合作式頻譜感測在多數決法則下次要使用者傳輸功率配置之研究

Abstract

本篇論文研究在感知無線電網路中，透過調整次要使用者的傳輸功率以及門檻值，以達到最大化在融合中心之全域偵測機率的目地。本研究考慮硬決策合作式感測，次要使用者根據在監聽通道上的觀測資料使用似然比檢驗法產生二元決策，再將決策經由回傳通道傳送給融合中心。本研究假設次要使用者與融合中心之間的回傳通道為獨立同分佈的瑞利衰落通道，融合中心使用多數決法則產生最終決策。首先，在假設的系統模型下，本研究推導出在融合中心的全域偵測機率與全域誤警機率的準確公式。接著考慮回傳通道在高訊雜比環境下，得到融合中心的全域偵測機率與全域誤警機率的解析近似公式。為了提升系統偵測效能，在滿足全域誤警機率與傳輸功率限制條件的前提下，本研究提出最大化融合中心的全域偵測機率之最佳化問題。模擬結果顯示經由搜尋演算法設計次要使用者之傳輸功率與門檻值得到的系統偵測效能確實優於均勻分配次要使用者傳輸功率的系統偵測效能。

In a cognitive radio setting, this thesis studies transmit power allocation as well as local thresholds allocation for secondary users (SUs) with the aim of maximizing the global probability of detection at the fusion center (FC). This work considers hard decision (HD) based cooperative sensing (CS). Based on the received observations on the listening channel, each SU uses likelihood ratio test (LRT) to generate a binary decision and then sends to the FC over error-prone reporting channel. We assume the reporting channels are independent and identically distributed (i.i.d.) Rayleigh fading channels and the FC employs the majority rule to make a final decision. The exact formulae of the global probability of detection and global probability of false alarm at the FC are derived. Furthermore, when the reporting channels are in high SNR regime, we derived the analytic and approximate formulae of the global probability of detection and the global probability of false alarm at the FC. To improve detection performance in our system model, we formulate an optimization problem which could be described as follows: maximize the global probability of detection at the FC while satisfying global probability of false alarm and transmit power constraints. When comparing global probability of detection under uniform transmit power allocation with non-uniform transmit power allocation, simulation results show that improved detection performance at the FC is achieved.