Intra-frame frequency hopping carrier Doppler estimation compensation method for data link burst communication system

Abstract

The invention provides an intra-frame frequency hopping carrier Doppler estimation compensation method for a data link burst communication system. The intra-frame frequency hopping carrier Doppler estimation compensation method is used for carrier compensation in one-frame burst communication. One frame of burst communication comprises 1 + N frequency pulses, the first frequency pulse is a synchronous pulse, the other N frequency pulses are N data pulses, each pulse corresponds to a carrier frequency point, and the method comprises the steps of synchronous pulse Doppler frequency coarse estimation, data pulse Doppler frequency linear transformation estimation and coarse compensation. And carrying out precise estimation and precise compensation on data pulse Doppler frequency residues, and carrying out data pulse carrier phase estimation and compensation. Compared with the prior art, the method has the advantages that carrier frequency Doppler frequency hopping coarse estimation compensation is performed by combining the FFT algorithm, the Rife algorithm and the Doppler linear transformation algorithm, Doppler frequency compensation residual frequency offset fine compensation is performed by adopting data pulse internal discrete pilot frequency, the Doppler residual of frequency hopping pulse receiving data after compensation is smaller, and the demodulation performance is better.

Description

Technical Field

[0001] The invention belongs to the technical field of data link communication and relates to DZ data link communication technology. Background Art

[0002] The D-type carrier platform undertakes multiple military tasks such as air-to-ground reconnaissance, target search and strike, and plays an important role in the field of military applications. With the development of joint combat thinking and combat systems, the D-type carrier platform is required to have multi-D information interconnection and coordination capabilities in combat use. As a carrier device for information interconnection and coordination, the D-type carrier data link plays an important role in the multi-D coordinated combat system.

[0003] In order to realize multi-D cooperative networking, the data link terminal mostly adopts the time division multiple access (TDMA) access mode, that is, each terminal adopts the burst waveform to transmit information in the pre-planned transmission time slot; the transmission waveform is composed of a mixture of synchronization pulses and data pulses, and wide interval frequency hopping and channel coding technology are used for transmission within the frame, that is, a frame of information is divided into several data groups after channel coding, and each group of data carries a time pulse and a carrier frequency point. Due to the high-speed and high-maneuverability movement characteristics of the D-carrier platform, the frequency of the received signal of the D-carrier data link terminal is offset from the transmission frequency point, that is, Doppler frequency drift. In actual work, the frequency interval of each transmission pulse frequency hopping carrier is generally large. If the Doppler frequency obtained by the synchronization pulse capture is directly used to roughly compensate the frequency of each pulse, there may still be a large Doppler frequency residue after compensation. When the residual frequency after compensation exceeds the measurement range of the Doppler residual frequency precise estimation algorithm, it will cause the precise estimation compensation algorithm to fail, so that the pulse data information cannot be accurately restored, affecting the communication performance of the data link terminal.

[0004] Prior art 1 "A method for jointly estimating Doppler frequency offset and Doppler frequency offset acceleration CN 113965436B2023.06.27" shows a method for compensating Doppler frequency offset and frequency offset acceleration for a data link system. The method consists of square demodulation, two-dimensional frequency offset and acceleration estimation search, frequency offset compensation and demodulation, which solves the problem that the prior art cannot achieve fast open-loop carrier synchronization of burst signals.

[0005] Prior art 2 "A spectrum compensation system under a broadband fast frequency hopping system CN 107204948A2017.09.26" shows a frequency deviation compensation method for a fast hopping system, studies the frequency deviation relationship of different hops in the frequency hopping system, designs unique codes of long and short lengths, uses the long unique code of the pilot hop to estimate the frequency deviation, and compensates for the frequency deviation of the remaining hops except the pilot hop according to the frequency relationship of different hops to complete carrier synchronization. This method improves the noise resistance of the system and can adapt to a large frequency deviation range.

[0006] The first existing technology eliminates the influence of modulation data on frequency offset estimation by square demodulation, and estimates Doppler parameters by two-dimensional frequency offset and frequency offset acceleration search. The estimation accuracy of this method increases with the increase of calculation scale, but it has the disadvantages of large consumption of computing resources or large calculation delay.

