BMBenoit Miramond

Benoit Miramond

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

Benoit Miramond has a total of 62 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Beyond binary SNNs: a hardware-algorithm co-design and architectural evaluation for graded spike networks

    by Wenfei Song, Andrea Castagnetti, Pierre-Emmanuel Novac, Benoit Miramond, Alain Pegatoquet

    While Spiking Neural Networks (SNNs) offer a promising path toward energy-efficient edge intelligence, conventional Binary representations often suffer from high inference latency and discretization errors. Multi-level (graded) spike models address these limitations by increasing information density per pulse, yet they typically necessitate power-hungry Multiply-Accumulate units that nullify the energy advantages of neuromorphic hardware. In this work, we propose a hardware-algorithm co-design framework that leverages multi-level Integrate-and-Fire neurons coupled with a specialized multiplier-less architecture. By decomposing multi-level interactions into a cascaded shift-and-add logic, our hardware preserves the event-driven sparsity of SNNs while supporting high-precision activations. We evaluate our proposed accelerator on the Xilinx Zynq MPSoC ZCU102 platform using the Google Speech Commands (GSCs), CIFAR-10, and DVS-Gesture datasets. Experimental results demonstrate that the graded spike paradigm achieves up to a 6.7 × reduction in total spike activity through temporal information compression. Compared with the binary baseline benchmark on the DVS Gesture dataset, the proposed architecture reduces memory accesses by 68.59%, improves steady-state throughput from 33.08 FPS to 251 FPS, and increases energy efficiency from 9.01 to 54.96 GOPS W − 1 . Furthermore, by minimizing pipeline stalls and workload imbalance, our design achieves an Efficiency-per-Processing-Element of up to 3.05 GOPS W − 1 PE − 1 , thereby providing a competitive PE-level energy efficiency compared with representative SNN accelerators. This research provides a scalable pathway for deploying high-performance, ‘always-on’ neuromorphic systems in resource-constrained environments.

  • Predictive Data-Driven Energy Efficiency in a Field-Deployed Wireless Sensor Network

    by Imourane Abdoulaye, Anthony Vigneron, Laurent Rodriguez, Cécile Belleudy, Benoît Miramond

    The limited energy resources in wireless sensor networks (WSNs) is a critical challenge to long-term operation, particularly in remote areas where recharging batteries is not easily feasible. This paper proposes an energy-efficient approach based on Convolutional Neural Networks (CNNs) for on-device data prediction at the Cluster Head (CH) level. By substituting certain data transmissions with predicted sensor values, this method minimizes radio communication, the most energy-intensive task in sensor nodes, thereby extending the network’s lifetime. We evaluate the performance of this approach by analyzing two neural network structures and operational modes in an environmental monitoring context. Simulations with real-world sensor data show that the CNN model accurately predicts sensor values with minimal error. Additionally, the solution is validated through real-world deployment in a WSN, with real-time monitoring, energy-efficient communication, and frame management. A detailed energy profile and consumption assessment is provided, alongside an estimation of network autonomy. Results indicate a reduction in data transmissions by over 50%, enabling ≈ 34.3% of energy savings and ≈ 57.0% improvement in network autonomy without compromising accuracy. This solution offers a practical approach for sustainable, long-term WSN deployments.


  • Semi-Decentralized Prediction Method for Energy-Efficient Wireless Sensor Networks

    by Imourane Abdoulaye, Cécile Belleudy, Laurent Rodriguez, Benoît Miramond

    [graphicalabstract] Addressing key challenges in Wireless Sensor Networks (WSNs) such as network lifetime, and energy balance, this paper introduces the Semi-Decentralized Prediction Method (SDPM) for energy-efficient wireless sensor networks. This approach enhances energy efficiency by combining clustering principles with data prediction for smart Cluster-Head (CH) selection. SDPM facilitates the periodic election of an effective CH from among the cluster-nodes, who then predicts data for the nodes within the cluster, thereby reducing transmission and conserving energy. Our findings demonstrate SDPM's significant impact on reducing energy consumption, promising for real-world WSNs to achieve longer network lifetime and better energy management.


  • Open Access

    Pruning Self-Organizing Maps for Cellular Hardware Architectures

    by Andres Upegui, Bernard Girau, Nicolas Rougier, Fabien Vannel, Benot Miramond

    Self-organization is a bio-inspired feature that has been poorly developed when it comes to talking about hardware architectures. Cellular computing approaches have tackled it without considering input data. This paper introduces the SOMA architecture, which proposes an approach for self-organizing machine architectures. In order to achieve the desirable features for such machine, we propose PCSOM, a bio-inspired approach for self-organizing cellular hardware architectures in function of input data. PCSOM is a vector quantization algorithm defined as a network of neurons interconnected through synapses. Synapse pruning makes it possible to organize the cellular system architecture (i.e., topology and configuration of computing elements) in function of the content of input data. We present performance results of the algorithm and we discuss the benefits of PCSOM compared to other existing algorithms.


  • Confronting machine-learning with neuroscience for neuromorphic architectures design

    by Lyes Khacef, Nassim Abderrahmane, Benoit Miramond

    Artificial neural networks are experiencing today an unprecedented interest thanks to two main changes: the explosion of open data that is necessary for their training, and the increasing computing power of today's computers that makes the training part possible in a reasonable time. The recent results of deep neural networks on image classification has given neural networks the leading role in machine learning algorithms and artificial intelligence research. However, most applications such as smart devices or autonomous vehicles require an embedded implementation of neural networks. Their implementation in CPU/GPU remains too expensive, mostly in energy consumption, due to the non-adaptation of the hardware to the computation model, which becomes a limit to their use. It is therefore necessary to design neuromorphic architectures, i.e. hardware accelerators that fit to the parallel and distributed computation paradigm of neural networks for reducing their hardware cost implementation. We mainly focus on the optimization of energy consumption to enable integration in embedded systems. For this purpose, we implement two models of artificial neural networks coming from two different scientific domains: the multi- layer perceptron derived from machine learning and the spiking neural network inspired from neuroscience. We compare the performances of both approaches in terms of accuracy and hardware cost to find out the most attractive architecture for the design of embedded artificial intelligence.


  • Open Access

    Energy Saving in a Wireless Sensor Network by Data Prediction by using Self-Organized Maps

    by Adrien Russo, François Verdier, Benoît Miramond


  • Towards a Multi-mission QoS and Energy Manager for Autonomous Mobile Robots

    by Dinh-Khanh Ho, Karim Ben Chehida, Benoit Miramond, Michel Auguin

    An increasing demand on the autonomy of mobile robotic platforms requires adaptive on-line decisions reacting to the dynamic and unpredictable working environment with respect to the robotic mission requirements. These decisions are constrained by the limited computing resources and the energy that the mobile robot can carry. Moreover, in a multi-mission robotic context, these decisions must ensure a minimal quality of service (QoS) for each mission. This decision making problem leverages a characterization and monitoring phase of the system (computing and energy) resources and the robotic missions. In this article, we characterize, monitor and analyze our mobile robotic platform and an autonomous navigation mission as the principal mission of a mobile robot. The results of real experimentations on our robotic platform are analyzed and they highlight the necessity for an efficient mission manager.


  • Generation of schedule tables on multi-core systems for AUTOSAR applications

    by Wenhao Wang, Fabrice Camut, Benoit Miramond

    The recent migration from single-core to multi-core platforms in the automotive domain reveals great challenges for the legacy embedded software design flow. First of all, software designers need new methods to fill the gap between application description and tasks deployment. Secondly, the use of multiple cores has also to remain compatible with real-time and safety design constraints. Finally, developers need tools to assist them in the new steps of the design process. We propose in this paper a method integrated in the AUTOSAR design flow for the generation of schedule tables imposing the execution constraints onto a multi-core architecture. We consider an engine control application described in AUTOSAR as a set of periodic and dependent runnables. For that, we propose a formalization of periodic dependencies adapted to this automotive framework and present a scheduling algorithm taking into account this specificity. We leaded experiments with a complex and real world control application onto automotive multi-core systems. We present our working process enabling the generation of several execution schemes for this industrial application. The results show that our scheduling method benefits the multi-core migration to generate solutions best suited to safety standards by reducing the total jitter.