CN102737145B - Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component - Google Patents

Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component Download PDF

Info

Publication number
CN102737145B
CN102737145B CN201210187626.9A CN201210187626A CN102737145B CN 102737145 B CN102737145 B CN 102737145B CN 201210187626 A CN201210187626 A CN 201210187626A CN 102737145 B CN102737145 B CN 102737145B
Authority
CN
China
Prior art keywords
overbar
components
vector
port
parameter
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201210187626.9A
Other languages
Chinese (zh)
Other versions
CN102737145A (en
Inventor
谢树果
陈曦
孙宏涛
赵明敏
马超
张宇
刘亚奇
高娜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN201210187626.9A priority Critical patent/CN102737145B/en
Publication of CN102737145A publication Critical patent/CN102737145A/en
Application granted granted Critical
Publication of CN102737145B publication Critical patent/CN102737145B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a measurement-based modeling method by prediction on an electromagnetic emission broadband behavioral level of an electronic component. The method comprises the following steps of: firstly carrying out active and passive judgments on a measured object, thus acquiring a port scattering parameter of the measured object; then performing vector fitting algorithm on the mapped port scattering parameter to acquire a state space model; and then converting the state space model into an SPICE (Simulation Program with Integrated Circuit Emphasis) circuit to be finally applied to circuit simulation. According to the invention, the modeling method by the prediction on the behavioral level can accurately establish an equivalent broadband model of the complicated electronic component by measuring data within a certain bandwidth or under fewer sampling points without relying on a simulation model provided by an electronic component manufacturer, and can achieve electromagnetic emission simulation on the circuit.

