Jin-Suk Kang

Intrusion detection based on clustering a data stream

In anomaly intrusion detection, how to model the normal behavior of activities performed by a user is an important issue. To extract the normal behavior as a profile, conventional data mining techniques are widely applied to a finite audit data set. However, these approaches can only model the static behavior of a user in the audit data set. This drawback can be overcome by viewing the continuous activities of a user as an audit data stream. This paper proposes a new clustering algorithm, which continuously models a data stream. A set of features is used to represent the characteristics of an activity. For each feature, the clusters of feature values corresponding to activities observed so far in an audit data stream are identified by the proposed clustering algorithm for data streams. As a result, without maintaining any historical activity of a user physically, new activities of the user can be continuously reflected to the on-going result of clustering.

Implementation of embedded system for intelligent image recognition and processing

The main challenges for image recognition in an embedded system design are the uncertainty of the mobile user’s location and the limitation of measurement algorithm. This paper demonstrates an implementation of an embedded system which is used for vehicle tracking and license plate recognition. The image of the plate is scanned by the camera attached to a mobile device and then encoded data is transmitted along with the location information to a server that is located in a remote place through a wireless communication network. In order to track a user’s location, the image recognition is estimated with spatial and attribute information receiving priority. In this paper, we present an embedded system for intelligent interface between vehicle and users, employing spatial relative distance algorithm. Comparing the image and location information provides comparable results, and confirms that the embedded system is fairly evaluated by the intelligent image recognition technique.

Energy-efficient Sensor Circuit Design for Space Applications

This paper presents energy-efficient, safety-critical wireless sensor circuits for next-generation space satellites and space robots. The hardware design and implementation proposed here exploits a novel control algorithm and ensures that appropriate reconstructed sensing qualities can be achieved. In this paper we demonstrate that focus on preserving the clarity of important power consumption features, such as dynamic power dissipation, at peak voltage ratios of 120 W. With this approach, features in the circuit that may be important for recognition are well preserved, even at high energy dissipation -a maximum power rate of about 125 W can be sustained.

Anomaly intrusion detection based on clustering a data stream

In anomaly intrusion detection, how to model the normal behavior of activities performed by a user is an important issue. To extract the normal behavior as a profile, conventional data mining techniques are widely applied to a finite audit data set. However, these approaches can only model the static behavior of a user in the audit data set. This drawback can be overcome by viewing the continuous activities of a user as an audit data stream. This paper proposes a new clustering algorithm which continuously models a data stream. A set of features is used to represent the characteristics of an activity. For each feature, the clusters of feature values corresponding to activities observed so far in an audit data stream are identified by the proposed clustering algorithm for data streams. As a result, without maintaining any historical activity of a user physically, new activities of the user can be continuously reflected to the on-going result of clustering.

Special Investigation of Variable Star Delta Cephei via Neural Network

We investigated that the brightness magnitude of the Del Cephei had been adapted to three different training values and each training had four different hidden layers in the back-propagation neural network (BPNN) models. The best forecasting mean through different approaches in the BPNN model was obtained at 15% training with four hidden layers. However, the difference in forecasting accuracies of the different levels was not statistically significant.

Optimum Sensing Technique and Approach for Low Power Consumption Network

This paper presents a technique to integrate energy efficient ubiquitous sensors to work in a network for optimum control. This optimization technique can reduce power consumption by using wireless communicating sensors with a supervisor control. This supervisory control of the sensors gives it an advantage of being controllable from any embedded, desktop or Internet platform. Sensors for different purposes can work together to form a network which can be implemented in many environment like homes, city, hospitals and remote locations like mountains, forests etc. We employed an optimized sensing system and algorithm which reduced the number of sensors and power consumption.

Implementation of Embedded System for Vehicle Tracking and License Plates Recognition using Spatial Relative Distance

This paper set out to use a camera attached to a mobile terminal in order to inquire a car and track its location anywhere. The technology based in the research is to automatically recognize and verify license plates on the front and back of a car. The plates are input through a camera attached to a mobile terminal. The technology enables us to detect a car registration number and to transmit it along with the location of the terminal to a server through a wireless communication network. The information of a car obtained through the terminal is encoded and transmitted to a server in a remote place through a wireless communication network. The car registration number and its location information are decoded and transmitted as a text to the server in a remote place. In order to track a users location through spatial relative distance estimated in real-time, the server uses the spatial and attribute information which are the most prior to the desired data value. With this property information, the right location can be calculated