CN104063539A - Band elimination filter diagnostic method - Google Patents

Abstract

The invention discloses a band elimination filter diagnostic method which comprises the following steps: firstly, the response result of the band elimination filter is educed, modeling is performed on the band elimination filter, numerator polynomials of a transmission function and a reflection function of the band elimination filter are worked out, and the mutual denominator polynomial of the transmission function and the reflection function is educed; then, according to the physical situation, the polynomials of the transmission function and the reflection function are corrected, a short-circuit admittance matrix is educed from a network feature functions, partial-fraction expansion is carried out, the short-circuit admittance matrix is compared with an equivalent circuit admittance matrix to extract an initial coupled matrix, and corresponding matrix change is carried out according to the topological structure of a to-be-diagnosed filter to obtain a diagnostic coupled matrix; finally, the diagnostic coupled matrix is compared with a design target coupled matrix to diagnose a coupling coefficient causing the band elimination filter to be non-ideal. With the adoption of the method, the diagnosis of the coupled matrix of the band elimination filter can be realized to effectively adjust the band elimination filter.

Description

A kind of diagnostic method of rejection filter

Technical field

The present invention relates to the diagnostic method of the rejection filter of communication technical field, specifically rejection filter is set up to model, substitution rejection filter response results realizes the diagnosis of rejection filter, the diagnosable coupled matrix that goes out rejection filter, make comparisons with the matrix of comprehensive Design, realize the optimization of rejection filter.

Background technology

Development along with modern communication technology, radio spectrum resources is crowded all the more, this has had higher requirement to the wave filter in wireless communication system, except making desired signal low-loss passes through, more want parasitic passband in effective filtering device stray wave and bandpass filter etc., this just needs rejection filter to realize.Yet traditional utilize 1/4 wavelength shorting stub to design rejection filter volume more very much not to meet system space requirement, although bandreject filtering comprehensive designing method is more direct, flexibly, but the optimizing process time is longer, regulate more complicated, therefore need to diagnose rejection filter, for optimize provider to, shorten optimization cycle.

Modern wave filter diagnosis mostly is the diagnosis of bandpass filter, apply wider and convenient effectively Cauchy method and realize the diagnosis of bandpass filter, document (A.Garc í a-Lamp é rez for example, S.Llorente-Romano, M.Salazar-Palma, and T.K.Sarkar, " Efficient electromagnetic optimization of microwave filters and multiplexers using rational models, " IEEE Trans.Microw.Theory Tech., vol.52, pp.508 – 521, Feb.2004) be exactly to utilize Cauchy's method to extract the multinomial coefficient of curve.Follow document (G.Macchiarella and D.Traina, " A formulation of the Cauchy method suitable for the synthesis of lossless circuit models of microwave filters from lossy measurements; " IEEE Microw.Wireless Compon.Lett., vol.16, no.5, pp.243 – 245 is May2006.) to extract transmission and the reflection multinomial coefficient in band pass filter circuit by improved Cauchy's method.But up to the present not yet see the diagnosis of rejection filter.Therefore, in order to realize the comprehensive Design of rejection filter, debug conveniently, accurately, need to realize the diagnosis of rejection filter.

Summary of the invention

Technical matters to be solved by this invention is for optimizing the realistic problems such as non-directional, process is complicated, the cycle is long in rejection filter comprehensive Design process, having proposed a kind of diagnostic method of rejection filter.

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

A diagnostic method for rejection filter, comprises the steps:

Step S1), derive the disconnected rejection filter measured data of follow-up, comprise amplitude and the phase place of transmission data, reflectance data;

Step S2), set up the expression formula model of rejection filter reflective function and transition function, utilize step S1) in the measured data of the derivation transition function of calculating rejection filter divide submultinomial and reflective function to divide submultinomial, and derive transition function and the total denominator polynomial expression of reflective function;

Step S3), to the transition function of rejection filter, divide submultinomial and reflective function denominator polynomial expression to revise, the transition function that obtains revised rejection filter divides submultinomial, reflective function to divide submultinomial and transition function and reflective function total denominator polynomial expression;

Step S4), from network characterization function, derive short-circuit admittance matrix, by its residue and with equivalent electrical circuit admittance matrix, contrast, extract initial coupled matrix;

Step S5), according to the topological structure of the disconnected rejection filter of follow-up, initial coupled matrix being done to corresponding matrix changes, obtain the diagnosis coupled matrix of this rejection filter, then compare with the coupled matrix of design object, determine that whether this rejection filter coupling element is consistent with target coupled matrix element, diagnose out the coupling coefficient that causes rejection filter undesirable.

