Cloud–Edge-Based Lightweight Temporal Convolutional Networks for Remaining Useful Life Prediction in IIoT

Ren, Lei; Liu, Yuxin; Wang, Xiaokang; Lü, Jinhu; Deen, M. Jamal

doi:10.1109/jiot.2020.3008170

“…where x and y are the input and output feature maps, t and m the output time-step and channel respectively, W the filter weights, C in the number of input channels, d the dilation factor, and K the filter size. In the original paper [8], TCNs were proposed as fully-convolutional architectures, but modern embodiments also include other common layers such as pooling and linear ones [10], [11].…”

Section: Background and Related Work A Temporal Convolutional Networkmentioning

confidence: 99%

Robust and Energy-efficient PPG-based Heart-Rate Monitoring

Risso,

Burrello,

Pagliari

et al. 2022

Preprint

0

View full text Add to dashboard Cite

A wrist-worn PPG sensor coupled with a lightweight algorithm can run on a MCU to enable non-invasive and comfortable monitoring, but ensuring robust PPG-based heart-rate monitoring in the presence of motion artifacts is still an open challenge. Recent state-of-the-art algorithms combine PPG and inertial signals to mitigate the effect of motion artifacts. However, these approaches suffer from limited generality. Moreover, their deployment on MCU-based edge nodes has not been investigated. In this work, we tackle both the aforementioned problems by proposing the use of hardware-friendly Temporal Convolutional Networks (TCN) for PPG-based heart estimation. Starting from a single "seed" TCN, we leverage an automatic Neural Architecture Search (NAS) approach to derive a rich family of models. Among them, we obtain a TCN that outperforms the previous state-of-theart on the largest PPG dataset available (PPGDalia), achieving a Mean Absolute Error (MAE) of just 3.84 Beats Per Minute (BPM). Furthermore, we tested also a set of smaller yet still accurate (MAE of 5.64 -6.29 BPM) networks that can be deployed on a commercial MCU (STM32L4) which require as few as 5k parameters and reach a latency of 17.1 ms consuming just 0.21 mJ per inference.

show abstract

“…where x and y are the input and output feature maps, t and m the output time-step and channel respectively, W the filter weights, C in the number of input channels, d the dilation factor, and K the filter size. In the original paper [8], TCNs were proposed as fully-convolutional architectures, but modern embodiments also include other common layers such as pooling and linear ones [10], [11].…”

Section: Background and Related Work A Temporal Convolutional Networkmentioning

confidence: 99%

Robust and Energy-Efficient PPG-Based Heart-Rate Monitoring

Risso

¹

,

Burrello

²

,

Pagliari

³

et al. 2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

A wrist-worn PPG sensor coupled with a lightweight algorithm can run on a MCU to enable non-invasive and comfortable monitoring, but ensuring robust PPG-based heart-rate monitoring in the presence of motion artifacts is still an open challenge. Recent state-of-the-art algorithms combine PPG and inertial signals to mitigate the effect of motion artifacts. However, these approaches suffer from limited generality. Moreover, their deployment on MCU-based edge nodes has not been investigated. In this work, we tackle both the aforementioned problems by proposing the use of hardware-friendly Temporal Convolutional Networks (TCN) for PPG-based heart estimation. Starting from a single "seed" TCN, we leverage an automatic Neural Architecture Search (NAS) approach to derive a rich family of models. Among them, we obtain a TCN that outperforms the previous state-of-theart on the largest PPG dataset available (PPGDalia), achieving a Mean Absolute Error (MAE) of just 3.84 Beats Per Minute (BPM). Furthermore, we tested also a set of smaller yet still accurate (MAE of 5.64 -6.29 BPM) networks that can be deployed on a commercial MCU (STM32L4) which require as few as 5k parameters and reach a latency of 17.1 ms consuming just 0.21 mJ per inference.

show abstract

“…Originally, TCNs have been proposed as fully-convolutional architectures that stacked multiple layers each implementing (1) [29]. However, more recent implementations also include other elements, such as pooling and fully-connected (FC) layers, which are analogous to those commonly found in standard CNNs [20], [31]. In our experiments, we consider TCN architectures that include all these types of layers.…”

Section: A Temporal Convolutional Networkmentioning

confidence: 99%

Q-PPG: Energy-Efficient PPG-Based Heart Rate Monitoring on Wearable Devices

Burrello

¹

,

Pagliari

²

,

Risso

³

et al. 2021

IEEE Trans. Biomed. Circuits Syst.

View full text Add to dashboard Cite

Hearth Rate (HR) monitoring is increasingly performed in wrist-worn devices using low-cost photoplethysmography (PPG) sensors. However, Motion Artifacts (MAs) caused by movements of the subject's arm affect the performance of PPG-based HR tracking. This is typically addressed coupling the PPG signal with acceleration measurements from an inertial sensor. Unfortunately, most standard approaches of this kind rely on hand-tuned parameters, which impair their generalization capabilities and their applicability to real data in the field. In contrast, methods based on deep learning, despite their better generalization, are considered to be too complex to deploy on wearable devices.In this work, we tackle these limitations, proposing a design space exploration methodology to automatically generate a rich family of deep Temporal Convolutional Networks (TCNs) for HR monitoring, all derived from a single "seed" model. Our flow involves a cascade of two Neural Architecture Search (NAS) tools and a hardware-friendly quantizer, whose combination yields both highly accurate and extremely lightweight models. When tested on the PPG-Dalia dataset, our most accurate model sets a new state-of-the-art in Mean Absolute Error. Furthermore, we deploy our TCNs on an embedded platform featuring a STM32WB55 microcontroller, demonstrating their suitability for real-time execution. Our most accurate quantized network achieves 4.41 Beats Per Minute (BPM) of Mean Absolute Error (MAE), with an energy consumption of 47.65 mJ and a memory footprint of 412 kB. At the same time, the smallest network that obtains a MAE < 8 BPM, among those generated by our flow, has a memory footprint of 1.9 kB and consumes just 1.79 mJ per inference.

show abstract

Cloud–Edge-Based Lightweight Temporal Convolutional Networks for Remaining Useful Life Prediction in IIoT

Cited by 94 publications

References 26 publications

Robust and Energy-efficient PPG-based Heart-Rate Monitoring

Robust and Energy-efficient PPG-based Heart-Rate Monitoring

Robust and Energy-Efficient PPG-Based Heart-Rate Monitoring

Q-PPG: Energy-Efficient PPG-Based Heart Rate Monitoring on Wearable Devices

Contact Info

Product

Resources

About