A standard cell switching current automatic fitting method and system based on a mathematical model

Abstract

The present application relates to a kind of standard cell switch current automation fitting method and system based on mathematical model, the present application analyzes the simulation result of mainstream EDA simulation tool for transistor level standard cell switch current, determines the relationship of current variation key node and input pin conversion time and load capacitance, and uses a general current formula to express simulation result, provides a kind of standard cell switch current modeling method that can reach higher accuracy and also has efficiency for power consumption simulation core IR Drop simulation.The present application uses mathematical function segmented fitting simulation result of mature EDA tool to transistor level standard cell switch current, improves the accuracy and speed of subsequent power consumption analysis and IR Drop analysis.

Description

Technical Field

[0001] This invention relates to an automated fitting method and system for the switching current of standard cells in digital integrated circuits, belonging to the technical field of EDA algorithms and device models. Background Technology

[0002] Power consumption analysis and IR drop analysis are crucial steps in the back-end work of digital integrated circuit design. Power consumption analysis includes analyzing the circuit's static power consumption, internal power consumption, and switching power consumption, while IR drop analysis checks whether standard cells in the circuit can obtain normal supply voltage. Excessive power consumption can lead to increased chip temperature, reducing chip stability and lifespan; excessive IR drop can cause timing violations and chip failure. Therefore, power consumption analysis and IR drop analysis are indispensable in circuit design.

[0003] Mainstream signoff tools support both static and dynamic simulation analysis modes. In static simulation mode, the user specifies parameters such as the toggle rate of the chip's input pins and the chip's operating frequency. The EDA analysis tool calculates the toggle rate of each instance based on a propagation algorithm and reads power consumption information from a timing library file. Based on the power consumption information, it calculates the average current of each instance and treats it as a constant current source, replacing the instances in the power network. Using Kirchhoff's voltage and current laws, it establishes the matrix equations of the power network and solves for the node voltages to obtain the steady-state IR drop of the entire chip's power network. In dynamic simulation mode, the EDA analysis tool needs to match the toggle situation of each instance to the simulation waveform file generated by the front-end design and treat each instance as a current source, replacing the instances in the power network. It then establishes the matrix equations for the power network and solves for the transient voltage changes at the nodes to obtain the dynamic IR drop of the entire chip's power network. Therefore, in dynamic simulation mode, the accuracy of the analysis tool's equivalent current source for each instance is crucial. The closer the current waveform provided by the analysis tool is to the actual current of the instance, the better the final IR Drop simulation result will reflect the actual IR Drop of the chip. This allows chip designers to detect more potential problems in the early stages of the design process.

[0004] Currently, in dynamic simulation, EDA analysis tools first calculate the energy generated by each instance during signal transitions based on information in the simulation waveform file, and then evenly distribute the energy over a period of time. During the time when the instance signal does not transition, the power consumption waveform only records leakage power, which is relatively low. During the time when the instance signal transitions, the power consumption waveform records not only leakage power but also internal power and switching power, resulting in higher values. Therefore, the power consumption waveform obtained through this equivalent simulation differs from the power consumption waveform under actual instance transitions, making it difficult for the final simulation results to reflect the chip's true dynamic IR drop. While transistor-level simulation in dynamic simulation can provide high-precision results, it significantly increases simulation time. Therefore, a standard cell switching current calculation method is needed to make the voltage waveform on the power network obtained during dynamic IR drop calculation more closely resemble the chip's actual operating conditions without increasing time costs, thereby improving the accuracy of the final simulation results. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes an automated fitting method for standard cell switching current based on a mathematical model. This invention analyzes and models the simulation results of standard cell switching current at the transistor level using mainstream EDA analysis tools. A general current formula with time, input pin transition time, and load capacitance as independent variables is used to express the simulation results. The signoff tool can directly call the current formula of the standard cell to obtain the magnitude of the switching current during dynamic simulation.

[0006] This invention also proposes an automated fitting system for standard unit switch current based on a mathematical model.

[0007] Terminology Explanation:

[0008] 1. EDA: Electronic Design Automation.

[0009] 2. Signoff: In the back-end of digital integrated circuit design, signoff refers to the process of reviewing the design data before it is handed over to the chip manufacturer for production, to confirm that the design data meets the delivery standards. These checks and confirmations are collectively referred to as signoff.

[0010] 3. Input pin transition time: The time it takes for an input pin to rise from a low level to a high level or fall from a high level to a low level.

[0011] 4. Output load capacitance: The capacitor driven by the output pin, which usually consists of three parts: the parasitic capacitance of the output pin of the standard unit, the parasitic capacitance of the network connected to the output pin, and the parasitic capacitance of the input pin of the driven gate unit.

[0012] 5. SPICE Simulator: SPICE (Simulation program with integrated circuit emphasis) is a powerful analog circuit simulator.

[0013] 6. IR Drop: IR drop refers to the phenomenon of voltage drop and rise on the power and ground networks in integrated circuits. With the continuous evolution of semiconductor technology, the width of metal interconnects is becoming narrower and the resistance value is becoming larger (while the supply voltage is becoming smaller), and the effect of IR drop is becoming more and more obvious.

[0014] 7. Static power consumption: also known as leakage power consumption, it refers to the power consumption generated by the leakage current of transistors in a circuit.

[0015] 8. Internal power consumption: also known as short-circuit power consumption, it refers to the power consumption generated by the short-circuit current from the power supply to ground flowing through the standard cell when PMOS and NMOS are turned on at the same time.

[0016] 9. Switching power consumption: The power consumption generated by the circuit charging and discharging the load capacitor of the output node during the switching process.

[0017] 10. Flip-over rate: The average number of times a gate cell output pin changes from high to low and from low to high within one clock cycle.

[0018] 11. Transient Simulation: This is a method of simulating the changes in voltage and current of various components in a circuit over time.

[0019] 12. Parametric Analysis: This is a method in circuit simulation used to study the impact of one or more parameters on changes in circuit performance.

[0020] The technical solution of this invention is as follows:

[0021] An automated fitting method for standard unit switch current based on a mathematical model includes:

[0022] Two sets of standard unit simulation circuits were built, and the input pin transition time and output load capacitance were selected as circuit parameters respectively. Transient simulation and parameter analysis were performed on the standard unit simulation circuits to obtain the power supply terminal switching current curve data with input pin transition time and time as independent variables and output load capacitance and time as independent variables.

[0023] Determine the corresponding curve model based on the changing trend of the power supply terminal switching current curve data;

[0024] Based on the influence of input pin switching time and output load capacitance on current, the power supply switching current curve is divided into two parts; the current change of the first part of the curve is affected by the input pin switching time, and the independent variables of the curve function for this part are time and input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, and the independent variables of the curve function for this part are time and output load capacitance.

[0025] Read the power supply switching current curve data with time and input pin switching time as independent variables, and with time and output load capacitance as independent variables;

[0026] The current curve data is piecewise fitted using a polynomial function to obtain a current curve function with time as the independent variable, and the input pin switching time and output load capacitance as constants.

[0027] A linear function is used to fit the relationship between the piecewise time value and the input pin switching time and the output load capacitance, and an exponential function is used to fit the relationship between the piecewise current value and the input pin switching time and the output load capacitance.

[0028] Based on the relationship between the fitted segmented current values, input pin switching time, and output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin switching time, and output load capacitance as independent variables, thereby achieving automated fitting of the standard unit switching current.

[0029] According to a preferred embodiment of the present invention, the curve model includes four types. In the first curve model, the current value first gradually increases with time, reaches its maximum value when the time reaches the input pin switching time, and then gradually decreases, with only one maximum value. In the second curve model, the current value first decreases, then gradually increases to reach its maximum value, with only one maximum value. In the third curve model, the current first gradually increases with time, reaches its first maximum value when the time reaches the input pin switching time, then gradually decreases and then increases to the second maximum value, and then gradually decreases again, with a total of two maximum values. In the fourth curve model, the current first decreases, then gradually increases to the first maximum value, then gradually decreases and then increases to the second maximum value, and then gradually decreases again, with a total of two maximum values.

[0030] According to a preferred embodiment of the present invention, the current curve function is as shown in equation (1):

[0031]

[0032] Where t is time, i is the current magnitude, and f n(t) is the current fitting expression for the nth segment, transition_time is the input pin transition time, c_load is the output load capacitance, tr1 and c1 are constants, and t1…t n The time values ​​at the segmentation points of the piecewise function.

[0033] According to a preferred embodiment of the present invention, a linear function is used to fit the relationship between the piecewise time value and the input pin switching time and the output load capacitance, and an exponential function is used to fit the relationship between the piecewise current value and the input pin switching time and the output load capacitance; as shown in equations (2) and (3):

[0034]

[0035]

[0036] Where transition_time is the input pin transition time, c_load is the output load capacitance, and t i t j Let i be the time value at the piecewise point of the piecewise function. i i j Let m be the current value at the piecewise point of the piecewise function. i m j n i n j a i a j b i b j c i c j d i d j These are constants determined during fitting.

[0037] According to a preferred embodiment of the present invention, based on the relationship between the fitted segmented current values ​​and the input pin switching time and the output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin switching time and output load capacitance as independent variables, as shown in equation (4):

[0038]

[0039] Here, k represents the time scaling ratio resulting from the different input pin switching times and output load capacitances, which lead to different time nodes in each stage.

[0040] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of an automated fitting method for standard unit switching current based on a mathematical model.

[0041] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of an automated fitting method for standard unit switching currents based on a mathematical model.

[0042] An automated fitting system for standard unit switch current based on a mathematical model includes:

[0043] The standard cell simulation circuit building module is configured to: build two sets of standard cell simulation circuits, selecting the input pin transition time and output load capacitance as circuit parameters respectively; perform transient simulation and parameter analysis on the standard cell simulation circuits to obtain power supply terminal switching current curve data with input pin transition time and time, and output load capacitance and time as independent variables;

[0044] The curve model determination module is configured to: determine the corresponding curve model based on the changing trend of the power supply switching current curve data; and divide the power supply switching current curve into two parts based on the influence of the input pin switching time and the output load capacitance on the current; the current change of the first part of the curve is affected by the input pin switching time, and the independent variables of the curve function of this part are time and the input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, and the independent variables of the curve function of this part are time and the output load capacitance.

[0045] The curve data reading module is configured to read power supply switching current curve data with time and input pin switching time as independent variables, and with time and output load capacitance as independent variables.

[0046] The standard unit switch current automatic fitting module is configured to: use a polynomial function to piecewise fit one of the current curve data to obtain a current curve function with time as the independent variable and input pin transition time and output load capacitance as constants; use a linear function to fit the relationship between the piecewise point time value and the input pin transition time and output load capacitance, and use an exponential function to fit the relationship between the piecewise point current value and the input pin transition time and output load capacitance; based on the fitted piecewise point current value and the relationship between the input pin transition time and output load capacitance, adjust the piecewise nodes in the current curve function, and compress or stretch the current curve function to obtain a current curve function with time, input pin transition time and output load capacitance as independent variables, thereby realizing the automatic fitting of the standard unit switch current.

[0047] The beneficial effects of this invention are as follows:

[0048] This invention proposes an automated method for fitting standard cell switching current based on a mathematical model. It uses mathematical functions to piecewise fit the simulation results of standard cell switching current transistor level using mature EDA tools, thereby improving the accuracy and speed of subsequent power consumption analysis and IR drop analysis. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of a curve model for an automated fitting method for standard unit switch current based on a mathematical model.

[0050] Figure 2 This is a schematic diagram of curve model two for an automated fitting method of standard unit switch current based on a mathematical model.

[0051] Figure 3 This is a schematic diagram of a curve model for an automated fitting method for standard unit switch current based on a mathematical model.

[0052] Figure 4 This is a schematic diagram of a curve model for an automated fitting method of standard unit switch current based on a mathematical model.

[0053] Figure 5 This is a schematic diagram of an automated fitting process for the standard unit switch current based on a mathematical model.

[0054] Figure 6 This is a circuit diagram of the CKINVM8T simulation circuit in the embodiment.

[0055] Figure 7 This is a schematic diagram of the current segmentation and fitting of the CKINVM8T in the embodiment.

[0056] Figure 8(a) is a schematic diagram of fitting the first segment time of the CKINVM8T current with the input pin conversion time in the embodiment.

[0057] Figure 8(b) is a schematic diagram of the fitting of the second segment time of the CKINVM8T current with the input pin conversion time in the embodiment.

[0058] Figure 9(a) is a schematic diagram of the fitting of the current at the first segment point of the CKINVM8T current and the input pin switching time in the embodiment.

[0059] Figure 9(b) is a schematic diagram of the fitting of the current at the second segment point of the CKINVM8T current and the input pin switching time in the embodiment.

[0060] Figure 10(a) is a schematic diagram of the fitting of the third segment point of the CKINVM8T current with the output load in the embodiment.

[0061] Figure 10(b) is a schematic diagram of the fitting of the current at the third segment point of the CKINVM8T current and the output load in the embodiment.

[0062] Figure 11(a) is a schematic diagram of the fitting results when the input pin switching time of CKBUFM12T is 10ps in the embodiment.

[0063] Figure 11(b) is a schematic diagram of the fitting results when the input pin switching time of CKBUFM12T is 30ps in the embodiment.

[0064] Figure 11(c) is a schematic diagram of the fitting results when the input pin switching time of CKBUFM12T is 50ps in the embodiment.

[0065] Figure 11(d) is a schematic diagram of the fitting results when the input pin switching time of CKBUFM12T is 80ps in the embodiment. Detailed Implementation

[0066] The present invention will be further described below with reference to the accompanying drawings and embodiments, but is not limited thereto.

[0067] Example 1

[0068] An automated fitting method for standard cell switching current based on a mathematical model is proposed. This method analyzes the simulation results of standard cell switching current at the transistor level using mainstream EDA simulation tools, determines the relationship between key nodes of current change and input pin switching time and load capacitance, and expresses the simulation results using a general current formula. Figure 5 As shown, it includes:

[0069] Two sets of standard cell simulation circuits were built, with the input pin transition time and output load capacitance selected as circuit parameters respectively. Using mainstream EDA simulation tools, transient simulation and parameter analysis were performed on the standard cell simulation circuits to obtain power supply switching current curve data with input pin transition time and time as independent variables and output load capacitance and time as independent variables.

[0070] Determine the corresponding curve model based on the changing trend of the power supply terminal switching current curve data;

[0071] Based on the influence of input pin switching time and output load capacitance on the current, the power supply switching current curve is divided into two parts; the current change of the first part of the curve is affected by the input pin switching time, therefore, the independent variables of the curve function for this part are time and input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, therefore, the independent variables of the curve function for this part are time and output load capacitance.

[0072] Use a C language program to read the power supply terminal switching current curve data with time and input pin switching time as independent variables, and time and output load capacitance as independent variables;

[0073] The current curve data is piecewise fitted using a polynomial function to obtain a current curve function with time as the independent variable, and the input pin switching time and output load capacitance as constants.

[0074] A linear function is used to fit the relationship between the piecewise time value and the input pin switching time and the output load capacitance, and an exponential function is used to fit the relationship between the piecewise current value and the input pin switching time and the output load capacitance.

[0075] Based on the relationship between the fitted segmented current values, input pin switching time, and output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin switching time, and output load capacitance as independent variables, thereby achieving automated fitting of the standard unit switching current.

[0076] Summarizing a large number of simulation results, the curves can be roughly divided into four types, such as... Figure 1 , Figure 2 , Figure 3 , Figure 4 As shown, there are four curve models. In the first curve model, the current value gradually increases with time, reaches its maximum value when the input pin transitions, and then gradually decreases, with only one maximum value. In the second curve model, the current value first decreases, then gradually increases to its maximum value, with only one maximum value. In the third curve model, the current first gradually increases with time, reaches its first maximum value when the input pin transitions, then gradually decreases and then increases to its second maximum value, and then gradually decreases again, with a total of two maximum values. In the fourth curve model, the current first decreases, then gradually increases to its first maximum value, then gradually decreases and then increases to its second maximum value, and then gradually decreases again, with a total of two maximum values.

[0077] The current curve function is shown in equation (1):

[0078]

[0079] Where t is time, i is the current magnitude, and f n (t) is the current fitting expression for the nth segment, transition_time is the input pin transition time, c_load is the output load capacitance, tr1 and c1 are constants, and t1…t n The time values ​​at the segmentation points of the piecewise function.

[0080] The relationship between the piecewise time value and the input pin switching time and the output load capacitance is fitted using a linear function, and the relationship between the piecewise current value and the input pin switching time and the output load capacitance is fitted using an exponential function; as shown in equations (2) and (3):

[0081]

[0082]

[0083] Where transition_time is the input pin transition time, c_load is the output load capacitance, and t i t j Let i be the time value at the piecewise point of the piecewise function. i i j Let m be the current value at the piecewise point of the piecewise function. i m j n i n j a i a j b i b j c i c j d i d j These are constants determined during fitting.

[0084] Based on the fitted segmented current values ​​and the relationship between the input pin switching time and the output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin switching time and output load capacitance as independent variables, as shown in equation (4):

[0085]

[0086] Here, k represents the time scaling ratio resulting from the different input pin switching times and output load capacitances, which lead to different time nodes in each stage.

[0087] Example 2

[0088] The difference between the standard unit switch current automatic fitting method based on a mathematical model described in Example 1 and the method described in Example 1 is as follows:

[0089] Step 1: Use simulation software to build a CKINVM8T simulation circuit based on the standard cell technology of 55nm. The circuit diagram of the CKINVM8T simulation circuit is shown below. Figure 6 As shown, parameter analysis is performed using the input pin switching time as a parameter. The input pin switching time parameters are 10ps, 20ps, 30ps, 40ps, 50ps, 60ps, 70ps, 80ps, 90ps, and 100ps, and the output load capacitance is 10fF. The power supply switching current curve data with the input pin switching time and time as independent variables are obtained. Parameter analysis is also performed using the output load capacitance as a parameter. The output load capacitance parameters are 2fF, 4fF, 6fF, 8fF, 10fF, 12fF, 14fF, 16fF, 18fF, and 20fF, and the input pin switching time is 50ps. The power supply switching current curve data with the output load capacitance and time as independent variables are obtained.

[0090] Step 2: Determine the corresponding curve model according to the variation trend of the curve, as Figure 1 shown. The current curve is divided into two parts with the maximum current as the boundary according to the influencing factors. For the first part, the independent variables are time and the transition time of the input pin, and a two-segment function is used for fitting, which are respectively the current from the start to the current turning point and the current from the current turning point to the maximum value; for the second part, the independent variables are time and the output load capacitance, where the current is the current when it drops from the maximum value to the order of 1μA.

[0091] Step 3: Use a C language program to read the power supply terminal switching current curve data taking time and input pin transition time, as well as time and output load capacitance as independent variables;

[0092] Step 4: Use piecewise polynomial to fit the current curve data with the input pin transition time of 50ps and the output load capacitance of 10fF, so as to obtain a piecewise current curve function where the independent variable is time, and the input pin transition time and output load capacitance are constants.

[0093] Wherein, the first stage is: when 0<t≤t1, i1=f1(t); when t1<t≤t2, i2=f2(t);

[0094] The second stage is: when t2<t≤t3, i3=f3(t);

[0095] The fitting result is as Figure 7 shown.

[0096] Step 5: Use a linear function to fit the relationship between the first time point of the piecewise function and the second time point of the piecewise function, which take the input pin transition time as the independent variable. Use an exponential function to fit the relationship between the current value of the first segment of the piecewise function and the current value of the second segment of the piecewise function, which take the input pin transition time as the independent variable, where the output load capacitance value is a constant of 10fF.

[0097]

[0098]

[0099] Figure 8(a) is a schematic diagram of fitting the time of the first current segment point of CKINVM8T and the input pin transition time. Figure 8(b) is a schematic diagram of fitting the time of the second current segment point of CKINVM8T and the input pin transition time. Figure 9(a) is a schematic diagram of fitting the current of the first current segment point of CKINVM8T and the input pin transition time. Figure 9(b) is a schematic diagram of fitting the current of the second current segment point of CKINVM8T and the input pin transition time.

[0100] Using a linear function to fit the output load capacitance, the current-time relationship in the third segment is shown. At this point, the input pin switching time is a constant 50ps.

[0101] t3 = m3·c_load + n3

[0102]

[0103] Figure 10(a) shows the fitting diagram of the time and output load at the third segment point of the CKINVM8T current. Figure 10(b) shows the fitting diagram of the current and output load at the third segment point of the CKINVM8T current.

[0104] Step 6: Adjust the segmented nodes in the current curve function obtained in Step 4 according to the relationship between the fitted nodes and the input pin switching time and the output load capacitance in Step 5. Then, compress or stretch the current polynomial function obtained in Step 4 according to the relationship between the input pin switching time and the output load capacitance to obtain the current curve function with time, input pin switching time and output load capacitance as independent variables, as shown below:

[0105]

[0106] Where, k1 = t1 / t 1|transitiob_time＝50ps k2 = t2 / t 2|transition_time＝50ps k3 = t3 / t 3|c_load＝10fF .

[0107] Example 3

[0108] The difference between the standard unit switch current automatic fitting method based on a mathematical model described in Example 1 and the method described in Example 1 is as follows:

[0109] Step 1: Use simulation software to build a 55nm standard cell CKBUFM12T simulation circuit. Perform parameter analysis using the input pin transition time as a parameter. The input pin transition time parameters are 10ps, 20ps, 30ps, 40ps, 50ps, 60ps, 70ps, 80ps, 90ps, and 100ps, and the output load capacitance is 10fF. Obtain the power supply switching current curve data with the input pin transition time and time as independent variables. Perform parameter analysis using the output load capacitance as a parameter. The output load capacitance parameters are 2fF, 4fF, 6fF, 8fF, 10fF, 12fF, 14fF, 16fF, 18fF, and 20fF, and the input pin transition time is 50ps. Obtain the power supply switching current curve data with the output load capacitance and time as independent variables.

[0110] Step 2: Determine the corresponding curve model based on the curve's changing trend. Figure 2as shown. According to the influencing factors, the current curve is divided into two parts with the maximum current as the boundary. The independent variables of the first part are time and the transition time of the input pin, which is fitted by two segments of functions, respectively the current from the start to the minimum current and the current from the minimum current to the boundary where the current is affected by the transition time of the input pin; the independent variables of the second part are time and the output load capacitance, which is fitted by two segments of functions, respectively the current from the boundary where the current is affected by the transition time of the input pin to the maximum current and the current from the maximum current to the order of 1 μA.

[0111] Step 3: Use a C language program to read the power supply terminal switching current curve data with time and input pin transition time, as well as time and output load capacitance as independent variables;

[0112] Step 4: Use piecewise polynomial to fit the current curve data with the input pin transition time of 50ps and the output load capacitance of 10fF, so as to obtain a piecewise current curve function with time as the independent variable, and the input pin transition time and output load capacitance as constants,

[0113] wherein the first stage is: when 0<t≤t1, i1=f1(t); when t1<t≤t2, i2=f2(t);

[0114] The second stage is: when t2<t≤t3, i3=f3(t); when t3<t≤t A , i A =f A (t);

[0115] Step 5: Use a linear function to fit the relationship between the first time point of the piecewise function with the input pin transition time as the independent variable and the second time point of the piecewise function, and use an exponential function to fit the relationship between the current value of the first segment of the piecewise function with the input pin transition time as the independent variable and the current value of the second segment of the piecewise function, at this time the output load capacitance value is a constant of 10fF:

[0116]

[0117]

[0118] Use a linear function to fit the relationship between the third segment current time of the piecewise function and the fourth segment current time of the piecewise function with the output load capacitance as the independent variable, and use an exponential function to fit the relationship of the third segment current value of the piecewise function with the input pin transition time as the independent variable, at this time the input pin transition time is a constant of 50ps:

[0119]

[0120]

[0121] Step 6: Adjust the segmented nodes in the current curve function obtained in Step 4 according to the relationship between the fitted nodes and the input pin switching time and the output load capacitance in Step 5. Then, compress or stretch the polynomial function obtained in Step 4 according to the relationship between the input pin switching time and the output load capacitance to obtain the current curve function with time, input pin switching time and output load capacitance as independent variables, as shown below:

[0122]

[0123] Where, k1 = t1 / t 1|transition_time＝50ps k2 = t2 / t 2|transition_time＝50ps k3 = t3 / t 3|c_load＝10fF ;k A =t A / t A|c_load＝10fF .

[0124] The fitting results are shown in Figures 11(a), 11(b), 11(c), and 11(d) when the input pin switching time is changed.

[0125] Example 4

[0126] A computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the standard unit switch current automatic fitting method based on a mathematical model as described in any of Embodiments 1-3.

[0127] Example 5

[0128] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the standard unit switch current automatic fitting method based on a mathematical model as described in any of Embodiments 1-3.

[0129] Example 6

[0130] An automated fitting system for standard unit switch current based on a mathematical model includes:

[0131] The standard cell simulation circuit building module is configured to: build two sets of standard cell simulation circuits, selecting the input pin transition time and output load capacitance as circuit parameters respectively; perform transient simulation and parameter analysis on the standard cell simulation circuits to obtain power supply terminal switching current curve data with input pin transition time and time, and output load capacitance and time as independent variables;

[0132] The curve model determination module is configured to: determine the corresponding curve model based on the changing trend of the power supply switching current curve data; and divide the power supply switching current curve into two parts based on the influence of the input pin switching time and the output load capacitance on the current; the current change of the first part of the curve is affected by the input pin switching time, therefore, the independent variables of the curve function of this part are time and the input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, therefore, the independent variables of the curve function of this part are time and the output load capacitance.

[0133] The curve data reading module is configured to use a C language program to read the power supply switching current curve data with time and input pin conversion time as independent variables, and time and output load capacitance as independent variables.

[0134] The standard unit switch current automatic fitting module is configured to: use a polynomial function to piecewise fit one of the current curve data to obtain a current curve function with time as the independent variable and input pin transition time and output load capacitance as constants; use a linear function to fit the relationship between the piecewise point time value and the input pin transition time and output load capacitance, and use an exponential function to fit the relationship between the piecewise point current value and the input pin transition time and output load capacitance; based on the fitted piecewise point current value and the relationship between the input pin transition time and output load capacitance, adjust the piecewise nodes in the current curve function, and compress or stretch the current curve function to obtain a current curve function with time, input pin transition time and output load capacitance as independent variables, thereby realizing the automatic fitting of the standard unit switch current.

[0135] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0136] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. An automated fitting method for standard unit switch current based on a mathematical model, characterized in that, include: Two sets of standard unit simulation circuits were built, and the input pin switching time and output load capacitance were selected as circuit parameters respectively. Transient simulation and parameter analysis were performed on the standard unit simulation circuit to obtain power supply terminal switching current curve data with input pin switching time and time as independent variables, and output load capacitance and time as independent variables, respectively. Determine the corresponding curve model based on the changing trend of the power supply terminal switching current curve data; Based on the influence of input pin switching time and output load capacitance on current, the power supply switching current curve is divided into two parts; the current change of the first part of the curve is affected by the input pin switching time, and the independent variables of the curve function for this part are time and input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, and the independent variables of the curve function for this part are time and output load capacitance. Read the power supply switching current curve data with time and input pin switching time as independent variables, and with time and output load capacitance as independent variables, respectively; The current curve data is piecewise fitted using a polynomial function to obtain a current curve function with time as the independent variable, and the input pin switching time and output load capacitance as constants. A linear function is used to fit the relationship between the piecewise time values ​​and the input pin switching time and the output load capacitance, and an exponential function is used to fit the relationship between the piecewise current values ​​and the input pin switching time and the output load capacitance. Based on the relationship between the fitted segmented current values ​​and the input pin switching time and output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin switching time and output load capacitance as independent variables, thereby realizing the automatic fitting of the standard unit switching current. The current curve function is shown in equation (1): （1） Where t is time and i is the magnitude of the current. This is the current fitting expression for the nth segment, where transition_time is the input pin transition time, c_load is the output load capacitance, tr1 and c1 are constants, and t1…t n The time values ​​at the segmentation points of the piecewise function; The relationship between the piecewise time values ​​and the input pin transition time and the output load capacitance is fitted using a linear function, and the relationship between the piecewise current values ​​and the input pin transition time and the output load capacitance is fitted using an exponential function; as shown in equations (2) and (3): （2） （3） Where transition_time is the input pin transition time, c_load is the output load capacitance, and t i t j Let i be the time value at the piecewise point of the piecewise function. i i j Let m be the current value at the piecewise point of the piecewise function. i m j n i n j a i a j b i b j c i c j d i d j These are constants determined during fitting; Based on the fitted segmented current values ​​and the relationship between the input pin transition time and the output load capacitance, the segmented nodes in the current curve function are adjusted, and the current curve function is compressed or stretched to obtain a current curve function with time, input pin transition time and output load capacitance as independent variables, as shown in equation (4): （4） Where k is the time scaling ratio caused by the different input pin switching times and output load capacitance, which result in different time nodes at each stage.

2. The method for automatically fitting standard unit switch current based on a mathematical model according to claim 1, characterized in that, There are four types of curve models. In the first curve model, the current value gradually increases with time, reaches its maximum value when the input pin transitions, and then gradually decreases, with only one maximum value. In the second curve model, the current value first decreases, then gradually increases to its maximum value, with only one maximum value. In the third curve model, the current first gradually increases with time, reaches its first maximum value when the input pin transitions, then gradually decreases and then increases to its second maximum value, and then gradually decreases again, with a total of two maximum values. In the fourth curve model, the current first decreases, then gradually increases to its first maximum value, then gradually decreases and then increases to its second maximum value, and then gradually decreases again, with a total of two maximum values.

3. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the standard unit switch current automatic fitting method based on a mathematical model as described in claim 1 or 2.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the standard unit switch current automatic fitting method based on a mathematical model as described in claim 1 or 2.

5. An automated fitting system for standard unit switch current based on a mathematical model, used to implement the automated fitting method for standard unit switch current based on a mathematical model as described in claim 1 or 2, characterized in that, include: The standard cell simulation circuit building module is configured to build two sets of standard cell simulation circuits, selecting the input pin switching time and output load capacitance as circuit parameters respectively. Transient simulation and parameter analysis were performed on the standard unit simulation circuit to obtain power supply terminal switching current curve data with input pin switching time and time as independent variables, and output load capacitance and time as independent variables, respectively. The curve model determination module is configured to determine the corresponding curve model based on the changing trend of the power supply switching current curve data. Based on the influence of input pin switching time and output load capacitance on current, the power supply switching current curve is divided into two parts; the current change of the first part of the curve is affected by the input pin switching time, and the independent variables of the curve function for this part are time and input pin switching time; the current change of the second part of the curve is affected by the output load capacitance, and the independent variables of the curve function for this part are time and output load capacitance. The curve data reading module is configured to read the power supply terminal switching current curve data with time and input pin switching time as independent variables, and time and output load capacitance as independent variables, respectively. The standard unit switch current automatic fitting module is configured to: use a polynomial function to piecewise fit one of the current curve data to obtain a current curve function with time as the independent variable and the input pin transition time and output load capacitance as constants; use a linear function to fit the relationship between the piecewise time value and the input pin transition time and output load capacitance, and use an exponential function to fit the relationship between the piecewise current value and the input pin transition time and output load capacitance; based on the fitted relationship between the piecewise current value and the input pin transition time and output load capacitance, adjust the piecewise nodes in the current curve function, and compress or stretch the current curve function to obtain a current curve function with time, input pin transition time and output load capacitance as independent variables, thus realizing the automatic fitting of the standard unit switch current.

Images