• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.10
• /
• pp.901-906
• /
• 2013

Chaff is a kind of matter spreading atmosphere with the purpose of preventing aircraft from detecting by radar. The chaff is commonly composed of small aluminum pieces, metallized glass fiber, or other lightweight strips which consists of reflecting materials. The chaff usually appears on the radar images as narrow bands shape of highly reflective echoes. And the chaff echo has similar characteristics to precipitation echo, and it interrupts weather forecasting process and makes forecasting accuracy low. In this paper, the chaff echo recognizing and removing method is suggested using Bayesian network. After converting coordinates from spherical to Cartesian in UF (Universal Format) radar data file, the characteristics of echoes are extracted by spatial and temporal clustering. And using the data, as a result of spatial and temporal clustering, a classification process for analyzing is performed. Finally, the inference system using Bayesian network is applied. As a result of experiments with actual radar data in real chaff echo appearing case, it is confirmed that Bayesian network can distinguish between chaff echo and non-chaff echo.

• Transactions on Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.17-20
• /
• 2024

This paper presents a novel lightweight object detection model tailored for low-powered edge devices, addressing the limitations of traditional resource-intensive computer vision models. Our proposed detector, inspired by the Single Shot Detector (SSD), employs a compact yet robust network design. Crucially, it integrates an 'enhancer block' that significantly boosts its efficiency in detecting smaller objects. The model comprises two primary components: the Light_Block for efficient feature extraction using Depth-wise and Pointwise Convolution layers, and the Enhancer_Block for enhanced detection of tiny objects. Trained from scratch on the Udacity Annotated Dataset with image dimensions of 300x480, our model eschews the need for pre-trained classification weights. Weighing only 5.5MB with approximately 0.43M parameters, our detector achieved a mean average precision (mAP) of 27.7% and processed at 140 FPS, outperforming conventional models in both precision and efficiency. This research underscores the potential of lightweight designs in advancing object detection for edge devices without compromising accuracy.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.168-173
• /
• 2024

Speaker recognition refers to a technology that analyzes voice frequencies that are different for each individual and compares them with pre-stored voices to determine the identity of the person. Deep learning-based speaker recognition is being applied to many fields, and pet robots are one of them. However, the hardware performance of pet robots is very limited in terms of the large memory space and calculations of deep learning technology. This is an important problem that pet robots must solve in real-time interaction with users. Lightening deep learning models has become an important way to solve the above problems, and a lot of research is being done recently. In this paper, we describe the results of research on lightweight speaker recognition for pet robots by constructing a voice data set for pet robots, which is a specific command type, and comparing the results of models using residuals. In the conclusion, we present the results of the proposed method and Future research plans are described.

• The KIPS Transactions:PartC
• /
• v.18C no.2
• /
• pp.119-126
• /
• 2011

In wireless sensor networks Time synchronization used to be performed after routing tree is constructed. It results in increasing the number of packets and energy consumption. In this paper, we propose a time synchronization algorithm combined with flooding routing tree construction algorithm, which applies LTS (Lightweight Time Synchronization) information packed into the forwarding and backward routing packets. Furthermore, the proposed algorithm compensates the time error due to clock drift using the round time with fixed period. We prove that the proposed algorithm could synchronize the time of among sensor nodes more accurately compared to TSRA (Time Synchronization Routing Algorithm) using NS2 simulation tool.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39B no.12
• /
• pp.822-830
• /
• 2014

Recently, A Security service in Internet environments has been more important and the use of SSL & TLS is increasing for the personel homepage as well as administrative institutions. Also, IETF suggests using DTLS, which can provide a security service to constrained devices with lower CPU power and limited memory space under IoT environments. In this paper, we implement LEA(Lightweight Encryption Algorithm) algorithm and apply it to OpenSSL. The implemented algorithm is compared with other symmetric encryption algorithms such as AES etc, and it shows the superior performance in calculation speed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.3
• /
• pp.1212-1228
• /
• 2016

Secure communication is an important aspect of today's interconnected environments and it can be achieved by the use of cryptographic algorithms and protocols. However, many existing cryptographic mechanisms are tightly integrated into communication protocols. Issues emerge when security vulnerabilities are discovered in cryptographic mechanisms because their replacement would eventually require replacing deployed protocols. The concept of cryptographic agility is the solution to these issues because it allows dynamic switching of cryptographic algorithms and keys prior to and during the communication. Most of today's secure protocols implement cryptographic agility (IPsec, SSL/TLS, SSH), but cryptographic agility mechanisms cannot be used in a standalone manner. In order to deal with the aforementioned limitations, we propose a lightweight cryptographically agile agreement model, which is formally verified. We also present a solution in the Agile Cryptographic Agreement Protocol (ACAP) that can be adapted on various network layers, architectures and devices. The proposed solution is able to provide existing and new communication protocols with secure communication prerequisites in a straightforward way without adding substantial communication overhead. Furthermore, it can be used between previously unknown parties in an opportunistic environment. The proposed model is formally verified, followed by a comprehensive discussion about security considerations. A prototype implementation of the proposed model is demonstrated and evaluated.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.901-923
• /
• 2017

Efficient key distribution and management mechanisms as well as lightweight ciphers are the main pillar for establishing secure wireless sensor networks (WSN). Several symmetric based key distribution protocols are already proposed, but most of them are not scalable, yet vulnerable to a small number of compromised nodes. In this paper, we propose an efficient and scalable key management and distribution framework, named KMMR, for large scale WSNs. The KMMR contributions are three fold. First, it performs lightweight local processes orchestrated into upward and downward tiers. Second, it limits the impact of compromised nodes to only local links. Third, KMMR performs efficient secure node addition and revocation. The security analysis shows that KMMR withstands several known attacks. We implemented KMMR using the NesC language and experimented on Telosb motes. Performance evaluation using the TOSSIM simulator shows that KMMR is scalable, provides an excellent key connectivity and allows a good resilience, yet it ensures both forward and backward secrecy. For a WSN comprising 961 sensor nodes monitoring a 60 hectares agriculture field, KMMR requires around 2.5 seconds to distribute all necessary keys, and attains a key connectivity above 96% and a resilience approaching 100%. Quantitative comparisons to earlier work show that KMMR is more efficient in terms of computational complexity, required storage space and communication overhead.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.586-589
• /
• 2017

The vibration monitoring system using the sensor network is used in various fields. However, in case of the vibration of the existing products, the size of the monitoring system is increased due to the separation of the sensor data collection function and the communication function. In this paper, we design and implement a lightweight vibration monitoring system using the MQTT protocol, which is oneM2M device standard protocol for the Arduino and Ethernet modules, to monitor frequent earthquakes and vibrations in narrow places.

• Convergence Security Journal
• /
• v.8 no.4
• /
• pp.21-30
• /
• 2008

This study researched into the method that allows only the certified user and computational engineer to possibly use network resources and computing resources by implementing the system of the intensified certification and security based on LDAP(Lightweight Directory Access Protocol) directory service, that copes with incapacitation in security program due to making the security program forcibly installed, and that can correctly track down the industrial-technology exporter along with applying the user-based security policy through inter-working with the existing method for the protection of industrial technology. Through this study, the intensified method for the protection of industrial technology can be embodied by implementing the integrated infra system through strengthening the existing system of managing the protection of industrial technology, and through supplementing vulnerability to the method for the protection of industrial technology.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.2
• /
• pp.51-58
• /
• 2023

This paper proposes a deep learning method for estimating the heart rate from facial videos. Our proposed method estimates remote photoplethysmography (rPPG) signals to predict the heart rate. Although there have been proposed several methods for estimating rPPG signals, most previous methods can not be utilized in low-power single board computers due to their computational complexity. To address this problem, we construct a lightweight student model and employ a knowledge distillation technique to reduce the performance degradation of a deeper network model. The teacher model consists of 795k parameters, whereas the student model only contains 24k parameters, and therefore, the inference time was reduced with the factor of 10. By distilling the knowledge of the intermediate feature maps of the teacher model, we improved the accuracy of the student model for estimating the heart rate. Experiments were conducted on the UBFC-rPPG dataset to demonstrate the effectiveness of the proposed method. Moreover, we collected our own dataset to verify the accuracy and processing time of the proposed method on a real-world dataset. Experimental results on a NVIDIA Jetson Nano board demonstrate that our proposed method can infer the heart rate in real time with the mean absolute error of 2.5183 bpm.