MinSeok Jang

Test case generation of a communication protocol by an adaptive state exploration

State exploration, which is widely used for generation of executable test cases for conformance test of a protocol, often causes state explosion problem. This paper proposes an adaptive state exploration method to overcome the problem for a single module protocol. Since state explosion arises in relation to the predicate, it tries to find out efficiently predicated transitions with priority by emulating the normal and erroneous communication environments closely related to the predicates. A test case generation method using the method is also proposed, which uses an Subdivided Extended Finite State Machine (SEFSM) as a protocol model to reduce semantic loss by modeling and adjusted UIO sequences for the protocol model. The empirical results of the proposed methods show their efficacy.

Active Rule System Based on User`s Emotional Margin for Power Saving Control

Abstract In this paper, we propose the active rule system applying with emotional margin for power saving control. The proposed system in this paper is a part of the system which derives smart power saving by adjusting the illuminance using active rules within compromising the user’s emotion. For this, we set the specific range of standard illuminance and the lower bound of user’s emotional margin of illuminance based on measurements and analysis, and use these data in design of active rules. And then, we design and implement the active rule system using mobile agent. The mobile agent in the proposed system migrates to the destination sensor nodes, acquires and transmits sensor data according to the purpose and needs through the active rules, and directly executes the actions corresponding to the optional events(changed sensor data and/or time etc.). And then, we show the potential applicability of the proposed active rule system in various active sensor network applications through the interaction with the rule base and mobile sensor network middleware system.

Design and Implementation of the Active Multi-Agent Middleware for the Sensor Network Application

In this paper, we suggest the active multi-agent middleware for the sensor network application. For this, firstly we design and implement the active rule based mobile agent middleware. The mobile agent in the proposed system visits the destination sensor nodes according to the migration list offered by the meta table in the name space of the naming agent, acquires and transmits sensor data according to the purpose and needs through the active rules, and directly executes the actions corresponding to the optional events(changed sensor data and/or time etc.). And then, we show the potential applicability of the active rule based mobile agent middleware in various active sensor networks through the interaction with the rule base system and context database system.

Traversal pattern mining in web environment

There have been researches about analyzing the information retrieval patterns of log file to obtain users’ information search patterns in web environment. Algorithms that find the frequently traversed path pattern from search path inputs are suggested mainly. But one of the existing works’ problems is to provide inadequate solution for complex, that is, general topological patterns. This paper suggests an efficient algorithm for deriving the maximal frequent traversal pattern from general paths.

Unsupervised speaker adaptation for phonetic transcription based voice dialing

Since the speaker independent phoneme HMM based voice dialing system uses only the phoneme transcription of the input sentence, the storage space could be reduced greatly. However, the performance of the system is worse than that of the speaker dependent system due to the phoneme recognition errors generated when the speaker independent models are used. In order to solve this problem, a new method that jointly estimates the transformation vectors (bias) and transcriptions for the speaker adaptation is presented. The biases and transcriptions are estimated iteratively from the training data of each user with maximum likelihood approach to the stochastic matching using speaker independent phoneme models. Experimental result shows that the proposed method is superior to the conventional method using transcriptions only.

Speaker adaptation techniques for speech recognition with a speaker-independent phonetic recognizer

We present a new method that improves the performance of the speech recognition system with the speaker-independent (SI) phonetic recognizer. The performance of the speech recognition system with the SI phonetic recognizer is worse than that of the speaker dependent system due to the mismatch between the training utterances and a set of SI models. A new training method that iteratively estimates the phonetic templates and transformation vectors is presented to reduce the mismatch using speaker adaptation techniques. The stochastic matching methods are used to estimate the transformation vectors for speaker adaptation. The experiment performed over actual telephone line shows that a reduction of about 40% in the error rates could be achieved as compared to the conventional method.

Iterative training techniques for phonetic template based speech recognition with a speaker-independent phonetic recognizer

This paper presents a new method that improves the performance of the speaker specific phonetic template based speech recognizer with the speaker-independent (SI) phoneme HMMs. Since the phonetic template based speech recognizer uses only the phoneme transcription of the input utterance, the performance of the system is worse than that of the speaker dependent system due to the mismatch between the training data and the SI models. In order to solve these problems, a new training method that iteratively estimates the phonetic templates and transformation vectors for the adaptation of the SI phoneme HMMs is presented. The phonetic class based and codebook-based stochastic matching methods are used to estimate the transformation vectors for speaker adaptation. Performance evaluation using the speaker dependent recognition experiments performed over actual telephone line showed a reduction of about 40% in the error rates when compare to the conventional speaker specific phonetic template based speech recognizer.