As the further prioritization scheme of diagnostic method of a kind of rejection filter of the present invention, described step S2) in the expression formula model of rejection filter reflective function and transition function be respectively:

$S_{11}=\frac{F\left(s\right)}{E\left(s\right)}=\frac{\underset{k=0}{\overset{\mathrm{np}}{\mathrm{\Σ}}}{a^{\left(2\right)}}_k{s}^k}{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k}---\left(1\right)$

$S_{21}=\frac{P\left(s\right)}{E\left(s\right)}=\frac{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{a^{\left(1\right)}}_k{s}^k}{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k}---\left(2\right)$

Wherein, S ₁₁represent reflective function, S ₂₁represent transition function, F (s) represents that reflective function divides submultinomial, and P (s) represents that transition function divides submultinomial, the denominator polynomial expression that E (s) expression transition function and reflective function are total;

N represents the exponent number of rejection filter, and np represents the outer limit of stopband, and the exponent number of polynomial expression E (s) and P (s) should be identical with the exponent number of rejection filter, and the exponent number of F (s) equals the number of the outer limit of stopband, a ⁽¹⁾the coefficient that represents P (s), a ⁽²⁾the coefficient that represents F (s), b represents the coefficient of E (s), s is Complex frequency variable, k be multinomial in every inferior number formulary.

As the further prioritization scheme of diagnostic method of a kind of rejection filter of the present invention, described step S2) in calculate rejection filter transition function divide submultinomial and reflective function to divide submultinomial and derive transition function and the polynomial detailed step of denominator that reflective function is total as follows:

Step S2.1), formula (1) and formula (2) are divided by, identical according to two expression formula denominator polynomial expressions, obtain:

$S_{11}\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{a_k}^{\left(1\right)}{s}^k-S_{21}\underset{k=0}{\overset{\mathrm{np}}{\mathrm{\Σ}}}{a_k}^{\left(2\right)}{s}^k=0---\left(3\right);$

Step S2.2), formula (3) is converted into matrix form and obtains overdetermined equation:

$[S_{11}{V}_n-S_{21}{V}_{\mathrm{np}}]\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]=0---\left(4\right)$

In formula,

a ⁽¹⁾＝[a ₀ ⁽¹⁾,a ₁ ⁽¹⁾,…,a _n-1 ⁽¹⁾,a _n ⁽¹⁾] ^T

a ⁽²⁾＝[a ₀ ⁽²⁾,a ₁ ⁽²⁾,…,a _np-1 ⁽²⁾,a _np ⁽²⁾] ^T，

S ₂₁＝diag{S ₂₁(s _i)} _{i＝1,2,…,N}

S ₁₁＝diag{S ₁₁(s _i)} _i＝1,2…N，

Diag{S ₂₁(s _i) _i=1,2,, _nand diag{S ₁₁(s _i) _{i=1,2 ... N}from described step S1) the matrix of N frequency formation extracting in the disconnected rejection filter measured data of follow-up, V is Fan Demeng matrix, Fan Demeng matrix and a _k ⁽¹⁾, a _k ⁽²⁾corresponding, Fan Demeng matrix element is herein v _ik=(j ω _i) ^k-1, k=1 ... m+1; M is the frequency number in stopband in N frequency, and gained multinomial coefficient is pressed ascending power and arranged;

Step S2.3), adopt step S2.2 described in unusual decomposition method solution) in overdetermined equation:

$[S_{11}{V}_n-S_{21}{V}_{\mathrm{np}}]\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]={\mathrm{U\ΣV}}^H\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]=0$

