Advances in Embedded Artificial Intelligence and Internet-of-Things

A special issue of Journal of Low Power Electronics and Applications (ISSN 2079-9268).

Deadline for manuscript submissions: closed (28 February 2023) | Viewed by 15662

Special Issue Editors

Department of Computer Science, National Yang Ming Chiao Tung University, Hsinchu 300, Taiwan
Interests: design of low-power/high-performance/cost-effective adaptive filter; computer arithmetic; independent component analysis (ICA); multi-dimensional filter; transform; 3-D graphics system; intelligent elevator system; UAV; wearable data fusion system
Adaptive Systems Laboratory, University of Aizu, Aizuwakamatsu 965-8580, Japan
Interests: three-dimensional integrated circuits; networks-on-chip; reliability; neuromorphic computing; computer architecture
Special Issues, Collections and Topics in MDPI journals
Department of Computer Science and Engineering, National Sun Yat-Sen University, Kaohsiung 80421, Taiwan
Interests: multiprocessor SoC (MPSoC) design; neural network learning algorithm design; reliable system design; VLSI/CAD design
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Since edge computing plays an important role in the computing communicating world, embedded artificial intelligence (AI) and embedded Internet of Things (IoT) are cutting-edge research topics. Embedded AI and IoT require lightweight computation and communication complexity and green power/energy with satisfactory accuracy and quality in terms of algorithm, architecture, integrated circuit, system, standard, and application levels. The embedded AI research topics can cover issues of lightweight machine learning, especially for state-of-the-art deep learning. Embedded IoT can include issues of green cyber-physical communications and network systems.

This Special Issue collaborates with the IEEE 15th International Symposium on Embedded Multicore/Manycore Systems-on-Chip (MCSoC), Malaysia, 19–22 December 2022 (https://www.mcsoc-forum.org/). Selected papers from IEEE MCSoC 2022 and the external submissions (not limited to IEEE MCSoC2022) related to the above research topics are welcome in this Special Issue.

Prof. Dr. Lan-Da Van
Dr. Khanh N. Dang
Dr. Kun-Chih Chen
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Journal of Low Power Electronics and Applications is an international peer-reviewed open access quarterly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Related Special Issue

Published Papers (6 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

16 pages, 2456 KiB  
Article
Efficient GEMM Implementation for Vision-Based Object Detection in Autonomous Driving Applications
by Fatima Zahra Guerrouj, Sergio Rodríguez Flórez, Mohamed Abouzahir, Abdelhafid El Ouardi and Mustapha Ramzi
J. Low Power Electron. Appl. 2023, 13(2), 40; https://doi.org/10.3390/jlpea13020040 - 06 Jun 2023
Cited by 1 | Viewed by 1581
Abstract
Convolutional Neural Networks (CNNs) have been incredibly effective for object detection tasks. YOLOv4 is a state-of-the-art object detection algorithm designed for embedded systems. It is based on YOLOv3 and has improved accuracy, speed, and robustness. However, deploying CNNs on embedded systems such as [...] Read more.
Convolutional Neural Networks (CNNs) have been incredibly effective for object detection tasks. YOLOv4 is a state-of-the-art object detection algorithm designed for embedded systems. It is based on YOLOv3 and has improved accuracy, speed, and robustness. However, deploying CNNs on embedded systems such as Field Programmable Gate Arrays (FPGAs) is difficult due to their limited resources. To address this issue, FPGA-based CNN architectures have been developed to improve the resource utilization of CNNs, resulting in improved accuracy and speed. This paper examines the use of General Matrix Multiplication Operations (GEMM) to accelerate the execution of YOLOv4 on embedded systems. It reviews the most recent GEMM implementations and evaluates their accuracy and robustness. It also discusses the challenges of deploying YOLOv4 on autonomous vehicle datasets. Finally, the paper presents a case study demonstrating the successful implementation of YOLOv4 on an Intel Arria 10 embedded system using GEMM. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

