KR101976489B1 - Apparatus and method for prediction of radiated electromagnetic waves from circuit - Google Patents

Abstract

An apparatus for predicting an electromagnetic wave according to a common mode of a circuit according to an embodiment of the present invention includes a sample extracting unit for extracting a certain number of samples based on a random variable having a probabilistic distribution of a process error in a transmission line, A simulation executing unit for modeling each of the predetermined number of samples to execute the electromagnetic field simulation, and a predicting unit for predicting the electromagnetic wave radiation characteristic value of the transmission line using the simulation result of each sample.

Description

[0001] APPARATUS AND METHOD FOR PREDICTION OF RADIATED ELECTROMAGNETIC WAVES FROM CIRCUIT [0002]

[0001] The present invention relates to an apparatus and method for predicting electromagnetic radiation by a common mode of a circuit, and more particularly, to an electromagnetic radiation simulation apparatus and method, To a device and a method capable of predicting the performance of the device.

When electronic products composed of various electric and electronic circuits such as communication devices, information devices, and automation devices are placed close to each other, they interact with each other electrically or electromagnetically to have undesirable effects. That is, when various spurious and noise components generated in a circuit or a system are radiated to the outside, it acts as a jammer to other electronic systems, and this is called Electro Magnetic Interference (EMI).

When electromagnetic interference is classified according to its type, there are conduction noise (Conducted Emission or Conducted Noise) going out to the outside through the power line of the product and radiated emission (radiated noise) radiated to the outside in normal operation of the electronic product. The present invention deals with a method of predicting radiated noise.

FIG. 1 shows radiation characteristics on a transmission line. When a radiation mode on a transmission line through which a high-speed digital signal is transmitted is classified into two types, differential mode radiation (DM radiation) and common mode radiation : CM Radiation). The radiation characteristics on the transmission line are radiated by a combination of the differential mode and the common mode, and it is difficult to accurately predict them. In particular, common-mode radiation is particularly difficult to predict because common mode is caused by fine asymmetry of the circuit, and this asymmetry is mainly caused by fabrication errors unrelated to the intention of the circuit designer and parasitic components with peripheral circuits It is very difficult to predict the electromagnetic radiation characteristics by checking the asymmetry in the design stage of the circuit.

In other words, assuming that the circuit is realized with accurate numerical values according to the designer's intention, the electromagnetic wave radiation characteristic is predicted. However, in the conventional method, it is impossible to predict the common mode current due to the asymmetry between the transmission lines due to the process error occurring during the process of manufacturing the circuit. Common mode currents require more care than electromagnetic radiation due to differential mode currents, since even small current differences cause large electromagnetic radiation phenomena.

Therefore, it is required to develop a technique for predicting the electromagnetic wave radiation caused by the common mode current that can occur in the circuit in the circuit designing stage.

Prior art related to this is Korea Unexamined Patent Publication No. 2012-0101873 entitled " Radiation electromagnetic wave prediction device and method of cable ".

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and method for predicting an electromagnetic wave radiated by a common mode of a circuit capable of predicting the electromagnetic wave radiation characteristic due to a common mode current that can occur in a circuit in a circuit design stage.

The apparatus for predicting electromagnetic radiation by a common mode of a circuit according to an embodiment of the present invention is a device for predicting electromagnetic waves radiated from a sample And a predictor for predicting an electromagnetic wave radiation characteristic value of a transmission line using the simulation results of each sample, wherein the simulation execution unit includes: a simulation unit The S-parameter is extracted by modeling the transmission line structure in which the random variable is reflected, and a circuit having the extracted S-parameter is simulated to extract a current source, and the extracted current source is applied to simulate the electromagnetic field emission .

Preferably, the sample extracting unit sets at least one element having a numerical value in the process error in the transmission line structure and a random variable such that the element has a specific stochastic distribution, and calculates a probability density function for the random variable , And a sample of a predetermined number ( K ) can be extracted by specifying a testing point of a certain number ( K ) by using a stochastic testing algorithm.

Preferably, the sample extracting unit may designate a predetermined number ( K ) of testing points by the following equation.

[Mathematical Expression]

Where d is the number of random variables and p is the order of the highest probability density function.

Preferably, the simulation executing unit includes: a structure modeling unit for modeling a transmission line structure reflecting the random variable to extract an S parameter; a source modeling unit for extracting a current source by simulating a circuit having the extracted S parameter; And a simulation unit for simulating the electromagnetic field emission by applying the extracted current source.

