• Home
  • Search
  • A Reconfigurable Deep Neural Network on Chip Design with Flexible Convolutional Operations
  • Cite Icon5
  • https://doi.org/10.1109/nocarc57472.2022.9911283Copy DOI Icon

A Reconfigurable Deep Neural Network on Chip Design with Flexible Convolutional Operations

  • Oct 2, 2022
  • Kun-Chih Chen +1 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

The designs of deep neural network (DNN) accelerators have gradually gained attention due to the increased demand for real-time AI applications. On the other hand, due to the diverse applications, kernel sizes and shapes for the involved convolutional operation in the target DNN model are not fixed. Therefore, it is necessary to design a reconfigurable DNN accelerator to cover different kernel sizes for convolutional operation in DNNs. However, due to the worst-case design policy, the designers usually select the largest kernel size as the design parameter to implement the DNN accelerator, which leads to lower hardware utilization. The reason is that the conventional array-based DNN design method restricts the efficiency of data delivery. Besides, the complicated data flow between neuron layers of DNN models counteracts the benefit of the involved data reuse method. To mitigate the design problems of complicated data flow on DNN accelerators, Network-on-Chip (NoC) interconnection has become an emerging technology to realize the Deep Neural Network on Chip (DNNoC). Compared with the conventional array-based DNN acceleration design, the DNNoC design supports flexible data flow, which leverages reconfigurable DNN accelerator implementations. In this work, we leverage the flexible NoC interconnection and propose a hybrid input/weight reuse method to reduce memory access. In addition, our proposed hybrid input/weight reuse method supports arbitrary kernel sizes for flexible convolutional operations. Compared with the related works, the proposed reconfigurable DNNoC with flexible convolutional operations helps to improve the utilization of computational capability in a PE by 1 % to 34 %, reduce memory access by 66% to 85%, which helps to improve 40% to 117% throughput.

Similar Papers
  • Conference Article
  • Citations6

Dynamic Mapping Mechanism to Compute DNN Models on a Resource-limited NoC Platform

  • Apr 19, 2021
  • Kun-Chih Jimmy Chen +3
  • Conference Article
  • Citations5

An Arbitrary Kernel-size Applicable NoC-based DNN Processor Design with Hybrid Data Reuse

  • Aug 09, 2021
  • Kun-Chih Jimmy Chen +2
  • Research Article
  • Citations5

NoCDAS: A Cycle-Accurate NoC-Based Deep Neural Network Accelerator Simulator

  • Jun 30, 2025
  • ACM Transactions on Modeling and Computer Simulation
  • Wenyao Zhu +2
  • Research Article
  • Citations12

An Error Compensation Technique for Low-Voltage DNN Accelerators

  • Dec 15, 2020
  • IEEE Transactions on Very Large Scale Integration (VLSI) Systems
  • Daehan Ji +2
  • Research Article

Securing DNN Acceleration From Off-Chip Memory Vulnerabilities With Low-Overhead Authenticated Encryption

  • Jan 01, 2026
  • IEEE Transactions on Very Large Scale Integration (VLSI) Systems
  • Kyungmi Lee +3
  • Research Article
  • Citations8

AutoAI2C: An Automated Hardware Generator for DNN Acceleration on Both FPGA and ASIC

  • Oct 01, 2024
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
  • Yongan Zhang +6
  • Conference Article
  • Citations2

An Energy-Efficient Deep Neural Network Accelerator Design

  • Nov 01, 2020
  • Jueun Jung +1
  • Research Article
  • Citations3

A Universal RRAM-Based DNN Accelerator With Programmable Crossbars Beyond MVM Operator

  • Jul 01, 2022
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
  • Zihan Zhang +5
  • Conference Article
  • Citations52

DNN-Chip Predictor: An Analytical Performance Predictor for DNN Accelerators with Various Dataflows and Hardware Architectures

  • Apr 02, 2020
  • Yang Zhao +5
  • Conference Article
  • Citations6

Fast FPGA-Based Emulation for ReRAM-Enabled Deep Neural Network Accelerator

  • May 01, 2021
  • Yongquan Shi +5
  • Research Article
  • Citations10

RiSA: A Reinforced Systolic Array for Depthwise Convolutions and Embedded Tensor Reshaping

  • Sep 17, 2021
  • ACM Transactions on Embedded Computing Systems
  • Hyungmin Cho
  • Conference Article
  • Citations34

Model Reverse-Engineering Attack using Correlation Power Analysis against Systolic Array Based Neural Network Accelerator

  • Oct 01, 2020
  • Kota Yoshida +4
  • Conference Article
  • Citations1

Simulators for Deep Neural Network Accelerator Design and Analysis: A Brief Review

  • Oct 29, 2021
  • Mijing Sun +6
  • Research Article
  • Citations3

Reliable Test Architecture With Test Cost Reduction for Systolic-Based DNN Accelerators

  • Mar 01, 2022
  • IEEE Transactions on Circuits & Systems II Express Briefs
  • Muhammad Ibtesam +4
  • Conference Article
  • Citations575

Timeloop: A Systematic Approach to DNN Accelerator Evaluation

  • Mar 01, 2019
  • Angshuman Parashar +9
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.