[0007] The second prior art designs an intra-frame frequency hopping frame format, estimates the Doppler frequency offset by using the pilot hopping long unique code, estimates and compensates for the remaining frequency hopping offsets based on the linear relationship between the carrier and Doppler, and the estimation accuracy is 10kHz. Then, the data intra-frame short unique code and the local pilot are used for conjugate cross-correlation accumulation, and the inverse tangent operation is performed on the correlation accumulation result to obtain the phase estimation result and compensate for the data phase offset. This method has the problem of insufficient frequency offset estimation accuracy, and the pilot hopping directly compensates for the frequency offset once, resulting in a large residual frequency offset, which affects the subsequent phase compensation effect and the final demodulation performance. Summary of the invention

[0008] In view of this, the present invention proposes a Doppler estimation and compensation method for an intra-frame frequency hopping carrier in a data link burst communication system, which is used to improve the frequency offset estimation accuracy and enhance the demodulation performance of the frequency hopping communication frame.

[0009] The specific technical solutions are as follows:

[0010] The invention discloses an intra-frame frequency hopping carrier Doppler estimation and compensation method for a data link burst communication system, which is used for carrier compensation within a frame burst communication. A frame burst communication includes 1+N frequency pulses, the first frequency pulse is a synchronization pulse, and the remaining N frequency pulses are N data pulses, each pulse corresponding to a carrier frequency point; wherein the synchronization pulse is composed of a pseudo-random pilot sequence, the pilot sequence contains the bit timing synchronization information of the transmission waveform, and the data pulse is composed of n-segment discrete pilots and n-1-segment data interleaving, including synchronization pulse Doppler frequency coarse estimation, data pulse Doppler frequency linear transformation estimation and coarse compensation, data pulse Doppler frequency residual fine estimation and fine compensation, and data pulse carrier phase estimation and compensation.

[0011] The synchronous pulse Doppler frequency rough estimation includes obtaining a complex exponential signal after de-piloting and decorrelation of the synchronous pulse, then converting the complex exponential signal into the frequency domain, and then completing the synchronous pulse Doppler frequency rough estimation through Rife interpolation.

[0012] The Rife interpolation is expressed as follows,

[0013]

[0014] Among them, f c is the rough estimation result of synchronous pulse Doppler frequency, X(k 0 ) is the maximum value of the spectrum line, X(k 0 +r) is the second largest value, |X(k0 +1)|≤|X(k 0 -r)|, r=-1, otherwise r=1, k 0 is the maximum spectral line position, T s is the symbol duration.

[0015] The linear transformation estimation of the data pulse Doppler frequency further includes: obtaining a linear relationship between different pulse Doppler frequencies based on the premise that the relative movement speed of the transceiver platform corresponding to a frame of burst communication remains unchanged; using this linear relationship, linearly transforming the rough estimation result of the synchronization pulse Doppler frequency to obtain a rough estimation of the data pulse Doppler frequency; the rough compensation refers to compensating the data pulses with the rough estimation result of the data pulse Doppler frequency to ensure that the frequency error of the data pulse Doppler frequency is constrained to within the carrier frequency fine compensation range.

[0016] The data pulse Doppler frequency residual accurate estimation refers to estimating the Doppler frequency residual error using the discrete pilots in the data pulse after the Doppler frequency coarse compensation; the Doppler frequency residual accurate estimation includes three sub-steps: the first step is to perform decorrelation processing on the discrete pilots in a data pulse; the second step is to perform symbol-by-symbol accumulation of the decorrelation results of each segment of the discrete pilots after the decorrelation is completed, and each segment of the pilots outputs a complex accumulated value; the third step is to perform conjugate complex multiplication on the accumulated results of the previous and subsequent discrete pilots, and then respectively calculate the inverse tangent and the derivative with respect to time of the complex multiplication results, that is, to obtain the Doppler frequency residual accurate estimation result of each data segment;

[0017] The data pulse Doppler frequency residual fine compensation refers to compensating the data pulse Doppler frequency after coarse compensation according to the fine estimation result, so as to bring the frequency error into the required range of phase compensation.

[0018] The data pulse carrier phase estimation and compensation is used to perform phase estimation and compensation processing on data after Doppler frequency residual precise compensation, and includes five sub-steps: the first step: after the Doppler frequency residual precise compensation, the discrete pilot in a data pulse is subjected to decorrelation processing; the second step: after the decorrelation is completed, the decorrelation result of each segment of discrete pilot is accumulated symbol by symbol, and each segment of pilot outputs a complex accumulated value; the third step: complex multiplication is performed on the accumulated results of the previous and subsequent discrete pilots; the fourth step: an inverse tangent operation is performed on the complex multiplication result to obtain a phase compensation result; the fifth step: phase compensation is performed on the data after the Doppler frequency residual precise compensation using the obtained phase compensation result of each segment of data, and the entire demodulation processing is completed.

