標題: 無基礎架構的無線網路下減少隱藏主機效應之加強型媒體存取控制協定
An Enhanced MAC Protocol for Minimizing Hidden Terminal Effects in Ad-Hoc Networks
作者: 王馥昶
Fu-Chang Wang
王國禎
Kuochen Wang
資訊科學與工程研究所
關鍵字: 無基礎架構的無線網路;無線媒體存取控制協定;隱藏主機問題;媒體存取控制;ad-hoc network;CSMA/CA;hidden terminal problem;medium access control
公開日期: 1999
摘要: 隨著電腦和通訊科技的進步,行動計算變得愈來愈重要了。人們希望無論在什麼時間、什麼地點都能夠獲得資訊。在一個沒有基礎架構網路的地方,人們使用無基礎架構的無線網路,以便不管在什麼時間、什麼地點都能得到各項計算服務。然而,無線網路的頻寬是有限的,而且隱藏主機的效應會減少傳輸量。為了減少隱藏主機效應的影響,以獲取較好的性能,我們提出了一個新的無線媒體存取控制協定。假如在一段短的時間內,有幾個行動主機發送RTS訊號給同一個目的地行動主機,依據原來IEEE 802.11之CSMA/CA協定,這些主機將會進入後退的程序,而且無法立即傳送資料。在我們提出來的方法中,若是接收到的RTS訊息沒有碰撞的話,那目的地主機就會接收到這些訊息,並給那些來源主機一個傳送的先後順序,而那些來源主機就不會進入後退的程序。我們把目的地主機當成在有基礎架構的網路□的基地台一樣,旨在減少那些來源主機花在後退程序的時間。隱藏主機的效應可以分成兩個情況,一種是隨意隱藏主機效應,另外一種是群集隱藏主機效應。我們使用ns-2模擬工具,針對這兩種情況去評估我們的方法。實驗的結果顯示,隨著不同的隱藏主機率,在群集隱藏主機效應的情況下我們方法的傳輸量比無線媒體存取控制協定(CSMA/CA)好2%到27%,而在隨意隱藏主機效應情形下則好1%到16%。
As computer and communication technologies advance, mobile computing is getting more and more important. People want to get information at anywhere and anytime. In an infrastructureless area, people may use an ad-hoc network to acquire computing services no matter where they are and what time it is. However, wireless bandwidth is limited and the hidden terminal effect will result in reduction of transmission throughput. We propose a new wireless MAC protocol (HT-MAC) to minimize the hidden terminal (HT) effect in order to achieve better performance. If several mobile terminals send request to send (RTS) messages to the same destination in a short period of time, in the original CSMA/CA of IEEE 802.11, the terminals will get into the backoff procedure and can not transmit data right away. In our method, if arrival RTS messages did not have collisions, the destination terminal will accept the RTS messages and give each source terminal a priority to transmit, and the source terminals will not get into the backoff procedure. We treat the destination terminal just like a base station in the infrastructured network and intend to reduce the time source terminals spent in the backoff procedure. The hidden terminal problem is classified into two models: random hidden terminal model and clustered hidden terminal model. A simulation tool, ns-2, is used to evaluate our HT-MAC under these two models. Experimental results show that, with the variations of hidden terminal rates, the throughput of our HT-MAC is 2% - 27% and 1% - 16% better than that of CSMA/CA under clustered hidden terminal model and random hidden terminal model, respectively.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT880394053
http://hdl.handle.net/11536/65550
顯示於類別:畢業論文