of different candidate signals, and the candidate signals whose peaks have the smallest average power ratio are transmitted. However, this method needs to transmit sideband information, so it has the problem that the aforementioned bandwidth is inefficiently used.

Since the above method has the disadvantages of transmitting sideband information, high bit error rate or high computational complexity, it needs to be improved to provide a sideband information without transmission, low bit error rate and computational complexity. A low method of reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system.

It is an object of the present invention to improve the above disadvantages to provide a method for reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system that does not require transmission of sideband information.

A second object of the present invention is to provide a method for reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system with a low bit error rate.

A method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system includes the steps of: a pilot signal placement step, a signal conversion step, a candidate signal generation step, and a transmission signal selection step. The pilot signal placement step starts with the first subcarrier position as a starting point, and the plurality of pilot signals are placed at equal intervals in the plurality of frequency domain signals, wherein the frequency domain signal includes a plurality of subcarrier frequency domain signals, each of the subcarriers The carrier frequency domain signal has several subcarriers and their locations. The signal conversion step converts the frequency domain signal into a time domain signal and filters out the number of subcarrier time domain signals. The candidate signal generating step is to input the sub-carrier time domain signal to generate a candidate displacement set, and the sub-carrier time domain signal except the first path is cyclically shifted, and is added to the first sub-carrier time domain signal to generate a number. Candidate signals. The transmission signal selection step selects a candidate signal having a minimum peak-to-average power ratio by using the time domain signal and the candidate signal, and sets the transmission signal as a transmission signal. Thereby, in the pilot signal assisted orthogonal frequency division multiplexing system, the peak-to-average power ratio is reduced by not transmitting the sideband information, and the judgment error rate is low.

A method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system includes the steps of: a pilot signal placement step, a signal conversion step, a group sequence generation step, a group candidate signal generation step, and a transmission signal Select the steps. The pilot signal placement step starts with the first subcarrier position as a starting point, and the plurality of pilot signals are placed at equal intervals in the plurality of frequency domain signals, wherein the frequency domain signal includes a plurality of subcarrier frequency domain signals, each of the subcarriers The carrier frequency domain signal has several subcarriers and their locations. The signal conversion step converts the frequency domain signal into a time domain signal and filters out the number of subcarrier time domain signals. The clustering sequence generation step is to generate a number of parity group Zadoff-Chu sequences. The grouping candidate signal generating step is to generate a Zadoff-Chu subcarrier time domain signal of the Zadoff-Chu sequence and the first subcarrier time domain signal respectively, and a candidate displacement set W and the first The subcarrier time domain signal outside the path performs a parity group cyclic shift to generate a plurality of cyclic shift subcarrier time domain signals, and the Zadoff-Chu subcarrier time domain signals are respectively added to the cyclic shift subcarrier time domain signals to generate Several candidate signals. The transmission signal selection step selects a candidate signal having a minimum peak-to-average power ratio from the candidate signals and sets it as a transmission signal transmission. Therefore, in the pilot signal-assisted orthogonal frequency division multiplexing system, the peak-to-average power ratio is reduced by not transmitting the sideband information, the judgment error rate is low, and fewer candidate signals are generated, which can reduce the complexity of the search. degree.

The above and other objects, features and advantages of the present invention will become more <RTIgt;

The 〝 coupling described in the full text of the present invention may substantially include a physical line connection or a wireless connection, and the detailed connection manner is understood by those skilled in the art.

The 〝Zadoff-Chu sequence 所述 described in the full text of the present invention has the following characteristics:

1. The Zadoff-Chu sequence is a sequence of equal amplitude autocorrelation Waveform (CAZAC);

2. The root index (Root Index) u of the Zadoff-Chu sequence is relatively prime with its sequence code length N _zc ;

3. Under the same root index u, the same Zadoff-Chu sequence has perfect periodic autocorrelation (Periodic Autocorrelation);

4. Under different root indexes u, the periodic crosscorrelation of the Zadoff-Chu sequences of different root indexes is low;

5. If the sequence code length N _zc is a prime number, then it has N _zc -1 periodic cross-correlation is 1/ the sequence of. It is understood by those of ordinary skill in the art.

Please refer to FIG. 1 , which is a schematic diagram of a transmission end architecture of a first embodiment of a method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to the present invention. It is a structure based on the selective mapping method. A transmitting device 1 includes: a pilot signal placing unit 11, a signal converting unit 12, a candidate signal generating unit 13, and a transmission signal selecting unit 14. The signal conversion unit 12 is coupled to the pilot signal placement unit 11 and the transmission signal selection unit 14. The candidate signal generation unit 13 is coupled to the signal conversion unit 12 and the transmission signal selection unit 14.

Among them, the pilot signal placement unit 11 will have several pilot signals (Pilot Signals) Equally placed in a frequency domain signal X , the frequency domain signal X includes a number of subcarrier frequency domain signals (Subcarrier Frequency Domain Signals) , U is a power of 2, and U≧2 (2 or 4 only), the subcarrier frequency domain signal Has several subcarriers and their locations The signal conversion unit 12 converts the frequency domain signal X into a time domain signal (Time Domain Signal) x and filters out the subcarrier time domain signal (Subcarrier Time Domain Signals). , U is a power of 2, and U ≧ 2 (only 2 or 4); the candidate signal generating unit 13 inputs the time-domain signal of the number of sub-carriers , generating a candidate displacement set (Candidate Shift Set) W, for the number of subcarrier time domain signals First subcarrier time domain signal Subcarrier time domain signal Performing a cyclic shift (Cyclic Shift) with the first subcarrier time domain signal Adding a candidate signal (Candidate Signals) y ; the transmission signal selecting unit 14 selects a Peak-to-Average Power Ratio (PAPR) by using the time domain signal x and the candidate signal y . The signal is set to a transmission signal r transmission. In this way, the peak-to-average power ratio can be reduced by the Side Information method in the Pilot Aided Orthogonal Frequency Division Multiplex System. In this embodiment, the U system is set to 4. The time domain signal x can be divided into four signals , , and , the four-way signal , , and Switch to quad-band signal , , and After the frequency domain signal , , and Each is a signal with a first interval (interval of 4), as shown in the following equations (1), (2), (3), (4):

Wherein, the frequency domain signal X = [ X [0], X [1], ..., X [ N -1]].

The pilot signal placement unit 11 is connected to the four-channel frequency domain signal , , and The pilot signal P _a is placed at equal intervals with the first subcarrier position L ₀ as a starting point. In detail, since the cyclic shift of the time domain signal is equivalent to the phase rotation of the effect of the progressive rotation angle of the subcarrier signal in the frequency domain, as shown in the following equations (5) and (6):

among them, Is an N × 1 vector, u = 0, 1, ..., U-1, The kth element in .

Assume that the pilot signal The number is p, the number of subcarriers is N, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , and the other pilot signals P ₁ , P ₂ , ..., P _{p -1} are equally spaced. Placing the remaining subcarrier locations Where the intervals are N / p +1 and p ² N. Please refer to FIG. 2, which is a schematic diagram of the pilot signal placement according to the first embodiment of the present invention. Among them, the pilot signal The number p is 4, and the 4-channel subcarrier frequency domain signal The number of subcarriers N is 16, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , the second pilot signal P _{1 is} placed at the sixth subcarrier position L ₅ , and the third pilot signal P _{1 is} placed. At the 11th subcarrier position L ₁₀ , and so on. To ensure that each pilot signal The interval is equally spaced and the pilot signal of each channel The same number and equal intervals, the pilot signal of the same road Is placed at equal intervals of 4×(N/p+1) subcarrier positions. If all subcarriers are used in the frequency domain signal, then these pilot signals are used. They are placed at equal intervals of (N/p+1) subcarrier positions.

The signal conversion unit 12 is provided with an inverse fast Fourier transformer 121 and a filter 122. The inverse fast Fourier transformer 121 is coupled to the filter 122. The inverse fast Fourier transformer 121 converts the frequency domain signal X into the time domain signal x , and transmits the signal to the filter 122 to filter the digital subcarrier time domain signal. . In detail, the inverse fast Fourier transformer 121 performs an inverse fast Fourier transform (IFFT) on the frequency domain signal X to generate the time domain signal x , as shown in the following equation (7):

Since the signals are equally spaced in the frequency domain, the time domain signal can be equivalently circularly convolved with a sequence, as shown in the following equation (8):

Therefore the time domain signal x and a filtering sequence As a circular convolution operation, the subcarrier time domain signal can be filtered out , as shown in the following formulas (9) to (13):

The candidate signal generating unit 13 is provided with a cyclic shifter 131 and an adder 132. The cyclic shifter 131 is coupled to the adder 132. The candidate signal generating unit 13 inputs the digital subcarrier time domain signal The cyclic shifter 131 generates the candidate displacement set W for the time domain subcarrier time domain signal First subcarrier time domain signal Subcarrier time domain signal Perform cyclic shift to generate digital subcarrier time domain signals Time domain signal with the first subcarrier The adder 132 is added to generate the candidate signal y .

The cyclic shifter 131 is the time domain signal of the digital subcarrier First subcarrier time domain signal Subcarrier time domain signal A cyclic shift is performed to generate the candidate displacement set W. In detail, in the case where the receiving channel is known, since the noise distribution on each subcarrier in the frequency domain can be regarded as an independent and identically distributed Gaussian distribution, the cyclic displacement of each channel of the transmitting end can be Equation (14) to decide:

Where p is the pilot signal Quantity; q {1,2,3} means the first, second or third way; d _q ( k ) is the kth pilot signal of the qth road The position of N is the number of subcarrier time domain signals; R _q and X _q are the receiving signals and pilot signals of the qth channel, respectively.

However, if all possible situations are considered, if Sq {0,1,..., N /4-1} and q {1, 2, 3}, the decision error rate (DER) at the receiving end will be high. In this embodiment, the number of cyclic shifts is selected such that the error rate is smaller than the original and lower than a certain value. If the number of displacements taken by each path is R, then S _q , q {1, 2, 3} are as shown in the following formulas (15), (16), and (17):

In this embodiment, in order to make each subcarrier time domain signal , and The judgment error rate is reduced, in the choice of s _{q , r} , q {1,2,3}, r =0,1,2,..., R -1, the cyclic displacements are assigned to the frequency domain, in the pilot signal The phase rotation indicated, select the displacement that meets the condition, as shown in the following equation (18):

Where q {1,2,3}, r =0,1,2,..., R -1;1 is the distance between any two displacements in the frequency domain constellation points. After obtaining the cyclic displacements of each channel, the displacement sets C ₁ , C ₂ and C _{3 of} each path can be known, as shown in the following equations (19), (20) and (21):

Suppose a candidate set of displacements W is M sets of displacements ( S _0,1 , S _0,2 , S _0,3 ), ( S _1,1 , S _1,2 , S _1,3 ), ... and ( A set consisting of S _{M -1,1} , S _{M -1,2} , S _{M -1,3} ) is as shown in the following equation (22):

Where M is the number of candidate signals. Since the number of elements of the displacement sets C ₁ , C ₂ and C ₃ of each path is R, each candidate displacement set w can have up to R ³ different displacement combinations. In the selective mapping method architecture, the displacement of the M group as the candidate signal can be arbitrarily selected from the R ³ combinations, and in order to further reduce the selection error rate, the displacement with the largest mutual difference can be selected as the candidate displacement. In this embodiment, from the displacement sets C ₁ , C ₂ and C ₃ of the respective paths, S _m,1 , S _m,2 , S _m,3 , m =0,1 of which the M elements do not intersect are selected. ..., M -1, since the number of elements of the displacement sets C ₁ , C ₂ and C ₃ of each path is R, at most R the most diverse displacement sets can be taken out, in order to avoid the signal after the displacement is completed. The peak-to-average power ratio is higher than the original signal, so the R candidate displacements need to include {0, 0, 0}. After obtaining the M sets of elements in the candidate displacement set W, how to determine the value of the candidate displacement is as shown in the following equation (23):

Where p is the pilot signal Quantity; q {1,2,3} means the first, second or third way; N is the number of subcarriers; d _q ( k ) is the kth pilot signal of the qth road The position of N is the number of subcarrier time domain signals; R _q and X _q are the receiving signals and pilot signals of the qth channel, respectively.

