電子鼻雖然已經行之有年，但其具體實現至今仍然停留在大型儀器。一方面，目前電子鼻系統對於單一已知氣體的辨識演算法已有相當程度的準確率，但是對於測試氣體中含有未知氣體的情形，則尚未有有效的演算法。然而，除了在實驗室中能產生單一氣體的環境，日常生活所遇到的應用，空氣中實際上都是包含了未知氣體成份的環境。本論文將未知氣體視為一個未歸類的氣體(不屬於訓練資料庫中的任何一個類別的氣體)，並設計一個排除機制以避免誤判，以提高氣體辨識正確率。
另一方面，現行技術主要是利用多組感測器的結合，以同時偵測多種氣體，但卻也使得整個偵測系統的體積相當大。本論文嘗試從規格與實際應用考量來設計可攜式電子鼻系統晶片，並實際驗證其可行性。

Electronic nose systems have been used to detect odorous molecules in industrial and environmental applications. Most existing algorithms for such applications are based on the concept of nearest-neighbor classification which computes the "distance" between the test odor and a set of known odors (also called training odors), and may erroneously classify an unknown odor as an odor in the training data set. In this thesis, we propose to improve
the situation by treating the unknown odor (if not belonging to the training data set) as an unclassified odor, and use an outlier rejection mechanism to
avoid misjudgment. We also designed and fabricated an integrated system chip (containing sensor interface circuitry, analog-to-digital converter, 8-bit
microcontroller, and memory) for use in a portable electronic nose system. Experimental results are shown to justify the effectiveness of our proposed algorithm and integrated chip.

1 緒論
1.1 研究動機
1.2 相關研究成果
1.2.1 主成分分析法(Principle Component Analysis, PCA)
1.2.2 線性識別分析法(Linear Discrimination Analysis, LDA)
1.2.3 K-最鄰近分類(K-Nearest Neighbor Classification, KNN)
1.2.4 局部加權最鄰近分類法(Locally Weighted Nearest Neighbor Classifier, LWNN)
1.2.5 氣體辨識結果
2 混合氣體分析法
2.1 氣體辨識方法之改進
2.1.1 以距離為基礎的異常偵測方法(Distance-based outlier
detection)
2.1.2 以中位數作為臨界值的K-最鄰近分類法(The
median threshold K-nearest-neighbors classification, MTKNN)
2.1.3 適用於可攜式電子鼻系統之氣體辨識方法
2.2 實驗方法與流程
2.2.1 實驗設計
2.2.2 實驗方法與步驟
2.2.3 氣體訊號擷取
2.3 結果與分析
2.3.1 氣體辨識正確率
2.3.2 拔靴法(Bootstrap Re-sampling)
3 電子鼻系統晶片之設計與整合測試
3.1 晶片規格與設計
3.2 晶片模擬方法
3.2.1 建立模擬環境
3.2.2 產生測試資料
3.2.3 驗證自動化
3.2.4 時序分析
3.2.5 測試考量
3.3 晶片測試方法
3.3.1 建立測試環境
3.3.2 測試流程
3.3.3 測試結果
3.4 電子鼻系統整合測試
3.4.1 測試環境
3.4.2 KNN測試
3.4.3 MTKNN測試
4 結論與未來研究方向

[1] T. Aishima. Aroma discrimination by pattern recognition analysis of re-sponses from semiconductor gas sensor arrays. In Journal of Agricultural and Food Chemistry, pages 752–756. ACS, 1991.
[2] L. D. and N. Ben-Arie. Olfactory receptors. pages 668–674. Current Biology, 1993.
[3] B. Efron. Bootstrap methods: Another look at the jackknife. In The Annals of Statistics, pages 1–26, 1979.
[4] B. Efron. Better bootstrap confidence intervals. In J. Amer. Statist. Assoc., 1987.
[5] A. F and P. C. Fast outlier detection in high dimensional spaces. In Pro-ceedings of the 6th European Conference on Principles of Data Mining and Knowledge Discovery, pages 15–26. ACM, 2002.
[6] D. Hawkins. Identification of outliers. London, 1980. Chapman and Hall.
[7] D. S. B. Jr., S. L. Rose, J. W. Grate, and H. Wohltjen. Correlationof surface acoustic wave device coating responses with solubility properties and chemical structure using pattern recognition. In Analytical Chemistry, pages 3058–3066, 1986.
[8] E. M. Knorr and R. T. Ng. Finding intensional knowledge of distance-based outliers. In Proceedings of 25th International Conference on Very Large Data Bases, pages 211–222, Edinburgh, Scotland, UK, 1999.
[9] Y.-S. Lin. Methods of odor analysis for electronic nose systems. MS Thesis, National Tsing Hua University Institutional Repository, 2009.
[10] A. M. Martinez and A. C. Kak. Pca versus lda. In IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 228–233, 2001.
[11] T. C. Pearce, J. W. Gardner, S. Friel, P. N. Bartlett, and N. Blair. Electronic nose for monitoring the flavour of beers. In Analyst, pages 371–377, 1986.
[12] R. Polikar, R. Shinar, V. Honavar, L. Udpa, and M. Porter. Detection and identification of odorants using an electronic nose. In Proceedings of the IEEE Conference on Acoustics, Speech, and Signal Processing (ICASSP), Salt Lake City, Utah, USA, 2001.
[13] R. R. Reed. Signaling pathways in odorant detection. pages 205–209. Neuron, 1992.
[14] R. S., R. R., and S. K. Efficient algorithms for mining outliers from large datasets. In Proceedings of the international conference on management of data(SIGMOD), pages 427–438. ACM SIGMOD, 2000.
[15] L. I. Smith. A tutorial on principal components analysis. 2002.