CN104914435B - A kind of wind profile radar phase coding method and circuit based on Fei Lanke codes - Google Patents

Abstract

The present invention relates to the Waveform Design of wind profile radar, signal transacting, Waveform generating method and frequency synthesis circuit technical field.Present invention firstly provides a kind of wind profile radar phase coding method and circuit based on Fei Lanke codes.The step of methods described, includes:Calculation code subpulse number M, calculates time domain accumulation number N_c, select code period number N, adjustment time domain accumulation number N_cWith code period number N, encoder matrix F is formed_N′_×M, use encoder matrix F_N′_×MCarry out the exomonental phase code of code period.Wind profile radar is used for the weather informations such as wind direction, wind speed and the echo strength of atmospheric sounding wind field, and prior art realizes the resolving power of the energy and burst pulse of broad pulse simultaneously using two-phase mutual-complementing code or using walsh function code equiphase coding techniques.The coding method of the present invention and circuit can adopt normal pulsed or burst waveforms, and with low distance side lobe, subpulse number selects flexible, can be used for antagonism point frequency radio interference, it is adaptable to various wind profile radars.

Description

Wind profile radar phase encoding method and circuit based on Frank code

Technical Field

The invention discloses a wind profile radar phase encoding method and circuit based on a Frank code, and relates to the technical field of waveform design, signal processing, waveform generation methods and frequency synthesis circuits of wind profile radars.

Background

The wind profile radar is mainly used for detecting turbulence in clear sky and is a meteorological radar. The wind profile radar mainly detects physical quantities such as an atmospheric wind field and the like by utilizing the scattering (Bragg scattering) effect of atmospheric turbulence on electromagnetic waves to acquire data such as wind profiles and the like. Compared with the traditional spherical wind measurement, the wind profile radar can obtain continuous real-time basic data with high space-time resolution.

The wind profile radar generally adopts a working system of a phased array full-coherent pulse radar, adopts five wave beams to detect in turn, measures the radial wind speed of each wave beam, and carries out vector projection calculation to obtain the wind direction and the wind speed as well as other data products and graphic products.

Wind profile radars are classified into boundary layer wind profile radars, troposphere wind profile radars, and stratosphere wind profile radars according to the difference in maximum detection height. The operating frequency of boundary layer is mostly L wave band, and the operating frequency of troposphere is mostly P wave band, and what the stratosphere is mostly the UHF frequency channel. The wind profile radar is divided into a fixed type and a movable type according to the installation mode, and the movable type is divided into a vehicle-mounted type and a shelter type.

The signal processing of the wind profile radar mainly completes time domain accumulation, pulse pressure, ground clutter removal, spectrum analysis and spectrum accumulation of atmospheric turbulence echo signals, and is used for obtaining wind field information such as wind direction and wind speed. The coherent processing time of the wind profile radar signal processing comprises a plurality of radar periods, the signal analysis of the radial velocity of a beam is generally carried out by adopting FFT of 512 points, the time domain accumulation number is generally between 30 and 200, the radar period is between 20 mu S and 240 mu S, the maximum detection height is between 2km and 25km, and the height resolution is generally between 60m and 480 m.

The wind profile radar is generally a full-coherent pulse wind profile radar, the Doppler frequency of an atmospheric target is low, the detection height is limited, the repetition period of the radar is short, the sampling rate is high, the distance mapping problem is not serious, and the distance mapping problem can be ignored frequently. Such fully coherent pulse wind profile radar is generally free of both range ambiguity and velocity ambiguity problems. The Doppler effect can be ignored in the matching receiving of the pulse waveform, namely Doppler information is extracted between radar periods without being extracted in one radar period, and the Doppler frequency of the atmospheric target is extracted by performing spectrum analysis by using multi-period echo signals.

Because the doppler frequency of the atmospheric target is low, the repetition period of the radar is short, and the sampling rate is high, the number of radar cycles for extracting doppler information in the coherent time is large, that is, the number of coherent samples is large. The large number of coherent samples results in a large amount of input data for the spectral analysis. In order to reduce the data volume and reduce the computation amount of spectral analysis, the signal processing of the wind profile radar generally adopts time domain accumulation processing (segmented accumulation and averaging), and after the time domain accumulation, each data segment (with the length being the time domain accumulation number N) is obtained_c) Is performed on these sample averages and a velocity FFT spectral analysis is performed on these sample averages. The number of periods in the coherent time is the time domain accumulation number N_cAnd the number N of FFT points_FFTProduct of (2) N_cN_FFT。

In this method of performing time domain accumulation first and then performing velocity FFT spectrum analysis, it can be understood as a fast algorithm for reducing the amount of calculation in calculation, and from the viewpoint of signal processing, it can be understood that one sample average value is extracted without extracting doppler information during the time of time domain accumulation, and doppler information is extracted using a plurality of sample average values.

The radar range unit samples collected by the radar period in the time domain accumulation time can be considered that the radar echo phase change caused by the Doppler effect is not large and is approximately equal, so that phase coding can be performed in a plurality of periods in the time domain accumulation time. Pulse compression cannot be realized by one period, and pulse compression can be realized by encoding a plurality of periods to form one encoding period. A radar code period refers to a group of periods consisting of multiple radar periods, and each radar period within a radar code period may employ a different phase encoding vector.

The detection mode of the wind profile radar can be divided into a low mode and a high mode according to the range of the detection height, and the high mode adopts phase coding. The pulsed radar transmitter of wind profile radar is typically limited to a maximum duty cycle, which is typically 10% to 20%. The radar period is required to meet the requirements of the detection height range and the duty ratio. In particular, for high mode detection, the duty cycle after phase encoding is generally smaller than but close to the maximum duty cycle in order to ensure the detection power of the radar.

The existing wind profile radar uses a phase encoding technology to improve the duty ratio and the average power, so that the high resolution of the radar and the maximum detection height of the radar are guaranteed. The existing wind profile radar phase encoding generally adopts two-phase complementary codes. The principle of two-phase complementary codes is to find pairs of phase-coded sequences, which are of the same length, where the side lobe of one autocorrelation function is the negative of the other, the coded outputs are summed, and the algebraic sum of the side lobes will be zero. Furthermore, on the waveform, two code vectors of two-phase complementary codes are required to be transmitted on two separate transmit pulses, separately detected, and then added. The transmit pulse is composed of immediately adjacent sub-pulses, each sub-pulse being subjected to complementary code phase modulation. The complementary code has zero range sidelobe of pulse compression under the condition of no Doppler effect.

Another wind profile radar phase encoding method uses walsh codes. The walsh code is an orthogonal code, and a walsh code matrix is formed by sampling consecutive walsh functions, and the walsh code matrix is orthogonal between column vectors (or row vectors). The Walsh code phase encoding is adopted, and the vector of an N-dimensional Walsh code matrix is used for carrying out phase encoding on the same sub-pulse position of N radar periods in a radar encoding period.

In such a multi-cycle orthogonal code phase encoding, the range sidelobe of pulse compression is zero as in the complementary code under the condition of no doppler effect, by utilizing the orthogonality of the code vectors. When the doppler effect is considered, the pulse compression range sidelobes of complementary codes and walsh codes are not zero but are small, and thus can be applied to wind profile radar.

The defects of the prior art are as follows:

(1) two-phase complementary codes, the pulse compression sub-pulse number is the power number of the base number 2, which is not convenient in design and limits the selection flexibility of the pulse compression sub-pulse number.

(2) Walsh codes require the radar cycle number in the radar coding cycle to be a power number of a base number 2, are also inconvenient in design, and limit the flexibility of selecting the signal processing time domain accumulation number.

(3) The two-phase complementary codes and walsh codes have insufficient ability to combat radio dot frequency interference.

Maximum Duty ratio Duty _ cycle of transmitted pulse of wind profile radar and radar period T_rSub-pulse width tau, radial velocity measurement range V_RmaxAnd working wavelength lambda, FFT point number N of speed FFT spectrum analysis_FFTThe index parameters are designed according to the technical requirements of the detection performance of the radar.On the basis, the method provides a wind profile radar phase encoding method and circuit based on the Frank code.

The wind profile radar phase encoding method and circuit based on the Frank code, disclosed by the invention, have no substantial difference from the prior art in terms of circuit structure, and the solidified or downloaded program in the circuit is completed according to the method disclosed by the invention. The invention discloses a wind profile radar phase encoding circuit based on a Frank code, which generally belongs to a part of a radar frequency integrated circuit and a part of a frequency integrated waveform generating circuit.