[0019] Beneficial Effects

[0020] 1. This method adopts the interpolation estimation method, which improves the coarse frequency offset estimation accuracy compared with the direct sampling FFT estimation result.

[0021] 2. The frequency offset residual precise estimation processing is introduced to compensate for the signal-to-noise ratio loss caused by the Doppler effect, which improves the demodulation performance of the frequency hopping communication frame as a whole.

[0022] 3. Compared with the existing technology, this method combines FFT algorithm, Rife algorithm and Doppler linear transformation algorithm to roughly estimate and compensate the carrier Doppler frequency hopping, and uses the internal discrete pilot of the data pulse to accurately compensate the residual frequency deviation of the Doppler frequency compensation. After compensation, the Doppler residual of the received data of the frequency hopping pulse is smaller and the demodulation performance is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 , a flow chart of the data frame frequency offset compensation method of the present invention;

[0024] Figure 2 , radiation pulse time-frequency distribution diagram;

[0025] Figure 3 , data frame structure diagram. DETAILED DESCRIPTION

[0026] The present invention proposes an intra-frame frequency hopping Doppler compensation method for a data link burst communication system. The physical layer waveform structure is shown in a data frame structure diagram and a radiation pulse time-frequency distribution diagram. A frame of burst communication contains a total of N+1 frequency pulses, the first frequency pulse is a synchronization pulse, and the remaining N frequency pulses are N data pulses, each pulse corresponds to a carrier frequency point; wherein the synchronization pulse is used to complete the bit timing capture processing of a single burst communication, the synchronization pulse is composed of a pseudo-random pilot sequence, and the pilot sequence contains the bit timing synchronization information of the transmission waveform; the data pulse is composed of n-segments of pilots and n-1-segments of data, the n-segments of pilots are evenly / non-uniformly discretely distributed between the beginning and end of the pulse and between the n-1-segments of data, the pilot data is composed of a known M sequence, and the front and rear two segments of discrete pilots are used to complete the Doppler frequency and carrier phase precise estimation and precise compensation of the data segment sandwiched therebetween.

[0027] The implementation process is divided into the following steps, as shown in the data frame frequency offset compensation algorithm flow chart:

[0028] 1) Synchronous pulse timing capture

[0029] This step is used to complete symbol timing processing and symbol phase estimation to ensure that the data bit timing extracted by the subsequent process estimation compensation algorithm is correct.

[0030] 2) Rough estimation of synchronous pulse Doppler frequency,

[0031] A. Decorrelation of synchronization pulses

[0032] According to the timing synchronization position, the synchronization pulse pilot symbol R is extracted symbol Local Pilot PNlocal Conjugate multiplication is performed, and the multiplication result is the complex exponential signal after the pilot is removed and decorrelated.

[0033] conj(PN local ).*R symbol

[0034] B. Perform FFT transformation on the synchronous pulse after decorrelation processing.

[0035] After decorrelation, the calculation result is subjected to FFT operation, that is, the obtained complex exponential signal is transformed into the frequency domain, and the peak value of the FFT result and the two next largest values ​​on the left and right of the peak value and their corresponding frequency index numbers are searched.

[0036] C. Rife frequency estimation

[0037] When the roughly estimated synchronous pulse Doppler frequency is not at the FFT quantization frequency point, directly estimating the signal frequency using the maximum spectrum line position of the FFT will result in quantization errors. Therefore, this method uses the Rife interpolation fitting algorithm to correct the frequency value of the FFT calculation result to improve the performance of frequency estimation. Specifically, interpolation is performed based on the FFT spectrum peak obtained in b) and the second largest value of the frequency points on both sides of the spectrum peak. The formula Get the synchronous pulse Doppler frequency rough estimation result, where X(k 0 ) is the maximum value of the spectrum line, X(k 0 +r) is the second largest value, |X(k 0 +1)|≤|X(k 0 -r)|, r=-1, otherwise r=1, k 0 is the maximum spectral line position, T s is the symbol duration, f c This is the rough estimation result of the synchronization pulse frequency. At this point, the rough estimation of the synchronization pulse Doppler frequency is completed.

