AFDM peak-to-average ratio suppression method based on pilot frequency position optimization and pre-chirp

Abstract

The invention discloses an AFDM peak-to-average ratio suppression method based on pilot frequency position optimization and pre-chirp. The AFDM peak-to-average ratio suppression method comprises the following steps: executing pilot frequency and data overall cyclic shift in a DAF domain for a pilot frequency frame; performing IDAFT on the processed pilot frequency frame to convert the pilot frequency frame into a time domain signal, calculating the PAPR of the time domain signal, and finding out a group of arrangement modes with the lowest PAPR as output; candidate pre-chirp parameters provided by the data frame are selected, and IDAFT is executed to convert the candidate pre-chirp parameters into time domain signals; performing amplitude limiting processing on a time domain signal of the data frame, performing optimization by using an iterative ACE algorithm, calculating the PAPR of the updated signal, and selecting a group with the lowest PAPR as the output of the data frame; and a receiving end respectively demodulates the pilot frequency frame and the data frame by using a maximum likelihood-minimum mean square error detector. According to the invention, the PAPR of the transmission signal of the AFDM system is effectively suppressed, and the transmission reliability of the system is improved.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to an AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp. Background Technology

[0002] Sixth-generation (6G) networks are expected to support reliable, high-data-rate, and low-latency data transmission to meet the demands of high-mobility scenarios such as unmanned aerial vehicles (UAVs) and high-speed trains. However, the high relative mobility in these scenarios introduces a significant Doppler effect, resulting in time-varying channel characteristics. In such communication environments, the classic multi-carrier modulation technique Orthogonal Frequency Division Multiplexing (OFDM) suffers from inter-carrier interference (ICI) due to Doppler frequency shift. To address the frequency spread effect in linear time-varying (LTV) channels, researchers have proposed several novel waveform design schemes, among which affine radio frequency division multiplexing (AFDM) is considered a promising modulation technique. This technique can adapt to channel characteristics by optimizing waveform parameters, thereby achieving channel path separation in the discrete affine Fourier (DAF) domain. This provides significant advantages in resisting delay spread and Doppler interference, improving system stability and spectral efficiency.

[0003] Similar to OFDM, AFDM is also a multi-carrier modulation technique, and therefore both face the problem of excessively high peak-to-average power ratio (PAPR), which leads to reduced power amplifier efficiency. However, research on the PAPR problem in AFDM systems is still in its early stages. Currently, some researchers have proposed a pre-chirp selection (GPS) algorithm to reduce PAPR, but this algorithm requires the transmission of additional sideband information during decoding at the receiver, and this additional overhead correspondingly reduces the system's spectral efficiency (PAPR Reduction With Pre-Chirp Selection for Affine Frequency Division Multiplexing). Researchers have also proposed a normalization method... The law companding transformation method is used for AFDM systems (PAPRinAFDMUpperBoundandReductionWithNormalizedLawCompanding), with the companding coefficient... As the peak-to-average power ratio (PAPR) increases, the PAPR decreases, but the bit error rate (BER) performance also declines accordingly. Furthermore, there is currently no PAPR reduction method specifically for AFDM pilot frames. Therefore, there is an urgent need for a solution that can reduce the PAPR of AFDM systems without sacrificing bit error rate and spectral efficiency. Summary of the Invention

[0004] To address the issue that existing techniques for reducing peak-to-average power ratio (PAPR) often result in reduced bit accuracy (BER) or decreased spectral efficiency, this invention proposes an AFDM PAPR suppression method based on pilot position optimization and pre-chirp combined with ACE. This method leverages the inherent characteristics of AFDM modulation—namely, the freedom of pilot position movement in the DAF domain and relatively relaxed pre-chirp parameter constraints—to reduce the system's PAPR, thereby improving the system's power amplifier efficiency.

[0005] The present invention is achieved by at least one of the following technical solutions.