Disclosure of Invention

The invention discloses a wind profile radar phase encoding method and circuit based on a Frank code aiming at the defects of the prior art, so that compared with the prior art, the method and circuit provided by the invention have better capability of resisting radio point frequency interference signals, and solve the problems that the pulse number of two-phase complementary code sub-pulses cannot be flexibly selected and the radar period number of a Walsh code in a radar encoding period cannot be flexibly selected, thereby being capable of obtaining wider application.

The Frank code encoding matrix is:

the technical scheme for realizing the method and solving the defects of the prior art is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

According to the maximum Duty ratio Duty _ cycle of the transmitted pulse and the radar period T_rAnd calculating the width tau of the sub-pulse to obtain each radar periodThe number of inner sub-pulses M.

The Duty cycle of the transmitted pulse is the ratio of the effective width of the transmitted pulse to the radar period, i.e., the Duty cycle is M τ/T_rIs provided with

M＝uint[Duty_cycle T_r/τ]uint[]Indicating unsigned integers.

Step 2, calculating the time domain accumulation number N_c。

Measuring range V according to working wavelength lambda and radial speed_RmaxAnd radar period T_rCalculating the time domain accumulation number N_c。

Radial velocity measurement range V_RmaxThe determined Doppler frequency range is 2V_RmaxAnd/lambda. According to the sampling theorem, the sampling rate should be 2 times of the maximum Doppler frequency, i.e. the sampling rate should be 4V_Rmaxλ, corresponding to a sampling time of λ/(4V)_Rmax). The sampling time of the signal processing speed FFT is T_rN_cThus having T_rN_c＝λ/(4V_Rmax) Therefore, it is

N_c＝uint[λ/(4V_RmaxT_r)]uint[]Indicating unsigned integers.

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c。

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×M。

Step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

With coding matrix F'_N×MCarry out N_cN_FFTPhase encoding of the transmitted pulses of N encoding periods, encoding matrix F'_N×MRespectively phase-encoding the transmit pulses of N radar periods within an encoding period, an encoding matrix F'_N×MThe M elements in each row vector phase encode the M sub-pulses in the corresponding period, respectively.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

According to the method, time domain accumulation processing is adopted for signal processing of the wind profile radar, and speed FFT spectrum analysis is carried out after the time domain accumulation. The method can simply and intuitively show that the phase change of the radar echo caused by the Doppler effect is not large and is approximately equal in the time domain accumulation period, so that phase coding and pulse compression can be carried out in a plurality of periods in the time domain accumulation period. N is a radical of_cN_FFTCan be divided by N to ensure an integer number of encoding cycles to achieve multi-cycle phase encoding.

The method has the beneficial effects that: frank code matrix F_N×NThe dimension N of (A) satisfies that M is less than or equal to N_cAnd N is_cN_FFTDivisible by N, coding matrix F_N×NIs an integer satisfying the above requirements, and F_N×NCan take a smaller value, allowing a time domain accumulation number N_cCan take a smaller number as the coherent time N of signal processing_cN_FFTT_rThe adjustment of (2) is convenient.

The dimension N of the coding matrix of the Walsh code must be a power of 2, where N is the base_cN_FFTMust be divided by N (due to N)_FFTIs a base number of a power of 2, and generally N_FFT> N), satisfies M ≦ N_cWhen N is present_cCannot take a value less than N, so that N is adjusted_cN_FFTIs not convenient enough, thereby making the signal processing coherent time N_cN_FFTT_rThe adjustment is not convenient enough.

Because the coherent time of signal processing sometimes needs to be adjusted according to weather conditions in the detection of the wind profile radar, the time domain accumulation number N is flexibly adjusted_cThe detection performance of the wind profile radar can be improved.

Furthermore, in some cases, wind profile radar requires N in the design of signal processing_cThe value is small, N is limited, and the Frank code matrix F is allowed to be used at the moment_N×NWhile walsh codes cannot be applied.

It should be noted that the time domain accumulation can be replaced by FFT with large number of points, such as N_cN_FFTPoint FFT spectral analysis. At this time, the spectrum data larger than the wind profile radar radial velocity measurement range is meaningless and should be discarded, and the spectrum data within the wind profile radar radial velocity measurement range is valid data. The accumulated time domain number is used as an important parameter of the radial velocity measuring range of the wind profile radar, and the radial velocity measuring range is determined by the accumulated time domain number and the radar period. Although the signal processing may not perform the actual time domain accumulation, the coding method designed according to the time domain accumulation parameter is still effective. The method of the invention is therefore described in terms of having time domain accumulation.

According to the wind profile radar phase encoding method, the transmitted pulse comprises M sub-pulses, the transmitted waveform is a pulse train waveform, and a determined time interval is arranged between the sub-pulses.

The transmitted pulse signal may be a normal pulse waveform, in which the sub-pulses are continuous in time, or a pulse train waveform, in which the sub-pulses have a certain time interval therebetween. Both the Frank code and the Walsh code are codes that can use burst waveforms.

The pulse train waveform has the advantages that the requirement on the bandwidth of a receiver is reduced, the pulse compression ratio reaches a theoretical calculation value, the bandwidth of a transmitted signal can also be reduced, and the transmitted frequency spectrum easily meets the requirement of radio management.

For the complementary codes, the close-range blind area is enlarged by adopting the pulse train waveform, so that the complementary codes generally adopt the common pulse waveform. Both the Frank code and the Walsh code are orthogonal codes, and since the short-range blind area can be reduced, the burst waveform can be practically applied.

The wind profile radar phase encoding method of the invention comprises the step 3, when the number N of encoding cycles is selected, the improvement is further carried out, so that N satisfies LM ≤ N_cK, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4, and step 5, forming a coding matrix F'_N×MIn a further development, the Frank code matrix F_N×NSelecting M column vectors according to an interval L and an arbitrary sequence to form an N × M-dimensional coding matrix F'_N×M。

The beneficial effects are illustrated as follows:

(1) frank code matrix F_N×NCorresponds to a sub-pulse position of N radar cycles within the coding period. One element in the column vector corresponds to one radar cycle of the N radar cycles. The multi-cycle phase encoding of one sub-pulse corresponds to a change of the carrier frequency. Frank code matrix F_N×NEach column vector in the array corresponds to a carrier frequency point, and the frequency point spacing between adjacent column vectors is (1/T)_r)/N。

(2) N satisfies LM ≤ and N ≤_cAnd/k, and selecting column vectors according to the frequency point interval L, so that the frequency point interval is increased. The frequency point spacing of the selected column vector is (1/T)_r) L/N. Because N is less than or equal to N_cK, between frequency pointsThe distance is greater than or equal to (1/T)_r)kL/N_ckL is more than or equal to 4, so the minimum value of the frequency point interval is 4/(T)_rN_c)。

(3) At the moment, the minimum value of the frequency point interval is the maximum value 1/(T) of the measurable Doppler frequency_rN_c) 4 times, the Doppler frequency in the velocity measurement range can not have great influence on the range side lobe performance. When the weighting is adopted in the velocity FFT in the signal processing, the analysis bandwidth is generally not more than 2 times of the original analysis bandwidth, so that the distance side lobe performance cannot be greatly influenced when the weighting is adopted in the signal processing. kL is more than or equal to 4, so that the frequency point spacing is increased, and the distance side lobe of phase encoding is reduced. The phase encoding side lobe characteristic is the key characteristic of phase encoding, and the reduction of the distance side lobe of the phase encoding has important significance for the wind profile radar.

The phase coding method of the invention is further improved in the step 3, when the number N of the coding cycles is selected, the N is selected to be (M +1) to N ≦ N_cThe value of N under the conditions.

The phase coding method of the invention is further improved in the step 3, when the number N of the coding cycles is selected, the N is selected to be L (M +1) to N_cN value under the condition of/k, and k and L meet the constraint of kL being more than or equal to 4, and step 5, forming a coding matrix F'_N×MIn a further development, the Frank code matrix F_N×NSelecting M column vectors according to an interval L and an arbitrary sequence to form an N × M-dimensional coding matrix F'_N×M。

The 2 items are further improved, and the selection of N and the realization mode, principle and beneficial effect thereof are described as follows:

(1) selecting dimension N of the coding matrix according to M +1 sub-pulses, in a Frank code matrix F_N×NSelecting M column vectors according to any sequence or selecting M column vectors according to interval L and any sequence to form a N × M-dimensional new coding matrix F'_N×MThen the selection of column vectors does not select all M +1 column vectors. I.e. designing the dimension N of the coding matrix according to M +1 sub-pulses, and addingThe frequency point of one sub-pulse is used for resisting the radio frequency point interference.