However, to obtain the candidate displacement with the largest difference in the M group, the displacement sets C _q , q {1, 2, 3} must have at least M elements, so when determining the displacement set C _q , calculate b distances, as shown in the following equation (24):

It can be seen from the above formula (24) that the computational complexity is very large. When determining the displacement set C _q of each path, the computational complexity is reduced and the judgment error rate is not lowered too much. Therefore, when determining the candidate displacement set W, The number of elements of the group of M elements can be reduced, for example, the candidate displacements in which the M elements are different from each other are reduced to the candidate displacements in which the M elements and the two elements are different. In this embodiment, the number of elements of the group of elements of the M group element is 2, and there are two elements between the candidate displacements of the candidate displacement set W. In each path displacement set C _q , only Each element can produce M qualified elements, so the above equation (24) is rewritten as shown in the following equation (25):

Calculating the set of displacements C _q , q In {1,2,3} After the elements, the calculation of the candidate displacement set W assumes three 1× The elements of the matrices c ₁ , c ₂ and c _{3 are} the values of the displacement sets C ₁ , C ₂ and C ₃ of the respective paths, as shown in the following equations (26), (27) and (28):

The above formula (28) can be rewritten as one Matrix c ₄ , indicating 1× Column vector c _4,1 , c _4,2 ,... and And each of the column vectors c _4,1 , c _4,2 ,... The cyclone displacement is made to the right by c _{3 of the} above formula (28), as shown in the following equation (29):

Each of the column vectors c _4,1 , c _4,2 ,... The system is arranged in a sequence of 1× The matrix c _{5 is} as shown in the following equation (30):

The matrices c ₁ , c ₂ and c ₅ can be calculated to obtain a 3× The matrix c _{6 is} as shown in the following equation (31):

among them, For the Kronecker product. The value of each of the displacement sets in the candidate displacement set W is the value of each row in the matrix c ₆ for the sub-carrier time domain signal Perform cyclic shift to generate digital subcarrier time domain signals .

The transmission signal selection unit 14 calculates and compares the peak-to-average power ratio of the time domain signal x and the candidate signal y by using the time domain signal x and the candidate signal y , and selects the signal with the minimum peak-to-average power ratio. And set to transmit the signal r .

Referring to FIG. 3, it is a schematic diagram of the receiving end architecture of the first embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to the present invention. The receiver device 2 includes a fast Fourier transformer 21, a channel estimation unit 22, a displacement selection unit 23, and a displacement compensation unit 24. The channel estimation unit 22 is coupled to the fast Fourier transformer 21 The displacement selecting unit 23 and the displacement compensating unit 24 are coupled to the channel estimating unit 22 and the displacement compensating unit 24.

The fast Fourier transformer 21 inputs the transmission signal r , performs Fast Fourier Transform (FFT), and converts the transmission signal r into a frequency domain signal R , as shown in the following formula (32):

The channel estimation unit 22 inputs the frequency domain signal R to the pilot signal Get a frequency domain signal The displacement selecting unit 23 obtains the displacement set according to the above formula (23) The displacement compensation unit 24 is based on the displacement set And the signal output by the channel estimation unit 22 , calculate a restore signal .

Referring to FIG. 4, it is a flowchart of a first embodiment of a method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to the present invention, the steps of which include a pilot signal placing step S1 and a signal. The converting step S2, a candidate signal generating step S3 and a transmission signal selecting step S4. The pilot signal placing step S1 takes a first subcarrier position L ₀ as a starting point and sets a number of pilot signals. Placed in a frequency domain signal X at equal intervals, the frequency domain signal X includes several subcarrier frequency domain signals , U is a power of 2, and U ≧ 2 (limited to 2 or 4), the number of subcarrier frequency domain signals Has several subcarriers and their locations The signal conversion step S2 converts the frequency domain signal X into a time domain signal x and filters out the digital subcarrier time domain signal , U is a power of 2, and U ≧ 2 (limited to 2 or 4); the cyclic shift step S3 is input to the digital subcarrier time domain signal , generating a candidate displacement set (Candidate Shift Set) W, for the number of subcarrier time domain signals Subcarrier time domain signal Subcarrier time domain signal Cyclic Shift, and then the first subcarrier time domain signal Adding to generate a plurality of candidate signals (Candidate Signals) y ; the transmission signal selection step S4 selects a minimum peak-to-average power ratio with the time domain signal x and the candidate signal y (Peak-to-Average Power The candidate signal of Ratio, PAPR) is set to be transmitted by a transmission signal r . In this embodiment, the U system is set to 4.

The pilot signal placing step S1 is based on the four-way frequency domain signal , , and In the first subcarrier position As the starting point, the pilot signal is placed at equal intervals. . In detail, since the cyclic shift of the time domain signal is equivalent to the phase rotation of the effect of the progressive rotation angle of the subcarrier signal in the frequency domain, as shown in the above equations (5) and (6), the pilot signal is assumed. The number is p, the subcarrier frequency domain signal The number of subcarriers is N, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , and the remaining pilot signals P ₁ , P ₂ , ..., P _{p -1} respectively place the remaining subcarriers at equal intervals. position Where the intervals are N / p +1 and p ² N. Please refer to Figure 2 again, where the pilot signal The number p is 4, and the 4-channel subcarrier frequency domain signal The number of subcarriers N is 16, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , the second pilot signal P _{1 is} placed at the 6th subcarrier position L ₅ , and so on. To ensure that each pilot signal The interval is equally spaced and the pilot signal of each channel The same number and equal intervals, the pilot signal of the same road Is placed at equal intervals of 4×(N/p+1) subcarrier positions. If all subcarriers are used in the frequency domain signal, then these pilot signals are used. They are placed at equal intervals of (N/p+1) subcarrier positions.

The signal conversion step S2 converts the frequency domain signal X into the time domain signal x , and filters out the digital subcarrier time domain signal. . In detail, the frequency domain signal X performs an inverse fast Fourier transform to generate the time domain signal x as shown in the above equation (7). Since the signals are equally spaced in the frequency domain, the time domain signal can be equivalently circularly convoluted with a sequence, as shown in the above equation. Therefore, the time domain signal x and a filtering sequence As a circular convolution operation, the subcarrier time domain signal can be filtered out , as shown in the above formulas (9) to (13).

The candidate signal generating step S3 is to input the digital subcarrier time domain signal And generating the candidate displacement set W for the number of subcarrier time domain signals First subcarrier time domain signal Time domain signal outside the subcarrier Cyclic displacement, generating several cyclic shift subcarrier time domain signals Time domain signal with the first subcarrier Adding, the candidate signal y is generated.