[0006] The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp processes pilot frames and data frames at the transmitter, including the following steps: Step 1: For the pilot frame, perform cyclic shift processing on the pilot and data as a whole in the discrete affine Fourier domain according to the set step size, perform inverse discrete affine Fourier transform on the processed pilot frame to convert it into a time domain signal, and calculate the peak-to-average power ratio (PAPR) of the signal. Step 2: Traverse all possible shift cases and repeat the calculation process in Step 1. After the traversal is completed, compare the PAPR values ​​of all records and output the set of signal data with the smallest PAPR as the pilot frame output. Step 3: Select candidate pre-chirp parameters provided by the data frame and perform inverse discrete affine Fourier transform to convert them into time-domain signals; Step 4: Perform amplitude limiting on the time-domain signal of the data frame and optimize it using the iterative SGP-ACE algorithm. Calculate the PAPR of the updated signal and select the set with the lowest PAPR as the output of the data frame. Step 5: The pilot frame output and the data frame output are arranged together and output as the transmitted signal. The receiver demodulates the pilot frame and the data frame using a maximum likelihood-minimum mean square error detector to obtain the final demodulated signal.

[0007] Further, in step 1, the data frame and pilot frame are processed separately at the transmitting end. For the pilot frame, the pilot and data are arranged in the DAF domain with an index of 0. That is, the pilot signal is placed at the beginning of the pilot frame, and Q zeros are placed on the left and right sides of the pilot signal as guard intervals. The remaining space is filled with data symbols, which can be described by the following formula:

[0008] in, It is the index value of the DAF domain signal. Indicates the number of zero-protection intervals. It is the frame length of an AFDM. It is the maximum delay value of the path. Therefore The normalized path maximum Doppler value; Indicates pilot symbol, Indicates the first A data symbol, Signals representing the DAF domain The A symbol.

[0009] Furthermore, in step 1, the pilot frame after each cyclic shift process is converted into a time-domain signal by performing an inverse discrete affine Fourier transform (IDAFT), and the peak-to-average power ratio (PAPR) of the signal is calculated.

[0010] Furthermore, in step 3, the selected pre-chirp parameter is an irrational number or a decimal less than 1 / 2N. It is the frame length of an AFDM.

[0011] Furthermore, step 4 includes the following steps: Record the positional information of the original constellation modulation symbols using the selected chirp parameters. IDAFT is performed on the position information of the original constellation modulation symbols to obtain the initialized time-domain signal. , the iteration parameters Initialize to 0; For the first The time-domain signal of the next iteration The peak-shaving portion of the signal is obtained by amplitude limiting. ; Using the peak-shaving portion of the signal The updated signal and iterative parameters yield a new time-domain signal. ; The SGP-ACE algorithm terminates when the PAPR value reaches an acceptable threshold, the maximum number of iterations is reached, or the step size parameter becomes negative. The PAPR at this point is recorded as follows. , To use the first Peak-to-average power ratio of signals using the SGP-ACE algorithm with pre-chirped parameters; Choose the chirp parameter with the lowest PAPR. The associated signals are output as data frames.

[0012] Furthermore, after the receiving end receives the signal, the pilot frame and the data frame are processed separately, including demodulation of the pilot frame and the data frame using a maximum likelihood-minimum mean square error detector.

[0013] Furthermore, after receiving the signal, the receiver performs channel estimation and equalization on the pilot frame, as follows: First, the maximum value of the received signal is indexed as... The pilot index is denoted as ,but , The possible spread region of the pilot signal is determined by its maximum value. It is determined by the cyclic shift step size A defined set of allowed pilot indices, The path is The normalized path maximum Doppler value, where Q is the number of zero-guard intervals. It is the frame length of an AFDM; Then, channel estimation is performed for each possible pilot location, and finally the signal is demodulated using a maximum likelihood-minimum mean square error detector.

[0014] Furthermore, for the data frames of the received signal, the channel information estimated from the pilot frames and all pre-chirp parameters are used to construct the DAF domain channel matrix, and the signal is demodulated using a maximum likelihood-minimum mean square error detector.

[0015] A computer device according to the present invention includes a memory and a processor, the memory being electrically connected to the processor, the memory storing a computer program, which, when executed by the processor, causes the processor to implement the method described herein.

[0016] The present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor implements the method described herein.

[0017] Compared with the prior art, the present invention has the following advantages and effects: This invention proposes for the first time a method to reduce PAPR of AFDM waveform pilot frames, changing the position of the pilot in the DAF domain, and using the guard interval of the pilot frame and the pilot diffusion law after passing through the dual-dispersion channel to perform correct demodulation at the receiving end, providing a solution with practical application value in real-world scenarios.

