基於階梯式頻道模型之多頻道無線存取協定

Multi-channel MAC Protocols by Applying Staggered Channel Model for Wireless Networks

王姿琳(Tzu-Lin Wang)

指導教授 ： 張志勇

淡江大學/工學院/資訊工程學系博士班/博士(2014年)

https://doi.org/10.6846/TKU.2014.00687

摘要

為了有效提升無線網路頻譜使用率，研發多頻道無線存取協定已成為無線網路眾多議題中的熱門選項之一。在多頻道環境中，送者如何得知收者所處頻道並切換至該頻道並與收者會面，進而交換控制封包，將是研發多頻道無線存取協定必定遭遇的挑戰之一，稱為”會面問題”。為了解決會面問題，目前多頻道模型可分為三種，分別為Control-Control Based(CCB)、Hopping Based(HB)以及Control-Period Based(CPB)。雖然上述三種頻道模型能解決會面問題，但仍各自存在一些缺點。在CCB頻道模型中，由於節點僅配置一根半雙工天線，因此當傳輸對傳輸資料時，便無法維護完整的頻道使用狀態，當傳輸對完成資料交換後，欲進行下一次資料傳輸時，可能因為離開控制頻道而導致擁有的頻道使用狀態資訊不足，而重覆預約正被其他傳輸對使用的頻道，造成”多頻道隱藏節點問題”。另一方面，在HB頻道模型中，每個節點依照特定跳頻順序進行跳頻，可解決會面問題。但當所有節點一同跳頻至某一頻道並交換控制封包時，將導致網路壅塞及封包碰撞問題。最後，在CPB頻道模型中，所有節點在每一段信標間隔(Beacon Interval)將切換至一預設頻道的控制視窗，以進行節點之間的會面與控制封包交換。但因為每一段信標間隔僅有一個控制視窗被節點使用，導致其他頻道的相同時段並無任何封包傳輸，進而造成網路效能低落。
有鑑於此，本論文提出一新穎的多頻道模型，稱為階梯式頻道模型 (Staggered Channel Model)。此階梯式頻道模型錯開每個頻道上的控制視窗，使其成為階梯形狀。藉由此頻道模型並使用本論文研發之基本傳輸規則，可有效解決會面問題、多頻道隱藏節點問題及低頻寬使用率等問題，進而提升頻道使用率及網路效能。此外，基於此階梯式頻道模型，本論文亦針對偕同網路 (Cooperative Networks)及無線感知網路 (Cognitive Radio Networks)提出相對應的多頻道無線存取協定。首先，無線網路環境由於訊號衰減及干擾等問題，造成資料傳輸成功率大幅下降。為了提升通訊可靠度，本論文採用偕同網路概念，進一步研發CM-MAC協定。在資料傳輸失敗時，CM-MAC利用階梯式多頻道模型的特性，使送者快速地找尋適當數量的合作節點，並與合作節點同時地將資料傳送至收者，以增加傳輸可靠度與網路效能。實驗結果顯示，相較於以往多頻道合作傳輸協定，本論文提出的CM-MAC協定除了大幅提升資料傳輸可靠度外，亦增加網路吞吐量及降低控制成本。而針對無線感知網路方面，本論文提出一SMC-CR-MAC協定，使次級使用者 (Secondary Users) 在不干擾付費使用者 (Primary Users)傳輸的前提下，依舊能有效利用未被付費使用者佔用的付費頻道進行資料輸。而因為付費頻道可能在任意時間與地點被付費使用者佔用，本論文提出之SMC-CR-MAC協定亦針對不同時間與地點出現付費使用者的情況，提出相對應的解決辦法，讓次級使用者在其他可使用的付費頻道繼續未完成的資料傳輸。實驗結果顯示SMC-CR-MAC協定可大幅提升次級使用者的會面成功機率、提升網路吞吐量以及有效降低次級使用者之封包傳輸延遲時間。