(2) Therefore when radio point-to-frequency interference is close to the code matrix F'_N×MWhen the frequency point corresponding to a certain column vector is selected and replaced, the frequency point is F'_N×MThe coding achieves both pulse compression capability and suppresses this radio dot frequency interference. The column vector closest to the frequency point of the interference signal is picked out for replacement, and the replacement can be regarded as avoidance and countermeasure of the frequency of the interference signal.

(3) Due to Ferlan code matrix F'_N×MEach column vector in the set corresponds to a frequency point, so that radio frequency interference avoidance and confrontation are easy to realize, while the frequency point corresponding to each column vector of the walsh code is not a frequency point but has frequency spectrum distribution, and cannot be completely avoided and confronted by a method of replacing the column vector.

The technical scheme for realizing the circuit and solving the defects of the prior art is as follows:

a wind profile radar phase encoding circuit based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and the circuit composition comprises: the device comprises a storage circuit 1, a time sequence circuit 2, a DDS circuit 3, an amplitude modulation circuit 4, a frequency conversion circuit 5 and a control computer 6.

The phase encoding circuit controls the computer 6 to execute the following steps:

step 1, calculating the number M of the encoding sub-pulses.

According to the maximum Duty ratio Duty _ cycle of the transmitted pulse and the radar period T_rAnd calculating the number M of the coded sub-pulses according to the sub-pulse width tau.

M＝uint[Duty_cycle T_r/τ]uint[]Indicating unsigned integers.

Step 2, calculating the time domain accumulation number N_c。

Measuring range V according to working wavelength lambda and radial speed_RmaxAnd radar period T_rCalculating the accumulation number N of the signal processing time domain_c。

N_c＝uint[λ/(4V_RmaxT_r)]uint[]Indicating unsigned integers.

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N in time domain_cSelecting the number N of coding cycles so that N satisfies LM ≤ N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected according to an interval L and combined according to an arbitrary order to form an N × M-dimensional coding matrix F'_N×M。

And 6, downloading the time sequence parameters of the time sequence circuit 2.

Downloading timing parameters to timing circuit 2 via control signal 105, these parameters including radar period T_rThe number M of sub-pulses, the sub-pulse pitch τ_rSub-pulse width tau, number of coding cycles N, number of time domain accumulations N_cSum velocity FFT spectral analysis point number N_FFT。

Step 7, downloading the coding matrix F'_N×MAnd starts the sequential circuit 2.

Encoding matrix F 'by control signal 103'_N×MThe matrix is downloaded to the memory circuit 1 and the timing circuit 2 is enabled by the control signal 105.

A memory circuit 1 for pre-storing baseband signal data of one radar cycle and downloaded coding matrix F'_N×MN is sent to the DDS circuit 3 by the clock signal 102 and the control signal 106 from the timing circuit 2_cN_FFTBaseband signals 107 and N for N coding periods_cN_FFTA coded signal 108 of/N coding cycles.

The baseband signal data of one radar cycle stored in advance is,l∈[0,T_r/t_s-1]，T_ris radar period, f_bIs the frequency, t, of the baseband signal_sIs the sampling time of the baseband signal, and is the sample number.

The baseband signal 107 in the coherent time output from the memory circuit 1 has the expression

The coded signal 108 in the coherent time output by the memory circuit 1 is expressed as

Wherein Q is N_cN_FFT/N

v₁(t) is the following pulse function

f′_n.,mIs F'_N×MOne element of (1). Design does not require radar period T_rAnd sub-pulse spacing τ_rIs the baseband signal sampling time t_sInteger multiples of.

And the sequential circuit 2 is used for outputting a control signal 106 to control the storage circuit 1 under the action of a clock signal 104 and a control signal 105 given by the control computer 6. The time sequence circuit 2 controls the address and data bus of the memory 1 according to the downloaded time sequence parameters, so that the memory circuit 1 generates N_cN_FFTA baseband signal 107 and a coded signal 108 for N coding periods, and generating N_cN_FFTThe pulse amplitude control signal 109 for the/N coding periods is sent to the amplitude modulation circuit 4.

The pulse amplitude control signal 109 in coherent time generated by the sequential circuit 2 has the expression

v₂(t) is the following pulse function

Tau is the sub-pulse width, tau is less than or equal to tau_rAnd the width of the sub-pulse is less than or equal to the space between the sub-pulses.

And a DDS circuit 3 for performing data synthesis and digital-to-analog conversion of the baseband signal 107 and the encoded signal 108 by the clock signal 101, and outputting an analog encoded intermediate frequency signal 110.

The DDS circuit 3 comprises two paths of digital multipliers, an adder, a DAC digital-to-analog converter and an analog filter. The clock signal 101 is used for clocking the DAC digital-to-analog converter. The memory circuit 1 outputs a baseband signal 107 and an encoded signal 108 to the DDS circuit 3, and the DDS circuit 3 outputs an encoded intermediate frequency signal 110. The DDS circuit 3 performs the following operations: the real part of the baseband signal u (l) is multiplied by the real part of the encoded signal p (l), the imaginary part of the baseband signal u (l) is multiplied by the imaginary part of the encoded signal p (l), and the two product terms are subtracted, resulting in:

wherein,

f′_n,mis a coding matrix F'_N×MArg () represents a function taking the phase angle of a complex number.

In the DDS circuit 3, the DAC and the analog filter perform digital-to-analog conversion on b (l), and the output analog signal is the encoded intermediate frequency signal 110, and the expression is:

wherein,

and an amplitude modulation circuit 4 for modulating the analog coded intermediate frequency signal 110 into an analog coded intermediate frequency pulse signal 111 by the pulse amplitude control signal 109 and outputting the analog coded intermediate frequency pulse signal to the frequency conversion circuit 5.

The output coded IF pulse signal 111 is expressed as

The frequency conversion circuit 5 is used for converting the local oscillator signal 112 and the coded intermediate frequency pulse signal 111 into a transmission pulse signal 113 of the wind profile radar.

The local oscillator signal 112 may be written asThe local oscillator signal 112 is multiplied by the coded intermediate frequency pulse signal 111 to filter the sum frequency or the difference frequency, and after amplification, the transmitted pulse signal 113 has the expression:

in the formula: f. of₀＝f_L±f_b，f₀In order to transmit the carrier frequency of the pulses,the pulse carrier frequency phase is transmitted. And a is the signal amplitude.

With the development of hardware technologies such as calculation, storage, waveform generation and the like, a Frank code matrix F is adopted_N×NIt is increasingly easy to perform phase encoding of multiple phases. The wind profile radar phase encoding method and circuit based on the Frank code can adopt common pulse or pulse train waveform, have low distance side lobe and flexible selection of sub-pulse number, can be used for resisting active interference signals, are widely applied to various wind profile radars, and have good application prospect.

Drawings

FIG. 1 is a schematic diagram of the radar transmitting pulse of the present invention

FIG. 2 is a schematic diagram of a radar transmission pulse encoding period

FIG. 3 is a schematic diagram of phase encoding of a transmit pulse for one encoding cycle with an encoding matrix

FIG. 4 is a schematic block diagram of a phase encoded signal generating circuit

FIG. 5 is a flow chart of the steps of the method of the present invention in embodiments 1, 2, 3, and 5

FIG. 6 is a flowchart of the steps of embodiments 4 and 6 of the method of the present invention

Fig. 7 is a flow chart of the calculation and control of the control computer 6 according to the embodiment 7 of the circuit of the present invention.

Fig. 1 shows a schematic waveform of a transmitted pulse in a radar cycle, showing a pulse train waveform. If sub-pulse interval τ_rEqual to the sub-pulse width τ, the sub-pulse is a wide normal pulse in waveform, and the phase encoding is performed on the equally divided sub-pulses within the wide normal pulse. And in the time period shown by the total width of the transmitted pulse, the receiving channel is closed by a receiving and transmitting switch of the pulse wind profile radar, and data is not collected. And in the time period shown by the range delay, the receiving channel is opened, and the intermediate frequency receiver of the radar acquires data.

