以P2P傳輸技術傳遞即時影音串流，一般而言將產生較大的現場延遲。這現象尤其在採用網狀(mesh-based)結構的P2P串流傳輸技術中更為明顯。其主要的原因來自於P2P串流傳輸技術比一
般串流技術使用較多的資料暫存空間，以使得客戶端間有較高的機率可進行分享。本論文主要分
析P2P串流傳輸技術在傳輸即時影音串流的現場延遲現象，進而探討現場延遲與P2P分享率間如何
達成平衡。同時藉由調整P2P串流傳輸之區塊請求方法而降低P2P串流協定控制封包之額外傳輸成
本。我們以實際實驗分析P2P串流區塊傳輸排程演算法，實驗結果證明在不同類型客戶端中，所
提演算法皆能達到穩定的現場延遲時間，同時達成P2P分享效果。

In general, using peer-to-peer transmission technology to deliver a live video will result in more live-delays. This phenomenon becomes more obvious in the case of mesh-based P2P streaming technology. The main reason comes from the P2P streaming technology to use more buffer to make the peers may have a higher chance to share with each other. This paper analyzes the reasons of causing live-delays in P2P streaming technology and provides an algorithm which can achieve a balance between live-delays and P2P sharing probability. In addition, the algorithm can also reduce control packet overheads by adjusting the number of requests of video segments. We provide an experimental result to show that our algorithm can have stable live-delays especially in heterogeneous environments and still have the effect of P2P sharing.