Description

Electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring
Technical field
The present invention relates to the prediction modeling method in a kind of electromagnetism field, more particularly, refer to a kind of electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring.
Background technology
In recent years, people were more and more higher to the attention rate of electromagnetic compatibility, from attention location system in the past, were externally transmitted into the Electromagnetic Launching that starts now to pay close attention to bottom.But several years ago often only focus is concentrated in the problems such as cabling on PCB (printed circuit board), crystal oscillator interconnection line, researcher found that the electronic devices and components on circuit board also can produce electromagnetic interference (EMI) in recent years, if control improper very serious electromagnetic compatibility (EMC) problem that also can produce.It in electronic devices and components, due to active device general work mechanism more complicated, is the main source of generation electromagnetic interference (EMI), before for the Electromagnetic Launching modeling of device be mainly the deviser of electronic devices and components in the situation that knowing device inside circuit to being applied to again after circuit abbreviation in PCB emulation, but this is unpractical for the user for this modeling, the method for setting up the model of device based on measurement is truly IBIS (Input/Output Buffer Information Specification) model reference document [the Hobbs W proposing the nineties; MuranyiA; Rosenbaum R IBIS:I/O buffer information specification, overview 1994], yet IBIS model is for input/output signal, and can not be satisfied with the emulation demand of electromagnetic compatibility.And the user who is concerned about electronic devices and components electromagnetic compatibility for major part, is in most of the cases difficult to find the IBIS model of electronic devices and components used, even if obtained IBIS model, inadequate often for research electromagnetic compatibility problem.
Therefore, by measuring the required information of electron gain components and parts modeling, set up the equivalent black box model of electronic devices and components, and be applied in whole artificial circuit and go tool to have very important significance: on the one hand, for Electronic Components Manufacturing business, do not need to divulge the circuit structure of electronic devices and components inside, can meet the demand of emulation yet simultaneously.On the other hand, user for electronic devices and components, even without the IBIS model of electronic devices and components, also can in certain frequency band or in the situation of less sampling number certificate, set up accurate broadband equivalent-circuit model and be applied in circuit simulation by measuring.
Summary of the invention
The object of the invention is to propose a kind of electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring, thereby the realistic model that needn't rely on Electronic Components Manufacturing producer to provide, only need by measuring in certain frequency band or in the situation of less sampling number certificate, set up exactly the equivalent wide band model of sophisticated electronic components and parts, and the modeling method of circuit being carried out to Electromagnetic Launching emulation, specifically, first by vector network analyzer, measure the scattering parameter of electronic devices and components port, obtain limited frequency point data under certain bandwidth, then data fitting is obtained the state-space model of system, again state-space model is converted to SPICE circuit, finally be applied in circuit simulation and go.
A kind of electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring of the present invention, it includes the following step:
Step 1: the judgement of measurand
Whether required voltage when measurand is carried out to work judges, thereby identify measurand, be active electronic components and parts, passive electronic components and parts;
Step 2: obtain scattering parameter
In setting frequency range, by microwave vector network analyzer, the measurand after step 1 identification is carried out to the measurement of port scattering parameter, and obtain port scattering parameter S (f, w); Described setting frequency range refers to the normal working frequency range according to measurand;
Step 3: scattering parameter mapping
The port scattering parameter S (f, w) that step 2 is obtained is converted to Y parameter form, and saves as .txt formatted file or Excel table format;
Described .txt formatted file can be opened by Microsoft Excel software, thereby forms form;
Step 4: obtain state-space model
The Y parameter form that step 3 is obtained is used vector fitting method to carry out data-switching in Matlab software, obtains the form of residual to the extreme H ‾ ‾ ( s ) = D ‾ ‾ + C ‾ ‾ ( s × 1 ‾ ‾ - A ‾ ‾ ) - 1 B ‾ ‾ ; The macro model more described limit residual form being characterized, converts state-space model to x · ‾ ‾ ( t ) = A ‾ ‾ × x ‾ ( t ) + B ‾ ‾ × u ‾ ( t ) y ‾ ( t ) = C ‾ ‾ × x ‾ ( t ) + D ‾ ‾ × u ‾ ( t ) ;
Step 5: obtain universal circuit net table
Step 501: the state-space model that opening steps four obtains in the software that can read SPICE net table x · ‾ ‾ ( t ) = A ‾ ‾ × x ‾ ( t ) + B ‾ ‾ × u ‾ ( t ) y ‾ ( t ) = C ‾ ‾ × x ‾ ( t ) + D ‾ ‾ × u ‾ ( t ) , Conversion obtains newly-generated SPICE net table;
Step 502: newly-generated SPICE net table is directed in the software that can read SPICE net table, replaces former modeling electronic devices and components, obtain Electromagnetic Launching behavioral scaling model net meter file;
Step 503: utilize computer simulation software ADS to import the Electromagnetic Launching behavioral scaling model net meter file producing, replace former modeling electronic devices and components, the equivalent electrical circuit spice net table obtaining, i.e. universal circuit net table.