Wherein, U and V are chief of a tribe's matrixes, and the diagonal entry of Σ matrix is singular value, obtain matrix of coefficients and be $\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]={\left[V^H\right]}_{n+\mathrm{np}+2};$

Step S2.4), calculate uncorrected transition function and divide submultinomial and the polynomial coefficient of reflective function molecule, use respectively P ' (s) and F ' (s) represents:

$F^{\′}\left(s\right)=\underset{k=0}{\overset{N}{\mathrm{\Σ}}}{a_k}^{\left(1\right)}{s}^k$

$P^{\′}\left(s\right)=\underset{k=0}{\overset{\mathrm{nz}}{\mathrm{\Σ}}}{a_k}^{\left(2\right)}{s}^k;$

Step S2.5), according to following formula, derive transition function and the total denominator polynomial expression E (s) of reflective function:

E(s)E ^*(-s)＝F′(s)F′ ^*(-s)+P′(s)P′ ^*(-s)

Wherein, s _ifor minus of above-mentioned real part, " * " represents to ask conjugation, and then tries to achieve multinomial coefficient b _k, with its representative polynomial E (s), be:

$E\left(s\right)=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k$

As the further prioritization scheme of diagnostic method of a kind of rejection filter of the present invention, step S3) described in the transition function of revised rejection filter divide submultinomial, reflective function to divide submultinomial to adopt respectively following formula:

P(s)＝P′(s)·j

F(s)＝F′(s)/j

P (s) wherein, F (s) represents that respectively revised transition function and reflective function divide submultinomial, and P ' (s) F ' (s) represents that respectively uncorrected transition function divides submultinomial, and j is that imaginary part is 1 pure imaginary number.

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

1. can directly diagnose out the coupling coefficient matrix of rejection filter;

2. the method for this diagnosis rejection filter is simpler, convenient, accurately;

3. can be applied in the optimizing process of rejection filter comprehensive Design, shorten the design cycle.

Accompanying drawing explanation

Fig. 1 is FB(flow block) of the present invention;

Fig. 2 is quadravalence rejection filter topology diagram in the invention process example;

Fig. 3 is the response diagram of quadravalence rejection filter in the invention process example;

Fig. 4 is the response diagram that in the invention process example, quadravalence rejection filter is diagnosed out.

Embodiment

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

Before the solution of the present invention is described in detail, first only consider lossless situation, the centre frequency of n rank rejection filter is f0, and bandwidth is BW, has np the outer limit of band.First according to the target coupled matrix of rejection filter, set up model, obtain the response results of rejection filter.

As shown in Figure 1, diagnosis rejection filter comprises the steps:

Step S1: derive the disconnected rejection filter measured data of follow-up, comprise S ₁₁and S ₂₁amplitude and phase place.

Step S2: the expression formula model of setting up rejection filter reflective function and transition function, derive and utilize the data of step S1 to calculate to such an extent that the transition function of rejection filter divides submultinomial P'(s) and reflective function divide submultinomial F'(s), and derive transition function and the total denominator polynomial expression E (s) of reflective function;

The expression formula model of model rejection filter reflective function and transition function, obtains S ₂₁and S ₁₁expression formula be

$S_{21}=\frac{P\left(s\right)}{E\left(s\right)}=\frac{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{a^{\left(1\right)}}_k{s}^k}{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k}$

$S_{11}=\frac{F\left(s\right)}{E\left(s\right)}=\frac{\underset{k=0}{\overset{\mathrm{np}}{\mathrm{\Σ}}}{a^{\left(2\right)}}_k{s}^k}{\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k}$

S in formula ₁₁represent reflective function, S ₂₁represent transition function, P (s) represents that transition function divides submultinomial, and F (s) represents that reflective function divides submultinomial, the denominator polynomial expression that E (s) expression transition function and reflective function are total.N represents the exponent number of rejection filter, and np represents the outer limit of stopband, and the exponent number of polynomial expression E (s) and P (s) should be identical with the exponent number of rejection filter, and the exponent number of F (s) equals the number of the outer limit of stopband, a ⁽¹⁾the coefficient that represents P (s), a ⁽²⁾the coefficient that represents F (s), b represents the coefficient of E (s), s=σ+j ω is Complex frequency variable.

