具忙音頻道多速率載波偵測無線環境之傳輸速度感知功率控制協定

Abstract

隨著無線行動網路快速的發展，使用無線網路的設備數量有爆炸性的成長，使得無線環境中傳播著非常巨大的資料流量，如何避免傳送的封包發生碰撞成為了非常重要的議題。IEEE 802.11使用CSMA/CA機制來避免碰撞發生，更進一步使用了RTS/CTS來保護傳輸中的資料。然而無線網路的環境中，傳輸的成功與否和接收端所收到的訊號強度及所受干擾(SINR)有關，大多數成功的傳輸對於使用最大傳輸功率是不需要的，使用多餘的發射功率在進行傳輸時，反而造成了資源的浪費，更造成鄰近節點的干擾，再加上RTS/CTS裡的virtaul carrier sense的特性，可能造成嚴重的暴露節點問題，所以應當考慮接收端的狀態來限制傳輸功率。 因此在本論文中，我們提出了一個速率及功率控制演算法，打破傳統傳與不傳的二元決策(Binary Decision)，以接收端為中心，根據各傳送速率所需的SINR及目前所收到的訊號SINR，藉由busy-tone來限制周圍的傳輸，另一傳送端接收來自busy tone的能量訊號後，推算出可使用多少功率傳輸而不影響其他node的接收，最後根據傳輸的情形調整速率，增加頻道的空間使用率，進而增加整體網路的吞吐量。 我們使用NS3中的 Wi-Fi model，以IEEE 802.11ac為基礎模擬我們設計的演算法，分析不同節點密度下網路的吞吐量，並將結果與IEEE 802.11ac無線網路環境進行比較。經由實驗結果可以發現，我們所提出的演算法可以隨著現在的環境變化來動態地調整傳輸功率，並維持良好的網路吞吐量及碰撞率，同時具有可增加多組同時傳輸的特性。

With the rapid development of wireless mobile networks, the number of wireless network equipment also grows exponentially. Considering an enormous amount of data traffic, packet collision avoidance has become an important yet challenging issue. IEEE 802.11 uses CSMA/CA mechanism to avoid collisions, and RTS/CTS is further used to protect the data being transmitted. However, in such wireless networks, successful transmission depends heavily on the perceived SINR (Signal-to-Interference-and-Noise Ratio) at the receiver. We observe that using maximum transmission power doesn’t always yield the best result. Rather, it leads to redundant transmitter power, wasting energy resources, and even causing a greater interference to neighboring nodes. Coupled with RTS/CTS virtual carrier sense, the IEEE 802.11 protocol suffers from serious exposed terminal problems. As such, we suggest to take the environmental status of the receiver into account, and properly limit transmission power that affects neighboring nodes. In this thesis, unlike traditional binary decision of transmitting or deferring, we propose a rate-aware power control algorithm which works as follows. Receiver uses the current transmission rate (from transmitter) to calculate the minimum required SINR for a successful transmission, and sends busy-tone signal to the surrounding nodes. The surrounding nodes utilize the busy-tone signal to estimate how much power can be used without affecting the reception of other nodes and adjust their transmission power accordingly. Our proposed algorithm attempts to maximize the channel’s spatial usage, thus effectively increasing the overall network throughput. We adopt the Wi-Fi model in NS3 to simulate our proposed algorithm based on IEEE 802.11ac. We conduct experiments for different node densities, analyze the network throughput and compare the performance with other existing algorithms. The experimental results show that our proposed algorithm can dynamically adjust the transmission power to adapt to the environmental changes while maintaining excellent network throughput and low collision rate. Furthermore, it effectively increases the number of simultaneous successful transmissions.