In detail, in the case where the receiving channel is known, since the noise distribution on each subcarrier in the frequency domain can be regarded as an independent and identically distributed Gaussian distribution, the cyclic displacement of each channel of the transmitting end can be obtained by Determine by equation (14). However, if all possible situations are considered, if S _q {0,1,..., N /4-1} and q {1, 2, 3}, the decision error rate (DER) at the receiving end will be high. In this embodiment, the number of cyclic shifts is selected such that the error rate is smaller than the original and lower than a certain value. If the number of displacements taken by each path is R, then S _q , q {1, 2, 3} is as shown in the above formulas (15), (16), and (17). In this embodiment, in order to make each subcarrier time domain signal , and The judgment error rate is reduced, in selecting s _q,r , q {1,2,3}, r =0,1,2,..., R -1, the cyclic displacements are assigned to the frequency domain, in the pilot signal The phase rotation indicated is selected to match the conditional displacement as shown in the above equation (18). The displacement sets C ₁ , C ₂ and C ₃ of the respective paths can be obtained by obtaining the cyclic displacements of the respective paths, as shown in the following equations (19), (20) and (21).

Suppose a candidate set of displacements W is M sets of displacements ( S _0,1 , S _0,2 , S _0,3 ), ( S _1,1 , S _1,2 , S _1,3 ), ... and ( A set consisting of S _{M -1,1} , S _{M -1,2} , S _{M -1,3} ) is as shown in the above formula (22). Since the number of elements of the displacement sets C ₁ , C ₂ and C ₃ of each path is R, each candidate displacement set w can have up to R ³ different displacement combinations. In the selective mapping method architecture, the displacement of the M group as the candidate signal can be arbitrarily selected from the R ³ combinations, and in order to further reduce the selection error rate, the displacement with the largest mutual difference can be selected as the candidate displacement. In this embodiment, from the displacement sets C ₁ , C ₂ and C ₃ of the respective paths, S _{m , 1} , S _{m , 2} , S _{m , 3} , m =0, ₁ in which the M elements do not intersect are selected. ..., M -1, since the number of elements of the displacement sets C ₁ , C ₂ and C ₃ of each path is R, at most R the most diverse displacement sets can be taken out, in order to avoid the signal after the displacement is completed. The peak-to-average power ratio is higher than the original signal, so the R candidate displacements need to include {0, 0, 0}. After obtaining the M sets of elements in the candidate displacement set W, how to determine the value of the candidate displacement is as shown in the above equation (23).

However, to obtain the candidate displacement with the largest difference in the M group, the displacement sets C _q , q {1, 2, 3} requires at least M sets of elements. Therefore, when determining the displacement set C _q of each path, b distances need to be calculated, as shown in the above equation (24). It can be seen from the above formula (24) that the computational complexity is very large. When determining the displacement set C _q of each path, the computational complexity is reduced and the judgment error rate is not lowered too much. Therefore, when determining the candidate displacement set W, The number of elements of the group of M elements can be reduced, for example, the candidate displacements in which the M elements are different from each other are reduced to the candidate displacements in which the M elements and the two elements are different. In this embodiment, the number of elements of the group of elements of the M group element is 2, and there are two elements between the candidate displacements of the candidate displacement set W. In each path displacement set C _q , only The elements can generate M qualified elements, so the above formula (24) is rewritten as shown in the above formula (25).

Calculating the set of displacements C _q , q In {1,2,3} After the elements, the calculation of the candidate displacement set W assumes three 1× The elements of the matrices c ₁ , c ₂ and c _{3 are} the values of the displacement sets C ₁ , C ₂ and C ₃ of the respective paths, as shown in the above equations (26), (27) and (28). The above formula (28) can be rewritten as one Matrix c ₄ , indicating 1× Column vector c _4,1 , c _4,2 ,... and And each of the column vectors c _4,1 , c _4,2 ,... The cyclone shift is made to the right by c _{3 of the} above formula (28), as shown in the above formula (29).

Each of the column vectors c _4,1 , c _4,2 ,... The system is arranged in a sequence of 1× The matrix c _{5 is} as shown in the above formula (30). The matrices c ₁ , c ₂ and c ₅ can be calculated to obtain a 3× The matrix c _{6 is} as shown in the above formula (31). The value of each of the displacement sets in the candidate displacement set W is the value of each row in the matrix c ₆ for the sub-carrier time domain signal Perform cyclic shift to generate digital subcarrier time domain signals .

The transmission signal selecting step S4 lines to the time domain signal x and the candidate signal y, calculated and compared, and the peak value of the time domain signal x to the candidate signal y is on the average power ratio, select which has a minimum peak signal-to-average power ratio of And set to transmit the signal r .

Therefore, the present invention utilizes the cyclic shift of the time domain signal, which is equivalent to the phase rotation of the subcarrier signal in the frequency domain exhibiting a progressive rotation angle effect, and estimates the cyclic displacement value by using the pilot signal in the frequency domain. In the pilot-assisted orthogonal frequency division multiplexing system, the peak-to-average power ratio is reduced by not transmitting sideband information, and the judgment error rate is low.

In addition, please refer to Figure 1 again, because the first frequency domain signal In the first embodiment of the present invention, the transmission is performed without cyclic shift, so the first frequency domain signal Pilot signal It is not affected by the cyclic shift in the time domain. Therefore, the characteristics of the Zadoff-Chu sequence can be used to group the signals before the displacement, so that the displacement of the same group is demodulated at the receiving end, and the error rate is further reduced.

Referring to FIG. 5, it is a schematic diagram of a transmission end architecture of a second embodiment of a method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to the present invention. The second embodiment and the first embodiment The differences are as described later. The transmitting device 3 includes a pilot signal placing unit 31, a signal converting unit 32, a plurality of candidate signal generating units 33, and a transmission signal selecting unit 34. The pilot signal placement unit 31 is coupled to the signal conversion unit 32. The plurality of candidate signal generation units 33 are coupled to the signal conversion unit 32 and the transmission signal selection unit 34.

Wherein, the pilot signal placement unit 31 is to have several pilot signals Equally placed in a frequency domain signal X , the frequency domain signal X includes several subcarrier frequency domain signals , U is a power of 2, and U≧2 (ie, 2, 4, 8, 32, ..., and so on), the subcarrier frequency domain signal Has several subcarriers and their locations The signal conversion unit 32 converts the frequency domain signal X into a time domain signal x and filters out the number of subcarrier time domain signals. , U is a power of 2, and U ≧ 2 (ie, 2, 4, 8, 32, ..., and so on); each of the candidate signal generating units 33 inputs the digital subcarrier time domain signal To a Zadoff-Chu sequence Time domain signal with the number of subcarriers First subcarrier time domain signal Generating a parity packet of Zadoff-Chu subcarrier time domain signal And a candidate displacement set W and the number of subcarrier time domain signals First subcarrier time domain signal Subcarrier time domain signal Perform parity and group cyclic shift to generate several cyclic shift subcarrier time domain signals , the Zadoff-Chu subcarrier time domain signal Time domain signal with the cyclic shift subcarrier Add together to generate a candidate signal , K is the number of groups; the transmission signal selection unit 34 is the candidate signal Select a signal with a minimum peak-to-average power ratio and set it as a transmission signal r . In this way, the peak-to-average power ratio can be reduced in the pilot signal-assisted orthogonal frequency division multiplexing system without passing the sideband information. In this embodiment, the U system is set to 4.

The pilot signal placement unit 31 is connected to four frequency domain signals. , , and The pilot signal is placed at equal intervals with the first subcarrier position L ₀ as a starting point , where the pilot signal Placed in the first frequency domain signal The value is 0. In detail, assume the pilot signal The number is p, the number of subcarrier frequency domain signals is N, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , and the other pilot signals P ₁ , P ₂ , ..., P _{p -1} respectively Placed at the same subcarrier position at equal intervals Where the intervals are N / p +1 and p ² N. Due to the pilot signal Placed in the first frequency domain signal Each of the subcarrier positions has a value of 0 for the frequency domain signal When changing to a time domain signal, the pilot signal Place the position in the Zadoff-Chu sequence . Please refer to Figure 6, where the pilot signal The number p is 4, and the four-channel subcarrier frequency domain signal The number of subcarriers N is 16, the first pilot signal P ₀ =0 is placed at the first subcarrier position L ₀ , the second pilot signal P _{1 is} placed at the 6th subcarrier position L ₅ , and so on.

The signal conversion unit 32 is provided with an inverse fast Fourier transformer 321 and a filter 322. The inverse fast Fourier transformer 321 is coupled to the filter 322. The inverse fast Fourier converter 321 converts the frequency domain signal X into the time domain signal x , and transmits the signal to the filter 322 to filter the digital subcarrier time domain signal. .

Each of the candidate signal generating units 33 is provided with a Zadoff-Chu sequence generator 331, a first adder 332, a cyclic shifter 333 and a second adder 334. The Zadoff-Chu sequence generator 331 is coupled to the The first adder 332 is coupled to the first adder 332 and the cyclic shifter 333. The candidate signal generating unit 33 inputs the digital subcarrier time domain signal The Zadoff-Chu sequence generator 331 generates the Zadoff-Chu sequence The first adder 332 is the time domain signal of the digital subcarrier First subcarrier time domain signal And the Zadoff-Chu sequence Adding and outputting the Zadoff-Chu subcarrier time domain signal of the parity group The cyclic shifter 333 generates the candidate displacement set W in a parity group manner, and uses the digital subcarrier time domain signal First subcarrier time domain signal Subcarrier time domain signal Performing a parity group cyclic shift with the candidate displacement set W to generate the cyclic shift subcarrier time domain signal The second adder 334 is a time domain signal of the Zadoff-Chu subcarrier. And the cyclic shift subcarrier time domain signal Add together to generate the candidate signal . Where the Zadoff-Chu sequence The code length N _zc is p/4, as shown in the following equation (33):

Where N _{zc is} the Zadoff-Chu sequence Code length; u is the Zadoff-Chu sequence The root index (Root Index).

Since the N/4 points are too close in the frequency domain, the signal determination interval (Decision Region) is too small, and the minimum distance between the constellation points becomes small, which increases the error rate and makes the judgment error rate too high. . Therefore, if N/4 cyclic shifts are used for parity grouping, an even cyclic shift set C _e and an odd cyclic shift set C _o can be obtained, as shown in the following equations (34) and (35):

C _e ={0,2,4,..., N /4-2} (34)

C _o ={1,3,5,..., N /4-1} (35)

Therefore, S ₁ , S ₂ , S ₃ There are a total of ²³ combinations of { C _o , C _e }, as shown in the following equations (36) and (37):

Among them, the pilot signal of the first road at the transmitting end There are 8 sets of Zadoff-Chu sequences with a code length of p/4 to represent these 8 combinations. Thereby, the Zadoff-Chu sequence is placed on the pilot signal of the first road. The position, the cyclic displacement representing the 2nd, 3rd, and 4th signals belongs to the even cyclic displacement set C _e or the odd cyclic displacement set C _o .

Since a Zadoff-Chu sequence needs to have its code length N _zc and its root index u, when the code length N _zc and the root index u are mutually prime, the generated Zadoff-Chu sequence has a perfect periodic autocorrelation. And low periodic cross-correlation. Therefore, the number of subcarriers N is a power of 2; the number of subcarriers N must be used by the pilot signal The number p is divisible; p is a multiple of 4 and p/4 is also a power of 2 to make the pilot signals of each route The number is the same. Wherein, if the code length N _zc of the Zadoff-Chu sequence is p/4, the root index of the Zadoff-Chu sequence u having p/8 odd numbers is mutually prime with the code length N _zc , and the p/8 Zadoff- The Chu sequence can obtain p/4 Zadoff-Chu sequences with perfect periodic autocorrelation and periodic cross-correlation by cyclic shift. Therefore, if the code length N _zc is p/4, the Zadoff-Chu sequence The quantity nzc is as shown in the following equation (38):

Wherein, in the case of parity grouping, the pilot signal The number is p≧16.

Therefore, as shown in FIG. 5, each of the candidate signal generating units 33 generates a Zadoff-Chu sequence that is known by the system and is located at the first pilot signal P ₀ by the Zadoff-Chu sequence generator 331. Time domain signal, by the first adder 332, the digital subcarrier time domain signal in the time domain First subcarrier time domain signal And the Zadoff-Chu sequence Adding and outputting the Zadoff-Chu subcarrier time domain signal of the parity group The cyclic shifter 333 generates the candidate displacement set W in a parity group manner, and generates a time domain signal of the Zadoff-Chu subcarrier in the parity group. Then, the digital subcarrier time domain signal First subcarrier time domain signal Subcarrier time domain signal Performing (N/4) ^three possible parity group cyclic shifts with the candidate displacement set W, generating the cyclic shift subcarrier time domain signal The second adder 334 is the Zadoff-Chu subcarrier time domain signal And the cyclic shift subcarrier time domain signal Add together to generate the candidate signal And the transmission signal selection unit 34 is coupled to the candidate signal. Select a signal with a minimum peak-to-average power ratio and set it as a transmission signal r .

Please refer to FIG. 7, which is a schematic diagram of the receiving end architecture of the second embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to the present invention. The receiving device 4 includes a fast Fourier converter 41, a channel estimating unit 42, a correlation detecting unit 43, a displacement selecting unit 44, and a displacement compensating unit 45. The channel estimating unit 42 is The fast Fourier transformer 41, the correlation detecting unit 43, the displacement selecting unit 44, and the displacement compensating unit 45 are coupled to the correlation detecting unit 43 and the displacement compensating unit 45. .

The fast Fourier transformer 41 inputs the transmission signal r , performs fast Fourier transform, and converts the transmission signal r into a frequency domain signal R. The channel estimation unit 42 inputs the frequency domain signal R to the pilot signal. Get a frequency domain signal . The correlation detecting unit 43 determines the Zadoff-Chu sequence of the first path signal. The sub-group signal represented by ), which of the odd-even group combinations of k =0, 1, ..., 7 is Zadoff-Chu sequence , as shown in the following formula (39):

among them, ) is a Zadoff-Chu sequence for demodulation; p is the pilot signal The number; m is the mth pilot signal for each path; d ( m ) is the position of the mth pilot signal; Z ^{( k ) *} is the Zadoff-Chu sequence of the first signal Take a total of The signal signal is received for the first channel of the completion channel. The displacement selecting unit 44 outputs the correlation detecting unit 43 After that, you can use The signal judgment interval corresponding to each channel demodulates the displacement of each path, as shown in the following formula (40):

Where p is the pilot signal Number; N is the number of subcarriers; m is the mth pilot signal of each path; d _q ( m ) is the position of the mth pilot signal in the qth path; R _q is the receiving signal of the qth channel; X _q It is the pilot signal of the qth road; H _q is the frequency response of the qth channel. The displacement compensation unit 24 is based on the displacement set And the signal output by the channel estimation unit 42 , calculate a restore signal .

Please refer to FIG. 8 , which is a flowchart of a second embodiment of a method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to the present invention. The steps of the method include a pilot signal placing step S1 ′, and a method. a signal conversion step S2, a group candidate signal generating step S3', a transmission signal selecting step S4, and a grouping sequence generating step S5, wherein the grouping sequence generating step S5 is any time before the grouping candidate signal generating step S3' is performed. get on. The pilot signal placing step S1' starts with the first subcarrier position L _{0 and} starts several pilot signals. Placed in a frequency domain signal X at equal intervals, the frequency domain signal X includes several subcarrier frequency domain signals , U is a power of 2, and U ≧ 2 (ie 2, 4, 8, 32, ..., and so on), each subcarrier frequency domain signal Has several subcarriers and their locations The signal conversion step S2 converts the frequency domain signal X into a time domain signal x and filters out the number of subcarrier time domain signals. , U is a power of 2, and U ≧ 2 (ie, 2, 4, 8, 32, ..., and so on); the clustering sequence generates step S5 to generate a number of odd-even groups of Zadoff-Chu sequences The candidate signal generating step S3' is the Zadoff-Chu sequence Time domain signal with the number of subcarriers First subcarrier time domain signal Generating time-domain signals of Zadoff-Chu subcarriers with several parity groups And generating a candidate displacement set W and the subcarrier time domain signal in a parity group manner First subcarrier time domain signal Subcarrier time domain signal Perform parity and group cyclic shift to generate several cyclic shift subcarrier time domain signals , the subcarrier time domain signal Time domain signal with the cyclic shift subcarrier Add together to generate several candidate signals , K is the number of groups; the transmission signal selection step S4 is from the candidate signal Medium, select a signal with a minimum peak-to-average power ratio and set it as a transmission signal r . In this way, the peak-to-average power ratio can be reduced in the pilot signal-assisted orthogonal frequency division multiplexing system without passing the sideband information. In this embodiment, the U system is set to 4.

The pilot signal placing step S1' is tied to the four-way frequency domain signal , , and The pilot signal is placed at equal intervals starting from the first subcarrier position L ₀ , where the pilot signal Placed in the first frequency domain signal The value is 0. In detail, assume the pilot signal The number of the subcarriers is N, the first pilot signal P _{0 is} placed at the first subcarrier position L ₀ , and the other pilot signals P ₁ , P ₂ , ..., P _{p -1} are respectively Placing the remaining subcarrier positions at equal intervals Where the intervals are N / p +1 and p ² N. Due to the pilot signal Placed in the first frequency domain signal Each of the subcarrier positions has a value of 0 for the frequency domain signal When it is converted to a time domain signal, it can be used in the pilot signal. The corresponding position is filled in the Zadoff-Chu sequence . Please refer to Figure 6, where the pilot signal The number p is 4, and the 4-channel subcarrier frequency domain signal The number of subcarriers N is 16, the first pilot signal P ₀ =0 is placed at the first subcarrier position L ₀ , the second pilot signal P _{1 is} placed at the 6th subcarrier position L ₅ , and so on.

The signal conversion step S2 converts the frequency domain signal X into the time domain signal x , and filters out the number of subcarrier time domain signals. . In detail, the frequency domain signal X performs an inverse fast Fourier transform to generate the time domain signal x as shown in the above equation (7). Since the signals are equally spaced in the frequency domain, the time domain signal can be equivalently circularly convoluted with a sequence, as shown in the above equation. Therefore, the time domain signal x and a filtering sequence As a circular convolution operation, the subcarrier time domain signal can be filtered out , as shown in the above formulas (9) to (13).

The grouping sequence generating step S5 is to generate the Zadoff-Chu sequence The code length N _zc is p/4, as shown in the above formula (33). Since the N/4 points are too close in the frequency domain, the signal determination interval (Decision Region) is too small, and the minimum distance between the constellation points becomes small, which increases the error rate and makes the judgment error rate too high. . Therefore, if N/4 cyclic shifts are used for parity grouping, an even cyclic shift set C _e and an odd cyclic shift set C _o can be obtained, as shown in the above equations (34) and (35), respectively.

Therefore, S ₁ , S ₂ , S ₃ There are a total of ²³ combinations of { C _o , C _e }, as shown in the above equations (36) and (37), in which the pilot signal of the first road at the transmitting end There are 8 sets of Zadoff-Chu sequences with a code length of p/4 to represent these 8 combinations. Thereby, the Zadoff-Chu sequence is placed at the pilot position P _{a of} the first path, so that the cyclic displacement of the second, third and fourth signals belongs to the even cyclic displacement set C _e or the odd cyclic displacement set. C _o .

Since a Zadoff-Chu sequence needs to have its code length N _zc and its root index u, when the code length N _zc and the root index u are mutually prime, the generated Zadoff-Chu sequence has a perfect periodic autocorrelation. And low periodic cross-correlation. Therefore, the number of subcarriers N is a power of 2 (ie, 2, 4, 8, 32, ..., and so on); the number of subcarriers N must be the pilot signal The number p is divisible; p is a multiple of 4 and p/4 is also a power of 2 (ie 2, 4, 8, 32, ..., and so on), so that the pilot signals of each way The number is the same. Wherein, if the code length N _zc of the Zadoff-Chu sequence is p/4, the root index of the Zadoff-Chu sequence u having p/8 odd numbers is mutually prime with the code length N _zc , and the p/8 Zadoff- The Chu sequence can obtain p/4 Zadoff-Chu sequences with perfect periodic autocorrelation and periodic cross-correlation by cyclic shift. Therefore, if the code length N _zc is p/4, the Zadoff-Chu sequence The number nzc is as shown in the above formula (38), wherein in the case of a parity group, the pilot signal The number is p≧16.

The candidate signal generating step S3' is the Zadoff-Chu sequence. Fill in the digital subcarrier time domain signal separately First subcarrier time domain signal Pilot signal Position, generating a Zadoff-Chu subcarrier time domain signal of the parity group And generating the candidate displacement set w and the number of subcarrier time domain signals in a parity group manner First subcarrier time domain signal Subcarrier time domain signal Perform parity and group cyclic shift to generate several cyclic shift subcarrier time domain signals , the Zadoff-Chu subcarrier time domain signal Time-domain signal with the cyclic shift subcarrier Add together to generate the candidate signal . Among them, it is judged which Zadoff-Chu sequence is used as shown in the above formula (39), and is obtained. After that, you can use The signal judgment interval corresponding to each channel uses the following equation to demodulate the displacement of each path, as shown in the above equation (40).

The transmission signal selecting step S4 lines to the time domain signal x and the candidate signal y, calculated and compared, and the peak value of the time domain signal x to the candidate signal y is on the average power ratio, select which has a minimum peak signal-to-average power ratio of And set to transmit the signal r .

Thereby, a Zadoff-Chu sequence of the system known and located at the first pilot signal P ₀ is generated. Time domain signal, the subcarrier time domain signal in the time domain And the Zadoff-Chu sequence Adding to generate the subcarrier time domain signal In addition, the candidate displacement set w is generated, plus the Zadoff-Chu sequence Time domain signal after the time domain signal Performing (N/4) ³ possible cyclic shifts with the candidate displacement set w , generating the subcarrier time domain signal The subcarrier time domain signal And the subcarrier time domain signal Add together to generate the candidate signal And the transmission signal selection unit 34 is coupled to the candidate signal. Select a signal with a minimum peak-to-average power ratio and set the transmission signal r to transmit.

Therefore, the present invention utilizes the cyclic shift of the time domain signal, which is equivalent to the phase rotation characteristic of the subcarrier signal in the frequency domain exhibiting a progressive rotation angle effect, and estimates the cyclic displacement value by the pilot signal in the frequency domain, and is in time Several Zadoff-Chu sequences on the domain Doing parity grouping, in the pilot signal-assisted orthogonal frequency division multiplexing system, the peak-to-average power ratio is reduced by not transmitting sideband information, the judgment error rate is low, and fewer candidate signals are generated, which can lower the search. The complexity.

The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system is that the pilot signal is equally spaced at the frequency domain signal X , and the cyclic shift of the time domain signal is equivalent to the subcarrier in the frequency domain. The signal exhibits the characteristics of the phase rotation of the progressive rotation angle effect. Therefore, in the pilot-assisted orthogonal frequency division multiplexing system, the peak-to-average power ratio can be reduced without passing the side information.

The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system is that the pilot signal is equally spaced at the frequency domain signal X , and the cyclic shift of the time domain signal is equivalent to the subcarrier in the frequency domain. The signal exhibits the characteristics of the phase rotation of the progressive rotation angle effect, so that the effect of determining the error rate can be achieved in the pilot signal assisted orthogonal frequency division multiplexing system.

The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system is that the pilot signal is equally spaced at the frequency domain signal X , and the cyclic shift of the time domain signal is equivalent to the subcarrier in the frequency domain. The signal exhibits the phase rotation characteristic of the progressive rotation angle effect, so that the search complexity can be reduced in the pilot signal assisted orthogonal frequency division multiplexing system.

While the invention has been described in connection with the preferred embodiments described above, it is not intended to limit the scope of the invention. The technical scope of the invention is protected, and therefore the scope of the invention is defined by the scope of the appended claims.

[this invention]

1. . . Transmitter device

11. . . Pilot signal placement unit

12. . . Signal conversion unit

121. . . Anti-fast Fourier converter

122. . . filter

13. . . Candidate signal generation unit

131. . . Cyclic displacement

132. . . Adder

14. . . Transmission signal selection unit

2. . . Receiving device

twenty one. . . Fast Fourier transform unit

twenty two. . . Channel estimation unit

twenty three. . . Displacement selection unit

twenty four. . . Displacement compensation unit

3. . . Transmitter device

31. . . Pilot signal placement unit

32. . . Signal conversion unit

321. . . Anti-fast Fourier converter

322. . . filter

33. . . Candidate signal generation unit

331. . . Zadoff-Chu Sequence Generator

332. . . First adder

333. . . Cyclic displacement

334. . . Second adder

34. . . Transmission signal selection unit

4. . . Receiving device

41. . . Fast Fourier transform unit

42. . . Channel estimation unit

43. . . Correlation detection unit

44. . . Displacement selection unit

45. . . Displacement compensation unit

1 is a schematic diagram of a transmission end architecture of a first embodiment of a method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to the present invention.

Fig. 2 is a schematic diagram showing the navigation signal placement of the first embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to the present invention.

Figure 3 is a schematic diagram showing the architecture of the receiving end of the first embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system.

Figure 4 is a flow chart showing a first embodiment of the method for reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system.

Figure 5 is a schematic diagram showing the architecture of the transmitting end of the second embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system.

Fig. 6 is a schematic diagram showing the navigation signal placement of the second embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to the present invention.

Figure 7 is a schematic diagram showing the architecture of the receiving end of the second embodiment of the method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system.

Figure 8 is a flow chart of a second embodiment of the method of reducing the peak-to-average power ratio of an orthogonal frequency division multiplexing system in accordance with the present invention.

1. . . Transmitter device

11. . . Pilot signal placement unit

12. . . Signal conversion unit

121. . . Anti-fast Fourier converter

122. . . filter

13. . . Candidate signal generation unit

131. . . Cyclic displacement

132. . . Adder

14. . . Transmission signal selection unit

Claims (10)

A method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system, the method comprising: a pilot signal placing step of placing a plurality of pilot signals at equal intervals in a plurality of frequency domain signals, the frequency The domain signal includes a plurality of subcarrier frequency domain signals, each of the subcarrier frequency domain signals having a plurality of subcarriers and a position thereof, wherein when the plurality of pilot signals are placed in the frequency domain signals at the intervals, The first subcarrier position is a starting point; a signal conversion step is to convert the frequency domain signal into a time domain signal and filter out the plurality of subcarrier time domain signals; and a candidate signal generating step is to input the digital subcarrier time domain signal Generating a candidate displacement set for cyclically shifting subcarrier time domain signals other than the first subcarrier time domain signal of the time domain signal of the plurality of subcarriers, and adding the first subcarrier time domain signal to generate a plurality of a candidate signal; and a transmission signal selection step, using the time domain signal and the candidate signal, selecting a candidate signal having a minimum peak-to-average power ratio and setting it as a transmission No transfer. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to claim 1 of the patent application scope, wherein the number of subcarrier frequency domain signals is U, U is a power of 2 and U≧2 The number of the subcarriers is N, and the number of the pilot signals is p, and the intervals are N/p+1, and p ² N. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to claim 1 of the patent application scope, wherein the method for generating the candidate displacement set is as follows: , s _{q, r} is the rth candidate displacement of the qth road, and l is the distance between the two displacements in the frequency domain constellation points. A method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system according to claim 1, wherein the candidate displacement set is a set of M displacement sets, and the value of each of the displacement sets is The value of each row in a matrix c ₆ is as follows: , For Kronecker, , s _q,r is the rth candidate displacement of the qth path, q {1,2,3}, r =0,1,2,..., M -1. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to claim 1 of the patent application scope, wherein the value of the candidate displacement is determined as follows: , p is the number of the pilot signal, q {1,2,3} means the first, second or third way, d _q ( k ) is the position of the kth pilot signal of the qth path, N is the number of subcarrier time domain signals, R _q and X _q They are the receiving signal and pilot signal of the qth road. A method for reducing a peak-to-average power ratio of an orthogonal frequency division multiplexing system, the method comprising: a pilot signal placing step of placing a plurality of pilot signals at equal intervals in a plurality of frequency domain signals, the frequency The domain signal includes a plurality of subcarrier frequency domain signals, each of the subcarrier frequency domain signals having a plurality of subcarriers and a position thereof, wherein when the plurality of pilot signals are placed in the frequency domain signals at the intervals, The first subcarrier position is a starting point; a signal conversion step is to convert the frequency domain signal into a time domain signal and filter out the number of subcarrier time domain signals; a grouping sequence generating step is to generate a number of odd and even groups of Zadoff- a Chu sequence; a cluster candidate signal generating step, which generates a Zadoff-Chu subcarrier time domain signal of the Zadoff-Chu sequence and the first subcarrier time domain signal of the digital subcarrier time domain signal respectively, and Generating a candidate displacement set and a subcarrier time domain signal other than the first subcarrier time domain signal of the time domain signal in a parity group manner to perform a parity group cyclic shift Generating a plurality of cyclically shifted subcarrier time domain signals, wherein the Zadoff-Chu subcarrier time domain signals are respectively added to the cyclically shifted subcarrier time domain signals to generate a plurality of candidate signals; and a transmission signal selecting step is performed from the Among the candidate signals, a candidate signal having a minimum peak-to-average power ratio is selected and set as a transmission signal transmission. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to Item 6 of the patent application scope, wherein the number of subcarrier time domain signals is U, U is a power of 2 and U≧2 The number of the subcarriers is N, and the number of the pilot signals is p, and the intervals are N/p+1, and p ² N , the pilot signal is placed in the first frequency domain signal value of 0. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to claim 6 of the patent application scope, wherein the parity group cyclic displacement is divided into an even cyclic displacement set C _e and an odd cyclic displacement set C _o , They are as follows: C _e ={0,2,4,..., N /4-2}, C _o ={1,3,5,..., N /4-1}; and parity There are 8 kinds of permutation and combination of grouping cyclic displacement, as shown in the following formula: The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to Item 6 of the patent application scope, wherein the method for determining the Zadoff-Chu sequence is as follows: , p is the number of the pilot signals, m is the mth pilot signal of each road, d ( m ) is the position of the mth pilot signal, and Z ^{( k )*} is the Zadoff-Chu sequence of the first signal. , The signal signal is received for the first channel of the completion channel. The method for reducing the peak-to-average power ratio of the orthogonal frequency division multiplexing system according to item 6 of the patent application scope, wherein the manner of determining the displacement is as follows: , p is the number of pilot signals, N is the number of subcarriers, m is the mth pilot signal of each way, d _q ( m ) is the position of the mth pilot signal in the qth road, and R _q is the qth road The receiving signal, X _q is the pilot signal of the qth road, and H _q is the frequency response of the qth channel.