이 논문에서는 모바일 통신망에서 호 손실율의 가능성 분포에 기초하여 최대 손실률을 추정하는 방법을 제안한다. 호 손실률 가능성 분포는 관측된 호 손실률을 이용하여 퍼지추론으로 추정한다. 퍼지규칙의 소속 함수는 신경망의 EBP(error backpropagation ) 알고리즘으로 튜닝하고, 퍼지추론은 퍼지집합의 가중치 평균에 기초하여 호 손실율의 상한계를 추정한다. 이 방법은 과도한 CDR(Call Dropping Ratio)의 추정을 방지할 수 있고, 추정된 CDR 이 관측된 CDR보다 작을 때는 실시간적으로 자기보상을 실시하여 관측된 CDR이 추정된 CDR을 초과하는 경우가 없게 한다. 시뮬레이션을 통하여 제안된 방법이 관측된 호 손실률에 기초하여 상한계값을 잘 추정해냄을 보인다.

Association rules are the discovery of previously unknown, potentially useful and hidden knowledge in databases. Many algorithms have been proposed to find association rules in databases. Due to the diverse use's interest and preference to items, former algorithms do not work well in real world application. That is to say, in most algorithms of mining association rules, the items are considered to have equal time weight and are not dealt with quantitative attributes. Hence, to improve former algorithms, we propose an algorithm in this paper to mine fuzzy association rules considering time weight of each item and quantity of each item.

This study is to develop and operate the R-FCE(Real-time Fuzzy Control Expert System) model on the SBR(Sequency Batch Reactors) type sewage disposal facility which is for the environmental problem of future and the effective treatment. It is the system that regularly handles the input, output information in real time by the Fuzzy control IF-THEN rule. Nowadays the water pollution caused by the increase of population and the industrialization already exceeds the purifying ability of nature and is getting worse. The sewage disposal facility needs an expert having wide experience and know-how because the quality or quantity of inflow water is so irregular that the process of sewage disposal is difficult. In the nation, however, the accumulation of technology is so weak because field-operators avoid the long-term duty. So the accurate and speedy decision or control of field are difficult in the case of emergency situation.