Preferably, the predicting unit may be configured to generate a transform matrix using a polynomial corresponding to a random variable at each testing point and a simulation result, and to calculate a coefficient corresponding to each degree using the inverse matrix of the transform matrix A calculating unit, and a predicted value calculating unit that calculates the predicted value of the electromagnetic wave radiation using the calculated coefficient.

Advantageously, said coefficient calculator is capable of generating a transformation matrix in the formulas described below.

[Mathematical Expression]

는 시뮬레이션 결과, y(t)는 계수임.Here, [phi] is a polynomial corresponding to a random variable at each testing point,

Is the simulation result, and y (t) is the coefficient.

A method of predicting electromagnetic wave radiation by a common mode of a circuit according to another embodiment of the present invention is a method of predicting electromagnetic wave radiation by a common mode of a circuit in an electromagnetic wave radiation prediction apparatus, Extracting a certain number of samples based on a random variable set to have a probabilistic distribution, performing electromagnetic field simulation by modeling each of the extracted predetermined number of samples, and outputting simulation results of the samples to the transmission line Wherein the step of performing the simulation includes the steps of modeling a transmission line structure reflecting the random variable to extract an S parameter and simulating a circuit having the extracted S parameter to calculate a current source The extracted current source is applied, Characterized in that the calibration.

Preferably, the step of extracting the sample comprises setting at least one element having a numerical value in the process error in the transmission line structure and a random variable such that the element has a specific probability distribution, and calculating a probability Density function, and a sample of a certain number ( K ) can be extracted by designating a testing point of a certain number ( K ) by using a stochastic testing algorithm.

Preferably, the number of test points K can be calculated by the following equation.

 [Mathematical Expression]

Where d is the number of random variables and p is the order of the highest probability density function.

delete

Preferably, the predicting step may include: generating a transform matrix by using a polynomial corresponding to a random variable at each testing point and a simulation result, and calculating a coefficient corresponding to each degree using an inverse matrix of the transform matrix And calculating a predicted value of the electromagnetic wave radiation using the calculated coefficient.

According to the present invention, it is possible to efficiently predict an electromagnetic wave radiation characteristic due to a common mode current that can occur in a circuit in a circuit designing stage. Specifically, by simulating the electromagnetic field by reflecting the process errors that may occur during the manufacturing process of the circuit, the common mode radiation characteristic is predicted, and the performance can be predicted only after the manufacturing is completed and the real object is secured The electromagnetic wave radiation characteristics can be predicted in advance. However, the present invention can predict the amount of electromagnetic wave radiation due to the common mode in the designing stage, thereby shortening the development period and improving the circuit quality. You can contribute.

Also, according to the present invention, by estimating the common mode radiation characteristic by the generalized Polynomial Chaos method, only a very small number of samples can be extracted and compared with the conventional method, and the common mode radiation characteristic can be predicted in a short time.

In addition, the radiation characteristics of the common mode caused by the stochastic structural asymmetry can be conveniently and accurately predicted using a commercial numerical analysis tool.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood to those of ordinary skill in the art from the following description.

1 is a view for explaining radiation characteristics on a transmission line.
2 is a diagram for comparing a method of predicting the electromagnetic wave radiated by a common mode of a circuit according to an embodiment of the present invention and a conventional method.
3 is a diagram for explaining a circuit including a variable having a general probability distribution.
4 is a diagram for explaining a transmission line circuit including a random variable according to an embodiment of the present invention.
5 is a diagram for explaining an apparatus for predicting an electromagnetic wave according to a common mode of a circuit according to an embodiment of the present invention.
6 is a flowchart for explaining a method of predicting the electromagnetic wave radiation according to a common mode of a circuit according to an embodiment of the present invention.
FIG. 7 is a diagram illustrating a coplanar stripline model for predicting the electromagnetic wave radiation by the common mode of a circuit according to an embodiment of the present invention. Referring to FIG.
8 is a graph comparing measurement results of the electromagnetic waves radiated from the circuit shown in FIG. 7 and prediction results.
9 is a diagram illustrating a microstrip coupled line model for predicting the electromagnetic wave radiation by the common mode of the circuit according to an embodiment of the present invention.
10 is a graph showing a result of simulating electromagnetic wave radiation in the common mode using the Monte Carlo method in the model shown in FIG.
FIG. 11 is a graph showing a result of simulation of the common mode electromagnetic wave radiation using the generalized Polynomial Chaos method in the model shown in FIG.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all changes, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.

The terms first, second, A, B, etc. may be used to describe various elements, but the elements should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly indicates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method of predicting electromagnetic radiation by a common mode of a circuit according to an embodiment of the present invention, 4 is a diagram for explaining a transmission line circuit including a random variable according to an embodiment of the present invention.

2 (a) is a diagram for explaining a method of predicting the electromagnetic wave radiation of a circuit using a conventional deterministic solution. This method is a method of executing an electromagnetic field simulation using a circuit implemented with precise numerical values according to the designer's intention.

Specifically, deterministic solution is a method that takes into account the number of all cases that can occur in the range of process errors. In order to predict common mode radiation by asymmetry between transmission lines, And the number of branches is multiplied. That is, if the process error of each variable is calculated, and if it is possible to have N branches all within the error range, a simulation of the total ^number N ^{( the number of variables )} must be performed.

를 적용한다고 하면, 85um부터 115um까지 1um단위의 등간격으로 31개의 개수로 나눠서 해석할 수 있다. 이때 4개의 다른 변수들도 31개의 개수로 나눠서 해석한다고 하면, 총 5개의 변수에 대하여

만큼 시뮬레이션을 진행해야한다. 물론, 각 변수의 개수를 줄여서 시뮬레이션을 진행할 수는 있지만, 분석의 질이 현저히 떨어지게 되는 단점이 있다.For example, in a 100-μm line,

, It is possible to divide it into 31 numbers at regular intervals of 1um from 85um to 115um. In this case, assuming that 4 different variables are divided into 31 numbers,

As shown in FIG. Of course, it is possible to reduce the number of variables and proceed with the simulation, but the quality of the analysis is considerably deteriorated.

On the other hand, in general, the product manufacturing process generates a stochastic distribution, which includes a random variable having a stochastic distribution such as the circuit shown in Fig.

In order to solve the problem involving the random variable with the probabilistic distribution, the conventional deterministic problems should be redefined as stochastic problems, and the variables present in the problem should be changed to random variables do.

Therefore, there is a method of probabilistically predicting the electromagnetic wave radiation characteristic by the common mode of the transmission line. In this way, there are (b), (c), (d), etc. shown in FIG.

First, (b) is a method for predicting the electromagnetic radiation by the common mode of the circuit by using the Monte-Carlo method. This method is a method of constructing random N sample circuits within a process error and simulating the electromagnetic field of each of the N sample circuits to predict the electromagnetic field radiation characteristic. This method has a similar prediction level as the number of random samples ( N ) increases, but has a drawback in that it is necessary to perform simulation repetition in order to obtain reliable data. In addition, the Monte-Carlo (MC) method has a problem that the number of execution increases suddenly according to the number of variables to which the stochastic tolerance should be applied in order to obtain a reliable prediction result by converging over a certain level, , It can be used only in a very narrow range based on the experience of the engineer.

Next, (c) is a method of predicting the electromagnetic radiation by the common mode of the circuit by using the generalized Polynomial Chaos (gPC) method. A detailed description thereof will be given later.

In addition to Monte-Carlo and generalized Polynomial Chaos method shown in (d), we can predict the electromagnetic radiation by the common mode of circuits by using Unscented Transformation, Stochastic Galerkin Method, Polynomial Chaos Method, Stochastic Collocation Method and Sensitivity Analysis There is a way. This method can appropriately or dramatically reduce the number of branches to be computed according to the stochastic technique to be applied. However, the Stochastic Collocation or Stochastic Galerkin method is not an engineer with a complex understanding of the basic knowledge and a complex process of deriving Partial Differential Equation (PDE) for a given statistical approach. It is difficult to carry out.

To overcome these shortcomings of the probabilistic prediction technique, there is a generalized Polynomial Chaos (gPC) method shown in (c). The generalized polynomial Chaos (gPC) method is a method of estimating the electromagnetic radiation by the common mode by extracting only a few samples using the stochastic testing algorithm when the process error appears as a certain probability distribution.

개의 샘플을 추출하는 방법으로

개의 샘플은

에 의해 산출된다. 이때 d는 랜덤변수의 개수이고, p는 최고차항의 차수를 의미한다. Specifically, generalized Polynomial Chaos method specifies a testing point of K (K << N) using a stochastic testing techniques, and the process proceeds to an electromagnetic field simulation by extracting only the K samples. Here, the stochastic testing method has a statistically significant

How to extract samples of

The samples of

Lt; / RTI &gt; Where d is the number of random variables and p is the order of the highest order.

Since the generalized polynomial Chaos method extracts only a very small number of samples and performs the simulation, it can predict the common mode radiation characteristics in a short time. In addition, the generalized Polynomial Chaos method can predict the common mode radiation due to the asymmetry of the circuit transmission line, which can occur during circuit manufacturing, through a probabilistic approach that the circuit is within a process range error. Also, the generalized polynomial Chaos (gPC) method can conveniently and accurately predict the radiation characteristics of common mode caused by stochastic structural asymmetry using a commercial numerical analysis tool.

Hereinafter, generalized Polynomial Chaos will be described.

(

는 독립적이고 이상분포를 가지는 다차원(multi-dimensional) 랜덤변수)로 표현할 수 있다고 하면, 임의의 노드 전압은 확률적 분포를 가지는 랜덤변수를 포함한 spectral 노드 전압

로 표현이 가능하다. 이때, 랜덤 변수를 포함한 전송선로 회로는 도 4와 같을 수 있다. According to the Polynomial Chaos framework, an arbitrary node voltage in a high-speed operation circuit including a simple transmission line (e.g., microstrip, stripline, etc.) and a general non-linear device is called v ( t ) Each of the elements in the random variable

(

Is an independent and multi-dimensional random variable having an ideal distribution, the arbitrary node voltage is expressed by a spectral node voltage including a random variable having a stochastic distribution

Can be expressed as. At this time, the transmission line circuit including the random variable may be as shown in FIG.

는 직교 다항식을 통해 랜덤변수와 기저함수를 이용한 전개식이 가능하다. 랜덤변수

와 직교 다항식간의 관계는 표 1과 같이 랜덤변수

의 확률밀도함수(Probability Density Function : PDF)에 관계가 있다. Spectral Node Voltage by Generalized Polynomial Chaos Concept

Can be expanded by using random variables and basis functions through orthogonal polynomials. Random variable

And the orthogonal polynomials are as shown in Table 1,

(Probability Density Function: PDF).

The spectral node voltage can be approximated again to a finite number of terms, which can be expressed as Equation 1 below.

Here,? Represents a k- th degree polynomial related to Table 1, v _k is a deterministic coefficient corresponding to each degree, and can be approximated by K terms of Equation (2) described below.

Where p is the order of the polynomial highest order and d is the number of random variables. It is generally known that accuracy is guaranteed even with p = 2 in a circuit circuit.

The inner product of two general functions Φ ₁ and Φ _{2 in} the stochastic space is shown in Equation 3 below.

는

의 확률 밀도 함수이다.here,

The

Is a probability density function of.

Since the basis functions of the normalized generalized polynomial Chaos are orthogonal to each other, they have the following Equation (4).

는 Kroneker-Delta 함수이다.here,

Is the Kroneker-Delta function.

Due to the characteristics of such an orthogonal function, it is possible to directly calculate the statistical indices regardless of the complex expansion form if the coefficients in Equation 1 can be obtained. The statistical index includes an average value, dispersion, and the like. The average value of the spectral node voltage can be calculated using Equation (5) described below, and the dispersion can be calculated using Equation (6) described below.

Hereinafter, a technique for predicting the electromagnetic wave radiation by the common mode of the circuit using the generalized Polynomial Chaos (gPC) will be described.

5 is a diagram for explaining an apparatus for predicting an electromagnetic wave according to a common mode of a circuit according to an embodiment of the present invention.

5, an apparatus 100 for predicting electromagnetic radiation by a common mode of a circuit according to an embodiment of the present invention includes a sample extracting unit 110, a simulation executing unit 120, and a predicting unit 130 .

