CN105204004A - Transmitting digital beam forming method based on digital delay and phase compensation - Google Patents

Abstract

The invention discloses a digital transmitting beam forming method based on digital delay and phase compensation. According to the method, a first DDS is designed by means of two accumulators and the CORDIC algorithm in an FPGA, and a simulative first intermediate-frequency signal is obtained; a second DDS is designed by means of the multiphase processing technique in an FPGA, a digital baseband signal of a second intermediate-frequency signal is obtained by means of two accumulators and the CORDIC algorithm, and a simulative second intermediate-frequency signal is obtained through interpolated filter digital frequency conversion; a frictional delay filter and an integral delay filter are designed, and a digital delay filter is realized in four phases; a first local oscillation signal and a second local oscillation signal are generated by means of output reference clock signals and frequency agility local oscillation signals, and a radio-frequency signal is generated through frequency mixing filtration; the radio-frequency signal is transmitted through an antenna array after being amplified, and beam forming is achieved in the space. According to the method, a narrow-band signal and a broad-band signal are generated by means of multiple DDS combinations, and narrow-band signal and broad-band signal transmitting beam forming is achieved by means of phase control and continuously variable digital delay.

Description

Based on the transmitting DBF method of digital delay and phase compensation

Technical field

The present invention relates to a kind of transmitting DBF method based on digital delay and phase compensation, belong to array signal process technique field.

Background technology

Phased-array radar can complete the multiple-tasks such as target search, tracking, guidance, is thus used widely in fields such as multiple goal trajectory measurement, air alert, multiple target tracking guided weapon systems.Large spatial domain, multiple goal, the demand such as anti-interference, impel radar to digitizing future development, and Digital Array Radar is the direction of following phased array technology development.Target is carried out to the requirement of imaging and identification, impel radar to adopt the signal waveform of wideband.The digital array of existing narrow band signal is by carrying out the control of digital beam froming based on the digital phase shift method of DDS, but launch the digital array of broadband signal, owing to there is aperture fill time, effectively can not form wave beam by means of only phase control, carry out the control of beam position.In order to solve the control problem of digital beam froming of digital array arrowband, broadband signal, pair array antenna element signal is needed to carry out digital delay and phase control, therefore, need to solve cell level signal digital production method, narrow band signal digit phase control, the problem of broadband signal digital delays time to control and phase compensation.

Between domestic existing Wide band array antenna unit, signal lag adopts the method for analog delay line, namely the analog delay line of K position state is utilized, signal lag between control module, as 5 analog delay lines, minimum time delay 1 wavelength, maximum delay 32 wavelength, analog delay line loss is large, volume is large, unstable, therefore, the control of the digital beam froming of digital array can not be used for.Digital delay can adopt the thinking of analog delay line, namely adopts the digital delay line of K mode bit, minimum time delay wavelength, maximum delay 2 ^kindividual wavelength, this is the method that studies in China unit mainly considers.We know the consistance of the Accuracy phase place of the minimum amount of delay of digital delay line and time delay, if delay precision is δ τ, then intermediate frequency phase error is:

Wherein, T is the signal period.Such as, T=170ps, the phase error of ± 7 °, the delay precision of requirement is 3.3ps.During wideband operation, by centre frequency computation delay and phase place, side frequency phase error is:

ΔP＝360°×δf×δτ

In formula, δ f is the difference on the frequency between side frequency and centre frequency, and τ is the amount of delay controlled.Therefore, the precision of digital delay line, affect the phase error of signal side frequency in array, time delay is more accurate, and phase error is less.The digital delay line of K mode bit is adopted to there is following points deficiency:

A. the resolution (minimum amount of delay) of time delay can not meet the phase accuracy requirement that wave beam controls, and needs to carry out time delay and phase control simultaneously;

B. limited delay resolution, the broadband signal side frequency phase error of generation, worsens the side lobe performance of broadband beams.

The research of continuously adjustabe digital delay line had made great progress in the world in recent years, propose a lot of theoretical and method for designing, be mainly used in the field such as communication modem, phonetic synthesis, signal sampling rate is relatively low, does not all consider the method for heterogeneous design.The digital direct of radar broadband signal is bonded into, and data transfer rate is higher, has exceeded the clock speed of digital circuit (as FPGA), need adopt heterogeneous synthetic method.Therefore, design heterogeneous continuously adjustabe digital delay wave filter, for the control of radar broadband signal transmitting DBF, become the key of broadband signal transmitting DBF.

Summary of the invention

Technical matters to be solved by this invention is to provide a kind of transmitting DBF method based on digital delay and phase compensation, its core technology is that multiple DDS combines and produces narrow band signal and broadband signal, utilize the digital delay of phase control and continuous variable, realize the launching beam formation of narrow band signal, broadband signal.

The present invention is for solving the problems of the technologies described above by the following technical solutions:

The invention provides a kind of transmitting DBF method based on digital delay and phase compensation, comprise following concrete steps:

Step 1, design the one DDS, produce the first intermediate-freuqncy signal, the first intermediate-freuqncy signal is single CF signal or narrowband linear FM signal, specifically comprises:

(1a) according to Digital Array Radar systematic parameter, in FPGA, two totalizer is utilized to produce the phase place of the first intermediate-freuqncy signal, and calculation compensation frequency and compensation of phase;

(1b) utilize cordic algorithm to carry out phase amplitude conversion to the phase place of the first intermediate-freuqncy signal, export and obtain digitizing first intermediate-freuqncy signal;

(1c) digitizing first intermediate-freuqncy signal is carried out digital-to-analog conversion through a D/A device, export and obtain simulated first intermediate-freuqncy signal;

Step 2, design the 2nd DDS, produce the second intermediate-freuqncy signal, the second intermediate-freuqncy signal is single CF signal or Wideband LFM Signals, specifically comprises:

(2a) digital baseband signal of the second intermediate-freuqncy signal is divided into two-phase, is 0 phase signals and 1 phase signals;

(2b) in FPGA, two totalizer is utilized to produce the phase place of 0 phase signals and the phase place of 1 phase signals respectively;

(2c) utilize cordic algorithm, phase amplitude conversion is carried out to the phase place of 0 phase signals and the phase place of 1 phase signals, export the digital baseband signal obtaining the second intermediate-freuqncy signal;

(2d) interpolation is carried out to the digital baseband signal of the second intermediate-freuqncy signal, and sent by interpolation result digital delay wave filter to carry out filtering;

(2e) Digital Up Convert module is sent in the output of digital delay wave filter, the real part that peek word up-converter module exports, then sends into the 2nd D/A device and carry out digital-to-analog conversion, exports and obtains simulated second intermediate-freuqncy signal;

Step 3, digital delay wave filter described in design procedure (2d), digital delay wave filter comprises mark filtering wave by prolonging time device and integer filtering wave by prolonging time device, is specially:

(3a) according to each unit of geometry determination radar of the aerial array amount of delay relative to reference unit, and integer amount of delay and mark amount of delay is decomposed into;

(3b) design the mark filtering wave by prolonging time device of Farrow structure, be specially:

The frequency response of mark filtering wave by prolonging time device is:

$H\left(e^{jw}\right)=\underset{l=0}{\overset{L}{\Σ}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\Σ}}{c}_{l,k}{\mathrm{\μ}}^l{e}^{-jwk}$

In formula, e represents natural logarithm; J is imaginary unit, w is digital angular frequency; L is fitting of a polynomial exponent number; n=max (N _l), l=0,1 ..., L, Nl are the exponent numbers of l subfilter; c _l,ka kth coefficient of l subfilter, k=0,1 ..., N _l; μ is mark amount of delay;

Use maxmin criterion, design factor c _{l, k}, make error ε reach minimum,

$\mathrm{\ϵ}=\underset{w\∈\mathrm{\Ω}}{\max }|\underset{l=0}{\overset{L}{\Σ}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\Σ}}{c}_{l,k}{\mathrm{\μ}}^l{e}^{-jwk}-e^{-jw(\mathrm{\μ}+\frac{N}{2})}|$

In formula, Ω represents the score filter set that each frequency band interval is formed in numerical frequency [0, π] scope;

(3c) design integer filtering wave by prolonging time device, namely utilize the register in FPGA to realize the time delay of integer amount of delay;

(3d) by mark filtering wave by prolonging time device and the combination of integer filtering wave by prolonging time device, four are divided to realize digital delay wave filter mutually;

Step 4, generation local oscillation signal, specifically comprise:

(4a) utilize the reference clock signal of outside input, produce the second local oscillation signal by frequency multiplier;

(4b) the frequency agility local oscillation signal of outside input and the second intermediate-freuqncy signal mixing and filtering is utilized to produce the first local oscillation signal, wherein, during narrow band signal work, the second intermediate-freuqncy signal is single CF signal, and during broadband signal work, the second intermediate-freuqncy signal is Wideband LFM Signals;

The frequency conversion amplification of step 5, signal and radiation, specifically comprise:

(5a) by after the first intermediate-freuqncy signal elder generation and the second local oscillation signal mixing and filtering, again with the first local oscillation signal mixing and filtering, produce radiofrequency signal, wherein, during narrow band signal work, the first intermediate-freuqncy signal is arrowband LFM signal, and during broadband signal work, the first intermediate-freuqncy signal is single CF signal;

(5b) radio frequency signal carries out power amplification, and is launched by aerial array;

The signal that step 6, aerial array are launched completes Beam synthesis in space, forms transmission digital beam.

As further prioritization scheme of the present invention, in step 1a, the phase place of the first intermediate-freuqncy signal is:

In formula, n represents the sampling number of the first intermediate-freuqncy signal; π represents the radian that semicircle is corresponding; f ₁it is the centre frequency of the first intermediate-freuqncy signal; γ ₁it is the chirp rate of the first intermediate-freuqncy signal; T ₁it is the pulse width of the first intermediate-freuqncy signal; T _s1it is the sampling interval of the first intermediate-freuqncy signal; Δ f _{α, β}it is compensating frequency; it is compensation of phase; The γ when the first intermediate-freuqncy signal is single CF signal ₁=0, when the first intermediate-freuqncy signal is arrowband LFM signal b ₁it is the bandwidth of arrowband LFM signal.

As further prioritization scheme of the present invention, in step 1a, compensating frequency and compensation of phase are:

During narrow band signal work,

Δf _α,β＝0

During broadband signal work

Δf _α,β＝γ ₂(η _α,β-τ _α,β)

In formula, f ₀for the carrier frequency of Digital Array Radar system; τ _{α, β}for (α, β) position units in aerial array is relative to the theoretical amount of delay of (0,0) reference by location unit, α, β are x direction and y direction unit coordinate respectively; η _{α, β}it is the amount of delay of working control; γ ₂it is the chirp rate of the second intermediate-freuqncy signal.

As further prioritization scheme of the present invention, in step 2a, the digital baseband signal of the second intermediate-freuqncy signal is divided into two-phase, i.e. 0 phase signals and 1 phase signals, is specially:

0 phase signals is:

$s_2\left(2m\right)=cos\{2\mathrm{\π}\[-\frac{{\mathrm{\γ}}_2{T}_2}{2}\left(2{\mathrm{mT}}_{s2}\right)+\frac{1}{2}{\mathrm{\γ}}_2{\left(2{\mathrm{mT}}_{s2}\right)}^2\]\}$

1 phase signals is:

$s_2(2m+1)=cos\{2\mathrm{\π}\[-\frac{{\mathrm{\γ}}_2{T}_2}{2}(2m+1)T_{s2}+\frac{1}{2}{\mathrm{\γ}}_2{(2m+1)}^2{T_{s2}}^2\]\}$

In formula, m represents the sampling number of 0 phase signals and 1 phase signals; T ₂it is the pulse width of the second intermediate-freuqncy signal; T _s2it is the sampling interval of the second intermediate-freuqncy signal; The γ when the second intermediate-freuqncy signal is single CF signal ₂=0, when the second intermediate-freuqncy signal is Wideband LFM Signals b ₂it is the bandwidth of Wideband LFM Signals.

As further prioritization scheme of the present invention, point digital delay wave filter that four phases realize in step 3d, is specially:

In formula, the p phase of representative digit filtering wave by prolonging time device represents; Z=e ^jw, I _ginteger amount of delay, represent that the p phase of l subfilter represents, represent and get the maximum integer be not more than.

The present invention adopts above technical scheme compared with prior art, has following technique effect:

(1) utilize heterogeneous treatment technology, adopt two totalizer and cordic algorithm, devise a DDS and the 2nd DDS, effectively produces many kinds of radar signal waveform;

(2) signal processing resources of FPGA is utilized, carry out the Integral design of signal generation, digital delay wave filter, analog delay line is substituted with digital delay wave filter, by carrying out phase control to a DDS, carry out digital delay to the digital baseband signal that the 2nd DDS exports, the transmitting DBF achieving narrow band signal and broadband signal controls.

