即時無線影像傳輸系統

Abstract

以往影像傳輸的瓶頸在於無線的傳輸速度,隨著無線寬頻時代來臨,無線網路系統利用多頻道方式來增加資料傳輸量,以現今的技術而言,要達到高畫質影像的傳輸頻寬已不困難,然而,我們現在所面臨的問題,就是如何設計出有效率,而且低成本的方式來處理大量的即時影像資料,並且透過多個頻道來傳送這些資料。 本論文主要的目的在於提出一嵌入式即時影像傳輸系統架構,利用硬體將欲傳輸的影像分割成許多小塊並平行壓縮和解壓縮,資料流很容易在多頻道無線網路下平行傳送,多頻道傳輸除了傳輸量大之外,還擁有較好的抗干擾和安全性。 我們利用標準的802.11無線網卡和FPGA 平台來實現上述的架構。根據實驗結果顯示,我們能夠正確的傳輸影像,即使在不穩定無線傳輸環境中也能正常工作,如災害現場的應用等。另外,當我們欲提高影像畫質,未來只需增加無線網卡的數量,提高傳輸率即可,因此,此架構很容易實現高畫質影像傳輸。

In the past, the challenge for wireless video transmission was mainly the datarate, which affected the video quality. However, with the advent of ultra-wide band (UWB) radio technologies, wireless network systems started to adopt multiple channels to increase the datarate. While it is possible to achieve satisfactory transmission datarates for high-definition wireless video transmission today, efficiently processing the large quantity of real-time video data and properly managing the multi-channel transmission have surfaced as major challenges in the design of a wireless video systems. In this thesis, we proposed a framework and built a prototype to achieve real-time video transmission. In the proposed framework, hardware components first divide the images that we intend to transmit into many small blocks of data, and then compress and decompress the blocks in parallel before and after wireless transmission, so that the data streams can also be transmitted in parallel through multiple channels to deliver video with higher datarate, better security, and more tolerance to interference. Our prototype employed a standard 802.11 wireless network card into an FPGA platform to implement the aforementioned framework. The results from our experiments showed that the transmission system was capable of delivering video frames even in a unreliable wireless environment, which made it an applicable design in disaster scenes. In addition, the proposed framework is extensible as the number of wireless network cards could be further increased to provide the datarate required by high-denition video transmission.