As seen in the figure, one radar period T_rWithin, there are M sub-pulses with sub-pulse width τ and sub-pulse interval τ_rRadar period T_r＝τ_PulseAll+τ_RmaxWherein the total width τ of the transmitted pulse_PulseAll＝τ_r(M-1) + tau, range delay tau_Rmax＝2R_max/C，R_maxFor the range, C represents the speed of light. Period T of wind profile radar_rIs mainly dependent on τ_Rmax，τ_Rmax＞＞τ_PulseAllConsidering the effect of reducing range mapping, the period T of the general radar_rDesigned to be much larger than the range delay tau_RmaxI.e. T_r＞＞τ_Rmax。

FIG. 2 shows a schematic diagram of the coding period of a transmitted pulse, one coding period NT_rOver time, there are N radar cycles, each of which T is shown in FIG. 2_rTotal width of transmitted pulse tau_PulseAllThis time period. In fig. 2 it is shown that the coding periods follow one another in time, for a total of N_cN_FFTOne radar period, N_cN_FFTN is an integer, the number of coding cycles is N_cN_FFTand/N is an integer.

Fig. 3 shows a schematic diagram of phase encoding of a transmit pulse for one encoding cycle with an encoding matrix. The time axis is divided into N segments in fig. 3. The period of time 1 represents [0, T_r) Representing the first radar cycle period, the period of time 2 representing [ T_r，2T_r) Representing a second radar cycle period, the period of time N representing [ (N-1) T-_r，NT_r) Representing the nth radar cycle time period.

The N radar periods constitute a radar coding period. There are M transmitted sub-pulses per radar cycle. The sub-pulse shown in the figure is a pulse train waveform, and if the sub-pulse interval is equal to the sub-pulse width, the sub-pulse is a wide normal pulse in waveform, and the phase encoding is performed on the equally divided sub-pulses in the wide normal pulse.

FIG. 3 shows a code matrix F 'for controlling phase encoding of N × M sub-pulses in one code period'_N×M，f′_n,mIs coding matrix F'_N×MOf (1).

F′_N×MOf (a) is the first line vector (f'_1,1,f′_1,2,…,f′_1,M) Phase encoding, f 'of a first radar period within a coding period'_1,1Corresponding to the first sub-pulse, f'_1,2Corresponding to the second sub-pulse, f'_1,MCorresponding to the mth sub-pulse. F'_N×MSecond row vector of (f'_2,1,f′_2,2,…,f′_2,M) For the second mine in the coding periodUp to period for phase encoding, f'_2,1Corresponding to the first sub-pulse, f'_2,2Corresponding to the second sub-pulse, f'_2,MCorresponding to the mth sub-pulse. F'_N×MN-th row vector (f'_N,1,f′_N,2,…,f′_N,M) Phase encoding, f 'of the Nth radar period within the encoding period'_N,1Corresponding to the first sub-pulse, f'_N,2Corresponding to the second sub-pulse, f'_N,MCorresponding to the mth sub-pulse.

F′_N×MOf (d) is the first column vector (f'_1,1,f′_2,1,…,f′_N,1)^TCorresponding to the first subpulse, F ', of each radar period in the coding period'_N×MThe second column vector of (f'_1,2,f′_2,2,…,f′_N,2)^TCorresponding to the second subpulse, F ', of each radar period in the coding period'_N×MM-th column vector (f'_1,M,f′_2,M,…,f′_N,M)^TCorresponding to the mth sub-pulse of each radar cycle in the coding period.

Fig. 4 is a schematic block diagram of a phase encoded signal generating circuit, which comprises: the device comprises a storage circuit 1, a time sequence circuit 2, a DDS circuit 3, an amplitude modulation circuit 4, a frequency conversion circuit 5 and a control computer 6. The input, output and connections in fig. 4 include:

clock signal 101: clock signal of DDS circuit 3

Clock signal 102: clock signal for memory circuit 1

Control signals 103: data download and control signals supplied by the control computer 6 to the memory circuit 1

Clock signal 104: clock signal of sequential circuit 2

Control signals 105: data download and control signals supplied by control computer 6 to sequential circuit 2

Control signals 106: control signal of sequential circuit 2 to memory circuit 1

Baseband signal 107: baseband signal output from memory circuit 1 to DDS circuit 3

The encoded signal 108: the coded signal output from the memory circuit 1 to the DDS circuit 3

Control signals 109: control signal of sequential circuit 2 to amplitude modulation circuit 4

Encoding the intermediate frequency signal 110: baseband coded intermediate frequency signal output by DDS circuit 3

Encoded intermediate frequency pulse signal 111: baseband coded intermediate frequency pulse signal output by amplitude modulation circuit 4

Local oscillator signal 112: local oscillator input signal of frequency conversion circuit 5 provided by radar frequency synthesizer

Transmission pulse signal 113: the output signal of the up-conversion circuit 5.

FIG. 5 is a flow chart of the steps of the method of the present invention in embodiments 1, 2, 3, and 5.

Step 201: calculating the number M of coded sub-pulses

Step 202: calculating the time domain accumulation number N_c

Step 203: selecting the number of coding cycles N

Step 204: adjusting the time-domain accumulation number N_cAnd the number of coding cycles N

Step 205: form a coding matrix F'_N×M

Step 206: with coding matrix F'_N×MPerforming phase encoding for a plurality of encoding periods to complete N_cN_FFTPhase encoding of/N encoding periods.

Fig. 6 is a flow chart of the steps of the method embodiments 4, 6 of the present invention.

Step 301: calculating the number M of coded sub-pulses

Step 302: calculating the time domain accumulation number N_c

Step 303: selecting the number of coding cycles N

Step 304: adjusting the time-domain accumulation number N_cAnd the number of coding cycles N

Step 305: form a coding matrix F'_N×M

Step 306: with coding matrix F'_N×MPerforming phase encoding for a plurality of encoding periods to complete N_cN_FFTPhase encoding of/N encoding periods.

Fig. 7 is a flow chart of the calculation and control of the control computer 6 according to the circuit embodiment 7 of the present invention. The steps for performing one calculation and control are shown in fig. 7.

Step 401: calculating the number M of coded sub-pulses

Step 402: calculating the time domain accumulation number N_c

Step 403: selecting the number of coding cycles N

Step 404: adjusting the time-domain accumulation number N_cAnd the number of coding cycles N

Step 405: form a coding matrix F'_N×M

Step 406: downloading timing parameters of timing circuit 2

Step 407: download coding matrix F'_N×MTo the memory circuit 1 and to start the sequential circuit 2.

In the drawings, the components represented by the respective reference numerals are listed below:

1. the device comprises a storage circuit 2, a time sequence circuit 3, a DDS circuit 4, an amplitude modulation circuit 5 and a frequency conversion circuit.

Detailed Description

The following description of the embodiments of the method and circuit of the present invention is provided for illustration, and the examples are provided to explain and fully illustrate the technical solution and applicability of the present invention.

Example 1:

movable L-band boundary layer wind profile radar with working frequency f₀1320MHz, maximum probe height R_maxThe wind speed measurement range is 0-60 m/s (the maximum wind speed V is 3 km)_max60m/s), the distance resolution of the high mode is D120 m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 10%, and the tilt angle α of the oblique wave beam is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_maxC is 20 mu S, C is light speed, and radar period T of wind profile radar is designed_r50 mu S, can meet the requirement of distance measurement range, and T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllThe allowable value is large, a pulse train waveform can be adopted, and the sub-pulse interval is 2 times of the sub-pulse width. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c).

FFT point number N for signal processing of wind profile radar_FFTTypically 512. Operating wavelength lambda according to operating frequency f₀Calculated λ ═ C/f₀0.2273 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C is 0.8 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝15.4m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 201 in fig. 5, the Duty _ cycle is 10%, T_rThe number of sub-pulses M can be calculated to be 6 when τ is 0.8 μ S at 50 μ S, and the actual duty ratio is not more than 10% at 9.6%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 202 in fig. 5, λ is 0.2273m, V_Rmax＝15.3m/S，T_rThe time domain accumulation number N can be calculated as 50 μ S_cCalculating to obtain N_c＝74。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy LM ≤ N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For this example, see step 203 in FIG. 5, 6 ≦ N ≦ 74. Further selected is N-24, where k-3 and L-4 are present, where N satisfies LM ≦ N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTDivisible by N. For this example, referring to step 204 in FIG. 5, N will be_cFrom 74 to 72, N is not adjusted to 24, N_cHas small adjustment amount, little influence on the coherent time, no influence on the detection performance, and N after adjustment_cN_FFT1536 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×MFurther, in selecting M column vectors, column vectors are selected at intervals L, and the selected column vectors may form a new coding matrix F 'of N × M dimension in an arbitrary order'_N×M。

For this example, see step 205 in fig. 5, where L is 4 and N is 24, then F is_24×24Selecting 6 column vectors according to the sequence with the interval L-4 to obtain a 24 × 6-dimensional Frank coding matrix F 'consisting of 6 column vectors'_24×6Comprises the following steps:

C′_n,m＝4nm n＝0,1,2,…,23 m＝0,1,2,3,4,5

step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 206 in FIG. 5, with the above-described encoding matrix F'_24×6Carry out N_cN_FFTand/N1536 code periods of the phase encoding of the transmitted pulse.