[0018] Secondly, by providing pre-chirp parameter selection for the AFDM waveform combined with an active constellation point extension method to reduce the PAPR of the data frame, the degrees of freedom of the waveform parameters are fully utilized. At the receiver, the data is processed using a maximum likelihood-minimum mean square error detector. Neither of these methods will impair the system's BER, and since side information does not need to be propagated, the system's spectral efficiency will not be reduced.

[0019] Therefore, this invention not only reduces the PAPR of the transmitted signal, but also provides a new direction for the optimization of AFDM systems by flexibly changing the pilot position and pre-chirp parameters. Attached Figure Description

[0020] Figure 1 The flowchart of the AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp provided in this application is shown.

[0021] Figure 2 This is a schematic diagram of the pilot frame transmission and reception processing method.

[0022] Figure 3 This is a block diagram of the method for reducing the peak-to-average power ratio of the data frame. Detailed Implementation

[0023] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0025] Reference Figure 2 and Figure 3 This embodiment considers an AFDM communication system with pilot frames and data frames. In this system, the pilot frames and data frames are processed separately at both the transmitting and receiving ends. To address the problem of excessively high peak-to-average power ratio (PAPR) in multi-carrier modulated waveforms (AFDM), this invention employs a method to change the position of the pilot in the discrete affine domain for the pilot frames; for the data frames, it utilizes the relatively relaxed pre-chirp parameter constraints, combined with an active constellation point extension method to reduce PAPR. At the receiving end, the pilot frames and data frames are processed separately, and a maximum likelihood-minimum mean square error detector is used to decode the received signal and recover the data. Since this invention performs direct decoding at the receiving end through an algorithm, it does not require sending side information to the receiving end, thus reducing PAPR without reducing spectral efficiency.

[0026] like Figure 2 , Figure 3 As shown in this embodiment, a PAPR suppression method for a simulated radio frequency multiplexing system based on pilot position optimization and pre-chirp parameter selection combined with active constellation extension includes the following steps: Step 1: At the transmitting end, perform a cyclic shift of the pilot and data in the Discrete Affine Fourier (DAF) domain for the pilot frame, shifting only one bit at a time. In one embodiment, during initialization, the pilot and data are shifted in the DAF domain... Figure 2 The pilots are arranged with an index of 0. The pilot signal is placed at the beginning of the pilot frame, with Q zeros placed on either side of the pilot signal as guard intervals. The remaining space is used for data symbols. This can be described by the following formula:

[0027] In the above formula, It is the index value of the DAF domain signal. Indicates the number of zero-protection intervals. It is the frame length of an AFDM. It is the maximum delay value of the path. Therefore The maximum Doppler value of the path after normalization. Indicates pilot symbol, Indicates the first A data symbol, Signals representing the DAF domain The A symbol.

[0028] Step 2: Perform an inverse discrete affine Fourier transform (IDAFT) on the pilot frame after each cyclic shift to convert it into a time-domain signal, and calculate the peak-to-average power ratio (PAPR) of the signal using the following formula:

[0029] in It is the peak power of the signal. It is the average power value of the signal. It is a time-domain signal The Values.

[0030] This process needs to be repeated. Next, among them This indicates rounding down to the nearest integer, ending the entire circular shift, and finally... The arrangement with the lowest PAPR among the candidate positions is selected as the pilot frame output. It should be noted that since the PAPR of the signal is random, quantitative analysis uses the complementary cumulative distribution (CCDF) statistical function, as shown below:

[0031] in This represents the probability distribution function, and PAPR0 represents the threshold of PAPR.

[0032] Step 3: After resolving the PAPR reduction issue in the pilot frames, we now focus on the PAPR issue in subsequent data frames, such as... Figure 3 As shown. This differs from traditional AFDM modulation which uses a fixed pre-chirp parameter. Unlike other methods, the Active Constellation Point Extension (SGP-ACE) algorithm of this invention provides... One candidate pre-chirp parameter Make a selection:

[0033] in Indicates the first One prechirp parameter. The prechirp parameter must be an irrational number or a decimal less than 1 / 2N; there are no other strict requirements. The parameters are selected sequentially, and IDAFT is performed on the data to convert it to a time-domain signal:

[0034] for The discrete Fourier transform matrix of a point. It is a DAF domain signal modulated by constellation. It is the frame length of an AFDM. It indicates that it was used The time-domain signal used as a pre-chirp parameter, It is the post-chirp parameter of AFDM. , These respectively indicate that parameters were used. and A diagonal matrix.

