FANN-on-MCU: An Open-Source Toolkit for Energy-Efficient Neural Network Inference at the Edge of the Internet of Things
Open accessThe DOI matches its Crossref record; 58 of the 59 references it cites were found in a registry or an index.
Abstract
The growing number of low-power smart devices in the Internet of Things is coupled with the concept of “edge computing” that is moving some of the intelligence, especially machine learning, toward the edge of the network. Enabling machine learning algorithms to run on resource-constrained hardware, typically on low-power smart devices, is challenging in terms of hardware (optimized and energy-efficient integrated circuits), algorithmic, and firmware implementations. This article presents a FANN-on-MCU, an open-source toolkit built upon the fast artificial neural network (FANN) library to run lightweight and energy-efficient neural networks on microcontrollers based on both the ARM Cortex-M series and the novel RISC-V-based parallel ultralow-power (PULP) platform. The toolkit takes multilayer perceptrons trained with FANN and generates code targeted to low-power microcontrollers. This article also presents detailed analyses of energy efficiency across the different cores, and the optimizations to handle different network sizes. Moreover, it provides a detailed analysis of parallel speedups and degradations due to parallelization overhead and memory transfers. Further evaluations include experimental results for three different applications using a self-sustainable wearable multisensor bracelet. The experimental results show a measured latency in the order of only a few microseconds and power consumption of a few milliwatts while keeping the memory requirements below the limitations of the targeted microcontrollers. In particular, the parallel implementation on the octa-core RISC-V platform reaches a speedup of 22× and a 69% reduction in energy consumption with respect to a single-core implementation on Cortex-M4 for continuous real-time classification.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The text we found for this paper looks like another document, so the self-check wasn’t run.
Citations
Cites
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 2×
“Recent results on machine learning models demonstrate impressive classification accuracy, in some cases even outperforming humans [1].”
BackgroundInternet of Things for Smart Cities
Zanella, A., Bui, N., Castellani, A., Vangelista, L., & Zorzi, M. (2014) · IEEE Internet of Things Journal · doi:10.1109/jiot.2014.2306328 · Cited by 6,292 · cited 1×
“One essential feature of machine learning and in particular neural networks (NNs) is their flexibility that makes them suitable for a wide range of applications, including computer vision [2], [3], natural language processing [4], [5], biomedical [6], [7], and several others [8]–[11].”
BackgroundEdge Computing for the Internet of Things: A Case Study
Premsankar, G., Di Francesco, M., & Taleb, T. (2018) · IEEE Internet of Things Journal · doi:10.1109/jiot.2018.2805263 · Cited by 697 · cited 1×
“The reasons are the massive workload on the IoT network, the cost of the communication infrastructure, the required energy for data transmission, and, more generally, reliability, latency, and privacy concerns [15], [16].”
BackgroundNear-Threshold RISC-V Core With DSP Extensions for Scalable IoT Endpoint Devices
Gautschi, M., Schiavone, P. D., Traber, A., Loi, I., Pullini, A., Rossi, D., Flamand, E., Gurkaynak, F. K., & Benini, L. (2017) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2017.2654506 · Cited by 507
BLE Beacons for Internet of Things Applications: Survey, Challenges, and Opportunities
Jeon, K. E., She, J., Soonsawad, P., & Ng, P. C. (2018) · IEEE Internet of Things Journal · doi:10.1109/jiot.2017.2788449 · Cited by 391 · cited 1×
“Thus, bringing intelligence to the edge is creating fascinating challenges for industrial and academic researchers [10], [12], [21].”
BackgroundCan, Y. S., Chalabianloo, N., Ekiz, D., & Ersoy, C. (2019) · Sensors · doi:10.3390/s19081849 · Cited by 349 · cited 1×
“Many studies identified MLPs to be the best or one of the best algorithms to solve tasks in the IoT domain using wearable devices [57]–[60].”
BackgroundCMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs
Oliveira, A., Costa, D., Moreira, G., Martins, J., & Pinto, S. (2018) · arXiv (Cornell University) · doi:10.4230/oasics.ng-res.2025.4 · Cited by 288 · cited 1×
“In parallel, ARM has been developing CMSIS-NN [48], an additional library for their Cortex Microcontroller Software Interface Standard (CMSIS), in order to provide optimal-performance implementations of layers commonly present in deep NNs with speedups in the order of 4.6 × over a simple baseline…”
MethodInfluentialA Robust and Energy-Efficient Classifier Using Brain-Inspired Hyperdimensional Computing
Rahimi, A., Kanerva, P., & Rabaey, J. M. (2016) · Proceedings of the 2016 International Symposium on Low Power Electronics and Design · doi:10.1145/2934583.2934624 · Cited by 276 · cited 1×
“Today’s IoT devices host microcontrollers (MCUs), especially from the ARM Cortex-M family, which are able to achieve power consumption in the order of mW and computational resources in the order of hundreds of MOPS [1], [26], [27].”
BackgroundYodaNN: An Architecture for Ultralow Power Binary-Weight CNN Acceleration
Andri, R., Cavigelli, L., Rossi, D., & Benini, L. (2018) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2017.2682138 · Cited by 247 · cited 1×
“The most promising approaches to improve the performance of ultra-low-power processors are parallelism, low-power fixed-function hardware accelerators [32], [33], memory access optimization [34], and near-threshold technology [35], [36].”
BackgroundSelf-citationFrom Cloud Down to Things: An Overview of Machine Learning in Internet of Things
Samie, F., Bauer, L., & Henkel, J. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2893866 · Cited by 240 · cited 1×
“to the sensors where the data are collected [13].”
BackgroundHuman body heat for powering wearable devices: From thermal energy to application
Thielen, M., Sigrist, L., Magno, M., Hierold, C., & Benini, L. (2017) · Energy Conversion and Management · doi:10.1016/j.enconman.2016.11.005 · Cited by 233 · cited 4×
“A dual-source energy harvester for solar and thermoelectric generator (TEG) [51] modules has been included in the design.”
MethodInfluentialSelf-citationDavide Schiavone, P., Conti, F., Rossi, D., Gautschi, M., Pullini, A., Flamand, E., & Benini, L. (2017) · 2017 27th International Symposium on Power and Timing Modeling, Optimization and Simulation (PATMOS) · doi:10.1109/patmos.2017.8106976 · Cited by 228 · cited 14×
“In this case, the IBEX core is as fast as the Cortex-M4.”
MethodInfluentialA comparison of wearable fitness devices
Kaewkannate, K., & Kim, S. (2016) · BMC Public Health · doi:10.1186/s12889-016-3059-0 · Cited by 214
YodaNN: An Ultra-Low Power Convolutional Neural Network Accelerator Based on Binary Weights
Andri, R., Cavigelli, L., Rossi, D., & Benini, L. (2016) · 2016 IEEE Computer Society Annual Symposium on VLSI (ISVLSI) · doi:10.1109/isvlsi.2016.111 · Cited by 211 · cited 1×
“Lots of research efforts towards specialized hardware [22]–[24] and optimized inference algorithms [25] to run such NNs on power-constrained devices have been made over the last few years.”
BackgroundSelf-citationA 64-mW DNN-Based Visual Navigation Engine for Autonomous Nano-Drones
Palossi, D., Loquercio, A., Conti, F., Flamand, E., Scaramuzza, D., & Benini, L. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2917066 · Cited by 186
An Analysis of Visitors' Behavior in the Louvre Museum: A Study Using Bluetooth Data
Yoshimura, Y., Sobolevsky, S., Ratti, C., Girardin, F., Carrascal, J. P., Blat, J., & Sinatra, R. (2014) · Environment and Planning B: Planning and Design · doi:10.1068/b130047p · Cited by 181 · cited 1×
“filtering and/or compression [12]–[14].”
Origami: A 803-GOp/s/W Convolutional Network Accelerator
Cavigelli, L., & Benini, L. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2592330 · Cited by 169 · cited 1×
“ing approaches to improve the performance of ultralow-power processors are parallelism, low-power fixed-function hardware accelerators [32], [33], memory access optimization [34], and near-threshold technology [35], [36].”
BackgroundSelf-citationMr.Wolf: An Energy-Precision Scalable Parallel Ultra Low Power SoC for IoT Edge Processing
Pullini, A., Rossi, D., Loi, I., Tagliavini, G., & Benini, L. (2019) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2019.2912307 · Cited by 161 · cited 2×
“Wolf are done considering both the SoC and cluster domains with the processors running at 100 MHz since it is shown that at this frequency, the energy efficiency is maximized [41].”
