本文選題：時域延續(xù)性　切入點：馬爾可夫轉移特性　出處：《華僑大學》2017年碩士論文　論文類型：學位論文

【摘要】：認知無線電是一種提高授權頻段頻譜利用率的新興技術。頻譜檢測是其關鍵任務之一。本文分析主用戶的行為特性,并研究基于該特性的頻譜檢測技術。本文首先介紹了認知無線電的背景、概念和關鍵技術,闡述了頻譜檢測技術。接著,對現(xiàn)有的幾種檢測技術進行分析和比較。其中,能量檢測法因為結構簡單、計算量小,得到廣泛使用。其次,本文分析了主用戶的時域延續(xù)性,介紹了兩時隙時域延續(xù)性模型,并研究基于該模型的快速能量檢測算法。該算法將前兩次能量檢測的檢測結果當作主用戶在前兩個時隙的實際狀態(tài),然后基于時域延續(xù)性預判主用戶的當前狀態(tài),從而減少檢測次數(shù),達到快速檢測的目的。其中,狀態(tài)預判可基于OR準則或者AND準則進行,分別對應OTPED算法和ATPED算法。本文在先驗等概的條件下推導了這兩種算法的檢測精度,分析了它們的檢測次數(shù)。和傳統(tǒng)算法相比,上述算法可以在檢測精度近似的條件下減少17~20%的檢測次數(shù)。再次,本文介紹了主用戶的馬爾可夫轉移特性,并基于該特性預測主用戶的當前狀態(tài),從而動態(tài)調整判決門限,提高頻譜檢測精度。首先假定主用戶前一時隙的實際狀態(tài)已知,在恒虛警條件下,推導出檢測概率的理論上界。然后,為了解決主用戶實際狀態(tài)未知的問題,利用檢測結果近似實際狀態(tài),提出了馬爾可夫恒虛警能量檢測(MCFED)算法和改進的MCFED(IMCFED)算法。MCFED算法在高信噪比情況下具有很高的檢測概率,但在低信噪比區(qū)域性能較差。IMCFED則始終優(yōu)于傳統(tǒng)的頻譜檢測算法。最后,為了兼顧虛警概率和漏檢概率的影響,本文研究利用主用戶馬爾可夫轉移特性減少貝葉斯代價。首先提出了馬爾可夫貝葉斯能量檢測(MBED)算法。其思想是將前一時隙的檢測結果看作主用戶的實際狀態(tài),并根據(jù)檢測結果選擇當前時隙的判決門限。該算法計算量小,信噪比較高時性能較好。為了克服MBED算法在低信噪比條件下檢測精度較差的問題,本文又提出了改進的MBED(IMBED)算法。該算法根據(jù)預測概率對判決門限進行動態(tài)調整,能夠進一步減少貝葉斯代價,且適用的信噪比區(qū)域更廣。
[Abstract]:Cognitive radio is a new technology to improve spectrum efficiency of authorized frequency band. Spectrum detection is one of its key Ren Wuzhi. Firstly, this paper introduces the background, concept and key technology of cognitive radio, and expounds the spectrum detection technology. Then, it analyzes and compares several existing detection techniques. The energy detection method is widely used because of its simple structure and low computational cost. Secondly, the time-domain continuity of the primary user is analyzed, and the time-domain continuity model of two time slots is introduced. The fast energy detection algorithm based on this model is studied, which regards the first two energy detection results as the actual state of the primary user in the first two time slots, and then prejudges the current state of the primary user based on the continuity of time domain. In order to reduce the number of times of detection and achieve the purpose of fast detection, the state prediction can be based on OR criterion or AND criterion, corresponding to OTPED algorithm and ATPED algorithm respectively. In this paper, the detection accuracy of these two algorithms is deduced under the condition of priori probability. The detection times of these algorithms are analyzed. Compared with the traditional algorithms, these algorithms can reduce the detection times by 17% or 20% under the condition of approximate detection accuracy. Thirdly, this paper introduces the Markov transfer characteristics of the primary users. Based on this characteristic, the current state of the primary user is predicted, and the decision threshold is dynamically adjusted to improve the accuracy of the spectrum detection. Firstly, the actual state of the previous slot of the primary user is assumed to be known, and under the condition of constant false alarm, The theoretical upper bound of detection probability is derived. Then, in order to solve the problem that the actual state of the primary user is unknown, the detection result is used to approximate the actual state. Markov constant false alarm energy detection (MCFED) and modified MCFED.MCFED have high detection probability in the case of high signal-to-noise ratio (SNR), but the performance of IMCFED in low signal-to-noise ratio (SNR) region is lower than that of the traditional spectrum detection algorithm. Finally, the MCFED algorithm is always superior to the traditional spectrum detection algorithm. In order to balance the influence of false alarm probability and missed detection probability, In this paper, we study how to reduce the Bayesian cost by using the Markov transfer characteristic of the primary user. Firstly, we propose the Markov Bayesian energy detection (MBED) algorithm. The idea is that the detection results of the previous time slot are regarded as the actual state of the host user. According to the detection results, the decision threshold of the current time slot is selected. The algorithm has less computation and better performance when the signal-to-noise ratio is high. In order to overcome the problem of poor detection accuracy of MBED algorithm under low SNR, In this paper, an improved MBED-IMBED-based algorithm is proposed, which dynamically adjusts the decision threshold according to the prediction probability, which can further reduce the Bayesian cost and has a wider range of applicable signal-to-noise ratio (SNR).
【學位授予單位】：華僑大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN925