[0035] Step 4: Amplify the time-domain signal of the data frame and optimize it using the iterative ACE algorithm to calculate the PAPR of the updated signal. In one embodiment, the SGP-ACE algorithm is used to optimize the PAPR of the AFDM, specifically including the following steps: Record the original constellation modulation symbols Location information, using the current location Parameter pair Perform IDAFT to obtain the initialized time-domain signal , the iteration parameters Initialize to 0; For the first The time-domain signal of the next iteration Amplitude limiting is applied:

[0036] in The signal after peak clipping, among which yes The One value, It is the symbol for a complex number. It is a complex number Angle, This is the threshold value for peak clipping; the signal of the peak-shaving portion is denoted as... : , yes The There are several values, and a DAFT is performed on these values ​​to obtain... Only keep Values ​​that conform to the constellation point expansion region constraints are corrected by setting non-constrained values ​​to zero. And then After performing an IDAFT, the corrected result is obtained. time domain value .

[0037] The signal is updated using the following formula. :

[0038] in It is the step size parameter, and it is updated. The step size parameter is obtained from the following formula:

[0039] in and They represent The maximum amplitude and its corresponding position index. express exist Projection in direction, Time domain value The One value, For index of The value of .

[0040] When the PAPR value reaches an acceptable threshold or Reaching the maximum number of iterations or step size parameter The SGP-ACE algorithm terminates when the value is negative, and the result at this point is recorded. .

[0041] Choosing and achieving the lowest PAPR Candidate-associated signals As the output of the data frame, the processed pilot frame and data frame are arranged and then output as the transmission signal.

[0042] Step 5: The receiver demodulates the pilot frame and data frame using a maximum likelihood-minimum mean square error detector (ML-MMSE) respectively. First, the pilot frame is processed; then, the received signal undergoes a DAFT to obtain the discrete affine domain signal. , will signal The index of the maximum amplitude is denoted as The index of the possible location of the pilot is... , The possible spread region of the pilot signal is determined by its maximum value. It is determined by the cyclic shift step size A defined set of allowed pilot indices, The path is The normalized path maximum Doppler value, where Q is the zero guard interval; Define the DAF domain channel matrix as follows ,but , For the channel information matrix, , This represents the conjugate transpose operation. The part related to channel estimation is extracted and denoted as For channel estimation, then ,in , , , , for The identity matrix, This indicates a modulo operation on N.

[0043] Signal The part related to channel estimation is extracted and denoted as :

[0044] in , For the extracted noise portion, , , , Corresponding to pilot position Path gain It can be obtained as follows:

[0045] Only the channel matrix will be retained. Middle and the first The value of the path-related value that is relevant to channel estimation is defined as follows: , for The The value of the column, This represents the number of multipath paths in a two-dispersion channel. It is a pilot symbol.

[0046] Generate DAF domain channel matrix using the obtained channel information The data was detected using an MMSE detector, and the results were recorded as follows: Constellation point correction is performed to obtain The final data was then determined using an ML detector:

[0047] Right now respectively let With minimum Euclidean distance and Value. Finally. go through The final output signal is obtained by a reverse cyclic shift.

[0048] Next, the data frame is processed: First, each prechirp parameter is... The corresponding DAFT is applied sequentially to the received signal to obtain As shown below:

[0049] in For noise vectors, , = This indicates that the pre-chirp parameter was used. The DAF domain channel matrix, , , This represents the conjugate transpose operation. The minimum mean square error detector is shown below:

[0050] in The results were obtained using the minimum mean square error detection method. Represents the identity matrix. This represents the average signal-to-noise ratio. The average power of the signal. This represents the average noise power. The obtained... Perform constellation point correction to obtain Then, by comparing different candidate parameters Decoding is performed using the MMSE detection results:

[0051] The final demodulated signal is obtained. ,Right now respectively let With minimum Euclidean distance and value.