26 pages, 18945 KiB  
Article
Nanomaterial-Based Sensor Array Signal Processing and Tuberculosis Classification Using Machine Learning
by Chenxi Liu, Israel Cohen, Rotem Vishinkin and Hossam Haick
J. Low Power Electron. Appl. 2023, 13(2), 39; https://doi.org/10.3390/jlpea13020039 - 29 May 2023
Viewed by 1718
Abstract
Tuberculosis (TB) has long been recognized as a significant health concern worldwide. Recent advancements in noninvasive wearable devices and machine learning (ML) techniques have enabled rapid and cost-effective testing for the real-time detection of TB. However, small datasets are often encountered in biomedical [...] Read more.
Tuberculosis (TB) has long been recognized as a significant health concern worldwide. Recent advancements in noninvasive wearable devices and machine learning (ML) techniques have enabled rapid and cost-effective testing for the real-time detection of TB. However, small datasets are often encountered in biomedical and chemical engineering domains, which can hinder the success of ML models and result in overfitting issues. To address this challenge, we propose various data preprocessing methods and ML approaches, including long short-term memory (LSTM), convolutional neural network (CNN), Gramian angular field-CNN (GAF-CNN), and multivariate time series with MinCutPool (MT-MinCutPool), for classifying a small TB dataset consisting of multivariate time series (MTS) sensor signals. Our proposed methods are compared with state-of-the-art models commonly used in MTS classification (MTSC) tasks. We find that lightweight models are more appropriate for small-dataset problems. Our experimental results demonstrate that the average performance of our proposed models outperformed the baseline methods in all aspects. Specifically, the GAF-CNN model achieved the highest accuracy of 0.639 and the highest specificity of 0.777, indicating its superior effectiveness for MTSC tasks. Furthermore, our proposed MT-MinCutPool model surpassed the baseline MTPool model in all evaluation metrics, demonstrating its viability for MTSC tasks. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

14 pages, 3125 KiB  
Article
Templatized Fused Vector Floating-Point Dot Product for High-Level Synthesis
by Dionysios Filippas, Chrysostomos Nicopoulos and Giorgos Dimitrakopoulos
J. Low Power Electron. Appl. 2022, 12(4), 56; https://doi.org/10.3390/jlpea12040056 - 17 Oct 2022
Cited by 1 | Viewed by 2333
Abstract
Machine-learning accelerators rely on floating-point matrix and vector multiplication kernels. To reduce their cost, customized many-term fused architectures are preferred, which improve the latency, power, and area of the designs. In this work, we design a parameterized fused many-term floating-point dot product architecture [...] Read more.
Machine-learning accelerators rely on floating-point matrix and vector multiplication kernels. To reduce their cost, customized many-term fused architectures are preferred, which improve the latency, power, and area of the designs. In this work, we design a parameterized fused many-term floating-point dot product architecture that is ready for high-level synthesis. In this way, we can exploit the efficiency offered by a well-structured fused dot-product architecture and the freedom offered by high-level synthesis in tuning the design’s pipeline to the selected floating-point format and architectural constraints. When compared with optimized dot-product units implemented directly in RTL, the proposed design offers lower-latency implementations under the same clock frequency with marginal area savings. This result holds for a variety of floating-point formats, including standard and reduced-precision representations. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

33 pages, 1927 KiB  
Article
Multi-Objective Resource Scheduling for IoT Systems Using Reinforcement Learning
by Shaswot Shresthamali, Masaaki Kondo and Hiroshi Nakamura
J. Low Power Electron. Appl. 2022, 12(4), 53; https://doi.org/10.3390/jlpea12040053 - 08 Oct 2022
Cited by 2 | Viewed by 2201
Abstract
IoT embedded systems have multiple objectives that need to be maximized simultaneously. These objectives conflict with each other due to limited resources and tradeoffs that need to be made. This requires multi-objective optimization (MOO) and multiple Pareto-optimal solutions are possible. In such a [...] Read more.
IoT embedded systems have multiple objectives that need to be maximized simultaneously. These objectives conflict with each other due to limited resources and tradeoffs that need to be made. This requires multi-objective optimization (MOO) and multiple Pareto-optimal solutions are possible. In such a case, tradeoffs are made w.r.t. a user-defined preference. This work presents a general Multi-objective Reinforcement Learning (MORL) framework for MOO of IoT embedded systems. This framework comprises a general Multi-objective Markov Decision Process (MOMDP) formulation and two novel low-compute MORL algorithms. The algorithms learn policies to tradeoff between multiple objectives using a single preference parameter. We take the energy scheduling problem in general Energy Harvesting Wireless Sensor Nodes (EHWSNs) as a case example in which a sensor node is required to maximize its sensing rate, and transmission performance as well as ensure long-term uninterrupted operation within a very tight energy budget. We simulate single-task and dual-task EHWSN systems to evaluate our framework. The results demonstrate that our MORL algorithms can learn better policies at lower learning costs and successfully tradeoff between multiple objectives at runtime. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

16 pages, 13796 KiB  
Article
BIoU: An Improved Bounding Box Regression for Object Detection
by Niranjan Ravi, Sami Naqvi and Mohamed El-Sharkawy
J. Low Power Electron. Appl. 2022, 12(4), 51; https://doi.org/10.3390/jlpea12040051 - 28 Sep 2022
Cited by 7 | Viewed by 3178
Abstract
Object detection is a predominant challenge in computer vision and image processing to detect instances of objects of various classes within an image or video. Recently, a new domain of vehicular platforms, e-scooters, has been widely used across domestic and urban environments. The [...] Read more.
Object detection is a predominant challenge in computer vision and image processing to detect instances of objects of various classes within an image or video. Recently, a new domain of vehicular platforms, e-scooters, has been widely used across domestic and urban environments. The driving behavior of e-scooter users significantly differs from other vehicles on the road, and their interactions with pedestrians are also increasing. To ensure pedestrian safety and develop an efficient traffic monitoring system, a reliable object detection system for e-scooters is required. However, existing object detectors based on IoU loss functions suffer various drawbacks when dealing with densely packed objects or inaccurate predictions. To address this problem, a new loss function, balanced-IoU (BIoU), is proposed in this article. This loss function considers the parameterized distance between the centers and the minimum and maximum edges of the bounding boxes to address the localization problem. With the help of synthetic data, a simulation experiment was carried out to analyze the bounding box regression of various losses. Extensive experiments have been carried out on a two-stage object detector, MASK_RCNN, and single-stage object detectors such as YOLOv5n6, YOLOv5x on Microsoft Common Objects in Context, SKU110k, and our custom e-scooter dataset. The proposed loss function demonstrated an increment of 3.70% at APS on the COCO dataset, 6.20% at AP55 on SKU110k, and 9.03% at AP80 of the custom e-scooter dataset. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

18 pages, 18002 KiB  
Article
LoRa-Based Wireless Sensors Network for Rockfall and Landslide Monitoring: A Case Study in Pantelleria Island with Portable LoRaWAN Access
by Mattia Ragnoli, Alfiero Leoni, Gianluca Barile, Giuseppe Ferri and Vincenzo Stornelli
J. Low Power Electron. Appl. 2022, 12(3), 47; https://doi.org/10.3390/jlpea12030047 - 07 Sep 2022
Cited by 14 | Viewed by 3630
Abstract
Rockfalls and landslides are hazards triggered from geomorphological and climatic factors other than human interaction. The economic and social impacts are not negligible, therefore the topic has become an important field in the application of remote monitoring. Wireless sensor networks (WSNs) are particularly [...] Read more.
Rockfalls and landslides are hazards triggered from geomorphological and climatic factors other than human interaction. The economic and social impacts are not negligible, therefore the topic has become an important field in the application of remote monitoring. Wireless sensor networks (WSNs) are particularly suited for the deployment of such systems, thanks to the different technologies and topologies that are evolving nowadays. Among these, LoRa modulation technique represents a fitting technical solution for nodes communication in a WSN. In this paper, a smart autonomous LoRa-based rockfall and landslide monitoring system is presented. The structure has been operating in Pantelleria Island, Sicily, Italy. The sensing elements are disposed in sensor nodes arranged in a star topology. Network access to the LoRaWAN and the Internet is provided through gateways using a portable, solar powered device assembly. A system overview concerning both hardware and functionality of the nodes and gateways devices, then a power analysis is reported, and a monthly recorded result is presented, with related discussion. Full article
(This article belongs to the Special Issue Advances in Embedded Artificial Intelligence and Internet-of-Things)
Show Figures

Figure 1

Back to TopTop