[[alternative]]Using Wavelet-based Denoising Method to Improve the Performance of Low-cost Ozone Sensor

[[abstract]]近年來，人們對於環境污染嚴重性的認知有所提升，開始試圖透過各種方式進行偵測和分析。現今除了公家機關設置的大型觀測站點外，隨著無線網路的涵蓋範圍逐漸完善，搭配上物聯網（IoT）應用倡導萬物皆可聯網的特性，微型自動化的感測系統已經開始陸續商業化且出現在我們周圍。這類系統的應用範圍，小至單純的居家環境和學校單位、大至工業區及機械化農場等。因此要如何針對各種環境下的需求，盡可能在低成本狀況下自行建置物聯網空汙設備以進行空汙資料蒐集、傳遞、保存以及後續分析，已成為當今重要的研究課題之一。本文提出一個小型且方便佈署的環境氣體感測器規劃，將資料透過 MQTT 傳輸協定進行資料傳遞。並利用 Python 帶入小波降噪法、k 最近鄰插補演算法等演算法應用，以減少存入後端資料庫時的資料雜訊。藉此提高資料的整體可用性和穩定度，以利後續製作更多加值應用。[[abstract]]These years, people have improved their awareness of serious environmental pollution,and started to detect and analyze it through many ways. Nowadays, espite the large observation station that set by government, with wireless network coverage gradually complete,match with the characteristic that IoT advcating every ccould connect with network,micro-automatic sensor systems have successively commercialized and shown up in our daily life.The system can be used from the home environment, school unit, industrial area,and mechanized farm, etc.Therefore,how to against with requirement in various environment,and self-building IoT air pollution equipments while reduce the cost,to deal with collecting,transferring,preserving,and analyzing data have become one of the crucial issues these days.In this paper, we propose a small and easy to deploy environmental gas sensor plan to transfer data through the MQTT transmission protocol. Use Python to bring in applications such as wavelet noise reduction and k-nearest neighbor interpolation algorithms, to reduce the data noise when depositing into the back-end database, and improve the overall availability and stability of the data for more value-added applications in the future