BackgroundOrigami: A Convolutional Network Accelerator
Cavigelli, L., Gschwend, D., Mayer, C., Willi, S., Muheim, B., & Benini, L. (2015) · Proceedings of the 25th edition on Great Lakes Symposium on VLSI · doi:10.1145/2742060.2743766 · Cited by 156
Self-citationRault, T., Bouabdallah, A., Challal, Y., & Marin, F. (2016) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2016.08.003 · Cited by 128 · cited 1×
“For instance, the IoT device may perform the pre-processing of the data and transmit the intermediate results to the fog where the rest of the processing is performed [18]– [20].”
MethodMSML: A Novel Multilevel Semi-Supervised Machine Learning Framework for Intrusion Detection System
Yao, H., Fu, D., Zhang, P., Li, M., & Liu, Y. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2018.2873125 · Cited by 126 · cited 2×
“However, there are many application scenarios, especially dealing with low-bandwidth time-series recordings from low-power sensors, where multi-layer fully-connected networks are just as effective [11], [19], [28], [29].”
BackgroundEfficient FPGA Implementation of Multilayer Perceptron for Real-Time Human Activity Classification
Gaikwad, N. B., Tiwari, V., Keskar, A., & Shivaprakash, N. C. (2019) · IEEE Access · doi:10.1109/access.2019.2900084 · Cited by 114 · cited 4×
“showcases found in the literature using an MLP with different network sizes [55]–[57] in order to demonstrate the usability and the power efficiency of our proposed framework with”
BackgroundMethodInfluentialAccelerating real-time embedded scene labeling with convolutional networks
Cavigelli, L., Magno, M., & Benini, L. (2015) · Proceedings of the 52nd Annual Design Automation Conference · doi:10.1145/2744769.2744788 · Cited by 106
Self-citationSmart environments and Cultural Heritage: a novel approach to create intelligent cultural spaces
Chianese, A., Piccialli, F., & Valente, I. (2015) · Journal of Location Based Services · doi:10.1080/17489725.2015.1099752 · Cited by 105 · cited 1×
“One essential feature of machine learning and in particular neural networks (NNs) is their flexibility that makes them suitable for a wide range of applications, including computer vision [2], [3], natural language processing [4], [5], biomedical [6], [7], and several others [8]–[11].”
BackgroundA Novel Human Activity Recognition Scheme for Smart Health Using Multilayer Extreme Learning Machine
Chen, M., Li, Y., Luo, X., Wang, W., Wang, L., & Zhao, W. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2018.2856241 · Cited by 104 · cited 1×
“However, there are many application scenarios, especially dealing with low-bandwidth time-series recordings from low-power sensors, where multilayer fully connected networks are just as effective [17], [27], [38], [39].”
BackgroundInfiniTime: Multi-sensor wearable bracelet with human body harvesting
Magno, M., Brunelli, D., Sigrist, L., Andri, R., Cavigelli, L., Gomez, A., & Benini, L. (2016) · Sustainable Computing Informatics and Systems · doi:10.1016/j.suscom.2016.05.003 · Cited by 100 · cited 2×
“This energy can be used to prolong the battery lifetime of InfiniWolf or to achieve an energy-autonomous smart watch [53].”
BackgroundMethodSelf-citationWearable-Sensor-Based Classification Models of Faller Status in Older Adults
Howcroft, J., Lemaire, E. D., & Kofman, J. (2016) · PLoS ONE · doi:10.1371/journal.pone.0153240 · Cited by 89 · cited 3×
“In this section, we present three application showcases found in the literature using an MLP with different network sizes [56]–[58] in order to demonstrate the usability and the power efficiency of our proposed framework with the supported MCUs.”
BackgroundMethodYoshimura, Y., Cai, B., Wang, Z., & Ratti, C. (2019) · Lecture notes in geoinformation and cartography · doi:10.1007/978-3-030-19424-6_14 · Cited by 86
Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
Dong, Y., Bao, F., Su, H., & Zhu, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.09035 · Cited by 74
Lemoyne, R. C., Heerinckx, F., Aranca, T., Jager, R. D., Zesiewicz, T. A., & Saal, H. J. (2016) · Wearable and Implantable Body Sensor Networks · doi:10.1109/bsn.2016.7516249 · Cited by 71
Power-Constrained Edge Computing With Maximum Processing Capacity for IoT Networks
Qin, M., Chen, L., Zhao, N., Chen, Y., Yu, F. R., & Wei, G. (2018) · IEEE Internet of Things Journal · doi:10.1109/jiot.2018.2875218 · Cited by 70 · cited 2×
“For instance, the IoT device may perform the preprocessing of the data and transmit the intermediate results to the fog where the rest of the processing is performed [12]– [14].”
BackgroundMethodPosture and Activity Recognition and Energy Expenditure Estimation in a Wearable Platform
Sazonov, E. E., Hegde, N., Browning, R. C., Melanson, E. L., & Sazonova, N. A. (2015) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2015.2432454 · Cited by 64
CBinfer: Change-Based Inference for Convolutional Neural Networks on Video Data
Cavigelli, L., Degen, P., & Benini, L. (2017) · International Conference on Distributed Smart Cameras · doi:10.1145/3131885.3131906 · Cited by 63 · cited 1×
“Thus, bringing intelligence to the edge is creating fascinating challenges for industrial and academic researchers [10], [12], [21].”
BackgroundSelf-citationXNORBIN: A 95 TOp/s/W Hardware Accelerator for Binary Convolutional Neural Networks
Bahou, A. A., Karunaratne, G., Andri, R., Cavigelli, L., & Benini, L. (2018) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000329926 · Cited by 49
Self-citationDistributed Neural Networks for Internet of Things: The Big-Little Approach
Coninck, E. D., Verbelen, T., Vankeirsbilck, B., Bohez, S., Simoens, P., Demeester, P. M. A., & Dhoedt, B. (2016) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-319-47075-7_52 · Cited by 48 · cited 1×
“the onset and, once the onset is detected, a deeper network is used for classification [53], both domains (SoC and cluster)”
MethodEEG signal analysis for epileptic seizures detection by applying Data Mining techniques
Zazzaro, G., Cuomo, S., Martone, A., Montaquila, R. V., Toraldo, G., & Pavone, L. (2019) · Internet of Things · doi:10.1016/j.iot.2019.03.002 · Cited by 48 · cited 2×
“and sensors for several application scenarios [7], [18].”
BackgroundEBPC: Extended Bit-Plane Compression for Deep Neural Network Inference and Training Accelerators
Cavigelli, L., Rutishauser, G., & Benini, L. (2019) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2019.2950093 · Cited by 46 · cited 1×
“The most promising approaches to improve the performance of ultra-low-power processors are parallelism, low-power fixed-function hardware accelerators [32], [33], memory access optimization [34], and near-threshold technology [35], [36].”
BackgroundSelf-citationSelf-Sustaining Acoustic Sensor With Programmable Pattern Recognition for Underwater Monitoring
Mayer, P., Magno, M., & Benini, L. (2019) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2018.2890187 · Cited by 45
Self-citationDeepEmote: Towards multi-layer neural networks in a low power wearable multi-sensors bracelet
Magno, M., Pritz, M., Mayer, P., & Benini, L. (2017) · 2017 7th IEEE International Workshop on Advances in Sensors and Interfaces (IWASI) · doi:10.1109/iwasi.2017.7974208 · Cited by 44 · cited 2×
“However, there are many application scenarios, especially dealing with low-bandwidth time-series recordings from low-power sensors, where multilayer fully connected networks are just as effective [17], [27], [38], [39].”
BackgroundSelf-citationConti, F., Cavigelli, L., Paulin, G., Susmelj, I., & Benini, L. (2018) · 2018 IEEE Custom Integrated Circuits Conference (CICC) · doi:10.1109/cicc.2018.8357068 · Cited by 44
Self-citationDas, A. K., Catthoor, F., & Schaafsma, S. (2018) · IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies · doi:10.1145/3278576.3278598 · Cited by 41 · cited 1×
“MLPs have been successfully used in a wide range of application scenarios, such as disease detection [55], activity recognition [56], and brain-machine interface [57].”
MethodTowards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
Dong, Y., Su, H., Zhu, J. J., & Bao, F. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1708.05493 · Cited by 39 · cited 1×
“much less demanding than deep convolutional networks in terms of memory and computing requirements, and yet are widely used and very effective in many application areas such as medical data analysis and in general applications where hand-engineered feature extractors are preferred for interpretability [43].”