Accompanying drawing explanation

Fig. 1 is method flow diagram of the present invention.

Fig. 2 is the present invention the one DDS schematic diagram.

Fig. 3 is the present invention the 2nd DDS schematic diagram.

Fig. 4 is the present invention's heterogeneous CORDIC shift theory figure.

Fig. 5 is array element coordinate geometry illustraton of model of the present invention.

Fig. 6 is digital delay wave filter four phase structure schematic diagram of the present invention.

Fig. 7 is the p phase structure figure of digital delay wave filter of the present invention.

Fig. 8 is the group delay frequency characteristic analogous diagram of mark filtering wave by prolonging time device of the present invention.

Fig. 9 is the group delay frequency characteristic error analogous diagram of mark filtering wave by prolonging time device of the present invention.

Figure 10 is wideband correlation baseband waveform time delay simulation figure of the present invention, and wherein, (a) is wideband correlation baseband waveform time delay simulation figure of the present invention, and (b) is its partial enlarged drawing.

Figure 11 is that broadband signal Wave beam forming of the present invention is in desired orientation frequency gain analogous diagram.

Figure 12 is broadband signal digital Wave beam forming analogous diagram of the present invention.

Embodiment

Below in conjunction with accompanying drawing, technical scheme of the present invention is described in further detail:

A kind of transmitting DBF method based on digital delay and phase compensation of the present invention, implementation step as shown in Figure 1.

Step 1, design the one DDS, produce the first intermediate-freuqncy signal, the first intermediate-freuqncy signal is single CF signal or narrowband linear frequency modulation (LFM) signal, as shown in Figure 2, specifically comprises:

(1a) according to Digital Array Radar systematic parameter, in FPGA, two totalizer is utilized to produce the phase place of the first intermediate-freuqncy signal, and calculation compensation frequency and compensation of phase.

The phase place of the first intermediate-freuqncy signal is:

In formula, n represents the sampling number of the first intermediate-freuqncy signal; π represents the radian that semicircle is corresponding; f ₁it is the centre frequency of the first intermediate-freuqncy signal; γ ₁it is the chirp rate of the first intermediate-freuqncy signal; T ₁it is the pulse width of the first intermediate-freuqncy signal; T _s1it is the sampling interval of the first intermediate-freuqncy signal; Δ f _{α, β}it is compensating frequency; it is compensation of phase; The γ when the first intermediate-freuqncy signal is single CF signal ₁=0, when the first intermediate-freuqncy signal is arrowband LFM signal b ₁it is the bandwidth of arrowband LFM signal.

With reference to Fig. 2, pass through frequency control word chirp rate control word and compensation of phase utilize two totalizer, the phase theta (n) of the first intermediate-freuqncy signal can be obtained.

Suppose that the carrier frequency of Digital Array Radar system is f ₀, theoretical radiofrequency signal s (t, the τ of radar emission _{α, β}) can be written as:

$s(t,{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})=\mathrm{Re}\[g(t-{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})e^{j2{\mathrm{\πf}}_0(t-{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})}\]$

In formula, real part computing is got in Re [] expression; G (t) represents the envelope of radiofrequency signal in t; τ _{α, β}for (α, β) position units in aerial array is relative to the theoretical amount of delay of (0,0) reference by location unit, α, β are x direction and y direction unit coordinate respectively; E represents natural logarithm, and j is imaginary unit,

When radiofrequency signal is narrow band signal, signal can be approximated to be:

$s(t,{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})\≈\mathrm{Re}\[g\left(t\right)e^{j2{\mathrm{\πf}}_0(t-{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})}\]=\mathrm{Re}\[g\left(t\right)e^{j2{\mathrm{\πf}}_{0}t}{e}^{-j2{\mathrm{\πf}}_0{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}}}\]$

Then for narrow radio frequency signal, compensating frequency and compensation of phase are respectively:

Δf _α,β＝0

When radiofrequency signal is broadband signal, according to institute of the present invention extracting method, the second intermediate-freuqncy signal is Wideband LFM Signals, and its digital baseband signal can be written as:

$s_2\left(n\right)=cos\{2\mathrm{\π}\[-\frac{{\mathrm{\γ}}_2{T}_2}{2}{\mathrm{nT}}_{s2}+\frac{1}{2}{\mathrm{\γ}}_2{\left({\mathrm{nT}}_{s2}\right)}^2\]\}$

In formula, γ ₂the chirp rate of the second intermediate-freuqncy signal, T ₂the pulse width of the second intermediate-freuqncy signal, T _s2it is the sampling interval of the second intermediate-freuqncy signal.

Radiofrequency signal s (t, the η of synthesis _{α, β}) can be written as:

In formula, η _{α, β}it is the amount of delay of working control.

Theoretical broadband rf signal s (t, τ _{α, β}) can be expressed as:

$s(t,{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})=cos\{2\mathrm{\π}\[(f_0-\frac{{\mathrm{\γ}}_2{T}_2}{2})(t-{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})+\frac{1}{2}{\mathrm{\γ}}_2{(t-{\mathrm{\τ}}_{\mathrm{\α},\mathrm{\β}})}^2\]\}$

So for broadband rf signal, compensating frequency and compensation of phase are respectively:

Δf _α,β＝γ ₂(η _α,β-τ _α,β)

(1b) utilize cordic algorithm to carry out phase amplitude conversion to the phase place of the first intermediate-freuqncy signal, export and obtain digitizing first intermediate-freuqncy signal.

