• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.6
• /
• pp.1089-1096
• /
• 2019

Recently, small-scale IoT services using a small amount of information through low-performance computing have been spread. It requires low cost, low-power, and long-distance communication technologies with wide communication radius, relatively low power consumption. This paper proposes a MAC layer and routing protocol that supports multi-hop transmission in small-scale IoT environment distributed over a large area based on LoRa communication and delivering a small amount of sensing data. The terminal node is mobile and the communication type provides bidirectional transmission between the terminal node and the network application server. By applying the proposed protocol, a production line monitoring system for smart factory was implemented. It was confirmed that the basic monitoring functions are normally performed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.693-698
• /
• 2018

This paper proposes a vehicle detection method for parking areas using unmanned aerial vehicles (UAVs) and using YOLOv2, which is a recent, known, fast, object-detection real-time algorithm. The YOLOv2 convolutional network algorithm can calculate the probability of each class in an entire image with a one-pass evaluation, and can also predict the location of bounding boxes. It has the advantage of very fast, easy, and optimized-at-detection performance, because the object detection process has a single network. The sliding windows methods and region-based convolutional neural network series detection algorithms use a lot of region proposals and take too much calculation time for each class. So these algorithms have a disadvantage in real-time applications. This research uses the YOLOv2 algorithm to overcome the disadvantage that previous algorithms have in real-time processing problems. Using Darknet, OpenCV, and the Compute Unified Device Architecture as open sources for object detection. a deep learning server is used for the learning and detecting process with each car. In the experiment results, the algorithm could detect cars in a dense area using UAVs, and reduced overhead for object detection. It could be applied in real time.

• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1126-1135
• /
• 2014

In recent years, RDF partitioning schemes have been studied for the effective distributed storage and management of large-scale RDF data. In this paper, we propose an RDF dynamic partitioning scheme to support load balancing in dynamic environments where the RDF data is continuously inserted and updated. The proposed scheme creates clusters and sub-clusters according to the frequency of the RDF data used by queries to set graph partitioning criteria. We partition the created clusters and sub-clusters by considering the workloads and data sizes for the servers. Therefore, we resolve the data concentration of a specific server, resulting from the continuous insertion and update of the RDF data, in such a way that the load is distributed among servers in dynamic environments. It is shown through performance evaluation that the proposed scheme significantly improves the query processing time over the existing scheme.

• Journal of KIISE
• /
• v.44 no.4
• /
• pp.363-373
• /
• 2017

Ext4 file system is widely used in various computing environments such as those of the PC, the server, and the Linux-based embedded system. Ext4, which uses a buffer for block I/O, provides fsync() system call to applications to guarantee the consistency of a specific file. A log of the analytical studies regarding the operation of Ext4 and the improvement of its performance has been compiled, but it has not been studied in detail in terms of kernel versions. We figure out that the behavior of fsync() system call is different depending on the kernel version. Between the kernel versions of 3.4.0 and 4.7.2, 3.4.0, 3.8.0, and 4.6.2 showed behavioral differences regarding the fsync() system call. The latency of fsync() in kernel 3.4.0 is longer than that of the more-advanced 3.7.10; meanwhile, the characteristics of 3.8.0 enabled the disruption of the Ext4 journaling order, but the ordered defect was solved with 4.6.2.

• Journal of Intelligence and Information Systems
• /
• v.15 no.1
• /
• pp.31-50
• /
• 2009

As the collected information which is static or dynamic is infinite in ubiquitous computing environments, information overload and invasion of privacy have been pressing issues in the recommendation service. In this study, we propose a recommendation service procedure through P2P, The P2P helps customer to obtain effective and secure product information because of communication among customers who have the similar preference about the products without connection to server. To evaluate the performance of the proposed recommendation service, we utilized real transaction and product data of the Korean mobile company which service character images. We developed a prototype recommender system and demonstrated that the proposed recommendation service makes an effect on recommending product in the ubiquitous environments. We expect that the information overload and invasion of privacy will be solved by the proposed recommendation procedure in ubiquitous environment.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.1
• /
• pp.75-82
• /
• 2015

Recently, many public auditing schemes have been proposed to support public auditability that enables a third party auditor to verify the integrity of data stored in the remote cloud server. To improve the performance of the auditor, several public auditing schemes support batch auditing which allows the auditor to handle simultaneously multiple auditing delegations from different users. However, when even one data is corrupted, the batch auditing will fail and individual and repeated auditing processes will be required. It is difficult to identify the corrupted data from the proof in which distinct data blocks and authenticators of distinct users are intricately aggregated. In this paper, we extend a public auditing scheme of Wang et al. to support batch auditing for multi-cloud and multi-user. We propose an identification scheme of the corrupted cloud when the data of a single cloud is corrupted in the batch auditing of multi-cloud and multi-user.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.704-711
• /
• 2003

Recently, the demands for real-time service and multimedia data are rapidly increasing. There are significant redundancies between header fields both within the same packet header and in consecutive packets belonging to the same packet stream. But there are many overheads in using the current UDP/IP protocol. Header compression is considered to enhance the transmission efficiency for small size of payload. By sending the static field information only once initially and by utilizing dependencies and predictability for other fields, the header size can be significantly reduced for most packets. This work describes an implementation for header compression of the headers of U/UDP protocols to reduce overhead on Ethernet network. Typical UDP/IP Header packets can be compressed down to 7 bytes and the header compression system is designed and implemented on the Linux environment. Using the designed Header compression system between a server and a client have the advantage of effective data throughput in network.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.11
• /
• pp.2435-2441
• /
• 2010

A global healthcare system based on M2M technology is proposed to support a good mobility, flexibility and scalability to the patients in 6LoWPAN. Sensor nodes integrated with wearable sensors are linked to gateway with IEEE 802.15.4 protocol and 6LoWPAN protocol for data acquisition and transmission purpose via external network. In the server, heart rate variability signals are obtained by signal processing and used for time and frequency domain performance analysis to evaluate the patient's health status. Our approach for global healthcare system with non-invasive and continuous IP-based communication is managed to process large amount of biomedical signals in the large scale of service range accurately.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.5
• /
• pp.979-985
• /
• 2016

In this paper, a distributed instant message system with multiple servers architecture which can distribute system load effectively using an extended media control channel has been presented. A media control channel provides establishing a reliable control channel and also keeping a reliable control channel between SIP server and client in the field of real-time media transport area. In this study, a new instant message system architecture which can distribute massive instant message including multimedia data to multiple servers has been presented. The presented instant message system architecture can distribute system load by extending media control channel. For this purpose, media control channel messages, which distribute system load to multiple servers dynamically according to increasing number of users, have been designed in our presented system. And, in our research, an exchanging procedures of media control channel messages between servers have also been presented. The performance of the proposed system has been analysed by simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.12
• /
• pp.2981-2986
• /
• 2015

In this paper, there are m servers to carry out broadcasts and the scheduling problem to serve the requests with deadlines is studied. If a server broadcasts a page, then all the requests which require the page are satisfied. A scheduling algorithm shall determine which pages are broadcasted on servers at a time. Its goal is to maximize the sum of weights of requests satisfied within their deadlines. The performance of an on-line algorithm is compared with that of the optimal off-line algorithm which can see all the inputs in advance. In general, the off-line algorithms outperform the on-line algorithms. So we will use the resource augmentation analysis in which the on-line algorithms can utilize more resources. We consider the case that the on-line algorithms can use more servers in this paper.