Referring to fig. 2 and its description, fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_24×6Respectively phase-encoding the transmit pulses of 24 radar periods within an encoding period, encoding matrix F'_24×6The 6 elements in each row vector phase encode the 6 sub-pulses in the corresponding period, respectively. Referring to FIG. 3 and its description, FIG. 3 represents a coding matrix F'_N×MOf element f'_n,mIn relation to the relationship between the radar cycle number n and the sub-pulse number m, the radar cycle number n is a number within one radar encoding cycle.

Example 2:

fixed L-band boundary layer wind profile radar with working frequency f₀1320MHz, maximum probe height R_maxThe wind speed measurement range is 0-60 m/s (the maximum wind speed V is 6 km)_max60m/s), the distance resolution of the high mode is D120 m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 10%, and the tilt angle α of the oblique wave beam is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_maxDesigning radar period T of wind profile radar, wherein,/C is 40 mu S, C is light speed_r80 mu S can meet the requirement of distance range time delay, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllAllowing for larger values, a pulse train waveform may be employed with sub-pulse spacing of 2 times the sub-pulse width. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength lambda is calculated according to the operating frequency f, and lambda is 0.2273 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C is 0.8 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝15.3m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 201 in fig. 5, the Duty _ cycle is 10%, T_rThe sub-pulse number 10 can be calculated at 80 μ S, τ 0.8 μ S, with a duty cycle of 10% not exceeding 10%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 202 in fig. 5, λ is 0.2273m, V_Rmax＝15.3m/S，T_rThe time domain accumulation number N can be calculated as 80 μ S_cCalculating to obtain N_c＝46。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy LM ≤ N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For this example, see step 203 in FIG. 5, 10 ≦ N ≦ 46. Further selected is N10, k 4.6, L1, where N satisfies LM N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation.

For this example, referring to step 204 in FIG. 5, the time domain accumulation number N_cAdjusted to 45 from 46, N is 10 and is not changed, N_cHas small adjustment amount, little influence on the coherent time, no influence on the detection performance, and N after adjustment_cN_FFT2304 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×MFurther, in selecting M column vectors, column vectors are selected at intervals L, and the selected column vectors may form a new coding matrix F 'of N × M dimension in an arbitrary order'_N×M。

For this example, see step 205 in fig. 5, where L is 1 and N is 10, then F is_10×1010 column vectors are selected in sequence to obtain a 10 × 10-dimensional Frank coding matrix F 'consisting of 10 column vectors'_10×10Comprises the following steps:

C′_n,m＝nm n＝0,1,2,…,9 m＝0,1,2,…,9

step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 206 in FIG. 5, with the above-described encoding matrix F'_10×10Carry out N_cN_FFTPhase encoding of the transmitted pulses for 2304 encoding periods.

Referring to fig. 2 and its description, fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_10×10Respectively phase-encoding the transmit pulses of 10 radar periods within an encoding period, encoding matrix F'_10×10The 10 elements in each row vector phase encode 10 sub-pulses in the corresponding period, respectively, see fig. 3 and its description. FIG. 3 shows a coding matrix F'_N×MOf element f'_n,mIn relation to the relationship between the radar cycle number n and the sub-pulse number m, the radar cycle number n is a number within one radar encoding cycle.

Example 3:

fixed L-band boundary layer wind profile radar with working frequency f₀1320MHz, maximum probe height R_maxThe wind speed is measured in the range of 0-80m/s (maximum wind speed V) 10km_max80m/s), the distance resolution of the high mode is 240m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 10%, and the inclined beam inclination angle α is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_maxDesigning radar period T of wind profile radar as 66.7 mu S/C and light speed C_r100 mu S, can meet the requirement of distance measurement range, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllThe maximum value is 33.3 muS, and the transmission pulse can adopt a pulse shape. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength lambda is calculated according to the operating frequency f, and lambda is 0.2273 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C1.6 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝20.4m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 201 in fig. 5, the Duty _ cycle is 10%, T_rThe number of sub-pulses can be calculated at 100 μ S, τ 1.6 μ S, and M6, with a duty cycle of 9.6% not more than 10%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 202 in fig. 5, λ is 0.2273m, V_Rmax＝20.4m/S，T_rThe time domain accumulation number N can be calculated as 100 μ S_cCalculating to obtain N_c＝28。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N in time domain_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy LM ≤ N_cK, L is a positive integer, k is a positive real number satisfying k ≧ 1, and kAnd L satisfies the constraint kL is more than or equal to 4.

For this example, referring to step 203 in fig. 5, 6 ≦ N ≦ 28, further selected N ≦ 6, and there are k ≦ 4.67 and L ≦ 1, where LM ≦ N is satisfied_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation.

For this example, referring to step 204 in FIG. 5, the time domain accumulation number N_cThe value is adjusted from 28 to 27, N is not changed to 6, the influence on the coherent time is very small, the detection performance is not influenced, and N is adjusted_cN_FFT2304 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×MFurther, in selecting M column vectors, column vectors are selected at intervals L, and the selected column vectors may form a new coding matrix F 'of N × M dimension in an arbitrary order'_N×M。

For this example, see step 205 in fig. 5, where L is 1 and N is 6, then F is_6×66 column vectors are selected in sequence to obtain a 6 × 6-dimensional Frank coding matrix F 'consisting of 6 column vectors'_6×6Comprises the following steps:

C′_n,m＝nm n＝0,1,2,…,5 m＝0,1,2,…,5

step 6: by braidingCode matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 206 in FIG. 5, with the above-described encoding matrix F'_6×6Carry out N_cN_FFTPhase encoding of the transmitted pulses for 2304 encoding periods.

See fig. 2 and its description. Fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_6×6Respectively phase-encoding the transmit pulses of 6 radar periods within an encoding period, encoding matrix F'_6×6The 6 elements in each row vector phase encode the 6 sub-pulses in the corresponding period, respectively, see fig. 3 and its description. FIG. 3 shows a coding matrix F'_N×MOf element f'_n,mIn relation to the relationship between the radar cycle number n and the sub-pulse number m, the radar cycle number n is a number within one radar encoding cycle.

For this example, the number of time-domain accumulations N of signal processing_c27, number of points of FFT N_FFTInstead of performing actual time domain accumulation, 512 may be implemented by performing doppler spectrum analysis using 27 × 512-13824 point FFT, discarding the samples representing values greater than V_RmaxReserving V or less for 20.4m spectrum data_RmaxThe spectrum data of 20.4m is used as effective data. Time domain accumulation number N at this time_cAs an important parameter in the design process 27.

Example 4:

fixed P-band troposphere wind profile radar with working frequencyf₀445MHz, maximum probe height R_maxThe wind speed is measured in the range of 0-100m/s (maximum wind speed V) at 12km_max100m/s), the distance resolution of the high mode is 240m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 10%, and the inclined beam inclination α is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_maxThe speed of light is 80 mu S, C is the speed of light, and the radar period T of the wind profile radar is designed_r120 mu S can meet the requirement of measuring range, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllThe allowable value is large, the emission pulse can adopt a pulse train waveform, and the sub-pulse interval is 2 times of the sub-pulse width. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength λ is calculated from the operating frequency f, and λ ═ C/f ═ 0.674 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C1.6 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝25.6m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 301 in fig. 6, the Duty _ cycle is 10%, T_rThe sub-pulse number can be calculated as 120 μ S, τ 1.6 μ S, and M7, in this caseThe duty ratio is 9.3% and not more than 10%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 302 in fig. 6, λ is 0.674m, V_Rmax＝25.6m/S，T_rThe time domain accumulation number N can be calculated as 120 μ S_cCalculating to obtain N_c＝54。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N in time domain_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy L (M +1) N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For this example, referring to step 303 in fig. 6, 7 ≦ N ≦ 54, further selected N-8, and there is k-6.75 and L-1, where L (M +1) ≦ N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation.