Vector rotation patten transformation is Iterative Shifts phase computation system by cordic algorithm, and the Representation Equation of each iteration is:

x ⁽ⁱ⁺¹⁾＝x ⁽ⁱ⁾-d _i(2 ^-iy ⁽ⁱ⁾)

y ⁽ⁱ⁺¹⁾＝y ⁽ⁱ⁾+d _i(2 ^-ix ⁽ⁱ⁾)

z ⁽ⁱ⁺¹⁾＝z ⁽ⁱ⁾-d _iθ ⁽ⁱ⁾

In formula, x ⁽ⁱ⁾, y ⁽ⁱ⁾, z ⁽ⁱ⁾represent the data before the i-th+1 time iteration; x ⁽ⁱ⁺¹⁾, y ⁽ⁱ⁺¹⁾, z ⁽ⁱ⁺¹⁾represent the data after the i-th+1 time iteration; θ ⁽ⁱ⁾=actan (2 ^-i), actan () represents arctan function; Symbol d _ia judgement operator, in order to determine sense of rotation, and $d_i=\left\{\begin{array}{c}1,z^{\left(i\right)}\&GreaterEqual;0\\ 0,z^{\left(i\right)}<0\end{array}\right..$

Obtain after M iteration:

x ^(M)＝K _M(x ⁽⁰⁾cosz ⁽⁰⁾-y ⁽⁰⁾sinz ⁽⁰⁾)

y ^(M)＝K _M(y ⁽⁰⁾cosz ⁽⁰⁾+x ⁽⁰⁾sinz ⁽⁰⁾)

By arranging y ⁽⁰⁾=0, z ⁽⁰⁾=θ (n), CORDIC export and obtain digitizing first intermediate-freuqncy signal, wherein, and K _mcontraction-expansion factor, $K_M=\underset{M}{\&Pi;}\left(\sqrt{1+2^{(-2i)}}\right),i=0,1,\mathrm{...},M.$

(1c) digitizing first intermediate-freuqncy signal is carried out digital-to-analog conversion through a D/A device, export and obtain simulated first intermediate-freuqncy signal.

First intermediate-freuqncy signal expression formula of simulation is written as:

Step 2, design the 2nd DDS, produce the second intermediate-freuqncy signal, the second intermediate-freuqncy signal is single CF signal or Wideband LFM Signals, as shown in Figure 3, specifically comprises:

(2a) digital baseband signal of the second intermediate-freuqncy signal is divided into two-phase, is 0 phase signals and 1 phase signals, is specially:

0 phase signals is:

$s_2\left(2m\right)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}\left(2{\mathrm{mT}}_{s2}\right)+\frac{1}{2}{\mathrm{\&gamma;}}_2{\left(2{\mathrm{mT}}_{s2}\right)}^2\&rsqb;\}$

1 phase signals is:

$s_2(2m+1)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}(2m+1)T_{s2}+\frac{1}{2}{\mathrm{\&gamma;}}_2{(2m+1)}^2{T_{s2}}^2\&rsqb;\}$

In formula, m represents the sampling number of 0 phase signals and 1 phase signals; T ₂it is the pulse width of the second intermediate-freuqncy signal; T _s2it is the sampling interval of the second intermediate-freuqncy signal; The γ when the second intermediate-freuqncy signal is single CF signal ₂=0, when the second intermediate-freuqncy signal is Wideband LFM Signals b ₂it is the bandwidth of Wideband LFM Signals.

(2b) in FPGA, two totalizer is utilized to produce the phase place of 0 phase signals and the phase place of 1 phase signals respectively.

With reference to Fig. 3, by frequency control word k _f0=-2 π γ ₂t ₂t _s2with chirp rate control word utilize two totalizer, obtain the phase place of 0 phase signals:

θ(2m)＝-2πγ ₂T ₂mT _s2+πγ ₂(2mT _s2) ²

By frequency control word k _f1=-2 π γ ₂t ₂t _s2+ 4 π γ ₂t _s2 ², chirp rate control word and stationary phase utilize two totalizer, obtain the phase place of 1 phase signals:

θ(2m+1)＝-πγ ₂T ₂(2m+1)T _s2+πγ ₂(2m+1) ²T _s2 ²。

(2c) utilize cordic algorithm, phase amplitude conversion is carried out to the phase place of 0 phase signals and the phase place of 1 phase signals, export the digital baseband signal obtaining the second intermediate-freuqncy signal.

Export the digital baseband signal obtaining the second intermediate-freuqncy signal with reference to Fig. 4, CORDIC, namely the phase theta (2m) of 0 phase signals is by cordic algorithm, produces in-phase component I ₀(2m) with quadrature component Q ₀(2m); The phase theta (2m+1) of 1 phase signals, by cordic algorithm, produces in-phase component I ₁(2m+1) with quadrature component Q ₁(2m+1).

(2d) interpolation is carried out to the digital baseband signal of the second intermediate-freuqncy signal, and sent by interpolation result digital delay wave filter to carry out filtering.

According to the sampling interval T of FPGA clock speed and the second intermediate-freuqncy signal _s2, need to carry out interpolation to the digital baseband signal of the second intermediate-freuqncy signal.In example of the present invention, adopt 2 times of null value interpolation, after interpolation, digital baseband signal in-phase component is:

I(4r)＝I ₀(2m),I(4r+1)＝0,I(4r+2)＝I ₁(2m+1),I(4r+3)＝0

After interpolation, digital baseband signal quadrature component is:

Q(4r)＝Q ₀(2m),Q(4r+1)＝0,Q(4r+2)＝Q ₁(2m+1),Q(4r+3)＝0

In formula, r represents that each road signal data is counted.

The unit impulse response of digital delay wave filter can be written as h (k, τ _{α, β}), for digital baseband signal in-phase component I (n) after interpolation, n=4r+v, v=0,1,2,3, wave filter Output rusults is:

x _I(n)＝I(n)*h(k,τ _α,β),n＝4r+v,v＝0,1,2,3

In formula, * represents convolution algorithm.

For digital baseband signal quadrature component Q (n) after interpolation, wave filter Output rusults is:

x _Q(n)＝Q(n)*h(k,τ _α,β),n＝4r+v,v＝0,1,2,3。

(2e) Digital Up Convert module is sent in the output of digital delay wave filter, the real part that peek word up-converter module exports, then sends into the 2nd D/A device and carry out digital-to-analog conversion, exports and obtains simulated second intermediate-freuqncy signal.

From step (2d), digital delay wave filter Output rusults is x _i(n)+jx _q(n), Digital Up Convert, gets real part, can be described as with mathematic(al) representation:

$s_{IF}\left(n\right)=\mathrm{Re}\&lsqb;\left(x_I\right(n)+{\mathrm{jx}}_Q(n\left)\right)\&times;e^{j2\mathrm{\&pi;}\frac{f_2}{f_{s2}}n}\&rsqb;$

In formula, f ₂the centre frequency of the second intermediate-freuqncy signal, f _s2the sample frequency of the second intermediate-freuqncy signal, in example of the present invention, then above formula can be reduced to further

s _IF(4r+v)＝Re{(x _I(4r+v)+jx _Q(4r+v))×(-j) ^v},v＝0,1,2,3

That is, real part result is got in Digital Up Convert is x _i(4r), x _q(4r+1) ,-x _i(4r+2) ,-x _q(4r+3), send the 2nd D/A device, carry out digital-to-analog conversion, export the second intermediate-freuqncy signal obtaining simulating, its expression formula is:

$s_{IF}\left(t\right)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})+\frac{1}{2}{\mathrm{\&gamma;}}_2{(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})}^2\&rsqb;+2{\mathrm{\&pi;f}}_{2}t\}.$

Step 3, digital delay wave filter described in design procedure (2d), digital delay wave filter comprises mark filtering wave by prolonging time device and integer filtering wave by prolonging time device, is specially:

(3a) according to each unit of geometry determination radar of the aerial array amount of delay relative to reference unit, and integer amount of delay and mark amount of delay is decomposed into.

(3a.1) with reference to Fig. 5, with (0,0) position units as reference unit, the steering vector of array element signal is:

$v\left(k\right)={\&lsqb;e^{-{\mathrm{jk}}^T{p}_{0,0}},e^{-{\mathrm{jk}}^T{p}_{0,1}},\mathrm{...},e^{-{\mathrm{jk}}^T{p}_{\mathrm{\&alpha;},\mathrm{\&beta;}}},\mathrm{...},e^{-{\mathrm{jk}}^T{p}_{A-1,B-1}}\&rsqb;}^T$

In formula, p _{α, β}=[α d _x, β d _y, 0] ^trepresent (α, β) cell position vector, d _x, d _ybe x direction and y direction unit interval respectively, Α, Β are x direction and y direction array element number respectively, u represents wave-number vector, and λ is the wavelength that transmits, u=[sin θ cos φ, sin θ sin φ, cos θ] ^tfor direction cosine vector, θ represents position angle, and φ represents the angle of pitch, [] ^trepresenting matrix transpose operation.

(3a.2) between the adjacent cells in array x direction and y direction, time delay is:

${\mathrm{\&Delta;\&tau;}}_x=\frac{u_x{d}_x}{c}$

${\mathrm{\&Delta;\&tau;}}_y=\frac{u_y{d}_y}{c}$

In formula, u _x=sin θ cos φ, u _y=sin θ sin φ, c is electromagnetic wave propagation speed.

(3a.3) in array (α, β) unit relative coordinate initial point unit between delay time τ _{α, β}for:

${\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}}=\frac{1}{c}({\mathrm{\&alpha;u}}_x{d}_x+{\mathrm{\&beta;u}}_y{d}_y)$

(3a.4) by delay time τ _{α, β}with the sampling interval T of the second intermediate-freuqncy signal _s2, obtain integer amount of delay with mark amount of delay in formula, represent the maximum integer being not more than x.

(3b) design the mark filtering wave by prolonging time device of Farrow structure, be specially:

The frequency response of idealized score filtering wave by prolonging time device is:

H _des(e ^jw)＝e ^-jw(μ+D)

In formula, w is digital angular frequency, n is the exponent number of wave filter.

By e ^{-jw μ}taylor series expansion is:

$e^{-jw\mathrm{\&mu;}}=R_L(\mathrm{\&mu;},w)+\underset{l=0}{\overset{L}{\&Sigma;}}\frac{{(-j\mathrm{\&mu;}w)}^l}{l!}$

In formula, R _l(μ, w) represents remainder, and L is fitting of a polynomial exponent number.

Design mark filtering wave by prolonging time device, makes its frequency response H (e ^jw) be approximately H _des(e ^jw):

$H\left(e^{jw}\right)=\underset{l=0}{\overset{L}{\&Sigma;}}{\mathrm{\&mu;}}^l{e}^{-{\mathrm{jwD}}_l}{H}_l\left(e^{jw}\right)$

In formula, H _l(e ^jw) be the frequency response of l subfilter, n _lthe exponent number of l subfilter,

If c _l,kbe a kth coefficient of l subfilter, then the frequency response of the mark filtering wave by prolonging time device of Farrow structure is:

$H\left(e^{jw}\right)=\underset{l=0}{\overset{L}{\&Sigma;}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\&Sigma;}}{c}_{l,k}{\mathrm{\&mu;}}^l{e}^{-jwk}.$

Use maxmin criterion, design factor c _l,k, make error ε reach minimum,

$\mathrm{\&epsiv;}=\underset{w\&Element;\mathrm{\&Omega;}}{\max }|\underset{l=0}{\overset{L}{\&Sigma;}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\&Sigma;}}{c}_{l,k}{\mathrm{\&mu;}}^l{e}^{-jwk}-e^{-jw(\mathrm{\&mu;}+\frac{N}{2})}|$

In formula, Ω represents the score filter set that each frequency band interval is formed in numerical frequency [0, π] scope, and max represents maximizing.

Definition error function H _e(e ^jw):

H _e(e ^jw)＝H(e ^jw)-H _des(e ^jw),w∈[0,w _c]

In formula, w _crepresent the cutoff frequency of wave filter.

The mark filtering wave by prolonging time device of design Farrow structure, that is, Selection parameter L, N _land c _l,k, make error function H _e(e ^jw) meet following requirements:

|H _e(e ^jw)|≤δ

In formula, δ represents wave filter tolerance.

Specific design step is: make q=1, ε _q=δ,

(1) basis

$\frac{{\left(0.5w_c\right)}^{L+1}}{(L+1)!}\&le;\frac{{\mathrm{\&epsiv;}}_q}{1+\sqrt{2}}$

Obtain the minimum L value met the demands, then basis

$\left|{{\mathrm{\&delta;}}_l}^{\left(q\right)}\left(w\right)\right|\&le;\frac{2^l{\mathrm{\&epsiv;}}_q}{C(1+\sqrt{2})}$

Calculate the tolerance of each subfilter in formula, represent the smallest positive integral being not less than x.According to design each optimum subfilter respectively, make meet

$\left|H_e^{\left(q\right)}\left(e^{jw}\right)\right|\&le;\mathrm{\&delta;}$

Finally calculate

${\mathrm{\&delta;}}_q=max\left|H_e^{\left(q\right)}\left(e^{jw}\right)\right|$

(2) using the result of step (1) as starting condition, devise optimum wave filter meet:

$min{\mathrm{\&delta;}}_{q,opt}subjectto\left|H_e^{(q,opt)}\left(e^{jw}\right)\right|\&le;{\mathrm{\&delta;}}_{q,opt}$

(3) if δ _{q, opt}≤ δ, then

q＝q+1,ε _q＝ε _q-1+Δ,Δ＞0

And turn back to (1).

