《ODFM系统中基于认知无线电的频谱检测的研究毕业论文外文翻译.doc》由会员分享，可在线阅读，更多相关《ODFM系统中基于认知无线电的频谱检测的研究毕业论文外文翻译.doc（17页珍藏版）》请在三一办公上搜索。
1、外文资料翻译Study of Cognitive Radio Spectrum Detection in OFDM SystemZengyou Sun, Qianchun Wang, Chenghua CheAbstractThis article introduces two typical means of spectrum sensing in CR: energy detection and cyclostationary detection. Two methods are used to verify the feasibility and detect authorized us
2、er in CR-OFDM system. It illustrates that how to choose parameters when detect the OFDM signals by MATLAB. It compares the superiorities and deficiencies of the two ways, then puts forward the improved solution.Keywords-Cognitive Radio; OFDM; Energy Detection; Cyclostationary Detection; MATLAB.INTRO
3、DUCTIONAs the development of wireless communication technology, the number of radio users increase sharply, spectrum resources decrease. In order to alleviate this contradiction, in one hand, people develop new valuable spectrum constantly, in another hand, people develop new wireless access technol
4、ogy and improve the coded modulation to improve spectral efficiency. However, huge waste of spectrum resources is caused due to the constraints of the size and power of mobile terminal antenna. Therefore, cognitive radio (CR) 1 technology is presented, which address these challenges effectively by u
5、sing the idle spectrum resources fully from time and space.CR is the most effective way to solve the shortage of spectrum resources by intelligent spectrum management. It senses the ambient idle channel in the RF system. Spectrum detection technology is the key technology of cognitive radio as well
6、as the basis and premise of cognitive radio application, and determines the channel conditions by intercepting signals of the primary users, usually in the case that the sensitivity of perception user is far greater than the primary user receiver. The basic idea of spectral detection techniques is t
7、hat the unoccupied bands are sensed, then the CR system can use these frequency bands for communication temporarily; when the CR system detects the legal users of the band to resume communications, then the equipment will give up the band, or continue occupying the band, but can not interfere the au
8、thorized users. It means that CR is able to use available spectrum dynamically and repeatedly by sensing interactively the wireless communications environment. Sensing spectrum dynamically, combined with sensing the surrounding environment, then a new transmitting program is required. Orthogonal fre
9、quency division multiplexing (OFDM) technology 2 is verified to be the best transmitting option of cognitive radio. The multi-carrier modulation technology and self-adaptive power allocation of OFDM bring a great flexibility to cognitive radio. OFDM uses multi-band modulation, and transmits the data
10、 to the given users in successive time by different subband, so non-artificial interfere in specific frequency band can be avoided without the radio frequency notch filter. An OFDM symbol consists of N parallel transmitted signals modulated on different carrier frequencies of interval f . The comple
11、x envelope of OFDM symbols can be expressed as follows.The spectrum of an OFDM symbol can be expressed as follows.There are three classical methods of spectrum detection in physical layer in CR system, such as matched filter detection, energy detection 3 and cyclostationary detection 4, which are al
12、so applicable to the cognitive OFDM system. But the matched filter detection needs priori information of primary users signals, and high simultaneous demanding, so it can be in line with the other, such as cyclostationary characteristics detection methods in order to achieve the optimal efficiency o
13、f the detection.ENERGY DETECTIONIf energy detection is used to sense CR-OFDM signals, spectrum detection of each sub-carrier and the entire signal can be achieved with lower complexity, judging and analyzing the channel state. Energy detection is a noncoherent detection means, which detects the sign
14、al by sensing the power of circuit signal. The detection to the signal of uncertain type possesses universality and timeliness as the energy detection does not need priori conditions. The method of period diagram is the classical power spectrum estimation method, with high computing efficiency and l
15、ow spectral resolution. Detection process is shown in Fig. 1.Figure 1. OFDM energy detection modelThe detection power of signal s(t) through above detecting process is as follows.Adequate number of samples N is needed to meet the performance goals of detector, as fixed size FFT is usually used to me
16、et expected resolution where presents that there is no primary user and corresponding presents that there is primary user. An OFDM signal of L periods length changes into frequency signal after a Fourier transformation. Whether sub-carrier channel existent authorized user signal or not is determined