【相似文獻】

相關期刊論文 前10條

1 曹慧;;瑞利信道下多天線能量檢測性能研究[J];電視技術;2011年11期

2 呂春英;周寧;趙偉;鞏艷;;基于小波包去噪的能量檢測方法研究[J];中國無線電;2012年03期

3 段棟棟;駱德淵;;基于信噪比墻的協(xié)作能量檢測算法[J];電子設計工程;2014年06期

4 林英沛;;一種自適應的快速能量檢測方法[J];科技視界;2014年25期

5 劉會衡;胡健;;實高斯信號下的能量檢測技術[J];通信技術;2011年08期

6 馬國棟;武穆清;徐春秀;;一種新的聯(lián)合頻譜能量檢測方法[J];通信技術;2011年04期

7 劉鑫;譚學治;徐貴森;;噪聲不確定下認知無線電能量檢測性能的分析[J];四川大學學報(工程科學版);2011年06期

8 王蒙;陳德章;楊晶晶;王魯嘉;黃銘;;基于蒙特卡洛方法的頻譜能量檢測研究[J];無線電工程;2013年01期

9 許建霞;劉會衡;劉克中;;認知無線電中一種雙門限能量檢測算法[J];武漢理工大學學報(信息與管理工程版);2011年04期

10 王曉侃;盧光躍;包志強;白輝;;一種新的分布式協(xié)作能量檢測算法[J];電訊技術;2012年09期

相關博士學位論文 前2條

1 楊志華;IR-UWB能量檢測接收機性能優(yōu)化與評估[D];哈爾濱工業(yè)大學;2010年

2 吳進波;感知無線電系統(tǒng)中能量檢測及MAC層調度技術的研究[D];北京郵電大學;2010年

相關碩士學位論文 前10條

1 王浩安;基于多層能量檢測的動物聲音檢測與識別[D];福州大學;2013年

2 王欣玉;認知無線電網(wǎng)絡中基于能量檢測的頻譜感知算法研究[D];哈爾濱工業(yè)大學;2016年

3 董淑雅;認知無線電中寬帶頻域能量檢測算法研究[D];海南大學;2016年

4 秦協(xié)安;基于USRP的認知無線電頻譜高效利用技術的研究和實現(xiàn)[D];湖南大學;2015年

5 滑偉;基于能量檢測的自適應頻譜檢測算法研究[D];西安電子科技大學;2015年

6 王隆前;穩(wěn)定分布噪聲中基于能量檢測的頻譜感知算法研究[D];大連海事大學;2017年

7 高仁陽;基于主用戶特性的頻譜檢測技術研究[D];華僑大學;2017年

8 胡小峰;基于噪聲不確定性和用戶狀態(tài)改變的能量檢測算法研究[D];重慶郵電大學;2014年

9 李粉;基于噪聲不確定度的能量檢測研究[D];煙臺大學;2012年

10 吳明清;鏈鋸反沖能量檢測方法研究[D];上海交通大學;2007年

，

本文編號：1625054