개의 샘플을 추출하는 방법으로

개의 샘플은 수학식 2에 의해 산출될 수 있다.The sample extracting unit 110 extracts a certain number of samples based on a random variable set so that at least one element constituting the transmission line has a probabilistic distribution. That is, the sample extracting unit 110 sets at least one element having a process error in a transmission line structure and a random variable such that the element has a certain probability distribution, sets a probability density function for the random variable, A sample of a predetermined number ( K ) is extracted by designating a testing point of a certain number ( K ) by using a stochastic testing algorithm. Here, the elements constituting the transmission line may include the width of the line, the distance between the lines, the thickness of the substrate, and the like in the structure of the transmission line. Therefore, the random variable is a design parameter having a random value in the process error during the manufacturing process of the transmission line, such as the width of the line, the spacing between the lines, And the like. The random variable has a stochastic distribution such as a Gaussian distribution, a gamma distribution, and a beta distribution. The types of probability density functions can include Hermite, Laguerre, Jacobi, and so on. Stochastic Testing algorithms are probabilistically meaningful

How to extract samples of

The number of samples can be calculated by the following equation (2).

For example, in the transmission line structure, the width of two lines and the interval between two lines are selected as main factors, and the order p of the probability density function maximum order is set to 2. (2! 3!) = 10 samples are extracted based on the equation (2) when Hermite is defined as a random density variable having a Gaussian distribution and the Hermite is defined as a probability density function. .

The simulation executing unit 120 models each of the predetermined number of samples extracted by the sample extracting unit 110 and executes electromagnetic field simulation.

The simulation execution unit 120 includes a structure modeling unit 122, a source modeling unit 124, and a simulation unit 126.

The structure modeling unit 122 extracts S parameters by modeling the transmission line structure reflecting the predefined random variables. That is, the structure modeling unit 122 reflects the random variables specified by the sample extracting unit 110 on a transmission line or a circuit, and then models the S parameters to extract S parameters of the transmission line. Here, the S parameter means the ratio of the incident voltage to the reflection voltage in the frequency distribution, and the transmission performance of the transmission line can be analyzed. That is, the insertion loss, transmission capacity, line coupling, etc. of each line can be checked through the S parameter.

The structure modeling unit 122 may use an electromagnetic field simulation S / W such as HFSS, CST, FEKO, etc. to extract the S-parameter of the transmission line.

The source modeling unit 124 constructs a circuit with S parameters extracted by the structure modeling unit 122 to extract a current source. That is, the source modeling unit 124 simulates a model having the S-parameter extracted by the structure modeling unit 122 to extract a current source at the input / output node to each transmission line. At this time, the source modeling unit 124 can extract a current source using a simulation S / W such as HSPICE, ADS, and Ansys Designer.

The simulation unit 126 applies the extracted current source from the source modeling unit 124 to simulate the electromagnetic field emission. That is, the simulation unit 126 applies the electric current source extracted by the source modeling unit 124 to the source of the electromagnetic field to be obtained from the electromagnetic field simulation S / W of HFSS, CST, FEKO, etc. to perform electromagnetic field emission simulation do.

The simulation execution unit 120 configured as described above executes simulations for each sample extracted by the sample extraction unit 110, so that it is possible to repeat the simulation for the number of samples.

The predicting unit 130 calculates the electromagnetic wave radiation predicted value of the transmission line using the simulation result of each sample. That is, the predicting unit 130 can calculate statistical indices of the electromagnetic wave radiation characteristics, that is, the average and the dispersion, using the simulation results of each sample and the conversion matrix.

The predicting unit 130 includes a coefficient calculating unit 132 and a predicted value calculating unit 134.

The coefficient calculator 132 generates a transform matrix using a polynomial corresponding to a random variable at each testing point and a simulation result, and calculates a coefficient corresponding to each degree using an inverse matrix of the transform matrix.

The key to the approach of the Polynomial Chaos framework is to find the coefficient v _k for each term in the approximation of equation (1). To obtain the coefficient v _k , a method of obtaining a testing node by a stochastic testing algorithm is used.

Thus, if an input variable having a stochastic distribution is expressed by an equation related to a standard random variable, Equation (7) described below is obtained.

Where X is the input variable, μ is the mean, and σ is the variance.

The output y of the system in which the input variables such as Equation (7) are considered is expressed as Equation (1) using the random variable as shown in Equation (8).

Equation (8) can be transformed into a multi-dimensional matrix equation (hereinafter referred to as a &quot; transformation matrix &quot;) as shown in Equation (9).

The transformation matrix may be a polynomial corresponding to a random variable at each testing point and a matrix representation of the input and output of the system.

에 대한 정규화된 직교 다항식은 아래 기재된 수학식 10 및 수학식 11과 같다. If the random variable is assumed to be a Gaussian random variable, the Hermite Polynomial can be used in the expansion formula of Equation (1) according to Table 1. One kind of random variable

The normalized orthonormal polynomials for Equation (10) and Equation (11) are as follows.