The described electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring, adopts wideband sampling oscillograph to measure the instantaneous voltage waveform parameter V of I/O port of the measurand of active electronic components and parts.
The described electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring, adopt microwave vector network analyzer for measuring the port scattering parameter S (f of the measurand of active/passive electronic devices and components, w), f represents frequency, and w represents scattered quantum corresponding under the frequency f of place.
The advantage that the present invention is based on the electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method of measurement is:
1. the production firm due to device does not provide the equivalent model of the device in broadband, even some device does not have equivalent model, this is just for emulation brings difficulty, device Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring can not rely on the model that device production manufacturer provides, do not need to know the physical circuit of device inside, utilize and measure the more accurate broadband device black-box model of limited frequency point data establishment.
2. because institute's established model is spice net sheet form, so model also can use on other circuit simulations, only model net table need be imported in circuit simulation and goes, and have versatility.
3. the input data of the device Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring are generally the limited frequency point data in arrowband, but the model of setting up is wide band model, and can be as accurate as any frequency, and to electromagnetic compatibility, prediction is significant.
Accompanying drawing explanation
Fig. 1 is active electronic components and parts port parameter measuring system block diagram.
Figure 1A is passive electronic components and parts port parameter measuring system block diagram.
Figure 1B is the test platform of tested device.
Fig. 2 is the process flow diagram that the present invention is based on the device Electromagnetic Launching broadband behavioral scaling prediction modeling method of measurement.
Fig. 3 is the SPICE net tableau format schematic diagram generating.
Fig. 3 A is SPICE circuit diagram.
Fig. 4 is the I/O end illustraton of model of building according to measuring I/O port.
Fig. 5 be after measurand B3732 installs, test S parameter testing result.
Fig. 6 A is the Y amplitude fitting result of B3732 vector fitting.
Fig. 6 B is the Y phase-fitting result of B3732 vector fitting.
Fig. 6 C is the model relative error percentage result of B3732 vector fitting.
Fig. 7 is the SPICE net tableau format schematic diagram that B3732 wave filter generates.
Embodiment
Below in conjunction with drawings and Examples, the present invention is described in further detail.
Referring to the electronic devices and components port parameter measuring system shown in Fig. 1, Figure 1A, this system includes measurand (active/passive electronic devices and components), microwave vector network analyzer, wideband sampling oscillograph and computing machine; Microwave vector network analyzer connects on computers by data conductor on the one hand, on the other hand by SMA pop one's head in (as shown in Figure 1B) be connected with measurand (active/passive electronic devices and components); Wideband sampling oscillograph connects on computers by data conductor on the one hand, is connected on the other hand by high-impedance probe with measurand (active/passive electronic devices and components); In described computing machine, store Matlab software, can read software, the Excel software of SPICE net table.
In the present invention, microwave vector network analyzer is used for measuring the port scattering parameter S (f, w) (f represents frequency, and w represents scattered quantum corresponding under the frequency f of place) of measurand (active/passive electronic devices and components).
In the present invention, wideband sampling oscillograph is used for measuring the instantaneous voltage waveform parameter V of the I/O port of active electronic components and parts (measurand).
In the present invention, computing machine is stored in the storer of computing machine with Excel form the described port scattering parameter S (f, w) receiving on the one hand.The row of Excel form is to recording frequency, amplitude and the phase place of the row of Excel form under to record be expert to frequency.
In the present invention, computing machine on the other hand to the described instantaneous voltage waveform parameter V receiving in can read the software of SPICE net table, after conversion, obtain equivalent electrical circuit.
In the present invention, the Matlab software in computing machine can carry out matching according to vector fitting theory to described port scattering parameter S (f, w), obtains state-space model.
In the present invention, Agilent refers to the english of Anjelen Sci. & Tech. Inc.Microwave vector network analyzer is that the model that Anjelen Sci. & Tech. Inc produces is the analyser of Agilent 8719D.
In the present invention, wideband sampling oscillograph is that the model that Anjelen Sci. & Tech. Inc produces is Agilent86100DCA wideband sampling oscillograph.
In the present invention, the measurand for active electronic components and parts adopts electronic devices and components port parameter measuring system as shown in Figure 1 to carry out parameter measurement.Measurand for passive electronic components and parts adopts electronic devices and components port parameter measuring system as shown in Figure 1A to carry out parameter measurement.
Shown in Figure 2, the present invention is a kind of electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring, and includes the following step:
Step 1: the judgement of measurand
Whether required voltage when measurand (as shown in Figure 1B) is carried out to work judges, thereby identify measurand, be active electronic components and parts, passive electronic components and parts; If desired supply voltage, is active electronic components and parts; Otherwise, do not need supply voltage, be passive electronic components and parts.
In the present invention, the I/O port wideband sampling oscilloscope measurement of active electronic components and parts, the power port of active electronic components and parts is measured with microwave vector network analyzer.Passive electronic components and parts are directly measured with microwave vector network analyzer.
Passive electronic components and parts: as resistance has two-port, and be passive electronic components and parts.As electric capacity has two-port, and be passive electronic components and parts.If wave filter is passive electronic components and parts.The logical SAW (Surface Acoustic Wave) filter B3732 of band is passive electronic components and parts.
Active electronic components and parts: as triode has three ports, and be active electronic components and parts.If amplifier chip is active electronic components and parts.
Step 2: obtain scattering parameter
In setting frequency range, by microwave vector network analyzer, the measurand (as shown in Figure 1B) after step 1 identification is carried out to the measurement of port scattering parameter, and obtain port scattering parameter S (f, w); Described setting frequency range refers to the normal working frequency range according to measurand.
During as the normal work of SAW wave filter, centre frequency is 433.92MHz, and passband is 0.72MHz.Should measure the S parameter in 380MHz~485MHz frequency range.
Step 3: scattering parameter mapping
The port scattering parameter S (f, w) that step 2 is obtained is converted to Y parameter form, and saves as .txt formatted file (being text document) or Excel table format.Described .txt formatted file can be opened by Microsoft Excel software, thereby forms form.
In the present invention, concrete input variable is defined as a K * 1 dimension complex frequency vector s and p * p * K dimension complex matrix row Y, wherein a s=j2 π f k,
Figure BDA00001739514400061
k=1,2 ... K.J is unit imaginary number, f kbe k the frequency that sampled point is corresponding, K is sampled point number, and p is the port number of passive electronic components and parts.
Y parameter refers to that < < Microwave Technique and Antenna (second edition) > > Wang Xin is steady, Li Ping, Li Yanping write, Electronic Industry Press, 2006.12ISBN7-121-02992-8.
In the present invention, Y parameter form form is table 1:
Table 1:Y parametric form
Y
f
Suppose to have passive electronic components and parts with three ports (the first port is designated as a, and the second port is designated as b, and the 3rd port is designated as c), Y parameter form is Y = Y 11 Y 12 Y 13 Y 21 Y 22 Y 23 Y 31 Y 32 Y 33 , Y 11the echo parameter that represents a port, Y 12represent that a port is to the transformation parameter of b port, Y 13represent that a port is to the transformation parameter of c port, Y 21represent that b port is to the transformation parameter of a port, Y 22the echo parameter that represents b port, Y 23represent that b port is to the transformation parameter of c port, Y 31represent that c port is to the transformation parameter of a port, Y 32represent that c port is to the transformation parameter of b port, Y 33the echo parameter that represents c port.
By Y parameter form, be Y = Y 11 Y 12 Y 13 Y 21 Y 22 Y 23 Y 31 Y 32 Y 33 Being converted to form is table 2.
Table 2: the Y parameter form of three port passive electronic components and parts
Figure BDA00001739514400064
Step 4: obtain state-space model
The Y parameter form that step 3 is obtained is used vector fitting method to carry out data-switching in Matlab software, obtains the form of residual to the extreme H &OverBar; &OverBar; ( s ) = D &OverBar; &OverBar; + C &OverBar; &OverBar; ( s &times; 1 &OverBar; &OverBar; - A &OverBar; &OverBar; ) - 1 B &OverBar; &OverBar; ; The macro model more described limit residual form being characterized, converts state-space model to x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) .
S represents a K * 1 dimension complex frequency vector;
Figure BDA00001739514400072
represent the admittance Y parameter under the dimension complex frequency vector s of K * 1;
Figure BDA00001739514400073
the real matrix that represents p * p;
Figure BDA00001739514400074
the real matrix that represents p * N, N represents matching exponent number;
Figure BDA00001739514400075
representation unit matrix;
Figure BDA00001739514400076
the real matrix that represents N * N;
Figure BDA00001739514400077
the real matrix that represents N * p;
T represents time domain;
Figure BDA00001739514400078
represent the vector that comprises state variable, and be the real vector of N * 1;
Figure BDA00001739514400079
represent input vector, and be the real vector of p * 1;
Figure BDA000017395144000710
it is right to represent differentiate;
Figure BDA000017395144000712
represent output vector, and be the real vector of p * 1.
In the present invention, vector fitting (Vector Fitting) is a kind of fitting technique of rational polynominal, and data are converted to limit residual form.
What in the present invention, state-space model was used is Jordan canonical form.
Step 5: obtain universal circuit net table
Step 501: the state-space model that opening steps four obtains in the software that can read SPICE net table x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) , Conversion obtains SPICE (Simulation Program with Integrated Circuit Emphasis) net table, is called newly-generated SPICE net table;
Step 502: newly-generated SPICE net table is directed in the software that can read SPICE net table, replaces former modeling electronic devices and components, obtain Electromagnetic Launching behavioral scaling model net meter file;
Step 503: utilize computer simulation software ADS to import the Electromagnetic Launching behavioral scaling model net meter file producing, replace former modeling electronic devices and components, the equivalent electrical circuit spice net table obtaining, i.e. universal circuit net table.
In the present invention, this newly-generated SPICE net table is the equivalent electric road network table of behavioral scaling model.
In the present invention, net table exists with .sp formatted file, and net sheet format as shown in Figure 3.
For example: the state-space model of a two-port network is x &CenterDot; 1 x &CenterDot; 2 = a 11 a 21 a 21 a 22 x 1 x 2 + b 11 b 12 b 21 b 22 v 1 v 2 With i 1 i 2 = c 11 c 12 c 21 c 22 x 1 x 2 + d 11 d 12 d 21 d 22 v 1 v 2 ;
Figure BDA00001739514400083
be equal to x &CenterDot; 1 x &CenterDot; 2 , be equal to a 11 a 12 a 21 a 22 ,
Figure BDA00001739514400087
be equal to x 1 x 2 , be equal to b 11 b 12 b 21 b 22 ,
Figure BDA000017395144000811
be equal to v 1 v 2 , be equal to i 1 i 2 ,
Figure BDA000017395144000815
be equal to c 11 c 12 c 21 c 22 ,
Figure BDA000017395144000817
be equal to d 11 d 12 d 21 d 22 .
X wherein 1, x 2for state variable,
Figure BDA000017395144000819
for the derivative of state variable, v 1, v 2for magnitude of voltage, v n1, v n2for the partial node magnitude of voltage definite according to state variable, i 1, i 2for current value, a 11, a 12, a 21, a 22for
Figure BDA000017395144000820
each element value, b 11, b 12, b 21, b 22for
Figure BDA000017395144000821
each element value, c 11, c 12, c 21, c 22for each element value, d 11, d 12, d 21, d 22for
Figure BDA000017395144000823
each element value, according to above formula, set up spice circuit diagram as shown in Figure 3A, wherein R d 1 = 1 d 11 , R d 2 = 1 d 22 , R n 1 = 1 a 11 , R n 2 = 1 d 22 , C n1=1,C n2=1。
G is current-controlled voltage source, and value is respectively: G c1-1=c 11v n1, G c1-2=c 12v n2, G 1-2=d 12v 2, G 2-1=d 21v 1, G c2-1=c 21v n1, G c2-2=c 22v n2, G a1-2=a 12v n2, G b1-1=b 11v 1, G b1-2=b 12v 2, G a2-1=a 21v n1, G b2-1=b 21v 1, G b2-2=b 22v 2.
Because two-port network has two state variables, so State Equation Coefficients
Figure BDA000017395144000828
be 2 * 2 matrix, a 11for
Figure BDA000017395144000829
first element value of matrix, a 12for second element value of matrix, a 21for
Figure BDA000017395144000831
the 3rd element value of matrix, a 22for
Figure BDA000017395144000832
the 4th element value of matrix, b 11for
Figure BDA000017395144000833
first element value of matrix, b 12for
Figure BDA000017395144000834
second element value of matrix, b 21for
Figure BDA000017395144000835
the 3rd element value of matrix, b 22for
Figure BDA000017395144000836
the 4th element value of matrix, c 11for
Figure BDA000017395144000837
first element value of matrix, c 12for
Figure BDA000017395144000838
second element value of matrix, c 21for
Figure BDA000017395144000839
the 3rd element value of matrix, c 22for
Figure BDA000017395144000840
the 4th element value of matrix, d 11for first element value of matrix, d 12for
Figure BDA000017395144000842
second element value of matrix, d 21for
Figure BDA000017395144000843
the 3rd element value of matrix, d 22for
Figure BDA000017395144000844
the 4th element value of matrix.
In the present invention, for passive electronic components and parts, resulting equivalent electrical circuit spice net table can directly be used in the emulation of system and go.
In the present invention, for active electronic components and parts, need to add equivalent source model and the I/O port model (as shown in Figure 4) of active device.
In Fig. 4, Ucc is fixing DC feedback voltage, i u(t) be upper trombone slide electric current, i l(t) be drop-down tube current, I prepresent the current value I on power supply catching diode grepresent ground clamp diode voltage value.C cfor input and output junction capacity, the encapsulation characteristic of device is by L p, R pand C prepresent.Then add lump unit prime model in order to describe encapsulation, the load on PCB cabling and installing plate.Finally build artificial circuit prediction circuit emission characteristics.
embodiment
A logical Surface Acoustic Wave Filter B3732 of band being installed on PCB, its centre frequency is 433.92MHz, passband is 0.72MHz.On the impact of performance of filter and well design front and back end match circuit, use the S parameter measurement of vowing after net is installed it when assessing actual installation.
Measurement result in 380MHz~485MHz frequency range is as shown in 5, and in figure, S21 curve is comparatively coarse, and has a large amount of extreme points.Therefore the scattering parameter amount of this B3732 wave filter is comparatively complicated, need to adopt the higher order pole residual model on 175 rank to approach it, obtain amplitude and phase place Approaching Results that result Fig. 6 A and Fig. 6 B as shown in Fig. 6 A, Fig. 6 B, Fig. 6 C are respectively Surface Acoustic Wave Filter B3732 port Y parameter, because this B3732 wave filter has reciprocity and structural symmetry, so in figure, only shown Y11 and the Y21 parameter in Y parameter form.Contrast Approaching Results and measurement result, known this 175 rank vector approximate model can well approach the port Y parameter of B3732 wave filter.Fig. 6 C is model relative error number percent.
Then in Matlab software, use vector fitting method to carry out data-switching the Y parameter form of B3732 wave filter, obtain the form of residual to the extreme H &OverBar; &OverBar; ( s ) = D &OverBar; &OverBar; + C &OverBar; &OverBar; ( s &times; 1 &OverBar; &OverBar; - A &OverBar; &OverBar; ) - 1 B &OverBar; &OverBar; ; The macro model more described limit residual form being characterized, converts state-space model to x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) ; And can read open mode spatial model in the software of SPICE net table x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) , Conversion obtains the SPICE net table of B3732 wave filter, as shown in Figure 7.
Finally the SPICE net table of B3732 wave filter is replaced to the wave filter of identical function, thereby obtain the universal circuit net table of general satisfied conventional wave filter.