The expression formula of above formula is multiplied each other, obtain

$S_{11}\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{a_k}^{\left(1\right)}{s}^k-S_{21}\underset{k=0}{\overset{\mathrm{np}}{\mathrm{\Σ}}}{a_k}^{\left(2\right)}{s}^k=0$

Be converted into matrix form and obtain overdetermined equation

$[S_{11}{V}_n-S_{21}{V}_{\mathrm{np}}]\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]=0$

A in formula ⁽¹⁾=[a ₀ ⁽¹⁾, a ₁ ⁽¹⁾..., a _n-1 ⁽¹⁾, a _n ⁽¹⁾] ^t

a ⁽²⁾＝[a ₀ ⁽²⁾,a ₁ ⁽²⁾,…,a _np-1 ⁽²⁾,a _np ⁽²⁾] ^T，

S ₂₁＝diagS ₂₁(s _i)} _{i＝1,2,…,N}

S ₁₁＝diag{S ₁₁(s _i)} _i＝1,2…N，

Diag{S wherein ₂₁(s _i) _i=1,2,, _nand diag{S ₁₁(s _i) _{i=1,2 ... N}it is the matrix of N frequency formation extracting from the measured data of the disconnected rejection filter of follow-up.V is Fan Demeng matrix.Be noted that especially Fan Demeng matrix is wanted and a _k ⁽¹⁾, a _k ⁽²⁾corresponding, Fan Demeng matrix element is herein v _ik=(j ω _i) ^k-1, k=1 ... m+1.; Gained multinomial coefficient is pressed ascending power and is arranged.The above-mentioned overdetermined equation of unusual decomposition method solution

$[S_{11}{V}_n-S_{21}{V}_{\mathrm{np}}]\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]={\mathrm{U\ΣV}}^H\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]=0$

U and V are chief of a tribe's matrixes, and the diagonal entry of Σ matrix is singular value, obtain matrix of coefficients and be

$\left[\begin{array}{c}{a}^{\left(1\right)}\\ a^{\left(2\right)}\end{array}\right]={\left[V^H\right]}_{n+\mathrm{np}+2}$

What now obtain is that uncorrected transition function divides submultinomial and the polynomial coefficient of reflective function molecule, uses respectively P ' (s) and F ' (s) represents:

$F^{\′}\left(s\right)=\underset{k=0}{\overset{N}{\mathrm{\Σ}}}{a_k}^{\left(1\right)}{s}^k$

$P^{\′}\left(s\right)=\underset{k=0}{\overset{\mathrm{nz}}{\mathrm{\Σ}}}{a_k}^{\left(2\right)}{s}^k;$

Deriving the total denominator polynomial expression E (s) of transition function and reflective function is determined by following formula

E(s)E ^*(-s)＝F′(s)F′ ^*(-s)+P′(s)P′ ^*(-s)

s wherein _ifor minus of above-mentioned real part, " * " represents to ask conjugation, and then the multinomial coefficient b asking _k, with its representative polynomial E (s), be:

$E\left(s\right)=\underset{k=0}{\overset{n}{\mathrm{\Σ}}}{b}_k{s}^k$

Step S3: divide submultinomial and reflective function denominator polynomial expression to revise to the transition function of rejection filter, the transition function that obtains revised rejection filter divides submultinomial P (s), reflective function to divide submultinomial F (s) and transition function and reflective function total denominator polynomial expression E (s);

Tried to achieve transition function and the reflective function polynomial expression of rejection filter above, but in application, can find to revise result, reason is scattering matrix and S ₁₁(s) the highest coefficient is real number, must guarantee P (s) before have the phase differential of individual 90 °.The response polynomial expression that polynomial expression of having tried to achieve is modified to final rejection filter is: P (s)=P'(s) j, F (s)=F'(s)/j.E (s) is constant.

Step S4: derive short-circuit admittance matrix [Y from network characterization function _n] _t, residue and with equivalent electrical circuit admittance matrix [Y _n] _ccontrast, extracts initial coupled matrix [M _n+2] (topological structure cannot be realized);

Rejection filter transition function P (s), reflection coefficient polynomial expression F (s) and polynomial expression E (s) from having tried to achieve above, can obtain short-circuit admittance matrix:

Wherein, m ₁=Re (e ₀+ f ₀)+jIm (e ₁+ f ₁) s+Re (e ₂+ f ₂) s ²+

n ₁＝jIm(e ₀+f ₀)+Re(e ₁+f ₁)s+jIm(e ₂+f ₂)s ²+…。

Due to e _iand f _ithe coefficient of E (s) and F (s), m ₁odd item coefficient be 0, n ₁even item coefficient be 0.Residue gained matrix and equivalent electrical circuit admittance matrix, as follows:

${\left[Y_N\right]}_T=j\left[\begin{array}{cc}0& K\\ K& 0\end{array}\right]+\underset{k=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{s-{\mathrm{j\λ}}_k}\left[\begin{array}{cc}{r}_{11k}& r_{12k}\\ r_{21k}& r_{22k}\end{array}\right].$

${\left[Y_N\right]}_C=j\left[\begin{array}{cc}0& M_{\mathrm{sL}}\\ M_{\mathrm{sL}}& 0\end{array}\right]+\underset{k=1}{\overset{N}{\mathrm{\Σ}}}\frac{1}{{\mathrm{sC}}_k+{\mathrm{jB}}_k}\left[\begin{array}{cc}{M^2}_{\mathrm{sk}}& M_{\mathrm{Lk}}{M}_{\mathrm{Lk}}\\ M_{\mathrm{sk}}{M}_{\mathrm{Lk}}& {M^2}_{\mathrm{Lk}}\end{array}\right]$

Wherein K is y in above formula ₂₁(s) the highest coefficient value of item, due to rejection filter y ₂₁(s) be improper fraction, so K is not 0.Each element that comparison two matrixes obtain coupling coefficient matrix is:

M _sL=K, C _k=1, B _k=-λ _k, initial coupled matrix [M _n+2] be:

So far the diagnosis of rejection filter completes substantially.

Step S5: according to the topological structure for the treatment of diagnosis filter, initial coupled matrix is done to corresponding matrix and change, obtain the diagnosis coupled matrix of this rejection filter, compare with design object coupled matrix, determine that whether this rejection filter coupling element is consistent with target coupled matrix element, diagnose out the coupling coefficient that causes rejection filter undesirable.

According to the topological structure of the rejection filter of institute's emulation, carry out accordingly matrix variation, as adopt method of elimination that matrix is become to corresponding coupled matrix of folded form topological structure etc., these can comparatively simply be realized by existing technology for a person skilled in the art, just repeat no more here.

By the above-mentioned description of this invention, the solution of the present invention has been proved to be to have exploitativeness, in order to enable those skilled in the art to get more information about the present invention program, has chosen quadravalence two limit rejection filters and has been further described.

Embodiment: the rejection filter of a quadravalence duopole, centre frequency is 1GHz, and bandwidth is 20MHz, and two limits are respectively-3,2, in stopband, decay to 24dB.The target coupled matrix (normalization) of rejection filter is:

$M=\left[\begin{array}{cccccc}0& 1.5805& 0& 0& 0& 1.0000\\ 1.5805& 0.0202& 0.9780& 0& 1.3945& 0\\ 0& 0.9780& 0.0706& -0.7916& -0.1692& 0\\ 0& 0& -0.7916& -0.2075& 0.9632& 0\\ 0& 1.3945& -0.1692& 0.9632& 0.0202& 1.5805\\ 1.0000& 0& 0& 0& 1.5805& 0\end{array}\right]$

Filter topology is shown in Fig. 2, obtains the response curve of rejection filter, as shown in Figure 3.The ideal response curve corresponding with target coupled matrix compared, and this response results is undesirable, therefore need to utilize the diagnostic method of rejection filter, obtains diagnosing coupling coefficient matrix, and the contrast of target coupled matrix, finds out the coupling coefficient of larger difference.Rejection filter response results data to be diagnosed are derived, use the present invention to carry out the diagnosis of rejection filter.According to step 2 and step 3, obtain respectively P'(s), F'(s), P (s), F (s) and E (s):