(4) optimum filter freguency response is:

$H\left(e^{jw}\right)=H_e^{(q-1,opt)}\left(e^{jw}\right).$

(3c) design integer filtering wave by prolonging time device, namely utilize the register in FPGA to realize the time delay of integer amount of delay.

The frequency response of integer filtering wave by prolonging time device is:

$H_g\left(e^{jw}\right)=e^{-{\mathrm{jwI}}_g}.$

(3d) by mark filtering wave by prolonging time device and the combination of integer filtering wave by prolonging time device, four are divided to realize digital delay wave filter mutually.

With reference to Fig. 6, digital delay wave filter, comprise mark filtering wave by prolonging time device and integer filtering wave by prolonging time device, the frequency domain response of digital delay wave filter is:

H _τ(z)＝H _g(z)H(z)

Wherein, z=e ^jw.

Four of digital delay wave filter is expressed as mutually:

$H_{\mathrm{\&tau;}}\left(z\right)=z^{-I_g}\underset{l=0}{\overset{L}{\&Sigma;}}{\mathrm{\&mu;}}^l{z}^{-D_l}{H}_l\left(z\right)=\underset{l=0}{\overset{L}{\&Sigma;}}{\mathrm{\&mu;}}^l{z}^{-(D_l+I_g)}\underset{p=0}{\overset{3}{\&Sigma;}}{z}^{-p}{H}_{l,4}^{\left(p\right)}\left(z^4\right)=\underset{p=0}{\overset{3}{\&Sigma;}}{z}^{-p}{H}_4^{\left(p\right)}\left(z^4\right)$

In formula, represent that the p phase of l subfilter represents, the p phase of representative digit filtering wave by prolonging time device represents,

Figure 7 shows that the p phase structure figure of digital delay wave filter.

By the x of step (2d) _i(n)=I (n) * h (k, τ _{α, β}), n=4r+v, v=0,1,2,3 is known,

$\begin{array}{c}{X}_I\left(z\right)=I\left(z\right)H_{\mathrm{\&tau;}}\left(z\right)=\left(I_0^{\left(4\right)}\right(z^4)+z^{-2}{I}_2^{\left(4\right)}(z^4\left)\right)\underset{p=0}{\overset{3}{\&Sigma;}}{z}^{-p}{H}_4^{\left(p\right)}\left(z^4\right)\\ =\left(I_0^{\left(4\right)}\right(z^4\left)H_4^{\left(0\right)}\right(z^4)+z^{-4}{I}_2^{\left(4\right)}(z^4\left)H_4^{\left(2\right)}\right(z^4\left)\right)+\\ z^{-1}\left(I_0^{\left(4\right)}\right(z^4\left)H_4^{\left(1\right)}\right(z^4)+z^{-4}{I}_2^{\left(4\right)}(z^4\left)H_4^{\left(3\right)}\right(z^4\left)\right)+\\ z^{-2}\left(I_0^{\left(4\right)}\right(z^4\left)H_4^{\left(2\right)}\right(z^4)+I_2^{\left(4\right)}(z^4\left)H_4^{\left(0\right)}\right(z^4\left)\right)+\\ z^{-3}\left(I_0^{\left(4\right)}\right(z^4\left)H_4^{\left(3\right)}\right(z^4)+I_2^{\left(4\right)}(z^4\left)H_4^{\left(1\right)}\right(z^4\left)\right)\end{array}$

In formula, X _iz () is x _in the frequency domain representation of (), I (z) is the frequency domain representation of data I (n).In like manner can obtain

$X_Q\left(z\right)=Q\left(z\right)H_{\mathrm{\&tau;}}\left(z\right)=\left(Q_0^{\left(4\right)}\right(z^4)+z^{-2}{Q}_2^{\left(4\right)}(z^4\left)\right)\underset{p=0}{\overset{3}{\&Sigma;}}{z}^{-p}{H}_4^{\left(p\right)}\left(z^4\right)$

In formula, X _qz () is x _qn the frequency domain representation of (), Q (z) is the frequency domain representation of data Q (n).

Step 4, generation local oscillation signal, specifically comprise:

(4a) utilize the reference clock signal of outside input, produce the second local oscillation signal by frequency multiplier;

(4b) the frequency agility local oscillation signal of outside input and the second intermediate-freuqncy signal mixing and filtering is utilized to produce the first local oscillation signal, wherein, during narrow band signal work, the second intermediate-freuqncy signal is single CF signal, and during broadband signal work, the second intermediate-freuqncy signal is Wideband LFM Signals.

First local oscillator number of signals expression formula can be written as:

$s_{lo1}\left(t\right)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})+\frac{1}{2}{\mathrm{\&gamma;}}_2{(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})}^2\&rsqb;+2\mathrm{\&pi;}(f_2+f_{lo1})t\}$

In formula, f _lo1it is the frequency of frequency agility local oscillation signal.

The frequency conversion amplification of step 5, signal and radiation, specifically comprise:

(5a) the first intermediate-freuqncy signal first with the second local oscillation signal mixing and filtering after, again with the first local oscillation signal mixing and filtering, produce radiofrequency signal, wherein, during narrow band signal work, the first intermediate-freuqncy signal is arrowband LFM signal, and during broadband signal work, the first intermediate-freuqncy signal is single CF signal.

First intermediate-freuqncy signal and the second local oscillation signal mixing and filtering output signal and are:

In formula, f _lo2it is the second local oscillation signal frequency.

During narrow band signal work, radiofrequency signal is:

$s(t,{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})=cos\{2\mathrm{\&pi;}\&lsqb;(f_0-\frac{B_1}{2})t+\frac{1}{2}{\mathrm{\&gamma;}}_1{t}^2\&rsqb;-2{\mathrm{\&pi;f}}_0{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}}\}$

Wherein, f ₀=f ₁+ f ₂+ f _lo1+ f _lo2.

During broadband signal work, radiofrequency signal is:

$s(t,{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})=cos\{2\mathrm{\&pi;}\&lsqb;(f_0-\frac{B_2}{2})(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})+\frac{1}{2}{\mathrm{\&gamma;}}_2{(t-{\mathrm{\&tau;}}_{\mathrm{\&alpha;},\mathrm{\&beta;}})}^2\&rsqb;\}.$

(5b) radio frequency signal carries out power amplification, and is launched by aerial array.

The signal that step 6, aerial array are launched completes Beam synthesis in space, forms transmission digital beam.

$Y=f\left(w\right)\underset{\mathrm{\&alpha;}=0}{\overset{A-1}{\&Sigma;}}\underset{\mathrm{\&beta;}=0}{\overset{B-1}{\&Sigma;}}{A}_{\mathrm{\&alpha;},\mathrm{\&beta;}}{e}^{j(k_s^T-k^T)\&CenterDot;p_{\mathrm{\&alpha;},\mathrm{\&beta;}}}=f\left(w\right)\underset{\mathrm{\&alpha;}=0}{\overset{A-1}{\&Sigma;}}\underset{\mathrm{\&beta;}=0}{\overset{B-1}{\&Sigma;}}{A}_{\mathrm{\&alpha;},\mathrm{\&beta;}}{e}^{-j\frac{2\mathrm{\&pi;}}{\mathrm{\&lambda;}}\&lsqb;(u_{xs}-u_x){\mathrm{\&alpha;d}}_x+(u_{ys}-u_y){\mathrm{\&beta;d}}_y\&rsqb;}$

Wherein, Y represents frequency-wavenumber response function, the frequency spectrum that f (w) is signal, k _srepresent wanted signal wave-number vector, u _xs=sin θ _scos φ _s, u _ys=sin θ _ssin φ _s, θ _srepresent wanted signal position angle, φ _srepresent the wanted signal angle of pitch, A _{α, β}for cell signal amplitude weighting, as A _{α, β}=a _αa _β, then:

$Y=f\left(w\right)\underset{\mathrm{\&alpha;}=0}{\overset{A-1}{\&Sigma;}}{a}_{\mathrm{\&alpha;}}{e}^{-j\frac{2\mathrm{\&pi;}}{\mathrm{\&lambda;}}(u_{xs}-u_x){\mathrm{\&alpha;d}}_x}\underset{\mathrm{\&beta;}=0}{\overset{B-1}{\&Sigma;}}{a}_{\mathrm{\&beta;}}{e}^{-j\frac{2\mathrm{\&pi;}}{\mathrm{\&lambda;}}(u_{ys}-u_y){\mathrm{\&beta;d}}_y}.$

Algorithm of the present invention and disposal route by checking, achieve satisfied effect:

1. experiment condition: C-band wideband digital array radar, array element is the linear array of Unit 64, and array element distance is the half of maximum wavelength, and wanted signal beam direction is θ _s=60 °, signal form is linear FM signal, and bandwidth is 400MHz, time wide be 20us, sampling rate is f _s2=1200MHz.

2. emulate content:

Emulation 1: based on following simulation parameter: fitting of a polynomial exponent number is L=11, and the exponent number of each subfilter is equal, N _l=11, l=0,1 ..., the mark filtering wave by prolonging time device of L, design Farrow structure.Fig. 8 gives the group delay frequency characteristic of mark filtering wave by prolonging time device, and mark amount of delay is μ=0.1, and Fig. 9 gives the group delay frequency characteristic error of mark filtering wave by prolonging time device.

Emulation 2: based on the wave filter of above-mentioned design, adopt even weighting to carry out broadband emission digital beam froming, in Figure 10, (a) gives time-domain diagram before and after the time delay of Wideband LFM Signals baseband waveform, and in Figure 10, (b) is its partial enlarged drawing.Figure 11 gives only phase-moving method and time-delay method frequency domain response figure in the desired direction.Figure 12 gives the beam pattern adopting institute of the present invention extracting method to be formed, and give also the beam pattern that desirable time-delay method obtains and the beam pattern obtained by means of only phase-moving method simultaneously.

3. analysis of simulation result:

As can be seen from Figure 8, the group delay frequency characteristic of the mark filtering wave by prolonging time device designed by the present invention is all very smooth in [0.0.4 π] scope, and can learn from Fig. 9 group delay frequency characteristic error, the precision of mark filtering wave by prolonging time device can reach

As can be seen from Figure 10, the difference before and after signal lag, due to MATLAB limited resolution, can only find out signal about time delay 4ns, and the amount of delay of mark filtering wave by prolonging time device is demonstrate the validity of mark filtering wave by prolonging time device.

Figure 11 frequency domain response figure describes only phase-moving method and the phase place between broadband signal side frequency can be caused inconsistent.

Wideband digital launching beam formation figure shown in Figure 12 shows, the only broadband beams figure main lobe broadening that formed of phase-moving method, and secondary lobe is also raised to some extent, and broadband signal base band time delay+phase compensating method that the present invention proposes effectively can form broadband digital beam figure, basic overlapping with desirable beam pattern, illustrate that method proposed by the invention is correct.

The above; be only the embodiment in the present invention; but protection scope of the present invention is not limited thereto; any people being familiar with this technology is in the technical scope disclosed by the present invention; the conversion or replacement expected can be understood; all should be encompassed in and of the present inventionly comprise within scope, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (5)

1., based on the transmitting DBF method of digital delay and phase compensation, it is characterized in that, comprise following concrete steps:

Step 1, design the one DDS, produce the first intermediate-freuqncy signal, the first intermediate-freuqncy signal is single CF signal or narrowband linear FM signal, specifically comprises:

(1a) according to Digital Array Radar systematic parameter, in FPGA, two totalizer is utilized to produce the phase place of the first intermediate-freuqncy signal, and calculation compensation frequency and compensation of phase;

(1b) utilize cordic algorithm to carry out phase amplitude conversion to the phase place of the first intermediate-freuqncy signal, export and obtain digitizing first intermediate-freuqncy signal;

(1c) digitizing first intermediate-freuqncy signal is carried out digital-to-analog conversion through a D/A device, export and obtain simulated first intermediate-freuqncy signal;

Step 2, design the 2nd DDS, produce the second intermediate-freuqncy signal, the second intermediate-freuqncy signal is single CF signal or Wideband LFM Signals, specifically comprises:

(2a) digital baseband signal of the second intermediate-freuqncy signal is divided into two-phase, is 0 phase signals and 1 phase signals;

(2b) in FPGA, two totalizer is utilized to produce the phase place of 0 phase signals and the phase place of 1 phase signals respectively;

(2c) utilize cordic algorithm, phase amplitude conversion is carried out to the phase place of 0 phase signals and the phase place of 1 phase signals, export the digital baseband signal obtaining the second intermediate-freuqncy signal;

(2d) interpolation is carried out to the digital baseband signal of the second intermediate-freuqncy signal, and sent by interpolation result digital delay wave filter to carry out filtering;

(2e) Digital Up Convert module is sent in the output of digital delay wave filter, the real part that peek word up-converter module exports, then sends into the 2nd D/A device and carry out digital-to-analog conversion, exports and obtains simulated second intermediate-freuqncy signal;

Step 3, digital delay wave filter described in design procedure (2d), digital delay wave filter comprises mark filtering wave by prolonging time device and integer filtering wave by prolonging time device, is specially:

(3a) according to each unit of geometry determination radar of the aerial array amount of delay relative to reference unit, and integer amount of delay and mark amount of delay is decomposed into;

(3b) design the mark filtering wave by prolonging time device of Farrow structure, be specially:

The frequency response of mark filtering wave by prolonging time device is:

$H\left(e^{jw}\right)=\underset{l=0}{\overset{L}{\&Sigma;}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\&Sigma;}}{c}_{l,k}{\mathrm{\&mu;}}^l{e}^{-jwk}$

In formula, e represents natural logarithm; J is imaginary unit, w is digital angular frequency; L is fitting of a polynomial exponent number; n=max (N _l), l=0,1 ..., L, N _lit is the exponent number of l subfilter; c _l,ka kth coefficient of l subfilter, k=0,1 ..., N _l; μ is mark amount of delay;

Use maxmin criterion, design factor c _l,k, make error ε reach minimum,

$\mathrm{\&epsiv;}=\underset{w\&Element;\mathrm{\&Omega;}}{max}|\underset{l=0}{\overset{L}{\&Sigma;}}{e}^{-{\mathrm{jwD}}_l}\underset{k=0}{\overset{N_l}{\&Sigma;}}{c}_{l,k}{\mathrm{\&mu;}}^l{e}^{-jwk}-e^{-jw\left(\mathrm{\&mu;}+\frac{N}{2}\right)}|$

In formula, Ω represents the score filter set that each frequency band interval is formed in numerical frequency [0, π] scope;

(3c) design integer filtering wave by prolonging time device, namely utilize the register in FPGA to realize the time delay of integer amount of delay;

(3d) by mark filtering wave by prolonging time device and the combination of integer filtering wave by prolonging time device, four are divided to realize digital delay wave filter mutually;

Step 4, generation local oscillation signal, specifically comprise:

(4a) utilize the reference clock signal of outside input, produce the second local oscillation signal by frequency multiplier;

(4b) the frequency agility local oscillation signal of outside input and the second intermediate-freuqncy signal mixing and filtering is utilized to produce the first local oscillation signal, wherein, during narrow band signal work, the second intermediate-freuqncy signal is single CF signal, and during broadband signal work, the second intermediate-freuqncy signal is Wideband LFM Signals;

The frequency conversion amplification of step 5, signal and radiation, specifically comprise:

(5a) by after the first intermediate-freuqncy signal elder generation and the second local oscillation signal mixing and filtering, again with the first local oscillation signal mixing and filtering, produce radiofrequency signal, wherein, during narrow band signal work, the first intermediate-freuqncy signal is arrowband LFM signal, and during broadband signal work, the first intermediate-freuqncy signal is single CF signal;

(5b) radio frequency signal carries out power amplification, and is launched by aerial array;

The signal that step 6, aerial array are launched completes Beam synthesis in space, forms transmission digital beam.

2. the transmitting DBF method based on digital delay and phase compensation according to claim 1, it is characterized in that, in step 1a, the phase place of the first intermediate-freuqncy signal is:

In formula, n represents the sampling number of the first intermediate-freuqncy signal; π represents the radian that semicircle is corresponding; f ₁it is the centre frequency of the first intermediate-freuqncy signal; γ ₁it is the chirp rate of the first intermediate-freuqncy signal; T ₁it is the pulse width of the first intermediate-freuqncy signal; T _s1it is the sampling interval of the first intermediate-freuqncy signal; Δ f _{α, β}it is compensating frequency; it is compensation of phase; The γ when the first intermediate-freuqncy signal is single CF signal ₁=0, when the first intermediate-freuqncy signal is arrowband LFM signal b ₁it is the bandwidth of arrowband LFM signal.

3. the transmitting DBF method based on digital delay and phase compensation according to claim 1, it is characterized in that, in step 1a, compensating frequency and compensation of phase are:

During narrow band signal work,

Δf _α,β＝0

During broadband signal work

Δf _α,β＝γ ₂(η _α,β-τ _α,β)

In formula, f ₀for the carrier frequency of Digital Array Radar system; τ _{α, β}for (α, β) position units in aerial array is relative to the theoretical amount of delay of (0,0) reference by location unit, α, β are x direction and y direction unit coordinate respectively; η _{α, β}it is the amount of delay of working control; γ ₂it is the chirp rate of the second intermediate-freuqncy signal.

4. the transmitting DBF method based on digital delay and phase compensation according to claim 1, it is characterized in that, in step 2a, the digital baseband signal of the second intermediate-freuqncy signal is divided into two-phase, i.e. 0 phase signals and 1 phase signals, is specially:

0 phase signals is:

$s_2\left(2m\right)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}\left(2{\mathrm{mT}}_{s2}\right)+\frac{1}{2}{\mathrm{\&gamma;}}_2{\left(2{\mathrm{mT}}_{s2}\right)}^2\&rsqb;\}$

1 phase signals is:

$s_2(2m+1)=cos\{2\mathrm{\&pi;}\&lsqb;-\frac{{\mathrm{\&gamma;}}_2{T}_2}{2}(2m+1)T_{s2}+\frac{1}{2}{\mathrm{\&gamma;}}_2{(2m+1)}^2{T_{s2}}^2\&rsqb;\}$

In formula, m represents the sampling number of 0 phase signals and 1 phase signals; T ₂it is the pulse width of the second intermediate-freuqncy signal; T _s2it is the sampling interval of the second intermediate-freuqncy signal; The γ when the second intermediate-freuqncy signal is single CF signal ₂=0, when the second intermediate-freuqncy signal is Wideband LFM Signals b ₂it is the bandwidth of Wideband LFM Signals.

5. the transmitting DBF method based on digital delay and phase compensation according to claim 1, is characterized in that, point digital delay wave filter that four phases realize in step 3d, is specially:

In formula, the p phase of representative digit filtering wave by prolonging time device represents; Z=e ^jw, I _ginteger amount of delay, represent that the p phase of l subfilter represents, represent and get the maximum integer be not more than.