近年來，巨量多人線上遊戲(massive multi-player online game, MMOG)已經成為網際網路中最熱門的應用之一。遊戲玩家們操作著各自的化身(avatar)，在巨大的虛擬遊戲世界中進行遊覽與互動以獲得樂趣及成就感。部分巨量多人線上遊戲透過主從式(client/server)架構讓使用者透過文字的方式進行交流，這樣的溝通方式不僅打擾使用者體驗遊戲內容，主從式架構的設計也因為伺服器計算能力以及頻寬的限制而導致較低的系統擴展性(scalability)。由於同儕式(peer-to-peer)架構一般而言具有較高的擴展性，這激發我們在本論文中提供一種基於人類行為模式，適用於巨量多人線上遊戲的浸入式語音交談 (Immersive Voice Communication, IVC) 技術。當使用者進入虛擬環境中通常僅會與互動範圍(area of interest, AOI)內的所謂鄰居使用者互動，而浸入式語音交談的目的在於玩家可以更輕鬆的與互動範圍中的鄰居使用者語音交談。我們觀察人類行為模式發現，當人類專注傾聽某一聲音時，對於環境中其它聲音較不在意。因此，我們假設一個使用者會專注傾聽一個鄰居使用者的聲音而成他的傾聽者(listener)，但較不在意其他使用者的聲音而成為他們的偶聽者(overhearer)。當使用者說話時會送出較佳音質的傾聽訊息(listening message)給傾聽者，而僅傳送一般音質的偶聽訊息(overhearing message)給偶聽者以節省頻寬的使用。所提技術使用網路座標系統(network coordinate system, NCS)以估計任意兩使用者的資料傳送延遲，並在轉傳語音資料時，優先選擇網路座標較接近的鄰居使用者轉傳資料以縮短語音延遲。所提技術並採用適應式k-ary樹(adaptive k-ary tree, Ak-tree)，依據使用者的上傳能力以決定一個使用者可以轉傳語音資料給多少數量的鄰居使用者點，以充分使用可用頻寬。在本論文中，我們並進行許多分析與模擬實驗以展示所提技術的效能。

In recent years, massively multiplayer online games (MMOGs) have become one of the most popular applications over internet. Players control their avatar to navigate the virtual world and interaction with other players to have fun and accomplishment. However, in most MMOGs, communication between players is based on text by client/server architecture, which is inconvenient and have scalability problem in computation power and bandwidth by using client/server architecture. Usually, peer-to-peer architecture have higher scalability, therefore, this paper proposes a human-behavior based immersive voice communication system for MMOGs. When players entered the virtual world, a player usually only interact with other players in the his/her area of interest (AOI) called Neighbor. Making the voice communication between Neighbor more convenient is the goal of IVC. When we observe human-behavior, when people focus on certain voice, they will feel careless about other voice in the environment. Therefore, we assume a player will become a listener when he focus on one of his neighbors, and become a overhearer when he careless about the others. Player will send voice data with better quality to listener and send voice data with normal quality, called overhear, to overhearer to save the consumption of bandwidth. The technique we proposed using network coordinate system (NCS) to evaluate the latency between any users. Player will prefer to send voice data to the neighbor with proximate position in NCS, to reduce latency when forwarding. The technique we proposed using adaptive k-ary tree (Ak-tree). According to user’s upload capacity to adjust how many neighbor to forward at the same time to use the bandwidth completely. We perform simulation experiments to show that proposed technique.

[1] World of warcraft, http://us.battle.net/wow/en/.
[2] Teamspeak, http://www.goteamspeak.com/.
[3] Skype, http://www.skype.com/.
[4] Ventrilo, http://www.ventrilo.com/.
[5] Ashwin Bharmbe, John R. Douceur, Jacob R. Lorch, Thomas Moscibroda, Jeffrey Pang, Srinivasan Seshan, Xinyu Zhuang, “Donnybrook: enabling large-scale, high-speed, peer-to-peer games,” Proceedings of SIGCOMM , August 17, 2008,
[6] Frank Dabek, Russ Cox, Frans Kaashoek, and Robert Morris, “Vivaldi: A Decentralized Network Coordinate System,” Proceedings of SIGCOMM, August 2004.
[7] Kai Chen, Klara Nahrstedt, “Effective Location-Guided Tree Construction Algorithms for Small Group Multicast in MANET,” INFOCOM 2002. Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies. Proceedings. IEEE.
[8] Shun-Yun Hu, Shao-Chen. Chang and Jehn-Ruey Jiang, “Voronoi State Management for Peer-to-Peer Massively MultiPlayer Online Games,” In Proceeding of the 5th IEEE Consumer Communication and Networking Conference (CCNC), p.1134-1138, 2008.
[9] Shun-Yun Hu, Jui-Fa Chen and Tsu-Han Chen, “VON: A Scalable Peer-to-Peer Network for Virtual Environments,” IEEE Network, vol.20, no. 4, Jul./Aug. 2006, p.22-31.
[10] I.T. Union. ITU-T Recommendation P.59, Artificial conversational speech, 1993.
[11] Tonio Triebel, Benjamin Guithier, Thomas Plotkowiak, Wolfgang Effelsberg, “Peer-to-peer Voice Communication for Massively Multiplayer Online Games,” In Proceeding of the 6th IEEE Consumer Communication and Networking Conference (CCNC) , p.1282-1286,2009.
[12] Jehn-Ruey Jiang, Hung-Shiang Chen, “Peer-to-Peer AOI voice chatting for massively multiplayer online games,” Proceeding of the 13th International Conference on Parallel and Distributed Systems, p.1-8, 2007.
[13] Nelson Cowan, “The magical number 4 in short-term memory﹕A reconsideration of mental storage capacity,” Behavioral and Brain Science 24, p.87-185.
[14] Roger Zimmermann and Ke Liang. “Spatialized Audio Streaming for Networked Virtual Environments,” In MM’08﹕Proceeding of the 16th ACM International Conference on Multimedia, p.209-308, 2008.
[15] Cisco, http://www.cisco.com/en/US/tech/tk652/tk698/technologies_tech_note09186a0080094ae2.shtml
[16] Ashwin R. Bharambe, Cormac Herley, and Venkata N. Padmanabhan, “Analyzing and Improving a BitTorrent Networks Performance Mechanisms,” In Proceedings of IEEE INFOCOM, Barcelona, Spain, Apr. 2006.
[17] Krishna P. Gummadi, Stefan Saroiu, and Steven D. Gribble. “King: Estimating latency between arbitrary Internet end hosts.” In Proc. of SIGCOMM IMW 2002, pages 5–18, November 2002.
[18] Chao-wei Hung. “Bandwidth- and Latency-Aware Peer-to-Peer FriendCast for Online Social Network.”, 2010.
[19] International Telecommunication Union. ITU-T Recommendation G.114 One-way transmission time, 2003.
[20] Luis Valente, Aura Conci, and Bruno Feijó, “Real time game loop models for single-player computer games.” In Proceedings of the IV Brazilian Symposium on Computer Games and Digital Entertainment, 89-99, 2005
[21] Wegel, R. L., and Lane, C.E.(1924). “The auditory masking of one sound by another and its probable relation to the dynamics of the inner ear,” Phys. Rev. 23, 266-285.