[0038] 3) Data pulse Doppler frequency linear transformation estimation and coarse compensation

[0039] The Doppler frequency is caused by the relative motion speed of the platform. The conversion formula between velocity and Doppler frequency is: where f d is the Doppler frequency, v is the relative speed of the platform, and λ is the carrier wavelength. Since the carrier frequencies of each pulse are different, that is, the carrier wavelengths are different, each pulse will cause a different Doppler frequency for the same relative speed. If the carrier interval is large, the Doppler frequency difference between different pulses will increase. If the Doppler frequency estimation result obtained by the synchronization pulse is directly used to perform coarse frequency offset compensation on the data pulse, a large compensation residue will be generated. However, for different pulses in a frame of data, it is assumed that the speed v remains basically constant. According to the formula, λ can be derived sync fdsync =λ data f ddata , where f dsync This is the rough estimate of the synchronization pulse frequency obtained in the previous step, f ddata The Doppler frequency of the data pulse, that is, the Dopplers corresponding to different pulses have a linear relationship. In the implementation of the scheme, according to the linear relationship between different pulse carrier frequencies, the Doppler parameters are obtained by estimating the synchronization frame, thereby achieving a rough estimation of the Doppler frequency of the data pulse, and then the data pulse is compensated with the result after linear transformation to ensure that the frequency error after rough frequency compensation is constrained to within the range of precise compensation of the Doppler frequency.

[0040] 4) Data pulse Doppler frequency residual accurate estimation and compensation

[0041] Since the accuracy of Doppler frequency coarse estimation is limited, Doppler frequency residues still exist after coarse Doppler frequency compensation. This frequency residue causes a large phase deviation before and after the data segment in a single data pulse. Therefore, it is necessary to further use the discrete pilot in the data frame after coarse Doppler frequency compensation to estimate the Doppler frequency residual error. This process is the precise estimation of Doppler frequency residues.

[0042] The steps for Doppler frequency residual accurate estimation are as follows:

[0043] Step 1: As in the first step of Doppler frequency rough estimation, decorrelation processing is performed on the discrete pilot in a data pulse, that is, the discrete pilot symbol R in the pulse is extracted. p_data_symbol Local Pilot PN p_data_local Perform conjugate multiplication to achieve decorrelation processing.

[0044] conj(PN p_data_local ).*R p_data_symbol

[0045] Step 2: After decorrelation is completed, the decorrelation results of each segment of the scattered pilot are accumulated symbol by symbol. Each segment of the pilot outputs a complex accumulated value. Each data pulse has a total of n accumulated values, among which the mth accumulated value is represented by corr i_m +1i*corr q_m , m is the sequence number index of the scattered pilot, corr i_m is the accumulated result part, corr q_m is the imaginary part of the accumulated result.

[0046] Step 3: Perform conjugate complex multiplication on the cumulative results of the previous and next discrete pilots to obtain the phase difference 2πf between the previous and next pilots. d'_m T , where f d ' _mis the Doppler frequency residual calculated for the mth and m+1th discrete pilots, and T is the time interval between two discrete pilots.

[0047]

[0048] Then, the arc tangent and the time derivative of the complex result are calculated to obtain the Doppler frequency residual accurate estimation result of each data segment.

[0049]

[0050] Finally, further parameter compensation is carried out on the roughly compensated data based on the precise estimation results to bring the frequency error into the required range of phase compensation.

[0051] 5) Data pulse carrier phase estimation and compensation

[0052] After Doppler frequency residual precise compensation, the frequency error is constrained to In the range of , T is the time interval between two discrete pilots in the data pulse. However, due to the uncertainty of the carrier phase of the transceiver and the uncertainty of the transmission link delay, the initial phase of the received signal has changed. Therefore, it is necessary to perform phase estimation and compensation processing on the data after the Doppler frequency residual precise compensation. The specific processing flow is as follows:

[0053] Step 1: As in the first step of Doppler frequency rough estimation, the discrete pilot in a data pulse after Doppler frequency residual fine compensation is de-correlated, that is, the discrete pilot symbol R' in the pulse is extracted. p_data_symbol With local pilot PN' p_data_local Perform conjugate multiplication to achieve decorrelation processing.

[0054] conj(PN' p_data_local ).*R' p_data_symbol

[0055] Step 2: After decorrelation is completed, the decorrelation results of each discrete pilot are accumulated symbol by symbol. Each pilot outputs a complex accumulated value, and each data pulse has a total of 5 accumulated values ​​corr' i_m +1i*corr' q_m , m is the sequence number index of the scattered pilot, corr' i_m is the cumulative result part, corr' q_m is the imaginary part of the accumulated result.

[0056] Step 3: Perform complex multiplication on the accumulated results of the previous and next discrete pilots. The calculation formula is as follows:

[0057]

[0058] in, is the phase compensation value of the mth segment data

[0059] Step 4: Obtain the phase compensation result through inverse tangent operation. The calculation formula is as follows:

[0060]

[0061] Step 5: Use the phase compensation value of each segment of data obtained above Phase compensation is performed on the data after the Doppler frequency residual precise compensation, thus completing the compensation process of the entire demodulation processing.

