CN115249005B - Method, system and related equipment for automatic layout of radio frequency front-end module - Google Patents
Method, system and related equipment for automatic layout of radio frequency front-end module Download PDFInfo
- Publication number
- CN115249005B CN115249005B CN202211149266.3A CN202211149266A CN115249005B CN 115249005 B CN115249005 B CN 115249005B CN 202211149266 A CN202211149266 A CN 202211149266A CN 115249005 B CN115249005 B CN 115249005B
- Authority
- CN
- China
- Prior art keywords
- layout
- parasitic effect
- preset
- neural network
- adjusted
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/392—Floor-planning or layout, e.g. partitioning or placement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/394—Routing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Architecture (AREA)
- Computer Networks & Wireless Communication (AREA)
- Design And Manufacture Of Integrated Circuits (AREA)
- Noise Elimination (AREA)
Abstract
The invention belongs to the field of wireless communication, and particularly relates to a method, a system and related equipment for automatic layout of a radio frequency front-end module, wherein the method comprises the following steps: acquiring an initial layout and a circuit schematic diagram; calculating parasitic effect indexes by taking the initial layout and the circuit schematic diagram as the input of a preset neural network; judging whether the parasitic effect index meets a preset parasitic effect threshold value, if so, moving the circuit components by adopting random action on the initial layout according to a preset reinforcement learning method to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout; judging whether the parasitic effect index meets a preset optimization threshold, wherein: if not, recalculating the parasitic effect index after the layout adjustment is carried out; and if so, outputting the adjusted layout as a final optimized layout. The invention realizes the automatic layout of the radio frequency front end layout based on the method combining the neural network and the reinforcement learning, thereby optimizing the performance of the radio frequency front end.
Description
Technical Field
The invention belongs to the field of wireless communication, and particularly relates to a method, a system and related equipment for automatic layout of a radio frequency front-end module.
Background
The radio frequency front end module is a small-volume independent function set formed by integrating four main discrete devices, namely a filter, a switch, a Power Amplifier (PA) and a Low Noise Amplifier (LNA), and can be further divided into DiFEM integrating the switch and the filter, LFEM integrating the switch, the filter and the LNA, L-PAMiF integrating the PA, the LNA, the switch and the filter and the like according to different integration modes and functions. In addition to the 4 discrete devices mentioned above, a small number of passive matching elements are also present in the rf front-end module.
In the process of drawing the radio frequency front-end module layout, layout positions of various devices need to be reasonably arranged, and on the premise of meeting design and inspection rules, the used area is reduced as much as possible, the wiring length is reduced, and the reasonable power tolerance in unit area is kept. In the process of layout from a schematic diagram to a layout, parasitic effects may be caused by adjusting the position, the details and the connection of the device, so that the overall electrical response is changed. In the field of current radio frequency front end design, the layout of a radio frequency front end layout is finally determined mainly by means of repeated manual iterative inspection, and the method is large in manpower requirement, long in time consumption and prone to errors.
Disclosure of Invention
The embodiment of the invention provides a method, a system and related equipment for automatic layout of a radio frequency front-end module, aiming at solving the problems of long time consumption and easy error existing in the existing radio frequency front-end design process depending on manual iterative inspection.
In a first aspect, an embodiment of the present invention provides a method for automatic layout of a radio frequency front end module, where the method includes the following steps:
acquiring an initial layout and a schematic circuit diagram which are to be optimized and provided with circuit components, wherein the circuit components comprise a radio frequency front end module;
calculating parasitic effect indexes in the initial layout by taking the initial layout and the circuit schematic as the input of a preset neural network;
judging whether the parasitic effect index meets a preset parasitic effect threshold value, wherein:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network;
judging whether the parasitic effect index meets a preset optimization threshold, wherein:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
Furthermore, the parasitic effect index is obtained by comparing the difference of electromagnetic simulation parameter curves of the initial layout and the circuit schematic diagram.
Furthermore, the preset neural network comprises a graph neural network and a multilayer perceptron, the graph neural network is used for extracting the electrical property and the coordinate of each circuit component in the initial layout, and the multilayer perceptron is used for calculating the parasitic effect index through regression.
Furthermore, the step of moving the circuit component of the initial layout by adopting random action according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic as the input of the preset neural network further comprises the steps of:
judging whether the random action meets a preset layout design rule, wherein:
if yes, setting a punishment mark for the random action;
if not, setting a reward mark for the random action.
Furthermore, the step of moving the circuit component of the initial layout by adopting random action according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic as the input of the preset neural network further comprises the steps of:
judging whether the parasitic effect index after the layout adjustment is carried out by adopting the random action is superior to the parasitic effect index before the adjustment, wherein:
if not, setting the penalty mark for the random action;
and if so, setting the reward mark for the random action.
Further, the reinforcement learning method preferentially selects the random action without the punishment mark to perform layout adjustment.
Furthermore, in the step of moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment, the random actions are forbidden so that the circuit components in the initial layout have a phenomenon of mutual overlapping.
In a second aspect, an embodiment of the present invention further provides a system for automatic layout of a radio frequency front end module, including:
the device comprises an initialization module, a first module and a second module, wherein the initialization module is used for acquiring an initial layout and a circuit schematic diagram which are to be optimized and are provided with circuit components, and the circuit components comprise radio frequency front end modules;
the parasitic effect calculation module is used for calculating a parasitic effect index in the initial layout by taking the initial layout and the circuit schematic diagram as the input of a preset neural network;
a reinforcement learning module, configured to determine whether the parasitic effect indicator satisfies a preset parasitic effect threshold, where:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network;
the layout output module is used for judging whether the parasitic effect index meets a preset optimization threshold, wherein:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
In a third aspect, an embodiment of the present invention further provides a computer device, including: a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor, when executing the computer program, implements the steps of the method for automatic layout of a radio frequency front end module as described in any one of the above embodiments.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method for automatically laying out a layout of a radio frequency front-end module according to any one of the foregoing embodiments is implemented.
The method has the advantages that the method based on the combination of the neural network and the reinforcement learning is used for optimizing the layout adjustment process in the radio frequency front end design process, so that the period from a schematic diagram to layout drawing can be reduced, parasitic effect indexes in the layout can be quickly judged through the regression task of the support vector machine, the layout is automatically adjusted on the premise of meeting design rules, the electromagnetic interference caused by layout is automatically reduced, and the performance of the radio frequency front end is optimized.
Drawings
Fig. 1 is a block diagram of a flow of steps of a method for automatic layout of a radio frequency front end module according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating parasitic effect indicator calculation according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of reinforcement learning according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a system for automatic layout of a radio frequency front-end module according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, fig. 1 is a flow chart illustrating steps of a method for automatic layout of a radio frequency front end module according to an embodiment of the present invention, where the method includes the following steps:
s101, obtaining an initial layout and a schematic circuit diagram which are to be optimized and provided with circuit components, wherein the circuit components comprise a radio frequency front end module.
And S102, calculating the parasitic effect index in the initial layout by taking the initial layout and the circuit schematic as the input of a preset neural network.
Furthermore, the preset neural network comprises a graph neural network and a multilayer perceptron, the graph neural network is used for extracting the electrical property and the coordinate of each circuit component in the initial layout, and the multilayer perceptron is used for calculating the parasitic effect index through regression.
Illustratively, in the embodiment of the present invention, the graph neural network adopts a GAT (self-attention-seeking neural network) structure, and in other possible neural network structures, GNN (graph neural network), GCN (graph convolution neural network) and other structures may also be adopted, the GAT used in the embodiment of the present invention has a self-attention-seeking mechanism, and the self-attention-seeking mechanism in the propagation process enables important features affecting parasitic effects in the layout to be extracted, for example, in an actual situation, if a power line in the layout is closer to a radio frequency line, relatively serious parasitic effects are likely to occur; embodiments of the present invention connect a multi-level perceptron (MLP) after the GAT for performing a regression task to predict the indicators of parasitic effects.
Furthermore, the parasitic effect index is obtained by comparing the difference of electromagnetic simulation parameter curves of the initial layout and the circuit schematic diagram.
For example, referring to fig. 2, fig. 2 is a schematic diagram illustrating calculation of a parasitic effect index according to an embodiment of the present invention, in the embodiment of the present invention, electromagnetic simulation data of HFSS is used as the circuit schematic diagram, in the diagram, an S parameter curve of the circuit schematic diagram is a thin curve, an S parameter curve of the initial layout is a thick curve, and the parasitic effect index can be obtained by solving an L2-norm value between the two curves.
S103, judging whether the parasitic effect index meets a preset parasitic effect threshold value.
Wherein:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network.
The reinforcement learning method is a self-adaptive iterative optimization algorithm, please refer to fig. 3, and fig. 3 is a schematic diagram of reinforcement learning provided by an embodiment of the present invention, and the reinforcement learning method includes the following concepts:
the state is as follows: describing a set of all possible scenarios in the environment;
the actions are as follows: all actions that a reinforcement learning agent can take can act on the environment and change states;
the reinforcement learning agent: the mechanism responsible for deciding what the next action is, generally speaking, needs to take into account exploration and conservation, wherein exploration refers to the mechanism trying to enter different new states, conservation refers to the mechanism which tends to take some actions under the condition that the actions are known to obtain larger income;
rewarding: given a state, taking an action leads to different results, so different rewards need to be set in the algorithm to encourage the use of an action;
environment: when an action is taken on one state, the environment will decide what the next state is, there will be some random factors in the environment, the same action is taken on the same state, and the same state will not necessarily be led to, and in one possible embodiment, a markov decision chain is needed to describe the environment.
In the embodiment of the present invention, the layout of all the electronic components in the initial layout is equivalent to an environment, and the action is equivalent to the initial action.
Furthermore, the step of calculating the parasitic effect index of the adjustment layout by adopting the adjustment layout and the circuit schematic diagram as the input of the preset neural network further comprises the steps of:
judging whether the random action meets a preset layout design rule, wherein:
if yes, setting a punishment mark for the random action;
if not, setting a reward mark for the random action.
Specifically, in the embodiment of the present invention, the preset layout design rule is used to avoid that the adjusted layout has an excessive influence on the whole radio frequency module after some components are moved.
Further, the reinforcement learning method preferentially selects the random action without the punishment mark to perform layout adjustment.
Furthermore, in the step of moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment, the random actions are forbidden to enable the circuit components in the initial layout to have a phenomenon of mutual overlapping.
And S104, judging whether the parasitic effect index meets a preset optimization threshold value.
Wherein:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
Specifically, two preset thresholds, namely the preset parasitic effect threshold and the preset optimization threshold, are set in the embodiment of the present invention, where the preset parasitic effect threshold is set with respect to the initial layout, and when the parasitic effect index of the initial layout does not meet the preset parasitic effect threshold, it indicates that the parasitic effect of the initial layout is relatively serious, and the time required for optimizing the initial layout is longer, or the parasitic effect index does not meet the electrical performance requirement that can be used as a final output, so in step S103, if the parasitic effect index does not meet the preset parasitic effect threshold, the initial layout may be discarded in the embodiment of the present invention.
Furthermore, the step of calculating the parasitic effect index of the adjusted layout by adopting the adjusted layout and the circuit schematic diagram as the input of the preset neural network comprises the following steps:
judging whether the parasitic effect index after the layout adjustment is carried out by adopting the random action is superior to the parasitic effect index before the adjustment, wherein:
if not, setting the penalty mark for the random action;
and if so, setting the reward mark for the random action.
The method has the advantages that the method based on the combination of the neural network and the reinforcement learning is used for optimizing the layout adjustment process in the radio frequency front end design process, so that the period from a schematic diagram to layout drawing can be reduced, parasitic effect indexes in the layout can be quickly judged through the regression task of the support vector machine, the layout is automatically adjusted on the premise of meeting design rules, the electromagnetic interference caused by layout is automatically reduced, and the performance of the radio frequency front end is optimized.
The embodiment of the present invention further provides a system 200 for automatic layout of a radio frequency front end module, please refer to fig. 4, where fig. 4 is a schematic structural diagram of the system for automatic layout of a radio frequency front end module according to the embodiment of the present invention, and the system includes:
the initialization module 201 is configured to obtain an initial layout and a schematic circuit diagram of a circuit component to be optimized, where the circuit component includes a radio frequency front end module;
a parasitic effect calculation module 202, configured to calculate a parasitic effect index in the initial layout by using the initial layout and the circuit schematic as inputs of a preset neural network;
a reinforcement learning module 203, configured to determine whether the parasitic effect indicator satisfies a preset parasitic effect threshold, where:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network;
a layout output module 204, configured to determine whether the parasitic effect index meets a preset optimization threshold, where:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
The system 200 for automatic layout of rf front-end module can implement the steps in the method for automatic layout of rf front-end module in the above embodiment, and can implement the same technical effects, which are not described herein again with reference to the description in the above embodiment.
Referring to fig. 5, fig. 5 is a schematic structural diagram of a computer device provided in an embodiment of the present invention, where the computer device 300 includes: a memory 302, a processor 301, and a computer program stored on the memory 302 and executable on the processor 301.
The processor 301 calls the computer program stored in the memory 302 to execute the steps in the method for automatic layout of the radio frequency front-end module according to the embodiment of the present invention, and with reference to fig. 1, the method specifically includes:
s101, obtaining an initial layout and a circuit schematic diagram of a circuit component used for optimization and comprising a radio frequency front end module, obtaining the initial layout and the circuit schematic diagram of the circuit component to be optimized and provided with the circuit component, wherein the circuit component comprises the radio frequency front end module.
And S102, calculating the parasitic effect index in the initial layout by taking the initial layout and the circuit schematic as the input of a preset neural network.
S103, judging whether the parasitic effect index meets a preset parasitic effect threshold value, wherein:
if yes, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjustment layout, and calculating the parasitic effect index of the adjustment layout by taking the adjustment layout and the circuit schematic diagram as the input of the preset neural network.
S104, judging whether the parasitic effect index meets a preset optimization threshold, wherein:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
Furthermore, the parasitic effect index is obtained by comparing the difference of electromagnetic simulation parameter curves of the initial layout and the circuit schematic diagram.
Furthermore, the preset neural network comprises a graph neural network and a multilayer perceptron, the graph neural network is used for extracting the electrical property and the coordinate of each circuit component in the initial layout, and the multilayer perceptron is used for calculating the parasitic effect index through regression.
Furthermore, the step of moving the circuit component of the initial layout by adopting random action according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic as the input of the preset neural network further comprises the steps of:
judging whether the random action meets a preset layout design rule, wherein:
if yes, setting a punishment mark for the random action;
if not, setting a reward mark for the random action.
Furthermore, the step of calculating the parasitic effect index of the adjustment layout by adopting the adjustment layout and the circuit schematic diagram as the input of the preset neural network further comprises the steps of:
judging whether the parasitic effect index after the layout adjustment is carried out by adopting the random action is superior to the parasitic effect index before the adjustment, wherein:
if not, setting the punishment mark for the random action;
and if so, setting the reward mark for the random action.
Further, the reinforcement learning method preferentially selects the random action without the penalty flag to perform layout adjustment.
Furthermore, in the step of moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment, the random actions are forbidden to enable the circuit components in the initial layout to have a phenomenon of mutual overlapping.
The computer device 300 according to the embodiment of the present invention can implement the steps in the method for automatic layout of rf front-end module in the above-mentioned embodiment, and can implement the same technical effects, which are described in the above-mentioned embodiment and will not be described herein again.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements each process and step in the method for automatic layout of a radio frequency front end module according to the embodiment of the present invention, and can implement the same technical effect, and in order to avoid repetition, the detailed description is omitted here.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of another like element in a process, method, article, or apparatus that comprises the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
While the present invention has been described in connection with the preferred embodiments of the present invention, as illustrated and described in the accompanying drawings, it is to be understood that the invention is not limited to the disclosed embodiments, but is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (9)
1. A method for automatic layout of a radio frequency front-end module is characterized by comprising the following steps:
acquiring an initial layout and a schematic circuit diagram of a circuit component to be optimized, wherein the circuit component is provided with a radio frequency front-end module;
calculating parasitic effect indexes in the initial layout by taking the initial layout and the circuit schematic diagram as input of a preset neural network, wherein the preset neural network comprises a graph neural network and a multilayer perceptron, the graph neural network is used for extracting the electrical attributes and the coordinates of each circuit component in the initial layout, and the multilayer perceptron is used for calculating the parasitic effect indexes through regression;
judging whether the parasitic effect index meets a preset parasitic effect threshold value, wherein:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network;
judging whether the parasitic effect index meets a preset optimization threshold, wherein:
if not, performing layout adjustment on the adjustment layout by using the preset reinforcement learning method and adopting the follow-up action, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
2. The method according to claim 1, wherein the parasitic effect indicator is obtained by comparing differences of respective electromagnetic simulation parameter curves of the initial layout and the schematic diagram of the circuit.
3. The method according to claim 1, wherein the step of calculating the parasitic effect index of the adjusted layout by moving the circuit components of the initial layout by random action according to a preset reinforcement learning method to adjust the layout to obtain an adjusted layout and using the adjusted layout and the schematic circuit diagram as the input of the preset neural network further comprises the steps of:
judging whether the random action meets a preset layout design rule, wherein:
if yes, setting a punishment mark for the random action;
if not, setting a reward mark for the random action.
4. The method according to claim 3, wherein the step of calculating the parasitic effect index of the adjusted layout by moving the circuit components of the initial layout by random action according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout and using the adjusted layout and the schematic circuit diagram as the input of the preset neural network further comprises the steps of:
judging whether the parasitic effect index after the layout adjustment is carried out by adopting the random action is superior to the parasitic effect index before the adjustment, wherein:
if not, setting the punishment mark for the random action;
and if so, setting the reward mark for the random action.
5. The method for automatic layout of a radio frequency front-end module according to claim 4, wherein said reinforcement learning method preferentially selects said random actions without said penalty flags set for layout adjustment.
6. The method according to claim 1, wherein in the step of moving the circuit components of the initial layout by random actions according to a preset reinforcement learning method to realize layout adjustment, the random actions are prohibited to cause the circuit components in the initial layout to have a phenomenon of mutual overlapping.
7. A system for automatic layout of a radio frequency front end module, comprising:
the device comprises an initialization module, a detection module and a control module, wherein the initialization module is used for acquiring an initial layout and a circuit schematic diagram which are to be optimized and are provided with circuit components, and the circuit components comprise a radio frequency front end module;
the parasitic effect calculation module is used for calculating a parasitic effect index in the initial layout by taking the initial layout and the circuit schematic diagram as input of a preset neural network, wherein the preset neural network comprises a graph neural network and a multilayer perceptron, the graph neural network is used for extracting the electrical property and the coordinate of each circuit component in the initial layout, and the multilayer perceptron is used for calculating the parasitic effect index through regression;
a reinforcement learning module, configured to determine whether the parasitic effect indicator satisfies a preset parasitic effect threshold, where:
if so, moving the circuit components of the initial layout by adopting random actions according to a preset reinforcement learning method to realize layout adjustment to obtain an adjusted layout, and calculating the parasitic effect index of the adjusted layout by taking the adjusted layout and the circuit schematic diagram as the input of the preset neural network;
the layout output module is used for judging whether the parasitic effect index meets a preset optimization threshold value, wherein:
if not, performing layout adjustment on the adjustment layout by using the random action by using the preset reinforcement learning method, and recalculating the parasitic effect index;
and if so, outputting the adjusted layout as a final optimized layout.
8. A computer device, comprising: memory, processor and computer program stored on the memory and executable on the processor, the processor implementing the steps in the method for automatic layout of a radio frequency front end module according to any of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for automatic layout of a radio frequency front end module as claimed in any one of claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211149266.3A CN115249005B (en) | 2022-09-21 | 2022-09-21 | Method, system and related equipment for automatic layout of radio frequency front-end module |
| PCT/CN2023/109777 WO2024060829A1 (en) | 2022-09-21 | 2023-07-28 | Automatic layout method and system for radio frequency front-end module, and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211149266.3A CN115249005B (en) | 2022-09-21 | 2022-09-21 | Method, system and related equipment for automatic layout of radio frequency front-end module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115249005A CN115249005A (en) | 2022-10-28 |
| CN115249005B true CN115249005B (en) | 2022-12-27 |
Family
ID=83699268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211149266.3A Active CN115249005B (en) | 2022-09-21 | 2022-09-21 | Method, system and related equipment for automatic layout of radio frequency front-end module |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115249005B (en) |
| WO (1) | WO2024060829A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115249005B (en) * | 2022-09-21 | 2022-12-27 | 深圳飞骧科技股份有限公司 | Method, system and related equipment for automatic layout of radio frequency front-end module |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012015702A1 (en) * | 2010-07-24 | 2012-02-02 | Cadence Design Systems, Inc. | Methods, systems, and articles of manufacture for implementing electronic circuit designs with simulation awareness |
| CN109684755A (en) * | 2018-12-28 | 2019-04-26 | 佛山中科芯蔚科技有限公司 | A kind of digital-analog mix-mode chip asynchronous circuit full custom method and system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007264184A (en) * | 2006-03-28 | 2007-10-11 | Toppan Printing Co Ltd | Transmission type rear screen and manufacturing method therefor |
| US8893066B2 (en) * | 2012-12-27 | 2014-11-18 | Taiwan Semiconductor Manufacturing Co., Ltd. | Parasitic component library and method for efficient circuit design and simulation using the same |
| CN105653744A (en) * | 2014-11-13 | 2016-06-08 | 中芯国际集成电路制造(上海)有限公司 | Method and device for designing layout |
| CN110998585B (en) * | 2017-06-22 | 2024-07-16 | 株式会社半导体能源研究所 | Layout design system and layout design method |
| KR20210136027A (en) * | 2019-03-20 | 2021-11-16 | 가부시키가이샤 한도오따이 에네루기 켄큐쇼 | Wiring layout design method, program, and recording medium |
| CN114595656A (en) * | 2022-03-02 | 2022-06-07 | 南京星火技术有限公司 | Method, apparatus, device and medium for generating algorithm model and layout of electronic device |
| CN114896937A (en) * | 2022-05-24 | 2022-08-12 | 广东工业大学 | Integrated circuit layout optimization method based on reinforcement learning |
| CN115000654B (en) * | 2022-08-05 | 2022-11-11 | 深圳飞骧科技股份有限公司 | Radio frequency module and packaging method thereof |
| CN115249005B (en) * | 2022-09-21 | 2022-12-27 | 深圳飞骧科技股份有限公司 | Method, system and related equipment for automatic layout of radio frequency front-end module |
-
2022
- 2022-09-21 CN CN202211149266.3A patent/CN115249005B/en active Active
-
2023
- 2023-07-28 WO PCT/CN2023/109777 patent/WO2024060829A1/en unknown
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012015702A1 (en) * | 2010-07-24 | 2012-02-02 | Cadence Design Systems, Inc. | Methods, systems, and articles of manufacture for implementing electronic circuit designs with simulation awareness |
| CN109684755A (en) * | 2018-12-28 | 2019-04-26 | 佛山中科芯蔚科技有限公司 | A kind of digital-analog mix-mode chip asynchronous circuit full custom method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024060829A1 (en) | 2024-03-28 |
| CN115249005A (en) | 2022-10-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115249005B (en) | Method, system and related equipment for automatic layout of radio frequency front-end module | |
| CN112257848B (en) | Method for determining logic core layout, model training method, electronic device and medium | |
| CN111260079B (en) | Electronic equipment and intelligent body self-training device | |
| WO2024093467A1 (en) | Automatic design method for surface acoustic wave filter, related system, and storage medium | |
| CN105896086B (en) | The matching process and device of impedance | |
| CN107943571B (en) | Background application control method and device, storage medium and electronic equipment | |
| CN117459889A (en) | Mobile terminal positioning method and device, storage medium and electronic equipment | |
| CN114925651A (en) | Circuit routing determination method and related equipment | |
| CN114168318A (en) | Training method of storage release model, storage release method and equipment | |
| CN115544953A (en) | Method, device, equipment and readable medium for simulating and optimizing PCB impedance | |
| CN111123728B (en) | Unmanned vehicle simulation method, device, equipment and computer readable medium | |
| CN114218887A (en) | Chip configuration design method, device and medium based on deep learning | |
| CN108681480B (en) | Background application program control method and device, storage medium and electronic equipment | |
| CN113596793B (en) | Bluetooth low-power-consumption equipment control method and device | |
| CN115328118A (en) | Path determining method and device, readable storage medium and electronic equipment | |
| CN116111984B (en) | Filter design optimization method and device, filter, equipment and medium | |
| CN113947056A (en) | Differential routing design method and differential line | |
| CN115223548B (en) | Voice interaction method, voice interaction device and storage medium | |
| JP2010177995A (en) | Communication control method of radio communication system, program therefor, and radio communication apparatus controller | |
| CN115980804B (en) | Indoor positioning method and device | |
| CN116992816A (en) | Circuit optimization method, device, equipment and storage medium based on reinforcement learning | |
| CN112578896B (en) | Frequency adjusting method, frequency adjusting device, electronic apparatus, and storage medium | |
| CN114417770A (en) | Printed circuit board checking method and device, terminal equipment and storage medium | |
| JP2005056431A (en) | Manufacturing line control device, and system and method for automatically changing measurement sequence including it | |
| CN111277962A (en) | Formation method and related equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |