黄砂や工事に由来するエアロゾル増加を検出する確率論的手法の検討

  • 岡田 匡史
    東京工業大学大学院情報理工学研究科 防衛省技術研究本部先進技術推進センター
  • 藤井 修二
    東京工業大学大学院情報理工学研究科
  • 林 譲
    帝京平成大学薬学部薬学科

書誌事項

タイトル別名
  • Stochastic Study of Detection of Aerosol Increase Due to Asian Dust and Outdoor Facility Construction
  • コウサ ヤ コウジ ニ ユライ スル エアロゾル ゾウカ オ ケンシュツ スル カクリツロンテキ シュホウ ノ ケントウ

この論文をさがす

抄録

The real-time detection of an increase in the atmospheric aerosol number concentration due to Asian dust and outdoor facility construction was studied with a stochastic method, called FUMI theory (function of mutual information). It is difficult to find this event in its early stage by means of instrumental analysis, since the instrumental output originating from the event is so small that it can almost be hidden under, or rather overlapped with, the omnipresent instrumental noise. Difficulty comes from the fact that the background and objective signal are both subject to random fluctuations. If an observed value exceeds a detection limit, the hypothesis that the observed one originates from a natural cause other than the Asian dust or outdoor facility construction would be denied. The following methods are elaborated to make the early detection of the event more practical with satisfactory precision: continuous shift of a time domain used for the detection-limit estimation; removal of spike noise and output judged as a signal in our method from the domain; classification of a signal in the scale of the detection limit.

収録刊行物

  • 分析化学

    分析化学 62 (1), 1-10, 2013

    公益社団法人 日本分析化学会

参考文献 (21)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