Claims

1. A method for estimating and compensating Doppler of frequency hopping carrier in a frame of data link burst communication system, which is used for carrier compensation in a frame of burst communication, wherein a frame of burst communication includes 1+N frequency pulses, the first frequency pulse is a synchronization pulse, and the remaining N frequency pulses are N data pulses, each pulse corresponding to a carrier frequency point; wherein, The synchronization pulse is composed of a pseudo-random pilot sequence, the pilot sequence contains the bit timing synchronization information of the transmission waveform, and the data pulse is composed of n segments of discrete pilots and n-1 segments of data interleaving, and is characterized by: including coarse estimation of the synchronization pulse Doppler frequency, linear transformation estimation and coarse compensation of the data pulse Doppler frequency, precise estimation and precise compensation of the data pulse Doppler frequency residual, and data pulse carrier phase estimation and compensation.

2. The method for estimating and compensating Doppler of an intra-frame frequency hopping carrier in a data link burst communication system according to claim 1, characterized in that: The synchronous pulse Doppler frequency rough estimation includes obtaining a complex exponential signal after de-piloting and decorrelation of the synchronous pulse, then converting the complex exponential signal into the frequency domain, and then completing the synchronous pulse Doppler frequency rough estimation through Rife interpolation.

3. The method for compensating Doppler estimation of an intra-frame frequency hopping carrier in a data link burst communication system according to claim 2, characterized in that: The Rife interpolation is expressed as follows, Among them, f c is the rough estimation result of the synchronous pulse Doppler frequency, X(k0) is the maximum value of the spectrum line, X(k0+r) is the second maximum value, when |X(k0+1)|≤|X(k0-r)|, r=-1, otherwise r=1, k0 is the position of the maximum spectrum line, T s is the symbol duration.

4. The method for estimating and compensating Doppler of an intra-frame frequency hopping carrier in a data link burst communication system according to claim 3, characterized in that: The data pulse Doppler frequency linear transformation estimation further includes obtaining a linear relationship between different pulse Doppler frequencies based on the premise that the relative motion speed of the transceiver platform corresponding to a frame of burst communication remains unchanged; By using this linear relationship, a rough estimation result of the synchronous pulse Doppler frequency is linearly transformed to obtain a rough estimation of the data pulse Doppler frequency; The coarse compensation refers to compensating the data pulses with the coarse calculation result of the data pulse Doppler frequency, so as to ensure that the frequency error of the data pulse Doppler frequency is constrained within the fine compensation range of the carrier frequency.

5. The method for estimating and compensating Doppler of an intra-frame frequency hopping carrier in a data link burst communication system according to claim 4, characterized in that: The data pulse Doppler frequency residual accurate estimation refers to estimating the Doppler frequency residual error using the discrete pilots in the data pulse after the Doppler frequency coarse compensation; the Doppler frequency residual accurate estimation includes three sub-steps: the first step is to decorrelate the discrete pilots in a data pulse; Step 2: After decorrelation is completed, the decorrelation results of each segment of the scattered pilot are accumulated symbol by symbol, and each segment of the pilot outputs a complex accumulated value; Step 3: Perform conjugate complex multiplication on the accumulated results of the previous and subsequent discrete pilots, and then calculate the arc tangent and the time derivative of the complex multiplication results, that is, obtain the Doppler frequency residual accurate estimation result of each data segment; The data pulse Doppler frequency residual fine compensation refers to compensating the data pulse Doppler frequency after coarse compensation according to the fine estimation result, so as to bring the frequency error into the required range of phase compensation.

6. The method for compensating Doppler estimation of an intra-frame frequency hopping carrier in a data link burst communication system according to any one of claims 1 to 5, characterized in that: The data pulse carrier phase estimation and compensation is used to perform phase estimation and compensation processing on the data after the Doppler frequency residual precise compensation, and includes five sub-steps: the first step: decorrelation processing of the discrete pilot in a data pulse after the Doppler frequency residual precise compensation; Step 2: After decorrelation is completed, the decorrelation results of each segment of the scattered pilot are accumulated symbol by symbol, and each segment of the pilot outputs a complex accumulated value; Step 3: Perform complex multiplication on the accumulated results of the previous and subsequent discrete pilots; Step 4: Perform an inverse tangent operation on the complex multiplication result to obtain the phase compensation result; Step 5: Use the phase compensation result of each segment of data to perform phase compensation on the data after Doppler frequency residual precise compensation to complete the entire demodulation process.

Images