Claims (3)

1. the electronic devices and components Electromagnetic Launching broadband behavioral scaling based on measuring is predicted a modeling method, it is characterized in that including the following step:
Step 1: the judgement of measurand
Whether required voltage when measurand is carried out to work judges, thereby identify measurand, be active electronic components and parts or passive electronic components and parts;
Step 2: obtain scattering parameter
When measurand is active electronic components and parts, the I/O port wideband sampling oscilloscope measurement of active electronic components and parts, the power port of active electronic components and parts is measured with microwave vector network analyzer;
When measurand is passive electronic components and parts, passive electronic components and parts are directly measured with microwave vector network analyzer;
In setting frequency range, by microwave vector network analyzer, the measurand after step 1 identification is carried out to the measurement of port scattering parameter, and obtain port scattering parameter S (f, w), f represents frequency, w represents scattered quantum corresponding under the frequency f of place; Described setting frequency range refers to the normal working frequency range of measurand; Step 3: scattering parameter mapping
The port scattering parameter S (f, w) that step 2 is obtained is converted to Y parameter form, and saves as .txt formatted file or Excel table format file;
Described .txt formatted file can be opened by Microsoft Excel software, thereby forms form;
Step 4: obtain state-space model
The Y parameter form that step 3 is obtained is used vector fitting method to carry out data-switching in Matlab software, obtains the form of residual to the extreme
Figure FDA0000418193170000011
the macro model more described limit residual form being characterized, converts state-space model to x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) ;
S represents a K * 1 dimension complex frequency vector, and K is sampled point number;
Figure FDA0000418193170000013
represent the admittance Y parameter under the dimension complex frequency vector s of K * 1;
Figure FDA0000418193170000014
the real matrix that represents p * p, the port number that p is electronic devices and components;
Figure FDA0000418193170000015
the real matrix that represents p * N, N represents matching exponent number;
Figure FDA0000418193170000016
representation unit matrix;
Figure FDA0000418193170000017
the real matrix that represents N * N;
Figure FDA0000418193170000018
the real matrix that represents N * p;
T represents time domain;
Figure FDA0000418193170000021
represent the vector that comprises state variable, and be the real vector of N * 1;
Figure FDA0000418193170000022
represent input vector, and be the real vector of p * 1;
Figure FDA0000418193170000023
it is right to represent
Figure FDA0000418193170000024
differentiate;
Figure FDA0000418193170000026
represent output vector, and be the real vector of p * 1;
Step 5: obtain universal circuit net table
Step 501: the state-space model that opening steps four obtains in the software that can read SPICE net table x &CenterDot; &OverBar; &OverBar; ( t ) = A &OverBar; &OverBar; &times; x &OverBar; ( t ) + B &OverBar; &OverBar; &times; u &OverBar; ( t ) y &OverBar; ( t ) = C &OverBar; &OverBar; &times; x &OverBar; ( t ) + D &OverBar; &OverBar; &times; u &OverBar; ( t ) , Conversion obtains newly-generated SPICE net table;
Step 502: newly-generated SPICE net table is directed in the software that can read SPICE net table, replaces former modeling electronic devices and components, obtain Electromagnetic Launching behavioral scaling model net meter file;
Step 503: utilize computer simulation software ADS to import the Electromagnetic Launching behavioral scaling model net meter file producing, replace former modeling electronic devices and components, the equivalent electrical circuit SPICE net table obtaining, i.e. universal circuit net table.
2. the electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring according to claim 1, is characterized in that: wideband sampling oscillograph is for measuring the instantaneous voltage waveform parameter V of I/O port of the measurand of active electronic components and parts.
3. the electronic devices and components Electromagnetic Launching broadband behavioral scaling prediction modeling method based on measuring according to claim 1, is characterized in that: the equivalent electric road network table that in step 5, newly-generated SPICE net table is a behavioral scaling model.
CN201210187626.9A 2012-06-07 2012-06-07 Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component Active CN102737145B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210187626.9A CN102737145B (en) 2012-06-07 2012-06-07 Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210187626.9A CN102737145B (en) 2012-06-07 2012-06-07 Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component

Publications (2)

Publication Number Publication Date
CN102737145A CN102737145A (en) 2012-10-17
CN102737145B true CN102737145B (en) 2014-03-26

Family

ID=46992643

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210187626.9A Active CN102737145B (en) 2012-06-07 2012-06-07 Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component

Country Status (1)

Country Link
CN (1) CN102737145B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103792435B (en) * 2013-12-30 2017-02-15 京信通信技术(广州)有限公司 Coupling component, and data measuring device and method for measuring scattering parameters
CN104008246A (en) * 2014-06-04 2014-08-27 北京航空航天大学 Vector fitting and balanced truncation method based electromagnetic compatible macro model modeling method
CN106372348B (en) * 2016-09-07 2018-03-02 北京航空航天大学 Vector fitting model order reducing method based on control errors in a kind of linear system
CN110188381B (en) * 2019-04-18 2023-04-18 中国北方车辆研究所 Method and system for constructing simulation model for electromagnetic interference prediction

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0642028B1 (en) * 1993-09-08 1999-05-26 Gunter Langer Measuring and experimenting system for sensing EMP susceptibility of electrical appliances and circuits
CN101629980B (en) * 2009-09-10 2011-04-27 南京师范大学 Method for testing performance of EMI filter based on scattering parameter
CN101714184B (en) * 2009-11-20 2011-08-17 北京航空航天大学 Behavioral level modeling system for circuit board level electromagnetic compatible sensitivity

Also Published As

Publication number Publication date
CN102737145A (en) 2012-10-17

Similar Documents

Publication Publication Date Title
Verspecht Large-signal network analysis
CN101629980B (en) Method for testing performance of EMI filter based on scattering parameter
CN104111435B (en) A kind of test fixture error elimination method
Cai et al. Padé-approximation-based behavioral modeling
CN102737145B (en) Measurement-based modeling method by prediction on electromagnetic emission broadband behavioral level of electronic component
CN102314532B (en) IBIS (Input/Output Buffer Information Specification) model verification method and system
US7295961B2 (en) Method for generating a circuit model
Morsey et al. PRIME: Passive realization of interconnect models from measured data
US11929902B2 (en) PCIe signal bandwidth determining method, apparatus and device
CN105572499A (en) Eye graph generating method based on vector network analyzer
CN106093749A (en) Embedding method and system are gone in a kind of millimeter wave/Terahertz ultra broadband open circuit short circuit
CN106405388B (en) A kind of digit chip function test method and system
Saraswat et al. Fast passivity verification and enforcement via reciprocal systems for interconnects with large order macromodels
Zhang et al. Behavioral modeling of power amplifier with long term memory effects using recurrent neural networks
Gubsky et al. Development a computer model of microstrip filter for virtual laboratory
WO2023124592A1 (en) Pt primary voltage reconstruction method based on inverse black box model and inverse electromagnetic duality model
CN103257269A (en) Harmonic source modeling method based on actual measured current characteristic curve fitting
CN106896270A (en) A kind of measuring method of transmission line impedance
US8593132B2 (en) Parameter calculating apparatus and simulation apparatus
Cai et al. X-parameter-based frequency doubler design
Horn et al. Measurement-based large-signal simulation of active components from automated nonlinear vector network analyzer data via X-parameters
CN102279896B (en) Hspice compatible equivalent circuit simulation system and method
Scott et al. Enhanced on-wafer time-domain waveform measurement through removal of interconnect dispersion and measurement instrument jitter
Baba et al. Noise Sources Extraction for Conducted Emission Modeling of IC’s using IBIS Models
Cai et al. An improved quadratic poly-harmonic distortion behavioral model

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant