Throughput Maximization for Cognitive Radio Networks with Wideband Spectrum Sensing

採用寬頻頻譜偵測之感知無線電網路的傳輸量最大化

蔡宗穎(Zong-Ying Tsai)

指導教授 ： 王晉良

國立清華大學/電機資訊學院/通訊工程研究所/碩士(2011年)

https://doi.org/10.6843/NTHU.2011.00462

摘要

Cognitive radio (CR) is an emerging technology for enhancing the efficiency of wireless spectrum utilization. In cognitive radio systems, secondary users (SUs) are allowed to access the frequency bands of primary users (PUs) if the interference with PUs is avoided. Thus, the spectrum sensing is an important issue in CR applications. In periodic sensing framework, the accuracy is getting better while we increase the sensing time. However, the longer the time it sensed, the lower the throughput that CR systems could be achieved.
A wideband spectrum sensing method has been proposed, which considers multiple bands jointly and solve the optimization problem to maximize the achievable throughput of CR networks with a tunable sensing time. However, the objective function in the previous works only considers the situation that the PUs are absent; it is not practical and could not achieve a real maximum throughput.
In this thesis, we reformulate a new objective function which considers all possible scenarios in which SUs can access the PUs’ spectrum to transmit data under the sufficient protection of PUs, and proposed a method to solve this optimization problem. In this problem, a two-stage iterative method is adopted. In the first stage, with a given sensing time, we determine the optimal decision thresholds for each subband to obtain the maximal throughput of CR networks. In the second stage, using the information obtained from the first stage, we evaluate a sensing time that maximized the achievable throughput of CR networks, and then substitute the updated sensing time into the first stage. By using the two-stage iterative method, the optimal solution can be obtained easily. Furthermore, the computational complexity of the proposed method achieves less than half of the complexity of existing optimization methods. In this work, we propose a more practical scenario as well as improve the achievable throughput of CR networks, and we though this work can be of enormous value to CR applications.

關鍵字

感知無線電 ； 多頻帶偵測 ； 傳輸量最大化

並列摘要

感知無線電(cognitive radio)技術具備有效提升無線頻譜使用效率之能力。在感知無線電系統中，次要使用者(secondary user)藉由頻譜偵測(spectrum sensing)技術探知主要使用者(primary user)之頻帶使用情形。在週期性頻譜偵測架構(periodic sensing framework)下，越長的偵測時間可以得到越準確的結果。然而，卻會導致感知無線電系統傳輸資料的時間變短，致使可提供之傳輸量效能不佳。
寬頻頻譜偵測(wideband spectrum sensing)概念考慮多個子頻帶同時進行頻譜偵測，利用可調式偵測時間與各個頻帶上的判斷門檻值來達到傳輸量的最大化。先前的傳輸量最大化研究之目標函數僅考慮主要使用者不存在時所能提供的傳輸量，然而，在實際的情況下，¬¬主要使用者並非一直不存在，以致此最佳化結果無法達到傳輸量最大值。
在此篇論文中，我們提出全新的目標函數，此目標函數考慮所有次要使用者與主要使用者之間交互影響之情形，透過新的目標函數可以得到感知無線電系統於真實環境應用所能達到的最大傳輸量。另一方面，我們提出兩階段遞迴式的方法來求得所需的偵測時間與所需的判斷門檻值。在第一階段中，利用固定偵測時間找出一組最佳的判斷門檻值，接著利用這一組最佳的判斷門檻值在第二階段求得對應的最佳偵測時間。隨後利用遞迴式的方法不斷逼近最佳的偵測時間與判斷門檻值，利用此方法可以求得最佳系統參數。由模擬結果得知，第三或第四個迴圈中即可逼近參數最佳解。而在兩階段遞迴式的方法中，複雜度主要來自於第一階段中寬頻頻譜偵測判斷位於邊界值的子頻帶。而對於此問題，我們提出了複雜度更低的演算法，與舊有的演算法相比足足降低一半的運算複雜度，而當寬頻頻譜偵測中的子頻帶越多時，我們提出的演算法能降低越多的運算複雜度，也使得寬頻頻譜偵測在實際應用上更容易實行。

並列關鍵字

cognitive radio ； multiband joint detection ； throughput ； MJD ； MSJD

參考文獻

[3] M. McHenry, E. Livsics, T. Nguyen, and N. Majumdar, “XG dynamic spectrum access field test results,” IEEE Commun. Mag., vol. 45, pp. 51-57, Jun. 2007.

[5] J. Mitola and G. Q. Maguire, “Cognitive radios: making software radios more personal,” IEEE Personal Communications, vol. 6, no. 4, pp. 13-18, Aug. 1999.

[8] IEEE 802.22, Working Group on Wireless Regional Area Networks (WRAN), http://grouper.ieee.org/groups/802/22/.

[9] S. Haykin, “Cognitive radio: Brain-empowered wireless communications,” IEEE J. Sel. Areas Commun., vol. 23, pp. 201-220, Feb. 2005.

[10] H. Kim and K. Shin, “Efficient discovery of spectrum opportunities with MAC-layer sensing in cognitive radio networks,” IEEE Trans. Mobile Comput., vol. 7, no. 5, pp. 533–545, May 2008.