BZBeining Zhao

Beining Zhao

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Beining Zhao has a total of 13 co-authors. Here are the top co-authors by number of shared publications:

  • Shan Cao

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    3 shared publications
  • Xinyu Chen

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    3 shared publications
  • Zhiyuan Jiang

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    3 shared publications
  • Yintao Liu

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    2 shared publications
  • Yu Li

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    2 shared publications
  • Limin Jiang

    Shanghai Key Laboratory of Chips and Systems for Intelligent Connected Vehicle and the School of Communication and Information Engineering, Shanghai University, Shanghai, China

    1 shared publication

Latest Publications

  • An Agile Systolic Array-Based Hardware Accelerator for Scalable Multi-Head Self-Attention

    by Xinyu Chen, Yu Li, Beining Zhao, Yintao Liu, Shan Cao, Zhiyuan Jiang

    As attention mechanisms find increasing application in computer vision, numerous neural networks are now integrating convolutional neural networks (CNNs) with attention layers to enhance performance. As a typical representative, the Bottleneck Transformer (BoT) network effectively balances the extraction of local features and modeling of global context, leading to significant performance improvements. However, the multi-head self-attention (MHSA) layer of BoT deviates from conventional MHSA by incorporating a wider range of matrix operations, posing challenges in computational complexity and memory bandwidth. This letter introduces a customized accelerator for the MHSA layer of BoT. The accelerator employs a configurable systolic array architecture, designed to support various matrix operations, including matrix transposition, addition, multiplication, and a hardware-optimized Softmax, facilitating hybrid scalable MHSA layers. Implemented on an FPGA, the proposed accelerator achieves a performance of 143.98 GOPS at a frequency of 320 MHz. When compared to CPU implementations, our design offers a 7.18× reduction in inference time.

  • Near-Sensor LiDAR and Visual Feature Extraction and Communication for Low-Latency Roadside Cooperative Perception

    by Wei Zhang, Yuhang Gu, Beining Zhao, Qingyu Deng, Xinyu Chen, Yi Shi, Limin Jiang, Shan Cao, Zhiyuan Jiang, Ruiqing Mao, et al.

    Autonomous driving technologies are swiftly evolving, characterized by two main strategies: 1) single-vehicle autonomous driving (SVAD) and 2) vehicle-infrastructure cooperative autonomous driving (VICAD). SVAD depends entirely on the vehicle’s internal sensors and processing capabilities, whereas VICAD benefits from a synergistic network combining roadside infrastructure, connected vehicles, and cloud services to boost safety and efficiency. Nevertheless, VICAD encounters challenges with high-bandwidth data transmission and perception latency. To mitigate these concerns, we introduce an innovative intelligent roadside unit (I-RSU) platform integrating perception, computing, and communication into one cohesive system. The platform features dual neural processing units (NPUs) for the effective extraction of images and LiDAR features, alongside a Cellular-V2X (C-V2X) communication module, all realized on a field-programmable gate array (FPGA). This setup minimizes latency and expenses by enabling computation near the sensors and facilitating selective data transmission. Our system also supports multimodal fusion, enhancing overall perception and safety. Through extensive real-world trials and simulations, our system demonstrates a substantial reduction in end-to-end latency, providing a scalable solution for VICAD scenarios.


  • A Heterogeneous CNN Compilation Framework for RISC-V CPU and NPU Integration Based on ONNX-MLIR

    by Shan Cao, Meiling Yang, Yintao Liu, Yu Li, Beining Zhao, Xinyu Chen, Zhiyuan Jiang

    With the continuous advancement of convolutional neural networks (CNNs), many neural network processing units (NPUs) have emerged in recent years. NPUs offer advantages, such as improved energy efficiency and low latency compared to traditional processors. However, NPUs often struggle to keep pace with the growing complexity of algorithmic models, which limits the implementation of certain AI applications. Central processing units (CPUs) are known for their versatility, but are computationally inefficient. Combining the strengths of both architectures could present a viable solution to these challenges. Despite this potential, there is currently insufficient academic focus on CPU/NPU co-computing, and few architectures or toolchains have been developed to support such integration. In this article, we propose a RISC-V CPU/NPU co-computing-based compilation framework to solve the problem of mapping from algorithms to heterogeneous architectures. To bridge the computational gap between heterogeneous platforms, within ONNX-MLIR, we developed three phases to support operator packing, algorithm adaption, and data coordination. To integrate heterogeneous compilation platforms, we introduce four additional functions to handle task scheduling, weight quantization, weight rearrangement, and memory management. In particular: 1) to optimize the transition overhead, an operator packing technique is introduced by extending the ONNX-MLIR framework with custom operators, which enhances NPU efficiency by reducing the number of transitions between operations; 2) at the aspect of task scheduling, a three-mode task scheduler is developed to allow users to customize task distribution for correctness verification, performance optimization, or detailed analysis on heterogeneous platforms, which demonstrates the versatility of our framework; and 3) In terms of memory management, an optimized scheme for CPU/NPU architectures is proposed to ensure accurate data access. This memory management employs a dual-end approach with self-checks and a release-after-use mechanism to further improve efficiency. Experimental results demonstrate that the framework effectively generates instructions for various CNNs, significantly enhancing the efficiency and performance of heterogeneous architectures. It achieves up to a $7.06\times $ reduction in code density and up to a $5.58\times $ improvement in memory usage, narrowing the computational gap between CPUs and NPUs.