For this example, see step 304 in FIG. 6, N_cAnd N is adjusted to 0, N_cN_FFT3456 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×M. Further in choosing the M column vectors according to the intervalColumn vectors are selected at intervals of L, and the selected column vectors can form a new coding matrix F 'with N × M dimensions in any order'_N×M。

For this example, see step 305 in fig. 6, where L is 1 and N is 8, then F is_8×88 column vectors are included, 7 column vectors are selected in sequence to obtain an 8 × 7-dimensional Franko code matrix F 'consisting of 7 column vectors'_8×7Comprises the following steps:

C′_n,m＝nm n＝0,1,2,…,7 m＝0,1,2,…,6

step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 306 in FIG. 6, with the above-described encoding matrix F'_8×7Carry out N_cN_FFTThe phase coding of the transmitted pulses for 3456 coding cycles.

Referring to fig. 2 and its description, fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_8×7Respectively phase-encoding the transmit pulses of 8 radar periods within an encoding period, encoding matrix F'_8×7The 7 elements in each row vector phase encode the 7 sub-pulses in the corresponding period, respectively, see fig. 3 and its description. FIG. 3 shows a coding matrix F'_N×MOf element f'_n,mWith respect to the radar cycle number n and the sub-pulse number m, where the radar cycle number n is defined asA sequence number within one radar encoding period.

N > M in this example, coding matrix F'_8×7Medium unselected matrix F_8×8If point-to-frequency interference occurs at a certain frequency point, selecting M-7 column vectors to form a coding matrix F 'of 8 × 7 dimensions'_8×7And when the point frequency interference is detected, the column vector corresponding to the point frequency interference is not selected, so that the point frequency interference is avoided and resisted.

Example 5:

fixed P-band high troposphere wind profile radar with working frequency f₀445MHz, maximum probe height R_maxThe wind speed is measured in the range of 0-100m/s (maximum wind speed V) at 16km_max100m/s), the distance resolution of the high mode is 480m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 20%, and the inclined beam inclination angle α is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_max106.6 mu S/C, C is light speed, and radar period T of the wind profile radar is designed_r160 mu S can meet the requirement of measuring range, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllThe maximum value is 53.4 mus, and the transmission pulse can adopt a pulse shape. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength λ is calculated from the operating frequency f, and λ ═ C/f ═ 0.674 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C is 3.2 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝25.6m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 201 in fig. 5, the Duty _ cycle is 20%, T_rThe sub-pulse number can be calculated at 160 μ S, τ 3.2 μ S, and M10, which is a duty ratio of 20%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 202 in fig. 5, λ is 0.674m, V_Rmax＝25.6m/S，T_r160 mus, the time domain accumulation number N can be calculated_cCalculating to obtain N_c＝41。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy LM ≤ N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For this example, referring to step 203 in fig. 5, 10 ≦ N ≦ 41, further selected N-20, and there are k-2 and L-2, where LM ≦ N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting time domainNumber of accumulations N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation. For this example, see step 204 in FIG. 5, N_cAdjusted to N41_c40, the influence on the coherent time is very small, the influence on the detection performance is not influenced, the adjustment amount of N is 0, and N is_cN_FFTand/N is an integer of 1024.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×MFurther, in selecting M column vectors, column vectors are selected at intervals L, and the selected column vectors may form a new coding matrix F 'of N × M dimension in an arbitrary order'_N×M。

For this example, see step 205 in fig. 5, where L is 2 and N is 20, then F is_20×20Selecting 10 column vectors according to the sequence with the interval L-2 to obtain a 20 × 10-dimensional Frank coding matrix F 'consisting of 10 column vectors'_20×10Comprises the following steps:

C′_n,m＝2nm n＝0,1,2,…,19 m＝0,1,2,…,9

step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 206 in FIG. 5, using the encoding matrix described aboveF′_20×10Carry out N_cN_FFTand/N is 1024 code periods of the phase code of the transmission pulse.

Referring to fig. 2 and its description, fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_20×10Respectively phase-encoding the transmit pulses of 20 radar periods within an encoding period, encoding matrix F'_20×10The 10 elements in each row vector phase encode 10 sub-pulses in the corresponding period, respectively, see fig. 3 and its description. FIG. 3 shows a coding matrix F'_N×MOf element f'_n,mIn relation to the relationship between the radar cycle number n and the sub-pulse number m, the radar cycle number n is a number within one radar encoding cycle.

Example 6:

fixed UHF band high troposphere wind profile radar with working frequency f₀52MHz, maximum probe height R_maxThe wind speed is measured in the range of 0-120m/s (maximum wind speed V) at 24km_max120m/s), the distance resolution of the high mode is 480m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 20%, and the inclined beam inclination angle α is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIs calculated τ_Rmax＝2R_max160 mu S/C, C is light speed, and radar period T of wind profile radar is designed_r240 mu S can meet the requirement of distance measurement range, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllThe maximum value is 80 mus, and the transmit pulse may take the form of a pulse. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength λ is calculated from the operating frequency f, λ ═ C/f ═ 5.77 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C is 3.2 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝30.6m/S。

For the embodiment, the specific implementation of the technical scheme of the method of the invention is as follows:

a wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and comprises the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

In this embodiment, referring to step 301 in fig. 6, the Duty _ cycle is 20%, T_rThe number of sub-pulses can be calculated at 240 μ S, τ 3.2 μ S, and M15, with a duty cycle of 20% not more than 20%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

In this embodiment, see step 302 in fig. 6, λ is 5.77m, V_Rmax＝30.6m/S，T_rThe time domain accumulation number N can be calculated as 240 μ S_cCalculating to obtain N_c＝196。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy L (M +1) N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For theIn this example, referring to step 303 of fig. 6, 15 ≦ N ≦ 196, further selected as N ≦ 16, where k is 12.25 and L is 1, where L (M +1) ≦ N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe FFT point number of FFT spectrum analysis after signal processing time domain accumulation. For this example, see step 304 in FIG. 6, N_cAnd N is adjusted by 0, N_cN_FFT6272 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected from the N-column vector group, and the M column vectors are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×MFurther, in selecting M column vectors, column vectors are selected at intervals L, and the selected column vectors may form a new coding matrix F 'of N × M dimension in an arbitrary order'_N×M。

For this example, see step 305 in fig. 6, where L is 1 and N is 16, then F is_16×16Selecting 15 column vectors according to the sequence with the interval L-1 to obtain a 16 × 15-dimensional Frank coding matrix F 'consisting of 15 column vectors'_16×15Comprises the following steps:

C′_n,m＝nm n＝0,1,2,…,15 m＝0,1,2,3,…,14

step 6: with coding matrix F'_N×MPhase encoding of the transmission pulses of the encoding period is performed.

Coding periods are temporally one after the other, with coding matrix F'_N×MCompleting phase encoding for multiple encoding periods cyclically one by one to complete N_cN_FFTPhase encoding of/N encoding periods.

For this example, see step 306 in FIG. 6, with the above-described encoding matrix F'_16×15Carry out N_cN_FFT6272 encoding periods of the phase encoding of the transmitted pulse.

Referring to fig. 2 and its description, fig. 2 shows that a coding cycle comprises N radar cycles, the coding cycles being one after the other in time.

Coding matrix F'_16×15Respectively phase-encoding the transmit pulses of 16 radar periods within an encoding period, encoding matrix F'_16×15The 15 elements in each row vector phase encode the 15 sub-pulses in the corresponding period, respectively, see fig. 3 and its description. FIG. 3 shows a coding matrix F'_N×MOf element f'_n,mIn relation to the relationship between the radar cycle number n and the sub-pulse number m, the radar cycle number n is a number within one radar encoding cycle.

N > M in this example, not selected to code matrix F'_16×15F in (1)_16×16If the dot frequency interference occurs on the corresponding frequency point of the last column vector of (1), the dot frequency interference can be suppressed.

Example 7:

fixed L-band boundary layer wind profile radar with working frequency f₀1320MHz, maximum probe height R_maxThe wind speed is measured in the range of 0-80m/s (maximum wind speed V) 10km_max80m/s), the distance resolution of the high mode is 240m, the high mode adopts phase coding, the maximum Duty cycle of the transmitted pulse is 10%, and the inclined beam inclination angle α is 14.8 °.

Range delay tau_RmaxAccording to the maximum detection height R_maxIt is calculated that the average value of the values,τ_Rmax＝2R_maxdesigning radar period T of wind profile radar as 66.7 mu S/C and light speed C_r100 mu S, can meet the requirement of distance measurement range, T_r＞＞τ_RmaxTotal width of emission pulse τ_PulseAllMaximum value of 33.3 muS, and pulse shape, tau, can be used for the emission pulse_rτ. Referring to FIG. 1 and its description, FIG. 1 shows the total width τ of the transmitted pulse in one radar cycle_PulseAllTime delay of measuring range tau_RmaxAnd radar period T_rThe relationship (2) of (c). FFT point number N for signal processing of wind profile radar_FFTTypically 512. The operating wavelength lambda is calculated according to the operating frequency f, and lambda is 0.2273 m. The sub-pulse width τ is calculated from the distance resolution D, and τ is 2D/C1.6 μ S. Tau is_rτ is 1.6 μ S. Maximum radial velocity V_RmaxAccording to the maximum wind speed V_maxAnd the oblique beam inclination α_Rmax＝V_maxsin(α),V_Rmax＝20.4m/S。

For the embodiment, the specific implementation of the circuit technical scheme of the invention is as follows:

a wind profile radar phase encoding circuit based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and the circuit composition comprises: the device comprises a storage circuit 1, a time sequence circuit 2, a DDS circuit 3, an amplitude modulation circuit 4, a frequency conversion circuit 5 and a control computer 6.

The phase encoding circuit controls the computer (6) to execute the following steps:

step 1, calculating the number M of the encoding sub-pulses.

M＝uint[Duty_cycle T_r/τ]

For this example, see step 401 in fig. 7, where Duty _ cycle is 10%, T_rThe number of sub-pulses can be calculated at 100 μ S, τ 1.6 μ S, and M6, with an actual duty cycle of 9.6% not exceeding 10%.

Step 2, calculating the time domain accumulation number N_c。

N_c＝uint[λ/(4V_RmaxT_r)]

For this example, see step 402 in fig. 7, λ ═ 0.2273m, V_Rmax＝20.4m/S，T_rThe time domain accumulation number N can be calculated as 100 μ S_c＝uint[λ/(4V_RmaxT_r)]Calculating to obtain N_c＝28。

And 3, selecting the number N of the coding cycles.

According to the number M of coded sub-pulses and the accumulated number N of signal processing time domains_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c. Further selecting N to make N satisfy LM ≤ N_cAnd k, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4.

For this example, referring to step 403 in fig. 7, N is selected to be 6, there is k 4.66 and L1, where LM ≦ N_cK, and satisfies the constraint kL is more than or equal to 4.

Step 4, adjusting the time domain accumulation number N_cAnd a number N of coding cycles.

Adjusting the time-domain accumulation number N_cAnd the number of coding cycles N, such that N_cN_FFTDivisible by N, i.e. Q ═ N_cN_FFTN is an integer, N_FFTThe number of FFT points of FFT spectrum analysis after time domain accumulation.

For this example, referring to step 404 in FIG. 7, the time domain accumulation number N_cThe value is adjusted from 28 to 27, N is not changed to 6, the influence on the coherent time is very small, the detection performance is not influenced, and N is adjusted_cN_FFT2304 is an integer.

Step 5, forming a coding matrix F'_N×M。

Selecting an N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected according to an interval L and combined according to an arbitrary order to form an N × M-dimensional coding matrix F'_N×M。

For this example, see step 405 in fig. 7, where L is 1 and N is 6, then F is_6×66 column vectors are selected in sequence to obtain a 6 × 6-dimensional Franko code matrix F 'consisting of 6 column vectors'_6×6Comprises the following steps:

C′_n,m＝nm n＝0,1,2,…,5 m＝0,1,2,…,5

and 6, downloading the time sequence parameters of the time sequence circuit 2.

For the present example, see step 406 in FIG. 7, these parameters include the radar period T_rThe number M of sub-pulses, the sub-pulse pitch τ_rSub-pulse width tau, number of coding cycles N, number of time domain accumulations N_cSum velocity FFT spectral analysis point number N_FFT。

Step 7, downloading the coding matrix F'_N×MAnd starts the sequential circuit 2. Encoding matrix F 'by control signal 103'_N×MThe matrix is downloaded to the memory circuit 1 and the timing circuit 2 is started, see step 407 in fig. 7.

A functional block diagram of the circuit of the present invention is shown in fig. 4.

A memory circuit 1 for storing baseband signal data of one radar cycle and a code matrix F'_N×MN is sent to the DDS circuit 3 by the clock signal 102 and the control signal 106 from the timing circuit 2_cN_FFTBaseband signals 107 and N for N coding periods_cN_FFTA coded signal 108 of/N coding cycles. In this example, the frequency of the clock signal 102 is designed to be 240MHz, and the sampling time of the baseband signal is the reciprocal of the frequency of the clock signal 102, t_sμ S (1/240). The frequency of the baseband signal is designed as f_b60 MHz. The phase of the baseband signal is not designed to,

the baseband signal 107 in the coherent time output from the memory circuit 1 has the expression

The coded signal 108 in the coherent time output by the memory circuit 1 is expressed as

Wherein Q is N_cN_FFT/N

v₁(t) is the following pulse function

f′_n.,mIs F'_N×MOne element of (1). Design does not require radar period T_rAnd sub-pulse spacing τ_rIs the baseband signal sampling time t_sInteger multiples of. In this example, T_r＝100μS，τ_r＝τ＝1.6μS，t_sAnd (1/240) mu S, and the integral multiple relation is satisfied.

A sequential circuit 2 for outputting a control signal 106 to control the memory circuit 1 under the action of a clock signal 104 and a control signal 105 provided by the control computer 6, so that the memory circuit 1 generates a baseband signal 107 and an encoding signal 108, and generates N_cN_FFTThe pulse amplitude control signal 109 for the/N coding periods is sent to the amplitude modulation circuit 4.

The pulse amplitude control signal 109 in coherent time generated by the sequential circuit 2 has the expression

v₂(t) is the following pulse function

τ is the sub-pulse width, for this example τ ═ τ_rThe sub-pulse width is equal to the sub-pulse spacing.

The sequential circuit 2 controls the address and data buses of the memory 1 according to the sequential parameters, and enables the memory 1 to output the baseband signal 107 and the coding signal 108 during normal operation. For this example, the frequency of the clock signal 104 is designed to be 10MHz, which can achieve the control accuracy requirement of 0.1 μ S for the pulse width.

And a DDS circuit 3, which is driven by the clock signal 101 to complete data synthesis and digital-to-analog conversion of the baseband signal 107 and the encoded signal 108, and output an analog encoded intermediate frequency signal 110.

The DDS circuit 3 comprises two paths of digital multipliers, an adder, a DAC digital-to-analog converter and an analog filter. The clock signal 101 is used for clocking the DAC digital-to-analog converter. The memory circuit 1 outputs a baseband signal 107 and an encoded signal 108 to the DDS circuit 3, and the DDS circuit 3 outputs an encoded intermediate frequency signal 110. The DDS circuit 3 performs the following operations: the real part of the baseband signal u (l) is multiplied by the real part of the encoded signal p (l), the imaginary part of the baseband signal u (l) is multiplied by the imaginary part of the encoded signal p (l), and then subtracted, the result being:

wherein,

for this example, the frequency of the clock signal 101 of the DDS circuit 3 is also designed to be 240MHz, which is the same as the frequency of the clock signal 101.

In the DDS circuit 3, the DAC and the analog filter perform digital-to-analog conversion on b (l), and the output analog signal is the encoded intermediate frequency signal 110, and the expression is:

wherein,

and an amplitude modulation circuit 4 for modulating the coded intermediate frequency signal 110 into a coded intermediate frequency pulse signal 111 by the pulse amplitude control signal 109 and outputting the coded intermediate frequency pulse signal to the frequency conversion circuit 5.

The output coded IF pulse signal 111 is expressed as

The frequency conversion circuit 5 is used for converting the local oscillator signal 112 and the coded intermediate frequency pulse signal 111 into a transmission pulse signal 113 of the wind profile radar.

