• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• Electronics and Telecommunications Trends
• /
• v.37 no.1
• /
• pp.32-41
• /
• 2022

With the development of computing technology, the convergence of ICT with existing traditional industries is being attempted. In particular, with the recent advent of 5G, connectivity with numerous AuT (autonomous Things) in the real world as well as simple mobile terminals has increased. As more devices are deployed in the real world, the need for technology for devices to learn and act autonomously to communicate with humans has begun to emerge. This article introduces "Device to the Edge," a new computing paradigm that enables various devices in smart spaces (e.g., factories, metaverse, shipyards, and city centers) to perform ultra-reliable, low-latency and high-speed processing regardless of the limitations of capability and performance. The proposed technology, referred to as EdgeCPS, can link devices to augmented virtual resources of edge servers to support complex artificial intelligence tasks and ultra-proximity services from low-specification/low-resource devices to high-performance devices.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.165-172
• /
• 2023

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1103-1110
• /
• 2022

Due to the rapid development of AI-related industries, countless edge devices are working in the real world. Since data generated within the smart space consisted of these devices is beyond imagination, it is becoming increasingly difficult for edge devices to process. To solve this issue, EdgeCPS has appeared. EdgeCPS is a technology to support harmonious execution of various application services including AI applications through interworking between edge devices and edge servers, and augmenting resources/functions. Therefore, we propose a knowledge-sharing graph-based componentized AI application support system applicable to the EdgeCPS platform. The graph is designed to effectively store information which are essential elements for creating AI applications. In order to easily change resource/function augmentation under the support of the EdgeCPS platform, AI applications are operated as components. The application support system is linked with the knowledge graph so that users can easily create and test applications, and visualizes the execution aspect of the application to users as a pipeline.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.159-164
• /
• 2023

Solar energy harvesting IoT devices prioritize maximizing the utilization of collected energy due to the periodic recharging nature of solar energy, rather than minimizing energy consumption. Meanwhile, research on edge AI, which performs machine learning near the data source instead of the cloud, is actively conducted for reasons such as data confidentiality and privacy, response time, and cost. One such research area involves performing various audio AI applications using audio data collected from multiple IoT devices in an IoT edge computing environment. However, in most studies, IoT devices only perform sensing data transmission to the edge server, and all processes, including data preprocessing, are performed on the edge server. In this case, it not only leads to overload issues on the edge server but also causes network congestion by transmitting unnecessary data for learning. On the other way, if data preprocessing is delegated to each IoT device to address this issue, it leads to another problem of increased blackout time due to energy shortages in the devices. In this paper, we aim to alleviate the problem of increased blackout time in devices while mitigating issues in server-centric edge AI environments by determining where the data preprocessed based on the energy state of each IoT device. In the proposed method, IoT devices only perform the preprocessing process, which includes sound discrimination and noise removal, and transmit to the server if there is more energy available than the energy threshold required for the basic operation of the device.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.11
• /
• pp.319-328
• /
• 2023

Unlike cloud computing, edge computing technology analyzes and judges data close to devices and users, providing advantages such as real-time service, sensitive data protection, and reduced network traffic. EdgeX Foundry, a representative open source of edge computing platforms, is an open source-based edge middleware platform that provides services between various devices and IT systems in the real world. EdgeX Foundry provides a service for handling camera devices, along with a service for handling existing sensed data, which only supports simple streaming and camera device management and does not store or process image data obtained from the device inside EdgeX. This paper presents a technique that can store and process image data inside EdgeX by applying some of the services provided by EdgeX Foundry. Based on the proposed technique, a service pipeline for object detection services used core in the field of autonomous driving was created for experiments and performance evaluation, and then compared and analyzed with existing methods.

• Electronics and Telecommunications Trends
• /
• v.37 no.5
• /
• pp.44-53
• /
• 2022

The Internet of Things (IoT) is growing exponentially, with the number of IoT devices multiplying annually. Accordingly, the paradigm is changing from cloud computing to edge computing and even tiny edge computing because of the low latency and cost reduction. Machine learning is also shifting its role from the cloud to edge or tiny edge according to the paradigm shift. However, the fragmented and resource-constrained features of IoT devices have limited the development of artificial intelligence applications. Edge MLaaS (Machine Learning as-a-Service) has been studied to easily and quickly adopt machine learning to products and overcome the device limitations. This paper briefly summarizes what Edge MLaaS is and what element of research it requires.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.42 no.12
• /
• pp.27-34
• /
• 2005

We have investigated the correlation between the active-layer uniformity and reliability of InGaN/GaN blue LEDs. According to initial characteristics, the devices are classified into two groups: group I devices of uniform light-emission and group II devices of non-uniform light-emission. The group II devices are more dependent on temperature and we have found two degradation mechanisms through reliability test. One is bulk degradation in which the degradation occurred over the entire chip and another one is edge degradation in which the degradation occurred from the edge of the chip. Bulk degradation caused by the nonradiative defects is found to be faster in group II devices while there is no difference in the rate of the much faster edge degradation, where darkening starts from the n-Ohmic contact edge. Therefore, more uniform active layer, more uniform current spreading, and the passivation of the dry-etched side-wall are essential for the high reliability of InGaN/GaN LEDs.

• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.23-29
• /
• 2022

Recently, due to the increase in the use of Internet of Things (IoT) devices, the amount of data transmitted and processed to cloud computing servers has increased rapidly. As a result, network problems (delay, server overload and security threats) are emerging. In particular, edge computing with lower computational capabilities than cloud computing requires a lightweight authentication algorithm that can easily authenticate numerous IoT devices.In this paper, we proposed a key-agreement protocol of a lightweight algorithm that guarantees anonymity and forward and backward secrecy between IoT and edge devices. and the proposed algorithm is stable in MITM and replay attacks for edge device and IoT. As a result of comparing and analyzing the proposed key-agreement protocol with previous studies, it was shown that a lightweight protocol that can be efficiently used in IoT and edge devices.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.952-973
• /
• 2021

In recent years, mobile devices have become an essential part of daily life. More and more applications are being supported by mobile devices thanks to edge computing, which represents an emergent architecture that provides computing, storage, and networking capabilities for mobile devices. In edge computing, heavy tasks are offloaded to edge nodes to alleviate the computations on the mobile side. However, offloading computational tasks may incur extra energy consumption and delays due to network congestion and server queues. Therefore, it is necessary to optimize offloading decisions to minimize time, energy, and payment costs. In this article, different offloading models are examined to identify the offloading parameters that need to be optimized. The paper investigates and compares several optimization techniques used to optimize offloading decisions, specifically Swarm Intelligence (SI) models, since they are best suited to the distributed aspect of edge computing. Furthermore, based on the literature review, this study concludes that a Cuckoo Search Algorithm (CSA) in an edge-based architecture is a good solution for balancing energy consumption, time, and cost.