| DC 欄位 |
值 |
語言 |
| DC.contributor | 光電科學與工程學系 | zh_TW |
| DC.creator | 余昆霖 | zh_TW |
| DC.creator | Kun-Lin Yu | en_US |
| dc.date.accessioned | 2021-7-8T07:39:07Z | |
| dc.date.available | 2021-7-8T07:39:07Z | |
| dc.date.issued | 2021 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=108226029 | |
| dc.contributor.department | 光電科學與工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 近年來發展室內定位已有許多方向的技術發展，有著各種的解決方法，而這些解決方法皆有各自的優點與侷限，但高精度的定位方式需要較高單價的設備去做佈署，本研究目的為設計一套新型定位演算法，在準確度上與設備成本取得最佳平衡。隨著許多新型個人移動設備如手機、穿戴裝置等，我們只要將設備的藍芽功能開啟，就能夠對其場域進行定位，感測位置的應用可更加廣泛。
本研究取用12個藍芽模組平均佈置於室內空間中，藉由人員在室內中影響到這些藍芽訊號做為機器學習的訓練依據，利用TensorFlow的學習框架，以卷積神經網路的方式來做訓練與預測，設計最佳的模型參數，大幅降低了訓練模型時間，在本研究的場域之中，藉由不同的藍芽訊號強弱判斷室內人員位置，定位精準度達到93.46%。 | zh_TW |
| dc.description.abstract | In recent years, there have been technological developments in many directions in the development of indoor positioning, and there are various solutions, and these solutions have their advantages and limitations, but high-precision positioning methods require higher unit price equipment for deployment. This study′s purpose is to design a new positioning algorithm to achieve the best balance between accuracy and equipment cost. With many new personal mobile devices such as mobile phones, wearable devices, etc., we can locate the field as long as the Bluetooth function of the device is turned on, and the application of sensing location can be more extensive.
This study uses 12 Bluetooth modules to be evenly arranged in the indoor space. These Bluetooth signals are used as the basis for machine learning training by personnel in the room. The TensorFlow learning framework is used to convolutional neural networks. The training and prediction methods are used to design the best model parameters, which greatly reduces the training model time. In the field of this research, the location of indoor personnel is judged by different Bluetooth signal strengths, and the positioning accuracy reaches 93.46%. | en_US |
| DC.subject | 機器學習 | zh_TW |
| DC.subject | 室內定位 | zh_TW |
| DC.subject | 低功耗藍芽 | zh_TW |
| DC.subject | Machine learning | en_US |
| DC.subject | Indoor positioning | en_US |
| DC.subject | Bluetooth Low Energy | en_US |
| DC.title | 以藍芽訊號強度指標及機器學習建構室內定位之研究 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Research on Bluetooth Signal Strength Indicator and Machine Learning to Construct Indoor Positioning | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |