10 results about "Scaling circuits" patented technology

The application discloses a circuit pre-training method and device based on a graph neural network and electronic equipment, comprising the following steps: for a large-scale circuit, a circuit network graph is constructed, a random walk road is generated, and the road is used as a sample for model training; a node2vec model is trained using the road sample; a model with the minimum loss function value is selected as an optimal model; and the optimal model is loaded, and representation vectors of all nodes are output and saved. In the application, the node2vec pre-training model of the graph neural network is used, the information of a long-distance standard unit is fully absorbed to absorb local circuit structure features, deep representation of nodes is automatically learned from the circuit structure, and tedious manual feature engineering is avoided, thereby providing high-quality representation vectors for a downstream model. The node representation learned by the node2vec model is a task-independent general feature, can be flexibly applied to various downstream tasks, and provides support for subsequent circuit analysis and optimization design work.

A radio frequency (RF) receiver is described, the RF receiver comprising: a plurality of analog mixing circuits including circuits for down-converting an input RF signal; a plurality of analog combining and scaling circuits operably connected to the analog mixing circuits and for generating a plurality of intermediate signals; a digital combining and scaling circuit operably connected to the analog combining and scaling circuits, the digital combining and scaling circuit comprising: in-phase signal component (I) circuitry for performing a first combining and scaling operation on a set of two or more of the plurality of intermediate signals to generate an in-phase signal component of a baseband signal; quadrature phase signal component (Q) circuitry for performing a second combining and scaling operation on another set of two or more of the plurality of intermediate signals to generate a quadrature phase signal component (Q) of the baseband signal.

This invention discloses a rapid assessment method for soft error rate caused by single-event effects based on an analytical function library. It addresses the problems of low efficiency and difficulty in supporting large-scale circuit evaluation in traditional TCAD simulation. The method constructs and calibrates device models based on process design kits to obtain accurate electrical characteristics of the calibration device model. Based on multidimensional key irradiation variables, dense TCAD sampling and parametric fitting are performed on the calibration device to construct a response function library. Based on the actual integrated circuit layout, particle transport simulation under radiation conditions is performed in Geant4 to extract the secondary particle set. The corresponding transient current is rapidly synthesized by querying the response function library and interpolating. The current is injected into the SPICE circuit netlist for simulation to determine whether single-event upset has occurred and to calculate the soft error rate. This method achieves an order-of-magnitude improvement in simulation efficiency while maintaining accuracy, making it suitable for rapid radiation resistance assessment and design hardening of complex integrated circuits.

The invention relates to the technical field of circuit simulation waveform data visualization, and particularly discloses a circuit simulation waveform sampling method based on MinMaxLTTB. The method aims at solving the problems of display lagging and waveform distortion caused by the fact that data volume simplification and key feature point reservation cannot be considered in mass waveform data visualization in a traditional sampling technology. The method comprises the following steps: (1) dynamic bucket division: adaptively creating data buckets with different sizes based on target resolution and simulation duration; (2) MinMax extremum pre-extraction: extracting extremum points in each bucket in a layered manner to form a simplified data set; (3) sampling LTTB characteristic values, and selecting an optimal characteristic point based on a triangle area maximum principle; and (4) integrating the sampling points in sequence to generate visual special data. According to the method, dynamic bucket division and a two-stage sampling strategy are combined, the data volume is greatly reduced, meanwhile, the waveform key features are accurately reserved, the real-time visualization performance and accuracy of large-scale circuit simulation waveforms are remarkably improved, and the method is suitable for waveform display modules of various EDA tools.

The invention discloses a self-adaptive hyperdimensional computing (HDC) accelerator supporting incremental learning, and belongs to the technical field of artificial intelligence hardware acceleration. The accelerator comprises a feature extraction and coding circuit, an incremental prototype-like storage resource pool, a parallel reasoning array and a confidence perception retraining circuit. Wherein the incremental prototype-like storage resource pool comprises a pre-allocated BRAM array and a write enable gating logic, and class expansion during operation is realized through logic-physical address decoupling; the confidence perception retraining circuit is integrated with a double-threshold comparator group and a barrel-shaped shift scaling circuit without a multiplier, and monitors the minimum Hamming distance in real time through a hardware comparator and drives a barrel-shaped shifter to dynamically generate an update weight. According to the method, low-overhead incremental learning is realized on the premise that hardware does not need to be synthesized again through resource isolation and adaptive weight adjustment of a pure hardware level, the disastrous forgetting phenomenon is effectively inhibited, and the retraining efficiency is remarkably improved.

The application discloses a process angle prediction method and system in static timing analysis based on ensemble learning, which combines multilayer perceptron and graph neural network for process angle prediction. Firstly, the MLP model is used for preliminary prediction, and then the prediction result of the MLP is input into the GNN model to optimize the prediction result in the form of graph structure data. The application also uses a generative adversarial network to generate synthetic data, thereby expanding the training set and improving the prediction ability of the model. The method of the application can significantly improve the prediction accuracy and is more efficient in running speed than the traditional method. The method not only reduces the calculation overhead of the traditional STA method, but also has good adaptability and high efficiency, and is particularly suitable for process angle prediction tasks of large-scale circuits. The system architecture of the application includes a data acquisition module, an MLP module, a GNN module and a GAN data generation module, and can realize efficient and accurate process angle prediction in industrial applications.

The invention discloses a scale-adaptive logic comprehensive optimization circuit representation learning method, and belongs to the technical field of integrated circuit design automation. Comprising the following steps: constructing circuit diagram data, and extracting node features, edge connection relationships and node depth information; a scale adaptive graph encoder is constructed, a double-layer GCN and a multi-scale feature fusion module are adopted, and scale adaptive node representation is generated through multi-hop neighborhood aggregation, circuit perception gating and attention fusion; performing hierarchical pooling based on the node depth to construct a hierarchical graph sequence; carrying out embedded coding on the optimized sequence and injecting position information to obtain sequence representation; establishing bidirectional association between a circuit structure and optimization steps through a dynamic graph-sequence adaptive interaction module, and generating a context sensing target sequence; and a Transform decoder is used for carrying out sequence modeling and predicting a quality result track. According to the method, through scale self-adaption and a dynamic interaction mechanism, the problems of representation learning and optimization dynamic modeling of a heterogeneous scale circuit are solved.

Embodiments herein relate to an interconnect architecture for a multi-transistor stack including channel structures in the form of nanoribbons or nanowires. In one aspect, a metal interconnect is routed between the transistors to provide between electrical connections for control gates and / or source / drain nodes of the transistors. The electrical connections can be provided between transistors in the same stack or in different stacks. In another aspect, control gates of transistors in a stack are independently controlled.

The application discloses a large-scale circuit partitioning method based on artificial intelligence adaptive weight distribution, and belongs to the technical field of integrated circuit computer-aided design, and specifically as follows: first, a graph structure is established to obtain data containing circuit topology characteristics; then, a cubic interpolation method is used to convert the input into a uniform length, improving the generality of the model; a gating mechanism is introduced to establish the structures of input gates, output gates and forget gates, helping the model better perceive the dependency relationship between inputs; the data with circuit topology characteristics obtained through the above processing are trained to obtain a weight distribution model, and the partitioning weight is updated; finally, the large-scale circuit is partitioned by using the updated weight, and multi-index balanced parallel system simulation is realized. The application helps to accelerate the circuit simulation process and improve the parallel benefit.

The invention discloses an electrical loop digital display method based on structured data, and the method comprises the steps: obtaining the structured data which comprises a primitive library, wiring point matrix data, relation set data, element mapping data and template configuration data; performing visual default parameter configuration, and drawing primitives, connecting lines and connecting points based on the structured data; and carrying out board card grouping and board card frame generation on the graphic elements of all the arranged elements, and displaying element names of the graphic elements of the elements. According to the method, the consistency of what you see is data is realized, the visualization of a large-scale circuit is supported, and the visualization quality and the engineering availability of electrical engineering are improved.