在無線通訊系統中應用於影像傳輸的適應性技術之優化

Abstract

本論文旨在研究無線系統中，應用於影像傳輸的適應性技術之聯合優化的方法，其包含了適應性的調變與編碼架構、封包長度動態調整、回傳機制、影像速率動態調整。本論文所提出的跨層整合的封包傳輸架構，尤其是針對於封包長度的優化，將在滿足訊框錯誤率的前提下，最小化期望的傳輸時間。根據無線通道的品質與影像的訊框大小，提出的架構可以用牛頓法或是幾何規劃來實現，且迅速得到可靠的傳輸參數。在以往的文獻中，平均吞吐量優化的研究很豐富，然而對於吞吐量變化量的控制則相對較少。本論文同時考量平均吞吐量與吞吐量的變化量來達到高畫質影像傳輸的服務品質。 此外，提出的聯合優化的方法，可應用於可調式影像串流，其在滿足訊框錯誤率與延遲時限的要求下，可以最小化系統的影像失真程度。根據無線通道的品質與影像的速率-失真曲線，並利用幾何規劃的數學工具，可以有效率的得到近似於廣域的優化參數。同時，我們針對空間多工模式下的多輸入多輸出無線通訊系統，亦提出了一個聯合優化的方法。其利用前述發展出的優化工具，進一步決定各空間通道的封包長度與調變，以最小化期望的傳輸時間。經由分析與模擬的結果顯示，對於低延遲、高傳輸量的無線高畫質影像串流，提出的優化架構是最小化期望傳輸時間的有效辦法。

In this dissertation, we investigate the problem of optimization of adaptive techniques for video transmission in wireless systems. The adaptive techniques include adaptive modulation and coding schemes (AMC), payload length adaptation, retransmission mechanism and video rate adaptation. We propose a new cross-layer packet optimization scheme to minimize the expected transmission time (ETT) at designated frame error rate (FER). While the proposed optimization scheme can be efficiently resolved by Newton’s method or geometry programing (GP), it is feasible to be implemented. In literatures, although systems to optimize average throughput has been extensively investigated, there have been very few efforts on the variation of throughput. In this dissertation, the average and variation of throughput are both considered to guarantee quality of service (QoS) of High-Definition (HD) video transmission. In addition, we investigate H.264/AVC scalable video coding (SVC) based traffic. Due to the layered characteristic of SVC traffic, the joint optimization scheme can minimize distortion of a video streaming under the given delay bound and target FER. Based on the channel quality and rate-distortion parameters, the proposed scheme leverages the theory of GP to resolve the parameters of adaptive techniques globally and efficiently. Furthermore, the joint optimization framework is proposed for wireless multiple-input-multiple-output (MIMO) spatial multiplexing (SM) system. By leveraging the aforementioned joint optimization schemes, it minimizes the expected transmission time of a video frame at designated FER. We show by analysis and simulation that the proposed scheme is a promising tool to minimize expected transmission time for low latency, high data rate wireless HD streaming.