Embodiment
Consulting shown in Fig. 1, is the applied environment figure of scattering parameter of the present invention (scattering parameters is called for short S parameter) passivity analysis system 31 (hereinafter to be referred as system 31) preferred embodiment.This system 31 is applied to calculation element 30.This calculation element 30 also comprises processor 33 and storer 34.As shown in Figure 1, this calculation element 30 is connected with measuring instrument 20, and this measuring instrument 20, for the signal of transmission in measurement circuit 10, obtains scattering parameter (scattering parameters, be called for short S parameter) file 32, and this S Parameter File 32 is stored in to storer 34.
This system 31 is for producing the Rational Matrix of S parameter to this S Parameter File 32, the Rational Matrix of S parameter is converted to state space matrices, and bring this state space matrices into Hamiltonian matrix and analyze, according to the eigenwert of Hamiltonian matrix, whether comprise pure imaginary number value judges whether the Rational Matrix of S parameter meets passivity, thereby whether the design of decision circuitry 10 meets passivity requirement.
Described storer 34 is also for the sequencing code of storage system 31.Processor 33 is carried out described sequencing code, and the above-mentioned functions of system 31 is provided.
In the present embodiment, the passivity circuit that this circuit 10 is design.For example, this circuit 10 may comprise difference transmission lines L1 and L2 as shown in Figure 4.This measuring instrument 20 is network analyzer.Described calculation element 30 can be PC, notebook, and server, workstation, or other has the electronic installation of data processing function.
Consulting shown in Fig. 2, is the functional block diagram of system 31.This system 31 comprises parameter read module 311, Vector Quasi compound module 312, matrix conversion module 313 and passivity analysis module 314.Function below in conjunction with the functional module of the method flow illustrative system 31 shown in Fig. 3.
As shown in Figure 3, be the process flow diagram of S parameter passivity analysis method of the present invention preferred embodiment.
Step S301, parameter read module 311 reads S Parameter File 32, and this original S Parameter File 32 is included in the S parameter value measuring from circuit 10 each ports under different frequency.For example, suppose that circuit 10 comprises difference transmission lines L1 and L2 as shown in Figure 4, this difference transmission lines L1 and L2 comprise four port ones, 2,3,4.When from a signal of port one transmitting, can reflect a signal from port one simultaneously, port 2,3,4 all can receive a signal, and the signal that port one reflects is denoted as S11 with the ratio of the signal that port one is launched, and is called again reflection parameters; The signal that port 2 receives is denoted as S12 with the ratio of the signal that port one is launched, and is called again insertion loss parameter; The signal that port 3 receives is denoted as S13 with the ratio of the signal that port one is launched, and is called again near-end cross parameter; The signal that port 4 receives is denoted as S14 with the ratio of the signal that port one is launched, and is called again far-end cross talk parameter.All these parameters include amplitude and phase place, and amplitude curve can change along with the variation of the frequency transmitting.Therefore, in the present embodiment, S parameter comprises: reflection parameters, insertion loss parameter, near-end cross parameter and far-end cross talk parameter etc.
In general, the rational function of the S parameter that vector matching produces has (1a) and (1b) two kinds of forms, wherein (1a) is common extreme value (common-pole) form, is (1b) non-common extreme value (non-common-pole) form:
Wherein, M represents control accuracy, and N represents port number, r
mrepresent residual value, p
mrepresent extreme value, s=ω=2 π f represents angular frequency, d
mfor constant.
As shown in Figure 5, be the S parameter of four ports through 8 the vectorial fitting result to (M=8) extreme value-residual value.Transverse axis in Fig. 5 represents frequency, and the longitudinal axis represents amplitude.By Fig. 5, can be learnt, use the rational function curve of the common extreme value form (By Common pole) of the S parameter that vectorial matching produces to depart from the original amplitude curve of S parameter far away, and use the rational function of the non-common extreme value form (By Non-common pole) of the S parameter that vectorial matching produces more to approach the original amplitude curve (Original System) of S parameter.That is to say, use the rational function error of common extreme value form of the S parameter that vectorial matching produces larger, use the rational function precision of non-common extreme value form of the S parameter that vectorial matching produces higher.Therefore the rational function that, the present invention is directed to non-common extreme value form is analyzed.
Step S303, Vector Quasi compound module 312 carries out respectively to each S parameter the Rational Matrix that vectorial matching produces the non-common extreme value form of S parameter, produces the equivalent electrical circuit 315 of circuit 10 according to this Rational Matrix.
Shown in (2), be that S parameter obtains the Rational Matrix of M to the non-common extreme value form of extreme value-residual value through vectorial matching as follows:
Wherein,
represent extreme value, its real number (real) form as shown in (3a),
represent residual value, its plural number (complex) form is as shown in (3b):
Wherein,
and
for real number.
Wherein,
u+2V=M, and
Step S305, matrix conversion module 313 is converted to state space matrices by the Rational Matrix of the non-common extreme value form of S parameter.For example, matrix conversion module 313 is converted to the state space matrices shown in following formula (4) by the Rational Matrix shown in formula (2) in conjunction with (3a) and expression way (3b):
jωX(jω)=AX(jω)+BU(jω) (4)
Y(jω)=CX(jω+DU(jω),
Wherein, A, B, C and D matrix represent with formula (5) respectively:
C=[C
r C
c],
A in formula (5)
r, B
rand C
rspace matrix for the real number form of extreme value-residual value in formula (3a), can represent by formula (6):
A wherein
rfor (NNU) * (NNU), B
rfor (NU) * N, C
rfor the sparse matrix of N * (NU), o.w represents other condition (otherwise).
A in formula (5)
c, B
cand C
cspace matrix for the plural form of extreme value-residual value in formula (3b), can represent by formula (7):
Wherein, (p-1)+2V (the q-1)+2v of Ψ (p, q, v)=(2VN), A
cfor (NN2V) * (NN2V), B
rfor (N2V) * N, C
rfor the sparse matrix of N * (N2V).
Matrix conversion module 313, in conjunction with above-mentioned formula (5), (6), (7), can obtain the A in the state space matrices shown in formula (4), B, C, the expression formula of D.
Step S307, passivity analysis module 314 is state space matrices substitution Hamilton (Hamiltonian) matrix being converted to, the Hamiltonian matrix H as shown in (8):
Wherein, R=D
td-IQ=DD
t-I, I is unit matrix.
Step S309,314 pairs of Hamiltonian matrixes of passivity analysis module are analyzed, and according to the eigenwert of Hamiltonian matrix, whether comprise that pure imaginary number value judges whether the Rational Matrix of the non-common extreme value form of S parameter meets passivity.
For example, if the Hamiltonian matrix H eigenwert of passivity analysis module 314 judgements as shown in (8) do not comprise pure imaginary number value, the Rational Matrix of the non-common extreme value form of the S parameter shown in passivity analysis module 314 judgment formulas (2) meets passivity.Otherwise, if the Hamiltonian matrix H eigenwert of passivity analysis module 314 judgements as shown in (8) comprises pure imaginary number value, the Rational Matrix of the non-common extreme value form of the S parameter shown in passivity analysis module 314 judgment formulas (2) does not meet passivity.
As shown in Figure 6, for four port S parameters shown in Fig. 4 obtain after the Rational Matrix of the non-common extreme value form as shown in formula (2) the vectorial matching of (M=8) extreme value-residual value through 8, in conjunction with substitution formula (8) after formula (3) to (7), the pure imaginary number eigenwert of resulting Hamiltonian matrix.Wherein,
Step S311, whether passivity analysis module 314 meets passivity according to the Rational Matrix of the non-common extreme value form of S parameter and judges whether equivalent electrical circuit 315 meets passivity requirement.
For example, if the Rational Matrix of the non-common extreme value form of S parameter does not meet passivity, passivity analysis module 314 judgement equivalent electrical circuit 315 do not meet passivity requirement, and the vectorial fitting result of Vector Quasi compound module 312 needs to adjust.If the Rational Matrix of the non-common extreme value form of S parameter meets passivity, equivalent electrical circuit 315 meets passivity requirement, can be applied to subsequent conditioning circuit sunykatuib analysis, for example, be applied to circuit simulation software Hspice and carry out time-domain analysis.