When two Gaussian random variables are included, the Hermite Polynomial used in one random variable should be used in the form of a product as shown in Equations (12) and (13) described below.

Therefore, the coefficient calculation unit 132 can generate a transform matrix using the polynomial equation corresponding to the random variable at each testing point and the simulation result, and calculate the K- th coefficient using the inverse matrix of the transform matrix .

The predictive value calculating unit 134 calculates the predicted value of the electromagnetic wave radiation using the coefficient calculated by the coefficient calculating unit 132. [ That is, the predictive value calculating unit 134 can calculate the statistical indices such as the mean and variance as the electromagnetic wave radiated predicted values by using Equation 5 and Equation 6 with the calculated k- th coefficient. Also, by using Equation 1 and the random variable, it is possible to easily calculate the statistical distribution using the equation instead of the repetitive simulation.

As described above, the predicting unit 130 may multiply the simulation result of each sample by the conversion matrix to predict the average value of the final results and the electromagnetic radiation characteristics such as the minimum / maximum boundary line.

6 is a flowchart for explaining a method of predicting the electromagnetic wave radiation according to a common mode of a circuit according to an embodiment of the present invention.

Referring to FIG. 6, the apparatus extracts a certain number of samples based on a predefined random variable (S610). At this time, the apparatus, define a type of random variable with probability density function applicable of the variables of Structural the transmission line, a predetermined number by using the Stochastic Testing algorithm specifies Testing point a predetermined number (for example, K pieces) (K pcs) .

When step S610 is performed, the apparatus models the transmission line structure reflecting the random variable to extract S parameters (S620). When the transmission line structure is modeled with random variables, the transmission line structure model as shown in (a) is simulated, and S parameters as shown in (b) are extracted through this model.

When step S620 is performed, the device constructs a circuit with a transmission line structural model to extract a current source (S630). Simulating a circuit with a transmission line composition model constitutes a circuit as in (c), and through this circuit, the current source at the input and output nodes to each transmission line can be extracted as shown in (d).

When step S630 is performed, the apparatus simulates the electromagnetic field emission by applying the extracted current source (S640). By simulating the electromagnetic field emission, the electromagnetic field intensity per frequency as shown in (e) can be output as a simulation result.

Steps S620 to S640 may be performed for each sample extracted in step S610.

When simulation for all the samples is performed, the apparatus predicts the electromagnetic wave radiation characteristic value of the transmission line using the simulation result of each sample (S650).

By using this procedure, the number of testing points is not significantly increased by the stochastic testing algorithm compared with other stochastic prediction techniques (Stochastic Galerkin method, etc.) even if the number of random variables increases by predicting the electromagnetic radiation characteristics of transmission lines. This time advantage makes it possible to predict more in a limited amount of time, which is very useful for predicting the effects of unpredictable process variations that may occur during product production or performance under various operating conditions.

Hereinafter, the effect of the electromagnetic wave radiation predicting method by the common mode of the circuit according to the present invention will be described with reference to experimental results.

FIG. 7 is a view illustrating a coplanar stripline model for predicting the electromagnetic wave radiation according to the common mode of the circuit according to the embodiment of the present invention. FIG. 8 is a graph showing the measurement result of the electromagnetic wave radiated from the circuit shown in FIG. Fig.

Referring to FIG. 7, the coplanar stripline was made of glass epoxy having a relative permittivity of 4.7, and the conductor was made of copper having a conductivity of 5.8 * 10 &lt; ^{7 &} gt ;. Here, the thickness of the conductor is 0.5 oz. The length of the coplanar stripline was 6 inches. A pulse waveform with amplitude 0-5V, frequency 10MHz, duty cycle 50%, rise time 4ns, fall time 2ns was applied to the input of the circuit, and the end of the load was opened. The width of the two lines of the coplanar stripline is 25 mils, and the spacing between the two lines (the distance from the center of one line to the center of the other) is 380 mils. In the transmission line model, the width of two lines and the distance between the two lines are set as design variables with random values within the process error during the manufacturing process. Three variables (the width of the two lines and the distance between the two lines) Simulation for predicting the common mode emission was performed by using the calculation method using deterministic parameters (hereinafter referred to as the conventional method) and the generalized Polynomial Chaos method.

안에 드는 결과 값을 보이며 측정값에 거의 일치하는 것을 알 수 있다.Referring to FIG. 8, the red line is a measurement result, the black O mark is a simulation result using a conventional method, and the blue X mark is a simulation result using a generalized polynomial Chaos method. The conventional method is difficult to predict the value of the common mode radiation when compared with the measurement result value, but the generalized polynomial Chaos method has a 3 standard deviation range

The results are shown in Fig.

9 is a diagram illustrating a microstrip coupled line model for predicting the electromagnetic wave radiation by the common mode of a circuit according to an embodiment of the present invention. FIG. 11 is a graph showing a simulation result of a common mode electromagnetic wave emission using the generalized Polynomial Chaos method in the model shown in FIG. 9. FIG.

인 구리를 사용했다. 여기서 도체의 두께는 0.5oz이다. microstrip coupled line의 길이는 150mm이며 회로의 입력단에 진폭 0-5V, 주파수 10MHz, 듀티사이클 50%, 상승 시간 4ns, 하강 시간 2ns인 펄스 파형을 인가하였고 전송선로의 종단의 저항은 50Ω이다. microstrip coupled line에는 4개의 선로가 있으며 각 선로의 폭은 4.65mm이고, 차동 쌍(differential pair)을 이루는 선로 사이의 간격은 9mm이다. 전송선로 모델에서 네 선로의 폭과 차동 쌍을 이루는 선로 사이의 간격, 그리고 기판의 높이를 제조 과정 중 공정 오차 내에서 랜덤한 수치를 가지는 설계 변수로 놓고 7가지 변수(네 선로의 폭, 차동 쌍을 이루는 선로 사이의 간격, 기판의 높이)에 대해 종래의 Monte Carlo 방법과 본 발명의 generalized Polynomial Chaos 방법을 사용하여 공통 모드 방사를 예측하기 위한 시뮬레이션을 진행하였다.Referring to FIG. 9, a microstrip coupled line has a dielectric with a substrate height of 1.57 mm and a relative dielectric constant of 2.2,

I used phosphorus. Here, the thickness of the conductor is 0.5 oz. The length of the microstrip coupled line is 150mm, and pulse waveform of amplitude 0-5V, frequency 10MHz, duty cycle 50%, rise time 4ns, fall time 2ns is applied to the input terminal of the circuit. The microstrip coupled line has four lines, the width of each line is 4.65mm, and the distance between the lines forming the differential pair is 9mm. In the transmission line model, the widths of the four lines, the spacing between the lines forming the differential pair, and the height of the substrate were set as design variables with random values within the process error during manufacturing. Seven variables (width of four lines, And the generalized polynomial chaos method of the present invention were used to perform the simulation for predicting the common mode radiation.

The simulation result of the common mode radiated emission using the Monte Carlo method is shown in FIG. At this time, random samples in the process error are 200 as shown in Table 2 below.

In this example, the simulation is performed only with 200 samples, but the higher the number of random samples, the higher the reliability.

The result of simulating the radiation of common mode radiation using the generalized Polynomial Chaos method is shown in FIG. Table 3 shows 36 samples extracted using the stochastic testing method.

개로 계산되고, 36개의 testing point로 지정된 샘플에 대해서만 전자기장 시뮬레이션을 진행한다. 이후 generalized Polynomial Chaos방법을 이용하여 36개의 결과에 Polynomial Chaos Approaches에 의해 얻어진 변환행렬(Transformation Matrix)을 역행렬 연산하면 최종 결과를 얻을 수 있다.Here, the number of random variables (d) in the 36 samples is 7, and the degree p of the highest order is 2, so that the K value ( K < N )

And the electromagnetic field simulation is performed only for the samples designated by 36 testing points. The final result can be obtained by inverting the Transformation Matrix obtained by Polynomial Chaos Approaches to 36 results using the generalized Polynomial Chaos method.

Comparing the results shown in FIG. 10 and FIG. 11, the generalized polynomial Chaos method extracts only a small number of samples as compared with the Monte Carlo method, and it can be confirmed that the two results are very similar although the simulation proceeds, Since the sample is extracted and the simulation is carried out, reliable results can be obtained even if only a small number of samples are extracted as compared with the Monte Carlo method.

The embodiments of the present invention described so far can be computer programmed and implemented in a computer. The computer may be a computing device including one or more processors and memories, and may be configured as a plurality of devices in a cloud format instead of a single device.

