Cooperative Asynchronous Sequential Sensing in Cognitive Radio Networks

感知型無線電網路( Cognitive Radio Networks )科技被視為可以有效的去解決 頻帶使用效率偏低的問題。為了實現感知無線電的功能，感知無線電使用者必須去對主要使用者的資訊去做頻帶偵測( Spectrum Sensing )。在本論文中，我們考慮了具有感知無線電基地台( Cognitive Radio Base Station )的集中式合作頻帶感測架構( Centralized Cooperative Spectrum Sensing )，我們提出了基於序列式檢測( Sequential Detection )的技術來特徵化以及分析感知無線電基地台在感知無線電網路下利用對數相似比率( Log-Likelihood Ratio Test )來對於感測資訊的收集和決策方式。基於這個模型，我們分析了在感知無線電使用者具有時間非同步感測情況下，去找到最佳的停止時間( Stop Time )來縮小整個感知無線電使用者的感測時間，以及當主要使用者在感測時間內有快速的變動產生，然後更敏捷的去偵測主要使用者狀態，有效的去減少感測時間延遲( Detection Delay )以及錯誤機率的發生。最後，我們強健新的機制在當主要使用者的資訊並不完整或未知狀況，如何去精準的猜測主要使用者的感測資訊範圍達到真實的情況。Cognitive Radio Networks (CRN) plays a critical role in efficient and reliable using spectrum. In order to achieve Cognitive Radio technology, secondary users need to sense primary users’ behavior at spectrum. In this paper, we consider a cognitive radio base station to presenting centralized cooperative spectrum sensing scheme. In the scheme, each secondary user computes the Log-Likelihood ratio for its every measurement, and the cognitive radio base station sequentially accumulates those Log-Likelihood statistics and determines whether to stop making sensing. Based on this model, we analyze the secondary users with a time of asynchronous sensing and the secondary users how to find the optimal stop time that let cognitive radio minimum totally sensing time. When the primary user is rapid changes in the sensing slot, we are more agile to detect the primary user’s behavior. We effectively reduce the detection delay and the error probability. Finally, we present a robust new mechanisms when the primary user of the information is incomplete or unknown status, how to accurately estimate the primary users of the sensing information to the real situation