Haibing Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Haibing Wang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Cong Shi
Chongqing University
6 shared publications - Liyuan Liu
Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China
6 shared publications - Tengxiao Wang
School of Microelectronics and Communication Engineering, Chongqing University, Chongqing, China
5 shared publications - Yingcheng Lin
School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, China
4 shared publications - Zhengqing Zhong
School of Microelectronics and Communication Engineering, Chongqing University, Chongqing, China
4 shared publications - Mingju Chen
School of Automation and Information Engineering, Sichuan University of Science and Engineering, Yibin 644005, China
3 shared publications
Latest Publications
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. PP (10638210) , 1-5
A Tiny Neuromorphic Processor Enabling On-Chip Spike-Based Learning for Edge Intelligent System
Neuromorphic processors deploying spiking neural networks (SNNs) hold great potential for edge intelligence applications. However, achieving high accuracy, high throughput, and highly energy-efficient on-chip learning under limited energy and resource budgets remains a critical challenge. To address this, we propose a tiny neuromorphic architecture to support a fully spike-based backpropagation learning algorithm tailored for a lightweight two-layer SNN aiming at edge scenarios. To enhance processing efficiency, this architecture tightly couples control and data flows, and features fully spike-based inference and learning. Furthermore, we propose a structured weight mapping scheme that satisfies different parallel memory access patterns during forward and backward passes, and introduce a group-based event-driven paradigm to exploit spike sparsity. A prototype was implemented on a very-low-cost Xilinx Zynq-7010 FPGA device. It achieved high on-chip learning accuracies comparable to state-of-the-art works across multiple benchmark datasets, while delivering a throughput of 2269/9082 frames/s and an energy efficiency of 56/ $14~\mu $ J/image for learning/inferring MNIST images, which outperforms existing approaches and shows its potential for wide deployments in extreme edge intelligence scenarios.
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. PP (10638210) , 1-5
A Tiny Neuromorphic Processor Enabling On-Chip Spike-Based Learning for Edge Intelligent System
Neuromorphic processors deploying spiking neural networks (SNNs) hold great potential for edge intelligence applications. However, achieving high accuracy, high throughput, and highly energy-efficient on-chip learning under limited energy and resource budgets remains a critical challenge. To address this, we propose a tiny neuromorphic architecture to support a fully spike-based backpropagation learning algorithm tailored for a lightweight two-layer SNN aiming at edge scenarios. To enhance processing efficiency, this architecture tightly couples control and data flows, and features fully spike-based inference and learning. Furthermore, we propose a structured weight mapping scheme that satisfies different parallel memory access patterns during forward and backward passes, and introduce a group-based event-driven paradigm to exploit spike sparsity. A prototype was implemented on a very-low-cost Xilinx Zynq-7010 FPGA device. It achieved high on-chip learning accuracies comparable to state-of-the-art works across multiple benchmark datasets, while delivering a throughput of 2269/9082 frames/s and an energy efficiency of 56/ $14~\mu $ J/image for learning/inferring MNIST images, which outperforms existing approaches and shows its potential for wide deployments in extreme edge intelligence scenarios.
- 20 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (12) , 2012-2016
MorphBungee-Tiny: An Ultra-Low-Cost FPGA Implementation of Neuromorphic Architecture With High-Accuracy On-Chip E-Prop Learning
Neuromorphic processors leverage brain-inspired spike-based sparse computation to gain energy efficiency for edge-node intelligence. Yet, achieving high-accuracy on-chip learning for in-situ self-adaptation at a low cost and a low latency poses a key challenge. To address this, we present the 4th generation of our MorphBungee processors. The processor adopts our integer e-prop algorithm for high-accuracy, low-cost on-chip SNN training. It employs a vector-scalar dual-core architecture with 2 levels of pipelines and the spike/error-event-driven paradigm to ensure a high sample throughput. A logically transposable memory enables the vector core’s flexible parallelism along the neuron- or the synapse-dimension in different processing phases. We prototyped our design on a very-low-cost Xilinx Zynq-7010 FPGA chip, and demonstrated its overall performance superiority over other related works, in terms of on-chip learning accuracy, processing speed, resource cost and energy efficiency.
- 20 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (12) , 2012-2016
MorphBungee-Tiny: An Ultra-Low-Cost FPGA Implementation of Neuromorphic Architecture With High-Accuracy On-Chip E-Prop Learning
Neuromorphic processors leverage brain-inspired spike-based sparse computation to gain energy efficiency for edge-node intelligence. Yet, achieving high-accuracy on-chip learning for in-situ self-adaptation at a low cost and a low latency poses a key challenge. To address this, we present the 4th generation of our MorphBungee processors. The processor adopts our integer e-prop algorithm for high-accuracy, low-cost on-chip SNN training. It employs a vector-scalar dual-core architecture with 2 levels of pipelines and the spike/error-event-driven paradigm to ensure a high sample throughput. A logically transposable memory enables the vector core’s flexible parallelism along the neuron- or the synapse-dimension in different processing phases. We prototyped our design on a very-low-cost Xilinx Zynq-7010 FPGA chip, and demonstrated its overall performance superiority over other related works, in terms of on-chip learning accuracy, processing speed, resource cost and energy efficiency.
- 19 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (7) , 943-947
A Visual-Cortex-Mimetic Tiny Neuromorphic Vision Processor Based on Reconfigurable Cortical Neuron Unit
Edge visual systems demand high energy-efficiency vision processors like neuromorphic hardware leveraging spikebased computations. But their disability of directly interacting with non-spike information in the real world requests additional components to execute image pre-processing, spike encoding and decoding, severely increasing overall system cost, energy and latency. To overcome such drawback, this brief proposes a tiny neuromorphic vision processor which emulates functional regions along the ventral pathway in the visual cortex. It performs image pre-processing, spike encoding, spike-based feature extraction and classification, spike decoding as well as decision making on a single chip. To reduce hardware resources, our processor builds on a reconfigurable cortical neuron (RCN) unit, which runs different neuron models for different visual cortex regions in a timemultiplexing fashion. It also embeds biological learning circuits to better adapt the processor to dynamic edge scenarios. Our neuromorphic processor was prototyped on a very-low-cost Xilinx Zynq-7010 device. On the MNIST dataset, it exhibited a real-time inference speed of 696 frame/s covering image pre-processing to final decision and a high on-chip learning accuracy of 97.12%, while only delivering a power consumption as low as 118 mW.
- 19 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (7) , 943-947
A Visual-Cortex-Mimetic Tiny Neuromorphic Vision Processor Based on Reconfigurable Cortical Neuron Unit
Edge visual systems demand high energy-efficiency vision processors like neuromorphic hardware leveraging spikebased computations. But their disability of directly interacting with non-spike information in the real world requests additional components to execute image pre-processing, spike encoding and decoding, severely increasing overall system cost, energy and latency. To overcome such drawback, this brief proposes a tiny neuromorphic vision processor which emulates functional regions along the ventral pathway in the visual cortex. It performs image pre-processing, spike encoding, spike-based feature extraction and classification, spike decoding as well as decision making on a single chip. To reduce hardware resources, our processor builds on a reconfigurable cortical neuron (RCN) unit, which runs different neuron models for different visual cortex regions in a timemultiplexing fashion. It also embeds biological learning circuits to better adapt the processor to dynamic edge scenarios. Our neuromorphic processor was prototyped on a very-low-cost Xilinx Zynq-7010 device. On the MNIST dataset, it exhibited a real-time inference speed of 696 frame/s covering image pre-processing to final decision and a high on-chip learning accuracy of 97.12%, while only delivering a power consumption as low as 118 mW.
- 15 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (3) , 509-513
An Edge Neuromorphic Processor With High-Accuracy On-Chip Aggregate-Label Learning
Neuromorphic computing with bio-inspired spiking neural networks (SNNs) offers an energy-efficient paradigm for edge intelligence. But, achieving practically high on-chip SNN learning accuracy with limited hardware resources still remains challenging. To tackle this issue, this brief proposes an edge neuromorphic processor architecture enabling computationally-simple optimized aggregate-label (AL) algorithm to realize high-accuracy on-chip learning. To maximize utilization of multicore resources and minimize processing latency, our processor adopts techniques including uniform neuron-core mapping, layer-alternating workflow, time-step pipeline and event-driven scheme. We benchmarked our processor on an FPGA device, and attained high on-chip learning accuracies of 97.21%, 88.1%, 90.92%, 100% and 99.22% on MNIST, Fashion-MNIST, ETH-80, ORL-10, and Yale-10 datasets, respectively, using a small 2-layer fully-connected (FC) SNN, with a moderate resource cost and a relatively high energy efficiency. These results indicate that our design is very useful for many self-adaptive edge systems.
- 15 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (3) , 509-513
An Edge Neuromorphic Processor With High-Accuracy On-Chip Aggregate-Label Learning
Neuromorphic computing with bio-inspired spiking neural networks (SNNs) offers an energy-efficient paradigm for edge intelligence. But, achieving practically high on-chip SNN learning accuracy with limited hardware resources still remains challenging. To tackle this issue, this brief proposes an edge neuromorphic processor architecture enabling computationally-simple optimized aggregate-label (AL) algorithm to realize high-accuracy on-chip learning. To maximize utilization of multicore resources and minimize processing latency, our processor adopts techniques including uniform neuron-core mapping, layer-alternating workflow, time-step pipeline and event-driven scheme. We benchmarked our processor on an FPGA device, and attained high on-chip learning accuracies of 97.21%, 88.1%, 90.92%, 100% and 99.22% on MNIST, Fashion-MNIST, ETH-80, ORL-10, and Yale-10 datasets, respectively, using a small 2-layer fully-connected (FC) SNN, with a moderate resource cost and a relatively high energy efficiency. These results indicate that our design is very useful for many self-adaptive edge systems.