It is to be understood that both the foregoing general description and the following detailed description of the present invention are illustrative and explanatory only and are intended to be illustrative of the invention and are not to be construed as limiting the scope of the invention as defined by the appended claims. It is not. Therefore, those skilled in the art will appreciate that various modifications and equivalent embodiments are possible without departing from the scope of the present invention. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

100: Electromagnetic wave radiation prediction device
110:
120: Simulation execution unit
122: Structure modeling unit
124: source modeling unit
126:
130:
132: coefficient calculating section
134: prediction value calculation unit

Claims (11)

A sample extracting unit for extracting a certain number of samples based on a random variable set so that at least one element constituting the transmission line has a probabilistic distribution;
A simulation executing unit for modeling each of the extracted predetermined number of samples to execute electromagnetic field simulation; And
And a predictor for predicting an electromagnetic wave radiation characteristic value of the transmission line using the simulation result of each sample,
The simulation executing unit extracts an S parameter by modeling a transmission line structure reflecting the random variable, extracts a current source by simulating a circuit having the extracted S parameter, applies the extracted current source, Wherein the electromagnetic wave radiated by the electromagnetic wave radiated from the electromagnetic wave radiated from the electromagnetic wave radiated from the electromagnetic wave radiated from the antenna is radiated. The method according to claim 1,
The sample extracting unit extracts,
A random variable is set to have at least one element having a numerical value within the process error in the transmission line structure and a specific probability distribution of the element, a probability density function for the random variable is set, Electromagnetic radiation prediction apparatus according to the common mode of the circuit, characterized in that by specifying the testing point (testing point) of the predetermined number (K) for extracting samples of a predetermined number (K). 3. The method of claim 2,
Wherein the sample extracting unit designates a predetermined number ( K ) of testing points by the following equation: &lt; EMI ID =
[Mathematical Expression]

Where d is the number of random variables and p is the order of the highest probability density function. The method according to claim 1,
Wherein the simulation executing unit includes:
A structure modeling unit for modeling a transmission line structure in which the random variable is reflected to extract an S parameter;
A source modeling unit for extracting a current source by simulating a circuit having the extracted S parameter; And
And a simulation unit for simulating the electromagnetic field emission by applying the extracted current source. The method according to claim 1,
The predicting unit,
A coefficient calculation unit for generating a transform matrix by using a polynomial corresponding to a random variable at each testing point and a simulation result and calculating a coefficient corresponding to each degree using an inverse matrix of the transform matrix; And
And a predicted value calculation unit for calculating an estimated value of the electromagnetic wave radiation using the calculated coefficient. 6. The method of claim 5,
Wherein the coefficient calculating unit generates the conversion matrix by the following equation
[Mathematical Expression]

Here, [phi] is a polynomial corresponding to a random variable at each testing point,

Is the simulation result, and y (t) is the coefficient. A method for predicting an electromagnetic wave radiation by a common mode of a circuit,
Extracting a certain number of samples based on a random variable set so that at least one element constituting a transmission line has a probabilistic distribution;
Performing electromagnetic field simulation by modeling each of the extracted predetermined number of samples; And
And estimating an electromagnetic wave radiation characteristic value of the transmission line using the simulation result of each sample,
The step of performing the simulation includes:
The S-parameter is modeled by modeling the transmission line structure reflecting the random variable, a circuit having the extracted S-parameter is simulated to extract a current source, and the extracted current source is applied to simulate electromagnetic field emission A method of predicting electromagnetic radiation according to a common mode of a circuit to be used. 8. The method of claim 7,
The step of extracting the sample comprises:
A random variable is set to have at least one element having a numerical value within the process error in the transmission line structure and a specific probability distribution of the element, a probability density function for the random variable is set, Electromagnetic radiation prediction method according to the common mode of the circuit, characterized in that by specifying the testing point (testing point) of the predetermined number (K) for extracting samples of a predetermined number (K). 9. The method of claim 8,
Wherein the number K of test points is calculated by the following equation:
[Mathematical Expression]

Where d is the number of random variables and p is the order of the highest probability density function. delete 8. The method of claim 7,
Wherein the predicting comprises:
Generating a transform matrix by using a polynomial corresponding to a random variable at each testing point and a simulation result, and calculating a coefficient corresponding to each degree using an inverse matrix of the transform matrix; And
And calculating a predicted value of the electromagnetic wave radiation using the calculated coefficient.

Images