BackgroundA wearable device for physical and emotional health monitoring
Murali, S., Rincón, F. J., & Atienza, D. (2015) · Computing in cardiology · doi:10.1109/cic.2015.7408601 · Cited by 33 · cited 1×
“gines available at the edge of the IoT [21]–[24].”
BackgroundCavigelli, L., & Benini, L. (2019) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2903421 · Cited by 32 · cited 1×
“Lots of research efforts towards specialized hardware [22]–[24] and optimized inference algorithms [25] to run such NNs on power-constrained devices have been made over the last few years.”
BackgroundSelf-citationHR-SAR-Net: A Deep Neural Network for Urban Scene Segmentation from High-Resolution SAR Data
Wang, X., Cavigelli, L., Eggimann, M., Magno, M., & Luca Benini, C. (2020) · IEEE Sensors Applications Symposium (SAS) · doi:10.1109/sas48726.2020.9220068 · Cited by 28
Self-citationTowards Deep Learning using TensorFlow Lite on RISC-V
Louis, M. S., Azad, Z., Delshadtehrani, L., Gupta, S., Warden, P., Reddi, V. J., & Joshi, A. (2019) · doi:10.13140/rg.2.2.30400.89606 · Cited by 27
Alfaro, J. G. C., Ibrahim, A. H., & Trejos, A. L. (2019) · International Conference on Rehabilitation Robotics · doi:10.1109/icorr.2019.8779533 · Cited by 26 · cited 2×
“MLPs have been successfully used in a wide range of application scenarios, such as disease detection [55], activity recognition [56], and brain-machine interface [57].”
MethodA RISC-V Based Open Hardware Platform for Always-On Wearable Smart Sensing
Eggimann, M., Mach, S., Magno, M., & Benini, L. (2019) · International Workshop on Advances in Sensors and Interfaces (IWASI) · doi:10.1109/iwasi.2019.8791364 · Cited by 26 · cited 3×
“This article presents FANN-on-MCU: an open-source framework for easy deployment of NNs trained with the FANN library on both ARM Cortex-M cores and new parallel ultralow-power (PULP) RISC-V-based processors [37].”
BackgroundMethodSelf-citationInfiniWolf: Energy Efficient Smart Bracelet for Edge Computing with Dual Source Energy Harvesting
Magno, M., Wang, X., Eggimann, M., Cavigelli, L., & Luca Benini, C. (2020) · Design, Automation, and Test in Europe · doi:10.23919/date48585.2020.9116218 · Cited by 25 · cited 2×
“This results in very few examples of NNs that are running on milliwatt-powered MCUs, which are the most common compute engines available at the edge of the IoT [6], [28]–[31].”
BackgroundSelf-citationExtended Bit-Plane Compression for Convolutional Neural Network Accelerators
Cavigelli, L., & Benini, L. (2019) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas.2019.8771562 · Cited by 24 · cited 1×
“filtering and/or compression [12]–[14].”
Self-citationHyperdrive: A Systolically Scalable Binary-Weight CNN Inference Engine for mW IoT End-Nodes
Andri, R., Cavigelli, L., Rossi, D., & Benini, L. (2018) · IEEE Computer Society Annual Symposium on VLSI · doi:10.1109/isvlsi.2018.00099 · Cited by 21
Self-citationMapping Biological Systems to Network Systems
Rathore, H. (2016) · Springer eBooks · doi:10.1007/978-3-319-29782-8 · Cited by 20 · cited 1×
“pute engines available at the edge of the IoT [6], [28]–[31].”
BackgroundRay, J. S., Johnny, O., Trovati, M., Sotiriadis, S., & Bessis, N. (2018) · Big Data and Cognitive Computing · doi:10.3390/bdcc2030022 · Cited by 20
Hyperdrive: A Multi-Chip Systolically Scalable Binary-Weight CNN Inference Engine
Andri, R., Cavigelli, L., Rossi, D., & Benini, L. (2019) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2019.2905654 · Cited by 19 · cited 1×
“The most promising approaches to improve the performance of ultra-low-power processors are parallelism, low-power fixed-function hardware accelerators [32], [33], memory access optimization [34], and near-threshold technology [35], [36].”
BackgroundSelf-citationMagno, M., Cavigelli, L., Mayer, P., von Hagen, F., & Benini, L. (2019) · IEEE World Forum on Internet of Things (WF-IoT) · doi:10.1109/wf-iot.2019.8767290 · Cited by 11 · cited 2×
“It does not support training on GPUs, which simplifies the framework and is not required for networks of a size range suitable for deployment on low-power MCUs [44].”
BackgroundMethodSelf-citationWang, X., Magno, M., Cavigelli, L., Mahmud, M., Cecchetto, C., Vassanelli, S., & Benini, L. (2018) · IEEE Biomedical Circuits and Systems Conference (BioCAS) · doi:10.1109/biocas.2018.8584830 · Cited by 10 · cited 2×
“This results in very few examples of NNs that are running on milliwatt-powered MCUs, which are the most common compute engines available at the edge of the IoT [6], [28]–[31].”
BackgroundSelf-citationLong-term monitoring of small-sized birds using a miniaturized bluetooth-low-energy sensor node
Magno, M., Vultier, F., Szebedy, B., & Luca Benini, C. (2017) · Proceedings of IEEE Sensors/Proceedings of IEEE Sensors ... · doi:10.1109/icsens.2017.8233883 · Cited by 9 · cited 1×
“MLPs are commonly used as the classifier after applying suitable nonlearnable feature extractors due to their relatively moderately resource demand [42].”
MethodSelf-citationWang, X., Magno, M., Cavigelli, L., et al. (2018) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-b-000291157 · Cited by 4
Self-citation
Research works printed but not linked: 3
- 2019, MYO Gesture Control Armband—Wearable Technology by Thalmic Labs
- 2017, CMSIS-Cortex Microcontroller Software Interface Standard
- kuyumcu, 2007, FANN Tool
Cited by
Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
Wang, X., Han, Y., Leung, V. C. M., Niyato, D., Yan, X., & Chen, X. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2970550 · Cited by 1,498
Self-citationA review on TinyML: State-of-the-art and prospects
Ray, P. P. (2022) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2021.11.019 · Cited by 460
Ingolfsson, T. M., Hersche, M., Wang, X., Kobayashi, N., Cavigelli, L., & Benini, L. (2020) · 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC) · doi:10.1109/smc42975.2020.9283028 · Cited by 353
“However, accurate networks such as the TPCT model have 7.78 M train-able parameters and require 1.73 billion multiply-accumulate (MAC) operations per inference, which is out of reach of a typical low-power MCU with few MB of Flash and few hundreds of kB of RAM [18].”
BackgroundInfluentialSelf-citationKhalil, R. A., Saeed, N., Masood, M., Fard, Y. M., Alouini, M.-S., & Al-Naffouri, T. Y. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3051414 · Cited by 318
Edge Computing on IoT for Machine Signal Processing and Fault Diagnosis: A Review
Lu, S., Lu, J., An, K., Wang, X., & He, Q. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3239944 · Cited by 295
“One of the main reasons is that edge computing is highly depended on both hardware and algorithm, and the implementation of fault diagnosis methods on the specific processors is still a difficulty and challenge [7].”
BackgroundSelf-citation
Metadata
- Authors
- Wang, X., Magno, M., Cavigelli, L., & Benini, L.
- Affiliations
- ETH Zurich (CH)
- Venue
- IEEE Internet of Things Journal
- Published
- 2020-05
- Volume
- 7
- Issue
- 5
- Pages
- 4403-4417
- ISSN
- 2327-4662 (ISSN-L)
- 2372-2541
- OpenAlex
- W3007729836
- Cited by
- 162
- References
- 59
- Citations by year
- 2026: 12 · 2025: 18 · 2024: 18 · 2023: 32 · 2022: 33 · 2021: 34 · 2020: 15
- FWCI
- 13.39
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Internet of Things Journal (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Archivio istituzionale della ricerca (Alma Mater Studiorum Università di Bologna) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Funders
- Swiss Data Science Center Ph.D. Fellowship (ID P18-04)
- U.S. Office of Naval Research Global “Zero-Power Sensing for Underwater Monitoring,” (ONRG - NICOP - N62909-19-1-2018)
- Armasuisse Science and Technology
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- edge computing, Internet of Things, microcontroller, ARM Cortex-M, RISC-V, multilayer perceptron, code generation, parallelization overhead, memory transfer, real-time classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.