關鍵字

多頻道 ； 媒體存取協定 ； 會面問題 ； 頻道模型 ； 偕同網路 ； 無線感知網路

並列摘要

Developing the multichannel Media Access Control (MAC) protocol is the key technique for improving the utilization of wireless spectrum. One of the major challenges to develop the multichannel MAC protocol is the well-known “Rendezvous Problem”. The rendezvous problem occurs since the sender cannot be aware of the channel where the receiver stayed. Therefore, the sender is unable to exchange control packets with its receiver. To overcome rendezvous problem, existing multichannel MAC protocols can be classified into three types, including Control-Control Based (CCB), Hopping Based (HB), and Control-Period Based (CPB) channel models. Although above-mentioned three channel models can resolve rendezvous problem, they still have some disadvantages. In CCB channel model, each node cannot maintain the complete channel usage information since it did not stay at the control channel. That is, when a pair of sender and receiver exchanges data on a data channel, they cannot overhear the control packets exchanged on the common control channel. The pair of sender and receiver may attempt to reserve a data channel where other pair occupied, and hence the Multichannel Hidden Terminal Problem (MHT-Problem) occurs. On the other hand, in HB channel model, each node follows a predefined hopping sequence, hoping to different channels at each time slot. Nevertheless, all nodes switching to a channel at the same time slot lead to the contention and collision phenomena. In CPB channel model, each node should switch to the negotiation window of a predefined channel in every beacon interval, aiming to rendezvous with its receiver and then exchange control packets. However, the low channel utilization problem will be arisen. This occurs because the negotiation windows of all channels other than the default channel will not be used.
This thesis firstly proposes a novel multichannel model, called Staggered Channel Model. In the proposed staggered channel model, the negotiation windows of all channels are arranged to be staggered. Moreover, by applying the proposed staggered channel model and basic transmission rules, each node overcomes rendezvous, MHT, and low channel utilization problems, improving the network capacity. In addition, this thesis proposes two multichannel MAC protocols based on the staggered channel model for cooperative networks and cognitive radio networks. In the signal-fading and interference-rich network environment, high probability of failure transmission results in low network performance. To improve communication reliability, the proposed CM-MAC protocol allows that each sender selects enough number of cooperative nodes as soon as possible when transmission fails. Then, the sender simultaneously transmit data with the selected cooperative node to the receiver. Simulation results show that the proposed CM-MAC protocol not only scientifically improves the packet delivery ratio and network throughput but also decreases the control overhead.
For cognitive radio networks, this thesis proposes the SMC-CR-MAC protocol to continue the transmission of secondary users (SUs) without interfering with the data exchange of primary users (PUs). Since the PU may appear at any time and location, the proposed SMC-CR-MAC protocol takes the timing and location of PU appearance into account and develops different policies to continue the unfinished data exchange of SU pairs. Performance evaluation shows that the proposed SMC-CR-MAC protocol can improve the successful rendezvous probability between SU sender and receiver, enhance network throughput, and reduce the transmission delay of SU pair when PU is detected.

並列關鍵字

Multichannel ； MAC protocol ； Rendezvous problem ； Channel model ； Cooperative networks ； Cognitive radio networks

參考文獻

[2]J. Mo, H. S. W. So, and J. Walrand, “Comparison of Multichannel MAC Protocols,” IEEE Transactions on Mobile Computing, vol. 7, no. 1, pp. 50-65, Jan. 2008.

[3]M. Natkaniec, K. Kosek-Szott, S. Szott, and G. Bianchi, “A Survey of Medium Access Mechanisms for Providing QoS in Ad-Hoc Networks,” IEEE Communication Survey & Tutorials, vol. 15, no. 2, pp. 592-620, second quarter 2013.

[4]H. J. Lei, C. Gao, Y. C. Guo, Z. Z. Zhang, “Survey of Multi-Channel MAC Protocols for IEEE 802.11-Based Wireless Mesh Networks,” The Journal of China Universities of Posts and Telecommunications, vol. 18, no. 2, pp. 33-34, Apr. 2011.

[5]J. Lee, J. Mo, T. M. Trung, J. Walrand, and H. S.W. So, “Design and Analysis of a Cooperative Multichannel MAC Protocol for Heterogeneous Networks,” IEEE Transactions on Vehicular Technology, vol. 59, no. 7, pp. 3536-3548, Sep. 2010.

[6]Y. Zhang and L. Lazos, “Countering Selfish Misbehavior in Multi-channel MAC Protocols,” IEEE Infocom, Apr. 2013.