無線感測網路(WSN)是近年來非常熱門的技術。隨著電子技術快速進展，體積、耗電、成本的持續改善，使得WSN可望在各種工商業環境下大量建置，例如︰溫、濕度、亮度、震動等環境監測應用。但是受限於無線感測網路低功耗、低頻寬、低儲存能力，多媒體的傳輸仍是極大的挑戰。本論文提出一個適應性的無線感測網路通訊方法，設計一個新的通訊協定架構，架構使用雙協定縮減檔頭，在大量資料傳輸時可以增加效率；同時我們實做了嵌入式JPEG影像編解碼，在硬體資源有限(hardware resources constrained)嵌入式環境下,提供一個高效能的無線影像通訊編碼方法。這個方法可以依照網路頻寬狀況和特定監控需求調整影像品質，以滿足品質/傳輸速率的適應性。 本論文以MIAT實驗室所研發的STM32模組板作為無線感測網路控制器晶片的驗證平台，搭配2.4GHz RF模組實作CMOS影像感測器的取像、JPEG壓縮和無線影像傳輸。實驗結果顯示，我們的實驗系統可以進行可靠的影像傳輸，可以即時調整影像品質，並且可以在多台從屬端的環境下進行資料即時通訊。本論文提出的適應性無線感測網路方法可以應用於小範圍、低功耗、硬體資源受限的分散式視訊監控應用。The WSN has been widely used in various applications of environmental information monitoring like of the temperature, humidity, brightness, and vibration etc., However, it is still a great challenge to transmit multimedia data in a WSN because of its low-power, low-bandwidth and low storage capacity.This thesis presents an adaptive wireless image sensor network. We propose a new communication protocol to increase the efficiency for large amount of image data transmission by using dual-protocol architecture. By implementing an embedded JPEG codec, we also provide an adaptive wireless image communication method which can be used in an un-reliable communication environment with extreme hardware resource constraints. According to the status of network bandwidth and specific monitoring requirement, the image quality is dynamically adjusted to find the balance between transmission rate and image quality.Our experiments show that our system achieves a reliable image transmission, real-time image quality adjustment and single-master/multiple-slave communication. The proposed image WSN is suitable to be applied in the small-scale distributed video surveillance with low power and hardware-constrained.