F'(s)＝j1.8760s ²+(-0.3837+j0.0264)s-0.1965+j12.8091

P'(s)＝s ⁴+(0.0667-j0.0950)s ³+(5.8040-j0.0047)s ²+(0.1934-j0.3761)s+7.9773-j0.0063

E(s)＝s ⁴+(2.4986-j0.095)s ³+(8.9233-j0.2863)s ²+(11.3216-j0.7947)s+15.0416-j1.2246

F(s)＝1.876s ²+(0.0264+j0.3837)s+(12.8091+j0.1965)

P(s)＝js ⁴+(0.095+j0.0667)s ³+(0.0047+j5.804)s ²+(0.3761+j0.1934)s+(0.0063+j7.9773)

By these polynomial transformations, be desirable admittance matrix, then carry out residue, with the contrast of equivalent electrical circuit admittance matrix, determine that the coupled matrix of diagnosing out is:

$M=\left[\begin{array}{cccccc}0& -1.0486& 1.0484& -0.3303& 0.4246& 1.0000\\ -1.0486& -2.6044& 0& 0& 0& 1.0299\\ 1.0484& 0& 2.5123& 0& 0& 1.0614\\ 0.3303& 0& 0& -2.0646& 0& 0.4001\\ -0.4246& 0& 0& 0& -2.0617& 0.3887\\ 1.0000& 1.0299& 1.0614& 0.4001& 0.3887& 0\end{array}\right]$

Because the topological structure of rejection filter is upper foldable structure, the matrix that utilizes method of elimination to obtain diagnosis is:

$M=\left[\begin{array}{cccccc}0& 1.5774& 0& 0& 0& 1.0000\\ 1.5774& 0.0179& \underset{\‾}{1.5055}& 0& \underset{\‾}{2.0035}& 0\\ 0& \underset{\‾}{1.5055}& 0.0670& \underset{\‾}{-1.5031}& -0.1670& 0\\ 0& 0& -\underset{\‾}{1.5031}& -0.1995& \underset{\‾}{1.4878}& 0\\ 0& \underset{\‾}{2.0035}& -0.1670& \underset{\‾}{1.4878}& 0.0195& 1.5807\\ 1.0000& 0& 0& 0& 1.5807& 0\end{array}\right]$

Diagnosis gained response curve is shown in Fig. 4, and response curve corresponding to coupling coefficient matrix of diagnosing out as can be seen from Figure 4 overlaps substantially with the test curve of this rejection filter, proved the correctness of this diagnostic method.

This matrix and target coupled matrix are compared, the coupling coefficient differing greatly is found out to (marking), exactly because the coefficient in these coupling coefficient and target coupled matrix exists larger difference to cause the undesirable of this rejection filter, therefore according to the physical parameter of adjustment rejection filter corresponding to the problem place of diagnosing out.So far proved the validity of diagnostic method shown in the present.

Those of ordinary skill in the art will recognize; example described here is in order to help the concrete implementation process of reader understanding the present invention; the protection domain that should be understood to invention is not limited to such special statement and example; those of ordinary skill in the art can make various other various concrete distortion and combinations that do not depart from essence of the present invention according to these technology enlightenments disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.

Claims (4)

1. a diagnostic method for rejection filter, is characterized in that comprising the steps:

Step S1), derive the disconnected rejection filter measured data of follow-up, comprise amplitude and the phase place of transmission data, reflectance data;

Step S2), set up the expression formula model of rejection filter reflective function and transition function, utilize step S1) in the measured data of the derivation transition function of calculating rejection filter divide submultinomial and reflective function to divide submultinomial, and derive transition function and the total denominator polynomial expression of reflective function;

Step S3), to the transition function of rejection filter, divide submultinomial and reflective function denominator polynomial expression to revise, the transition function that obtains revised rejection filter divides submultinomial, reflective function to divide submultinomial and transition function and reflective function total denominator polynomial expression;

Step S4), from network characterization function, derive short-circuit admittance matrix, by its residue and with equivalent electrical circuit admittance matrix, contrast, extract initial coupled matrix;

Step S5), according to the topological structure of the disconnected rejection filter of follow-up, initial coupled matrix being done to corresponding matrix changes, obtain the diagnosis coupled matrix of this rejection filter, then compare with the coupled matrix of design object, determine that whether this rejection filter coupling element is consistent with target coupled matrix element, diagnose out the coupling coefficient that causes rejection filter undesirable.

2. the diagnostic method of rejection filter according to claim 1, is characterized in that, described step S2) in the expression formula model of rejection filter reflective function and transition function be respectively:

Wherein, S ₁₁represent reflective function, S ₂₁represent transition function, F (s) represents that reflective function divides submultinomial, and P (s) represents that transition function divides submultinomial, the denominator polynomial expression that E (s) expression transition function and reflective function are total;

N represents the exponent number of rejection filter, and np represents the outer limit of stopband, and the exponent number of polynomial expression E (s) and P (s) should be identical with the exponent number of rejection filter, and the exponent number of F (s) equals the number of the outer limit of stopband, a ⁽¹⁾the coefficient that represents P (s), a ⁽²⁾the coefficient that represents F (s), b represents the coefficient of E (s), s is Complex frequency variable, k be multinomial in every inferior number formulary.

3. the diagnostic method of rejection filter according to claim 2, it is characterized in that described step S2) in calculate rejection filter transition function divide submultinomial and reflective function to divide submultinomial and derive transition function and the polynomial detailed step of denominator that reflective function is total as follows:

Step S2.1), formula (1) and formula (2) are divided by, identical according to two expression formula denominator polynomial expressions, obtain:

Step S2.2), formula (3) is converted into matrix form and obtains overdetermined equation:

In formula,

a ⁽¹⁾＝[a ₀ ⁽¹⁾,a ₁ ⁽¹⁾,…,a _n-1 ⁽¹⁾,a _n ⁽¹⁾] ^T

a ⁽²⁾＝[a ₀ ⁽²⁾,a ₁ ⁽²⁾,…,a _np-1 ⁽²⁾,a _np ⁽²⁾] ^T，

S ₂₁＝diag{S ₂₁(s _i)} _{i＝1,2,…,N}

S ₁₁＝diag{S ₁₁(s _i)} _i＝1,2…N，

Diag{S ₂₁(s _i) _{i=1,2 ..., N}and diag{S ₁₁(s _i) _{i=1,2 ... N}from described step S1) the matrix of N frequency formation extracting in the disconnected rejection filter measured data of follow-up, V is Fan Demeng matrix, Fan Demeng matrix and a _k ⁽¹⁾, a _k ⁽²⁾corresponding, Fan Demeng matrix element is herein v _ik=(j ω _i) ^k-1, k=1 ... m+1; M is the frequency number in stopband in N frequency, and gained multinomial coefficient is pressed ascending power and arranged;

Step S2.3), adopt step S2.2 described in unusual decomposition method solution) in overdetermined equation:

Wherein, U and V are chief of a tribe's matrixes, and the diagonal entry of Σ matrix is singular value, obtain matrix of coefficients and be

Step S2.4), calculate uncorrected transition function and divide submultinomial and the polynomial coefficient of reflective function molecule, use respectively P ' (s) and F ' (s) represents:

Step S2.5), according to following formula, derive transition function and the total denominator polynomial expression E (s) of reflective function:

E(s)E ^*(-s)＝F′(s)F′ ^*(-s)+P′(s)P′ ^*(-s)

Wherein, s _ifor minus of above-mentioned real part, " * " represents to ask conjugation, and then tries to achieve multinomial coefficient b _k, with its representative polynomial E (s), be:

。

4. the diagnosis of rejection filter according to claim 3, is characterized in that, step S3) described in the transition function of revised rejection filter divide submultinomial, reflective function to divide submultinomial to adopt respectively following formula:

P(s)＝P′(s)·j

F(s)＝F′(s)/j

P (s) wherein, F (s) represents that respectively revised transition function and reflective function divide submultinomial, and P ' (s) F ' (s) represents that respectively uncorrected transition function divides submultinomial, and j is that imaginary part is 1 pure imaginary number.