- 30 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (1) , 293-297
MorphBungee-Lite: An Edge Neuromorphic Architecture With Balanced Cross-Core Workloads Based on Layer-Wise Event-Batch Learning/Inference
Neuromorphic processors are promising candidates for energy-constrained intelligent systems, as they emulate cortical computations via spatiotemporally sparse binary spikes. However, achieving high-accuracy, high-throughput and cost-efficient neuromorphic processing remains challenging. To fully utilize hardware resources for performance improvement, we propose a multi-core neuromorphic architecture characteristic of a uniform neuron-core mapping scheme and a layer-wise event-batch-based parallel processing paradigm. These techniques ensure highly balanced cross-core workloads regardless of actual mapped neural network topologies as well as unpredictable input and internally generated spike counts varying from sample to sample. An FPGA prototype of our neuromorphic processor was implemented. It exhibited comparably high on-chip learning accuracies on various visual and non-visual benchmarks, high learning/inference frame rates (low processing latencies), with a moderate amount of logic and memory resource consumptions.
- 30 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 72 (1) , 293-297
MorphBungee-Lite: An Edge Neuromorphic Architecture With Balanced Cross-Core Workloads Based on Layer-Wise Event-Batch Learning/Inference
Neuromorphic processors are promising candidates for energy-constrained intelligent systems, as they emulate cortical computations via spatiotemporally sparse binary spikes. However, achieving high-accuracy, high-throughput and cost-efficient neuromorphic processing remains challenging. To fully utilize hardware resources for performance improvement, we propose a multi-core neuromorphic architecture characteristic of a uniform neuron-core mapping scheme and a layer-wise event-batch-based parallel processing paradigm. These techniques ensure highly balanced cross-core workloads regardless of actual mapped neural network topologies as well as unpredictable input and internally generated spike counts varying from sample to sample. An FPGA prototype of our neuromorphic processor was implemented. It exhibited comparably high on-chip learning accuracies on various visual and non-visual benchmarks, high learning/inference frame rates (low processing latencies), with a moderate amount of logic and memory resource consumptions.
- 11 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Circuits and Systems
- Vol. 19 (1) , 209-225
MorphBungee: A 65-nm 7.2-mm 2 27-µJ/Image Digital Edge Neuromorphic Chip With on-Chip 802-Frame/s Multi-Layer Spiking Neural Network Learning
This paper presents a digital edge neuromorphic spiking neural network (SNN) processor chip for a variety of edge intelligent cognitive applications. This processor allows high-speed, high-accuracy and fully on-chip spike-timing-based multi-layer SNN learning. It is characteristic of hierarchical multi-core architecture, event-driven processing paradigm, meta-crossbar for efficient spike communication, and hybrid and reconfigurable parallelism. A prototype chip occupying an active silicon area of 7.2 mm 2 was fabricated using a 65-nm 1P9M CMOS process. when running a 256-256-256-256-200 4-layer fully-connected SNN on downscaled 16 × 16 MNIST images. it typically achieved a high-speed throughput of 802 and 2270 frames/s for on-chip learning and inference, respectively, with a relatively low power dissipation of around 61 mW at a 100 MHz clock rate under a 1.0V core power supply, Our on-chip learning results in comparably high visual recognition accuracies of 96.06%, 83.38%, 84.53%, 99.22% and 100% on the MNIST, Fashion-MNIST, ETH-80, Yale-10 and ORL-10 datasets, respectively. In addition, we have successfully applied our neuromorphic chip to demonstrate high-resolution satellite cloud image segmentation and non-visual tasks including olfactory classification and textural news categorization. These results indicate that our neuromorphic chip is suitable for various intelligent edge systems under restricted cost, energy and latency budgets while requiring in-situ self-adaptative learning capability.
- 11 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Biomedical Circuits and Systems
- Vol. 19 (1) , 209-225
MorphBungee: A 65-nm 7.2-mm 2 27-µJ/Image Digital Edge Neuromorphic Chip With on-Chip 802-Frame/s Multi-Layer Spiking Neural Network Learning
This paper presents a digital edge neuromorphic spiking neural network (SNN) processor chip for a variety of edge intelligent cognitive applications. This processor allows high-speed, high-accuracy and fully on-chip spike-timing-based multi-layer SNN learning. It is characteristic of hierarchical multi-core architecture, event-driven processing paradigm, meta-crossbar for efficient spike communication, and hybrid and reconfigurable parallelism. A prototype chip occupying an active silicon area of 7.2 mm 2 was fabricated using a 65-nm 1P9M CMOS process. when running a 256-256-256-256-200 4-layer fully-connected SNN on downscaled 16 × 16 MNIST images. it typically achieved a high-speed throughput of 802 and 2270 frames/s for on-chip learning and inference, respectively, with a relatively low power dissipation of around 61 mW at a 100 MHz clock rate under a 1.0V core power supply, Our on-chip learning results in comparably high visual recognition accuracies of 96.06%, 83.38%, 84.53%, 99.22% and 100% on the MNIST, Fashion-MNIST, ETH-80, Yale-10 and ORL-10 datasets, respectively. In addition, we have successfully applied our neuromorphic chip to demonstrate high-resolution satellite cloud image segmentation and non-visual tasks including olfactory classification and textural news categorization. These results indicate that our neuromorphic chip is suitable for various intelligent edge systems under restricted cost, energy and latency budgets while requiring in-situ self-adaptative learning capability.