Automatic model parameter selection method and system based on hidden markov

Abstract

Disclosed are an automatic model parameter selection method and system based on hidden Markov. The method comprises the following steps: in response to a selection instruction input by a user in a visual operation interface, determining selected areas in the model and constructing a selected area set; constructing a hidden state set, wherein the hidden state set comprises adjustable parameters of the model; obtaining a state transition probability matrix and an observation probability matrix; and on the basis of the selected areas and the acquired state transition probability matrix and observation probability matrix, calculating an implicit state sequence by means of a Viterbi algorithm to obtain a predicted parameter combination. According to the technical solution of the present application, on the basis of a hidden Markov algorithm, expert experience is solidified in the state transition matrix and the observation probability matrix, and after a user selects some areas, a parameter combination can be automatically predicted and prompted, and the user is guided to use, thereby reducing the experience threshold for use of the user, and improving the efficiency of parameter selection of technicians.

Description

Automatic selection method and system of model parameters based on hidden Markov Technical Field

[0001] The present invention belongs to the technical field of integrated circuit computer-aided design, and in particular relates to a method and system for automatically selecting model parameters based on hidden Markov. Background Art

[0002] The continuous advancement of semiconductor and integrated circuit technology has made integrated circuit computer-aided design (CAD) or electronic design automation (EDA) platforms increasingly important. A fundamental function of EDA platforms is device model parameter extraction. This involves extracting model parameters for semiconductor devices manufactured using specific integrated circuit processes based on standard device models. Once these model parameters are extracted and combined with corresponding standard device models, the various operating characteristics of semiconductor devices can be mathematically depicted, enabling subsequent device simulation during circuit design.

[0003] The BSIM model, developed by the University of California, Berkeley, is a metal oxide field-effect transistor (MOSFET) model suitable for digital and analog circuit design and simulation. In actual parameter extraction, various BSIM models corresponding to actual MOSFET devices (such as BSIM4, BSIM-Bulk, and BSIM-CMG) can be selected to process MOSFET device test data (for example, IV curves and CV curves of MOSFETs of different sizes) to extract the model parameters of the MOSFET device.

[0004] Existing techniques for extracting parameters from various device models involve first selecting the region to be fitted, then manually entering parameters based on experience, and then modifying these parameters through an optimization algorithm to fit the selected region. However, this parameter selection method relies too much on experience and is not suitable for those with limited experience. Summary of the Invention

[0005] To solve the above problems, the purpose of the present invention is to provide a method and system for automatic selection of model parameters based on hidden Markov. The selection method and system solidify expert experience in the state transition matrix and observation probability matrix based on the hidden Markov algorithm. When the user selects certain areas, it will automatically predict and prompt parameter combinations to guide the user's use, reduce the user's experience threshold, and improve the efficiency of technical personnel in parameter selection.

[0006] The first technical solution provided in this application is: a method for automatic selection of model parameters based on hidden Markov, comprising the following steps: responding to a selection instruction input by a user in a visual operation interface to determine a selected area in the model and constructing a selected area set; constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model; obtaining a state transition probability matrix and an observation probability matrix; and using a Viterbi algorithm to calculate a hidden state sequence based on the selected area and the obtained state transition probability matrix and observation probability matrix to obtain a predicted parameter combination.

[0007] Preferably, in response to a selection instruction input by a user in the visual operation interface to determine the selected area in the model, constructing the selected area set further comprises: based on the representation graphic selected by the cursor in the current visual operation interface, obtaining the attribute information of the representation graphic; obtaining the parameter information of all points in the representation graphic in a marked state; determining the corresponding selected area in the model based on the attribute information of the representation graphic and the parameter information of all points in the representation graphic in a marked state; constructing the selected area set V = {v1, v2, ..., v m}, m represents the number of possible observations.

[0008] Preferably, constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model further comprises: constructing a hidden state set Q = {q1, q2, ..., q n}, n represents the number of possible states; the models include but are not limited to the following models: BSIM3, BSIM4, BSIM6, BSIM-CMG, BSIM-IMG, BSIMSOI, UTSOI, HiSIM2, HiSIM_HV, PSP, GP-BJT, RPITFT, each state q in the implicit state set n corresponds to an adjustable parameter in the model.

[0009] Preferably, constructing a hidden state set, wherein the hidden state set includes the adjustable parameters of the model further comprises: each state q in the hidden state set n Corresponding to one of the adjustable parameters or a combination of multiple adjustable parameters, the adjustable parameter is an adjustable parameter in the corresponding model.

[0010] Preferably, the method further comprises, after calculating the implicit state sequence using the Viterbi algorithm, performing a deduplication process on the implicit state sequence to obtain a predicted parameter combination.

[0011] Preferably, the state transition probability matrix A is: A=[a ij ] N×N , where a ij =P(it+1 =q j |i t =q i ), i=1,2,...,N; j=1,2,...,N; a ij Represents the state q at time t i State, transfer to q at time t+1 j The probability of the state.

[0012] Preferably, the observation probability matrix B is: B = [b j (k)] N×M , where b j (k)=P(o t =v k |i t =q j ), k=1,2,...,M; j=1,2,...,N; b j (k) represents the position in q at time t i State generation observation v k The probability of , also called generation probability and emission probability.

[0013] Preferably, the step of calculating the implicit state sequence using the Viterbi algorithm based on the selected region and the acquired state transition probability matrix and observation probability matrix to obtain the predicted parameter combination further includes: determining the observation sequence O=(o1, o2, ...o T ), where T is the number of selected regions; combining the state transition probability matrix A, the observation probability matrix B, and the number of selected regions T, the Viterbi algorithm is used to calculate k implicit state sequences that may produce the observation event sequence to obtain k sets of predicted parameter combinations.

[0014] Based on the same concept, the present invention also provides a model parameter automatic selection system based on hidden Markov, including: a selected area acquisition module, which is used to respond to the selection instruction input by the user in the visual operation interface to determine the selected area in the model and construct a selected area set; a parameter combination construction module, which is used to construct an implicit state set, wherein the implicit state set contains the adjustable parameters of the model; a matrix acquisition module, which is used to obtain a state transition probability matrix and an observation probability matrix; a prediction module, which is used to calculate the implicit state sequence based on the selected area and the acquired state transition probability matrix and observation probability matrix using the Viterbi algorithm to obtain a predicted parameter combination.

[0015] Based on the same concept, the present invention also provides an electronic device, characterized in that it includes: a memory, the memory is used to store a processing program; and a processor, the processor implements any one of the above-mentioned methods for automatically selecting model parameters based on hidden Markov when executing the processing program.

[0016] Based on the same concept, the present invention also provides a readable storage medium, characterized in that a processing program is stored on the readable storage medium, and when the processing program is executed by a processor, any one of the above-mentioned methods for automatically selecting model parameters based on hidden Markov is implemented.

[0017] Due to the adoption of the above technical solution, the present invention has the following advantages and positive effects compared with the prior art:

[0018] 1. The technical solution of this embodiment is based on the hidden Markov algorithm to predict the selected parameter combination according to the selected area. First, the expert experience is solidified in the state transfer matrix and the observation probability matrix. When the user selects the area, the system will automatically predict and prompt the parameter combination based on the constructed hidden Markov model and the state transfer matrix and observation probability matrix solidified by expert experience, guide the user to use, reduce the experience threshold for users, and improve the efficiency of technical personnel in parameter selection.

[0019] 2. The state transition probability matrix A and the observation probability matrix B described in the technical solution of this embodiment are means of solidifying expert experience. By solidifying expert experience and combining it with the hidden Markov algorithm, scientific prediction of parameter combinations can be achieved, and parameter combinations can be automatically predicted and prompted to guide user use, thereby lowering the user's experience threshold and improving the efficiency of technical personnel in parameter selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The specific embodiments of the present invention are further described in detail below with reference to the accompanying drawings, wherein:

[0021] FIG1 is a flow chart of a method for automatically selecting model parameters based on Hidden Markov Model of the present invention;

[0022] FIG2 is a schematic diagram of ID-Vg characterization of the present invention;

[0023] FIG3 is a schematic diagram showing Vtlin characterization of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and claims. It should be noted that the drawings are greatly simplified and not to exact ratios, and are intended solely to facilitate and clearly illustrate the embodiments of the present invention.

[0025] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0026] First embodiment

[0027] As shown in FIG1 , the technical solution provided by this embodiment of the present application is: a method for automatically selecting model parameters based on Hidden Markov Model, comprising the following steps:

[0028] S100: In response to a selection instruction input by a user in a visual operation interface, determining a selected area in the model and constructing a selected area set;

[0029] S200: Constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model;

[0030] S300: Obtaining a state transition probability matrix and an observation probability matrix;

[0031] S400: Based on the selected region and the acquired state transition probability matrix and observation probability matrix, a Viterbi algorithm is used to calculate an implicit state sequence to obtain a predicted parameter combination.

[0032] The technical solution of this embodiment is based on the hidden Markov algorithm to predict the selected parameter combination according to the selected area. First, the expert experience is solidified in the state transfer matrix and the observation probability matrix. When the user selects the area, the system will automatically predict and prompt the parameter combination based on the constructed hidden Markov model and the state transfer matrix and observation probability matrix solidified by expert experience, guide the user's use, reduce the user's experience threshold, and improve the efficiency of technical personnel in parameter selection.

[0033] Preferably, in response to a selection instruction input by a user in the visual operation interface to determine the selected area in the model, constructing the selected area set further comprises: based on the representation graphic selected by the cursor in the current visual operation interface, obtaining the attribute information of the representation graphic; obtaining the parameter information of all points in the representation graphic in a marked state; determining the corresponding selected area in the model based on the attribute information of the representation graphic and the parameter information of all points in the representation graphic in a marked state; constructing the selected area set V = {v1, v2, ..., v m}, m represents the number of possible observations.

[0034] The technical solution of this embodiment provides a method for obtaining selected areas. See Figures 2 and 3. All solid points in ID-Vg in Figure 2 represent selected areas. A point in the Vtlin kop in Figure 3 can also be a selected area. The set of all selected areas, V, is the set of visible states, where V = {v1, v2, …, vm}, where m represents the number of possible observations. The selected areas are selected by the user.

[0035] Preferably, constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model further comprises: constructing a hidden state set Q = {q1, q2, ..., q n}, n represents the number of possible states; the models include but are not limited to the following models: BSIM3, BSIM4, BSIM6, BSIM-CMG, BSIM-IMG, BSIMSOI, UTSOI, HiSIM2, HiSIM_HV, PSP, GP-BJT, RPITFT, each state q in the implicit state set n corresponds to an adjustable parameter in the model.

[0036] For example, taking the bsim6 model as an example, the model parameters include common adjustable parameters such as a0, a1, vth0, voff, nfactor, k2, cit, etc. Of course, other parameters in the model not listed here can also be included. This is only for illustrative purposes. The implicit state here is the parameter or parameter combination. For example, q1 = {a0}, q2 = {a1}, q3 = {vth0}, q4 = {voff}, etc.

[0037] Preferably, the devices used in the model of this embodiment include, but are not limited to, MOSFET transistors, silicon-on-insulator transistors (SOI), fin field-effect transistors (FinFETs), bipolar transistors (BJTs), heterojunction transistors (HBTs), thin-film transistors (TFTs), metal-semiconductor contact field-effect transistors (MESFETs), diodes, resistors, or inductors. The determined device model may be, but is not limited to, BSIM3, BSIM4, BSIM6, BSIM-CMG, BSIM-IMG, BSIMSOI, UTSOI, HiSIM2, HiSIM_HV, PSP, GP-BJT, or RPITFT. For example, for a MOSFET transistor, its corresponding device model may be BSIM3, BSIM4, BSIM6, or other known standard models or non-standard models. It will be understood that the above device models are merely exemplary, and in actual applications, a model corresponding to an integrated circuit device may be selected as needed. Based on the fact that MOSFET transistors are one of the most commonly used devices in integrated circuits, the integrated circuit device in this embodiment is described using MOSFET transistors as an example. However, those skilled in the art will appreciate that the application of this application is not limited thereto.

[0038] Preferably, constructing a hidden state set, wherein the hidden state set includes the adjustable parameters of the model further comprises: each state q in the hidden state set nCorresponding to one of the adjustable parameters or a combination of multiple adjustable parameters, the adjustable parameter is an adjustable parameter in the corresponding model.

[0039] For example, q1 = {a0}, q2 = {voff}, q3 = {a0, vth0, k2}, q4 = {voff, k2, cit}. The above adjustable parameter combination relationships can be configured in combination as needed.

[0040] Preferably, the method further comprises, after calculating the implicit state sequence using the Viterbi algorithm, performing a deduplication process on the implicit state sequence to obtain a predicted parameter combination.

[0041] In the technical solution of this embodiment, after the parameter combination is predicted, deduplication processing is performed to achieve the uniqueness of each parameter in the predicted parameter combination.

[0042] Preferably, the state transition probability matrix A is: A=[a ij ] N×N , where a ij =P(i t+1 =q j |i t =q i ), i=1,2,...,N; j=1,2,...,N; a ij Represents the state q at time t i State, transfer to q at time t+1 j The probability of the state.

[0043] The state transition probability matrix A in this embodiment is one of the means to solidify expert experience. By solidifying expert experience and combining it with the hidden Markov algorithm, scientific prediction of parameter combinations can be achieved, parameter combinations can be automatically predicted and prompted, user use can be guided, the user experience threshold can be lowered, and the efficiency of parameter selection by technical personnel can be improved.

[0044] Preferably, the observation probability matrix B is: B = [bj ( k)] N×M , where b j (k)=P(o t =v k |i t =q j ), k=1,2,...,M; j=1,2,...,N; b j (k) represents the position in q at time t i State generation observation v k The probability of , also called generation probability and emission probability.

[0045] The observation probability matrix B in this embodiment is one of the means to solidify expert experience. By solidifying expert experience and combining it with the hidden Markov algorithm, scientific prediction of parameter combinations can be achieved, parameter combinations can be automatically predicted and prompted, and user use can be guided, thereby lowering the user experience threshold and improving the efficiency of parameter selection by technical personnel.

[0046] Preferably, the step of calculating the implicit state sequence using the Viterbi algorithm based on the selected region and the acquired state transition probability matrix and observation probability matrix to obtain the predicted parameter combination further includes: determining the observation sequence O=(o1, o2, ...o T ), where T is the number of selected regions; combining the state transition probability matrix A, the observation probability matrix B, and the number of selected regions T, the Viterbi algorithm is used to calculate k implicit state sequences that may produce the observation event sequence to obtain k sets of predicted parameter combinations.

[0047] The technical solution of this embodiment specifically discloses constructing an observation sequence O based on a selected area selected by the user, and using the Viterbi algorithm to calculate the Viterbi path (i.e., k implicit state sequences) that may produce an observation event sequence based on the state transition probability matrix A, the observation probability matrix B, and the number of selected areas T. The sequences are sorted by probability, such as k1 = {q1, q3} = {a0, a0, vth0, k2}, k2 = {q3, q4} = {a0, vth0, k2, voff, k2, cit}. For the k sequences output, repeated parameters in the same sequence are excluded, i.e., k1 = {a0, vth0, k2}, k2 = {a0, vth0, k2, voff, cit}. Thus, k sets of predicted parameter combinations are obtained.

[0048] Second embodiment

[0049] Based on the same concept, the present invention also provides a model parameter automatic selection system based on hidden Markov, including: a selected area acquisition module, which is used to respond to the selection instruction input by the user in the visual operation interface to determine the selected area in the model and construct a selected area set; a parameter combination construction module, which is used to construct an implicit state set, wherein the implicit state set contains the adjustable parameters of the model; a matrix acquisition module, which is used to obtain a state transition probability matrix and an observation probability matrix; a prediction module, which is used to calculate the implicit state sequence based on the selected area and the acquired state transition probability matrix and observation probability matrix using the Viterbi algorithm to obtain a predicted parameter combination.

[0050] The technical solution of this embodiment is based on the hidden Markov algorithm to predict the selected parameter combination according to the selected area. First, the expert experience is solidified in the state transfer matrix and the observation probability matrix. When the user selects the area, the system will automatically predict and prompt the parameter combination based on the constructed hidden Markov model and the state transfer matrix and observation probability matrix solidified by expert experience, guide the user's use, reduce the user's experience threshold, and improve the efficiency of technical personnel in parameter selection.

[0051] Third embodiment

[0052] Based on the same concept, the present invention also provides an electronic device, characterized in that it includes: a memory, the memory is used to store a processing program; and a processor, the processor implements any one of the above-mentioned methods for automatically selecting model parameters based on hidden Markov when executing the processing program.

[0053] Based on the same concept, the present invention also provides a readable storage medium, characterized in that a processing program is stored on the readable storage medium, and when the processing program is executed by a processor, any one of the above-mentioned methods for automatically selecting model parameters based on hidden Markov is implemented.

[0054] If the method for automatically selecting parameters of a hidden Markov model is implemented in the form of program instructions and sold or used as an independent product, it can be stored in a computer-readable storage medium and perform the following steps:

[0055] S100: In response to a selection instruction input by a user in a visual operation interface, determining a selected area in the model and constructing a selected area set;

[0056] S200: Constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model;

[0057] S300: Obtaining a state transition probability matrix and an observation probability matrix;

[0058] S400: Based on the selected region and the acquired state transition probability matrix and observation probability matrix, a Viterbi algorithm is used to calculate an implicit state sequence to obtain a predicted parameter combination.

[0059] Based on this understanding, the technical solution of this embodiment, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of software. The computer software is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0060] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the identification content specifically executed by the above-described system and device can refer to the corresponding process in the aforementioned method embodiment.

[0061] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the scope of protection of the present invention.

Claims

1. A method for automatically selecting model parameters based on hidden Markov, characterized in that: The following steps are involved: In response to a selection instruction input by a user in the visual operation interface, determining a selected area in the model, and constructing a selected area set; Constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model; Obtain state transition probability matrix and observation probability matrix; Based on the selected region and the acquired state transition probability matrix and observation probability matrix, a Viterbi algorithm is used to calculate an implicit state sequence to obtain a predicted parameter combination.

2. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 1, characterized in that: In response to a selection instruction input by a user in the visual operation interface, determining a selected area in the model, constructing a selected area set further includes: Based on the representation graphic selected by the cursor in the current visual operation interface, obtaining attribute information of the representation graphic; Get parameter information of all points in the marked state in the representation graph; Determining a corresponding selected area in the model based on the attribute information representing the graphic and the parameter information of all points in the marked state in the graphic representing the graphic; Construct the selected region set V = {v1,v2,…,v m }, m represents the possible number of observations.

3. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 1, characterized in that: Constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model further comprising: Construct the implicit state set Q = {q1,q2,…,q n }, n represents the number of possible states; The models include but are not limited to the following models: BSIM3, BSIM4, BSIM6, BSIM-CMG, BSIM-IMG, BSIMSOI, UTSOI, HiSIM2, HiSIM_HV, PSP, GP-BJT, RPITFT, each state q in the implicit state set n corresponds to an adjustable parameter in the model.

4. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 1 or 3, characterized in that: Constructing a hidden state set, wherein the hidden state set includes adjustable parameters of the model and further comprises: Each state q in the implicit state set n Corresponding to one of the adjustable parameters or a combination of multiple adjustable parameters, the adjustable parameter is an adjustable parameter in the corresponding model.

5. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 1, characterized in that: The method further includes, after calculating the implicit state sequence using the Viterbi algorithm, performing a deduplication process on the implicit state sequence to obtain a predicted parameter combination.

6. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 1, characterized in that: The state transition probability matrix A is: A = [a ij ] N×N , Among them, a ij =P(i t+1 =q j |i t =q i ), i=1,2,…,N; j=1,2,…,N; a ij represents the state q at time t i State, transfer to q at time t+1 j The probability of the state.

7. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 6, characterized in that: The observation probability matrix B is: B = [b j (k)] N×M , Among them, b j (k) = P(o t =v k |i t =q j ), k=1,2,…,M; j=1,2,…,N; b j (k) represents the state of q at time t i State generation observation v k The probability of a given event is also called the generation probability and emission probability.

8. The method for automatically selecting model parameters based on Hidden Markov Model according to claim 7, characterized in that: Calculating the implicit state sequence using the Viterbi algorithm based on the selected region and the acquired state transition probability matrix and observation probability matrix to obtain the predicted parameter combination further includes: Based on the selected area, an observation sequence O=(o1, o2, ... o T ), where T is the number of selected regions; The state transition probability matrix A, the observation probability matrix B, and the number T of selected regions are combined to calculate k implicit state sequences that may generate observation event sequences using the Viterbi algorithm to obtain k sets of predicted parameter combinations.

9. A system for automatically selecting model parameters based on hidden Markov, characterized in that: include: A selected area acquisition module, used to determine the selected area in the model in response to a selection instruction input by a user in the visual operation interface, and to construct a selected area set; A parameter combination construction module, used to construct a hidden state set, wherein the hidden state set includes adjustable parameters of the model; Matrix acquisition module, used to obtain state transition probability matrix and observation probability matrix; The prediction module is used to calculate the implicit state sequence using the Viterbi algorithm based on the selected area and the acquired state transition probability matrix and observation probability matrix to obtain a predicted parameter combination.

10. An electronic device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the method for automatically selecting model parameters based on hidden Markov as described in any one of claims 1 to 8.

Images