• Electronics and Telecommunications Trends
• /
• v.38 no.5
• /
• pp.12-22
• /
• 2023

The development of neural networks is evolving towards the adoption of transformer structures with attention modules. Hence, active research focused on extending the concept of lightweight neural network algorithms and hardware acceleration is being conducted for the transition from conventional convolutional neural networks to transformer-based networks. We present a survey of state-of-the-art research on lightweight neural network algorithms and hardware architectures to reduce memory usage and accelerate both inference and training. To describe the corresponding trends, we review recent studies on token pruning, quantization, and architecture tuning for the vision transformer. In addition, we present a hardware architecture that incorporates lightweight algorithms into artificial intelligence processors to accelerate processing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3224-3242
• /
• 2020

Internet of Things (IoT) networks composed of a large number of sensors and actuators generate a huge volume of data and control commands, which should be enforced by strong data reliability. The end-to-end data reliability of IoT networks is an essential industrial enabler. Blockchain technology can provide strong data reliability and integrity within IoT networks. We designed a lightweight end-to-end blockchain network that applies to common IoT applications. Its enhanced modular architecture and lightweight consensus mechanism guarantee its practical applicability for general IoT applications. In addition, the proposed blockchain network is highly software compatible because it adopts the Hyperledger development environment. Directly embedding the proposed blockchain middleware platform in small computing devices proves its practicability.

• Computers and Concrete
• /
• v.25 no.5
• /
• pp.401-409
• /
• 2020

Artificial neural networks are used as a useful tool in distinct fields of civil engineering these days. In order to control long-term quality of Self-Compacting Semi-Lightweight Concrete (SCSLC), the 90 days compressive strength is considered as a key issue in this paper. In fact, combined artificial neural networks are used to predict the compressive strength of SCSLC at 28 and 90 days. These networks are able to re-establish non-linear and complex relationships straightforwardly. In this study, two types of neural networks, including Radial Basis and Multilayer Perceptron, were used. Four groups of concrete mix designs also were made with two water to cement ratios (W/C) of 0.35 and 0.4, as well as 10% of cement weight was replaced with silica fume in half of the mixes, and different amounts of superplasticizer were used. With the help of rheology test and compressive strength results at 7 and 14 days as inputs, the neural networks were used to estimate the 28 and 90 days compressive strengths of above-mentioned mixes. It was necessary to add the 14 days compressive strength in the input layer to gain acceptable results for 90 days compressive strength. Then proper neural networks were prepared for each mix, following which four existing networks were combined, and the combinatorial neural network model properly predicted the compressive strength of different mix designs.

• Computers and Concrete
• /
• v.26 no.4
• /
• pp.327-342
• /
• 2020

The objective of this study is to manufacture environmentally-friendly synthetic lightweight aggregates that may be used in the structural lightweight concrete production. The cold-bonding pelletization process has been used in the agglomeration of the pozzolanic materials to achieve these synthetic lightweight aggregates. In this context, it was aimed to recycle the waste fly ash by employing it in the manufacturing process as the major cementitious component. According to the well-known facts reported in the literature, it is stated that the main disadvantage of the synthetic lightweight aggregate produced by applying the cold-bonding pelletization technique to the pozzolanic materials is that it has a lower strength in comparison with the natural aggregate. Therefore, in this study, the metakaolin made of high purity kaolin and calcined kaolin obtained from impure kaolin have been employed at particular contents in the synthetic lightweight aggregate manufacturing as a cementitious material to enhance the particle crushing strength. Additionally, to propose a curing condition for practical attempts, different curing conditions were designated and their influences on the characteristics of the synthetic lightweight aggregates were investigated. Three substantial features of the aggregates, specific gravity, water absorption capacity, and particle crushing strength, were measured at the end of 28-day adopted curing conditions. Observed that the incorporation of thermally treated kaolin significantly influenced the crushing strength and water absorption of the aggregates. The statistical evaluation indicated that the investigated properties of the synthetic lightweight aggregate were affected by the thermally treated kaolin content more than the kaoline type and curing regime. Utilizing the thermally treated kaolin in the synthetic aggregate manufacturing lead to a more than 40% increase in the crushing strength of the pellets in all curing regimes. Moreover, two numerical formulations having high estimation capacity have been developed to predict the crushing strength of such types of aggregates by using soft-computing techniques: gene expression programming and artificial neural networks. The R-squared values, indicating the estimation performance of the models, of approximately 0.97 and 0.98 were achieved for the numerical formulations generated by using gene expression programming and artificial neural networks techniques, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.3421-3437
• /
• 2018

With the enlargement of wireless technology, vehicular ad-hoc networks (VANETs) are rising as a hopeful way to realize smart cities and address a lot of vital transportation problems such as road security, convenience, and efficiency. To achieve data confidentiality, integrity and authentication applying lightweight cryptosystems is widely recognized as a rather efficient approach for the VANETs. The Khudra cipher is such a lightweight cryptosystem with a typical Generalized Feistel Network, and supports 80-bit secret key. Up to now, little research of fault analysis has been devoted to attacking Khudra. On the basis of the single nibble-oriented fault model, we propose a differential fault analysis on Khudra. The attack can recover its 80-bit secret key by introducing only 2 faults. The results in this study will provides vital references for the security evaluations of other lightweight ciphers in the VANETs.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.1
• /
• pp.55-67
• /
• 2023

Lightweight face recognition models, as one of the most popular and long-standing topics in the field of computer vision, has achieved vigorous development and has been widely used in many real-world applications due to fewer number of parameters, lower floating-point operations, and smaller model size. However, few surveys reviewed lightweight models and reimplemented these lightweight models by using the same calculating resource and training dataset. In this survey article, we present a comprehensive review about the recent research advances on the end-to-end efficient lightweight face recognition models and reimplement several of the most popular models. To start with, we introduce the overview of face recognition with lightweight models. Then, based on the construction of models, we categorize the lightweight models into: (1) artificially designing lightweight FR models, (2) pruned models to face recognition, (3) efficient automatic neural network architecture design based on neural architecture searching, (4) Knowledge distillation and (5) low-rank decomposition. As an example, we also introduce the SqueezeFaceNet and EfficientFaceNet by pruning SqueezeNet and EfficientNet. Additionally, we reimplement and present a detailed performance comparison of different lightweight models on the nine different test benchmarks. At last, the challenges and future works are provided. There are three main contributions in our survey: firstly, the categorized lightweight models can be conveniently identified so that we can explore new lightweight models for face recognition; secondly, the comprehensive performance comparisons are carried out so that ones can choose models when a state-of-the-art end-to-end face recognition system is deployed on mobile devices; thirdly, the challenges and future trends are stated to inspire our future works.

• Journal of Internet Computing and Services
• /
• v.23 no.2
• /
• pp.1-7
• /
• 2022

Machine learning (ML) algorithms have been intended to seamlessly collaborate for enabling intelligent networking in terms of massive service differentiation, prediction, and provides high-accuracy recommendation systems. Mobile edge computing (MEC) servers are located close to the edge networks to overcome the responsibility for massive requests from user devices and perform local service offloading. Moreover, there are required lightweight methods for handling real-time Internet of Things (IoT) communication perspectives, especially for ultra-reliable low-latency communication (URLLC) and optimal resource utilization. To overcome the abovementioned issues, this paper proposed an intelligent scheme for traffic steering based on the integration of MEC and lightweight ML, namely support vector machine (SVM) for effectively routing for lightweight and resource constraint networks. The scheme provides dynamic resource handling for the real-time IoT user systems based on the awareness of obvious network statues. The system evaluations were conducted by utillizing computer software simulations, and the proposed approach is remarkably outperformed the conventional schemes in terms of significant QoS metrics, including communication latency, reliability, and communication throughput.

• Journal of Sensor Science and Technology
• /
• v.30 no.5
• /
• pp.286-289
• /
• 2021

Image classification is one of the fundamental applications of computer vision. It enables a system to identify an object in an image. Recently, image classification applications have broadened their scope from computer applications to edge devices. The convolutional neural network (CNN) is the main class of deep learning neural networks that are widely used in computer tasks, and it delivers high accuracy. However, CNN algorithms use a large number of parameters and incur high computational costs, which hinder their implementation in edge hardware devices. To address this issue, this paper proposes a lightweight image classifier that provides good accuracy while using fewer parameters. The proposed image classifier diverts the input into three paths and utilizes different scales of receptive fields to extract more feature maps while using fewer parameters at the time of training. This results in the development of a model of small size. This model is tested on the CIFAR-10 dataset and achieves an accuracy of 90% using .26M parameters. This is better than the state-of-the-art models, and it can be implemented on edge devices.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.105-110
• /
• 2022

Traffic sign recognition plays an important role in solving traffic-related problems. Traffic sign recognition and classification systems are key components for traffic safety, traffic monitoring, autonomous driving services, and autonomous vehicles. A lightweight model, applicable to portable devices, is an essential aspect of the design agenda. We suggest a lightweight convolutional neural network model with residual blocks for traffic sign recognition systems. The proposed model shows very competitive results on publicly available benchmark data.

• Journal of Information Processing Systems
• /
• v.18 no.1
• /
• pp.12-25
• /
• 2022

In recent years, deep convolutional neural networks have made significant progress in the research of single image super-resolution. However, it is difficult to be applied in practical computing terminals or embedded devices due to a large number of parameters and computational effort. To balance these problems, we propose CSRNet, a lightweight neural network based on channel split residual learning structure, to reconstruct highresolution images from low-resolution images. Lightweight refers to designing a neural network with fewer parameters and a simplified structure for lower memory consumption and faster inference speed. At the same time, it is ensured that the performance of recovering high-resolution images is not degraded. In CSRNet, we reduce the parameters and computation by channel split residual learning. Simultaneously, we propose a double-upsampling network structure to improve the performance of the lightweight super-resolution network and make it easy to train. Finally, we propose a new evaluation metric for the lightweight approaches named 100_FPS. Experiments show that our proposed CSRNet not only speeds up the inference of the neural network and reduces memory consumption, but also performs well on single image super-resolution.