[0052] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An AFDM peak-to-average power ratio (PAPR) suppression method based on pilot position optimization and pre-chirp, characterized in that, The pilot frames and data frames are processed at the transmitting end, including the following steps: Step 1: For the pilot frame, perform cyclic shift processing on the pilot and data as a whole in the discrete affine Fourier domain according to the set step size, perform inverse discrete affine Fourier transform on the processed pilot frame to convert it into a time domain signal, and calculate the peak-to-average power ratio (PAPR) of the signal. Step 2: Traverse all possible shift cases and repeat the calculation process in Step 1. After the traversal is completed, compare the PAPR values ​​of all records and output the set of signal data with the smallest PAPR as the pilot frame output. Step 3: Select candidate pre-chirp parameters provided by the data frame and perform inverse discrete affine Fourier transform to convert them into time-domain signals; Step 4: Perform amplitude limiting on the time-domain signal of the data frame and optimize it using the iterative SGP-ACE algorithm. Calculate the PAPR of the updated signal and select the set with the lowest PAPR as the output of the data frame. Step 5: The pilot frame output and the data frame output are arranged together and output as the transmitted signal. The receiver demodulates the pilot frame and the data frame using a maximum likelihood-minimum mean square error detector to obtain the final demodulated signal.

2. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, In step 1, the data frame and pilot frame are processed separately at the transmitting end. For the pilot frame, the pilot and data are arranged in the DAF domain with an index of 0. That is, the pilot signal is placed at the beginning of the pilot frame, and Q zeros are placed on the left and right sides of the pilot signal as guard intervals. The remaining space is filled with data symbols, which can be described by the following formula: in, It is the index value of the DAF domain signal. Indicates the number of zero-protection intervals. It is the frame length of an AFDM. It is the maximum delay value of the path. Therefore The normalized path maximum Doppler value; Indicates pilot symbol, Indicates the first A data symbol, Signals representing the DAF domain The A symbol.

3. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, In step 1, the pilot frame after each cyclic shift process is converted into a time-domain signal by performing an inverse discrete affine Fourier transform (IDAFT), and the peak-to-average power ratio (PAPR) of the signal is calculated.

4. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, In step 3, the selected pre-chirp parameter is an irrational number or a decimal less than 1 / 2N. It is the frame length of an AFDM.

5. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, Step 4 includes the following steps: Record the positional information of the original constellation modulation symbols using the selected chirp parameters. IDAFT is performed on the position information of the original constellation modulation symbols to obtain the initialized time-domain signal. , the iteration parameters Initialize to 0; For the The time-domain signal of the next iteration The peak-shaving portion of the signal is obtained by amplitude limiting. ; Using the peak-shaving portion of the signal The updated signal and iterative parameters yield a new time-domain signal. ; The SGP-ACE algorithm terminates when the PAPR value reaches an acceptable threshold, the maximum number of iterations is reached, or the step size parameter becomes negative. The PAPR at this point is recorded as follows. , To use the first Peak-to-average power ratio of signals using the SGP-ACE algorithm with pre-chirped parameters; Choose the chirp parameter with the lowest PAPR. The associated signals are output as data frames.

6. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, After receiving the signal at the receiving end, the pilot frame and the data frame are processed separately, including demodulation of the pilot frame and the data frame using a maximum likelihood-minimum mean square error detector.

7. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, After receiving the signal, the receiver performs channel estimation and equalization on the pilot frame, as follows: First, the maximum value of the received signal is indexed as... The pilot index is denoted as ,but , The possible spread region of the pilot signal is determined by its maximum value. It is determined by the cyclic shift step size A defined set of allowed pilot indices, The path is The normalized path maximum Doppler value, where Q is the number of zero-guard intervals. It is the frame length of an AFDM; Then, channel estimation is performed for each possible pilot location, and finally the signal is demodulated using a maximum likelihood-minimum mean square error detector.

8. The AFDM peak-to-average power ratio suppression method based on pilot position optimization and pre-chirp according to claim 1, characterized in that, For the data frame of the received signal, the channel information estimated from the pilot frame and all pre-chirped parameters are used to construct the DAF domain channel matrix, and the signal is demodulated using a maximum likelihood-minimum mean square error detector.

9. A computer device comprising a memory and a processor, the memory being electrically connected to the processor, the memory storing a computer program, characterized in that: When the computer program is executed by the processor, it causes the processor to implement the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor implements the method as described in any one of claims 1 to 8.

Images