The local oscillator signal 112 may be written asThe local oscillator signal 112 is multiplied by the coded intermediate frequency pulse signal 111 to filter the sum frequency or the difference frequency, and after amplification, the transmitted pulse signal 113 has the expression:

in the formula: f. of₀＝f_L±f_b，f₀In order to transmit the carrier frequency of the pulses,the pulse carrier frequency phase is transmitted. And a is the signal amplitude. For the present example, the frequency f of the local oscillator signal 112_L＝1260MHz，Q＝N_cN_FFT/N＝2304，N＝6，M＝6，

T_r＝100μS，τ_r＝1.6μS，τ＝1.6μS，f₀＝f_L+f_b＝1320MHz，N_c＝27，N_FFT＝512，t∈[0,N_cN_FFTT_r)，N_cN_FFTT_r＝1.3824S。

Defining a relative time t_q,n,m＝t-qNT_r-nT_r-mτ_rWhen t is equal to t_q,n,m+qNT_r+nT_r+mτ_rRelative time t_q,n,mExpress buttonThe waveforms of the sub-pulses are collected (described) with the starting time of each sub-pulse as the time zero.

Substituting the expression of t into the carrier part of the above formula due to 2 pi f₀(qNT_r+nT_r+mτ_r)＝2πl，l＝qN(f₀T_r)+n(f₀T_r)+m(f₀τ_r) Not requiring f in design₀T_rAnd f₀τ_rAre integers, then l is an integer. Each sub-pulse of the transmit pulse may be rewritten as:

in this embodiment, phase encoding of sub-pulsesThe values are tabulated as follows:

the above examples are only preferred embodiments of the present invention, and are not intended to limit the present invention. The method and the circuit are used for a frequency integrated circuit waveform generating part of the wind profile radar, provide a phase coding technology based on a Frank code for the wind profile radar, adopt a new code pattern, indicate the relation between the dimension of a coding matrix and a time domain accumulation number related to signal processing, indicate the constraint relation required for obtaining good distance side lobe performance, illustrate a realizing circuit, enumerate a plurality of embodiments, and fully illustrate that the method and the circuit can be applied to wind profile radars of various specifications and models in the field of the wind profile radar, and have wider applicability.

Claims (6)

1. A wind profile radar phase encoding method based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and is characterized by comprising the following steps:

step 1, calculating the number M of encoding sub-pulses according to the maximum Duty ratio Duty _ cycle of the transmitted pulse and the radar period T_rAnd calculating the number M of coded sub-pulses according to the sub-pulse width tau;

step 2, calculating the time domain accumulation number N_cMeasuring the range V according to the working wavelength lambda and radial velocity_RmaxAnd radar period T_rCalculating the time domain accumulation number N_c；

Step 3, selecting the number N of the coding cycles according to the number M of the coded sub-pulses and the time domain accumulation number N_cSelecting the number N of coding cycles such that N satisfies the condition that M is less than or equal to N_c；

Step 4, adjusting the time domain accumulation number N_cAnd the number N of coding cycles, adjusting the number N of accumulation in the time domain_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, where N_FFTThe number of FFT points for FFT spectrum analysis after time domain accumulation;

step 5, forming a coding matrix F'_N×MSelecting a N-dimensional Frank code matrix F_N×NIn a Frank code matrix F_N×NM column vectors are selected from the N column vectors, and are combined in an arbitrary order to form an N × M-dimensional coding matrix F'_N×M；

Step 6, using a coding matrix F'_N×MPhase encoding of the transmitted pulses of the encoding period is performed with a coding matrix F'_N×MCarry out N_cN_FFTPhase encoding of the transmitted pulses of N encoding periods, encoding matrix F'_N×MRespectively phase-encoding the transmit pulses of N radar periods within an encoding period, an encoding matrix F'_N×MThe M elements in each row vector phase encode the M sub-pulses in the corresponding period, respectively.

2. The wind profile radar phase encoding method of claim 1, wherein said transmitted pulses comprise M encoded sub-pulses, the transmitted waveform being a pulse train waveform, the encoded sub-pulses having a defined time interval therebetween.

3. The wind profile radar phase encoding method of claim 1, wherein in step 3, the improvement is further such that N satisfies LM ≦ N_cK, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, k and L satisfy constraint kL being more than or equal to 4, and a coding matrix F 'is formed in step 5'_N×MIn a further development, the Frank code matrix F_N×NIn accordance with the interval LM column vectors are intentionally selected to form an N × M-dimensional coding matrix F'_N×M。

4. The wind profile radar phase encoding method of claim 1, wherein in step 3, the improvement is further such that N satisfies (M +1) N_c。

5. The wind profile radar phase encoding method of claim 1, wherein in step 3, the improvement is further such that N satisfies L (M +1) N_cK, and k and L still meet the constraint of kL being more than or equal to 4, and forming a coding matrix F 'in step 5'_N×MIn a further development, the Frank code matrix F_N×NRandomly selecting M column vectors according to an interval L to form an N × M-dimensional coding matrix F'_N×M。

6. A wind profile radar phase encoding circuit based on a Frank code uses an encoding matrix to carry out phase encoding of a plurality of encoding period transmitting pulse signals, and the circuit composition comprises: a storage circuit (1), a time sequence circuit (2), a DDS circuit (3), an amplitude modulation circuit (4), a frequency conversion circuit (5) and a control computer (6), the phase coding circuit is characterized in that,

the control computer (6) executes the following steps:

step 1, calculating the number M of encoding sub-pulses according to the maximum Duty ratio Duty _ cycle of the transmitted pulse and the radar period T_rAnd calculating the number M of coded sub-pulses according to the sub-pulse width tau;

step 2, calculating the time domain accumulation number N_cMeasuring the range V according to the working wavelength lambda and radial velocity_RmaxAnd radar period T_rCalculating the time domain accumulation number N_c；

Step 3, selecting the number N of the coding cycles according to the number M of the coded sub-pulses and the time domain accumulation number N_cSelecting the number N of coding cycles so that N satisfies LM ≤ N_cK, L is a positive integer, k is a positive real number satisfying k being more than or equal to 1, and k and L satisfy constraint kL being more than or equal to 4;

step 4, adjusting the time domain accumulation number N_cAnd the number N of coding cycles, adjusting the number N of accumulation in the time domain_cAnd the number of coding cycles N, such that N_cN_FFTCan be divided by N, N_FFTThe number of FFT points of FFT spectrum analysis after time domain accumulation;

step 5, forming a coding matrix F'_N×MSelecting a N-dimensional Frank code matrix F_N×NAt F_N×NM column vectors are selected according to an interval L and combined according to an arbitrary order to form an N × M-dimensional coding matrix F'_N×M；

And 6, downloading the time sequence parameters of the time sequence circuit (2), namely downloading the time sequence parameters to the time sequence circuit (2) through a control signal (105), wherein the parameters comprise a radar period T_rThe number M of sub-pulses, the sub-pulse pitch τ_rSub-pulse width tau, number of coding cycles N, number of time domain accumulations N_cSum velocity FFT spectral analysis point number N_FFT；

Step 7, downloading the coding matrix F'_N×MAnd a sequence circuit (2) is started, wherein the coding matrix F 'is coded by a control signal (103)'_N×MDownloading data to a storage circuit (1) and starting a sequential circuit (2);

the storage circuit (1) is used for pre-storing baseband signal data of one radar period and the downloaded coding matrix F'_N×MData, under the action of a clock signal (102) and a control signal (106) given by a sequential circuit (2), sends N to a DDS circuit (3)_cN_FFTA baseband signal (107) of/N coding periods and N_cN_FFTA coded signal (108) of/N coding periods;

the sequential circuit (2) is used for outputting a control signal (106) to control the storage circuit (1) under the driving of a clock signal (104), so that the storage circuit (1) generates a baseband signal (107) and an encoding signal (108), generates a pulse amplitude control signal (109) and sends the pulse amplitude control signal to the amplitude modulation circuit (4);

the DDS circuit (3) is used for completing data synthesis and digital-to-analog conversion of a baseband signal (107) and a coding signal (108) under the drive of a clock signal (101), and outputting an analog coding intermediate frequency signal (110);

the amplitude modulation circuit (4) is used for modulating the analog coded intermediate frequency signal (110) into an analog coded intermediate frequency pulse signal (111) under the control of the pulse amplitude control signal (109) and sending the analog coded intermediate frequency pulse signal to the up-conversion circuit (5);

and the up-conversion circuit (5) is used for converting the received local oscillator signal (112) and the coded intermediate frequency pulse signal (111) into a wind profile radar transmission pulse signal (113).