17、 due to the orthogonal and non-relevant sub-carriers of single sub-carrier signal.If the channel is idle, what the detector obtained is only noise signal of channel, subjected to assumptions; on the contrary, if there is authorized users in the channel, what the detector obtained is authorized users
18、 signal and noise signal, subjected to assumptions. In this way, the setting noise signal obeys Gaussian distribution of mean zero and variance 2 according to the above detecting steps. Then statistical energy S is compared with setting threshold. If the measured value exceeds threshold, we consider
19、 it accords with , otherwise we consider there is no authorized user signal and it accords with . As the sub-carrier is to meet the orthogonal Gaussian function, so modulus square sample energy S obeys distribution. If the channel is Gaussian white noise, S obeys center distribution as its mean valu
20、e is zero; if there are authorized users in the channel, perceived signal is composite signal of noise signal and certain signal, whose mean is non-zero, and thus subjects to non-central distribution, the formula is as follows.Detection probability and false alarm probability of two assumptions are
21、as follows.where SNR is the signal to noise ratio. A string of OFDM signals is simulated using above formula, where the modulation mode is QPSK, the number of carrier is 128, the length of FFT is 128, the number of single sub-band symbols is 8, and the guard interval is 32.Figure 2. Relationship bet
22、ween energy detecting SNR and detecting probabilityFigure 3. Relationship between threshold and false alarm probabilityWe can see from Fig. 2 and Fig. 3 that accurate detection probability can be obtained only by choosing a suitable threshold. The smaller threshold is, the greater the probability is
23、. However, with the increase of SNR, wrong protection even no protection will appear; to false alarm probability, the larger of the length is, the smaller of the perceived error verdict is, and no protection will also appear if the threshold selection is inappropriate. For effective detection, appro
24、priate parameters need to be selected including spectral detection time and setting the threshold. 5 presented optimized compromised criterion, obtained high transmitting efficiency by traversing d, and made total false detecting probability minimum.III. CYCLOSTATIONARY DETECTIONCyclostationary sign
25、al is a random signal whose statistical characteristics cyclically changes over time, including first-order (mean), second order (correlating function) and higher-order (higher-order cumulants) cyclostationary.For the random signal X(t), when it meets cyclostationary properties, the mean and autocor
26、relation functions are both cyclical, and the cycle is the same as the signal cycles. The self-correlation function of s(t) is defined as , where T is the cycle. If it is shown into Fourier series, Fourier coefficient can be obtained as follows, which is called cyclic autocorrelation function (CAF).
27、According to Wiener-Khintchine relations, the Fourier transform of CAF can be obtained and defined as spectrum correlation function (SCF).In project, the discrete form of CAF is obtained using its discrete form as follows.Where is cycle frequency, which is important in cyclostationary detection. The
28、 cycle frequency of a cyclostationary signal includes zero-cycle frequency and non-zero cycle frequency. Where, the zero cycle frequency corresponds to the smooth part of signal, and only non-zero cycle frequency portrays the cyclostationarity of signal.The carrier of OFDM signal has multiple loop f
29、requency, and meets the second-order cyclostationarity properties. Cyclostationary detection of OFDM signals is done using the method of evolutionary optimal detection presented by Dandawate 6. A delayed matrix is constructed as statistical detecting amount, then correlating function matrix is estim
30、ated using matrix, which is appropriate for the situation that there is no priori information of primary user signal in the receiver, and cyclostationary feature of detecting signal is obtained to verdict whether there is protecting signal. However, the method can be optimized in terms of its basic
31、assumptions due to its high complexity, chooses and obtains the most effective time delay parameter, then obtains the maximum amplitude . According to Dandawate theory, detecting probability detects all of the cycle frequency and sums the detectable quantity by the sum criteria, where M is the numbe
32、r of random variables, and meets the distribution.The simulated signal above-mentioned is detected circularly, then the coordinate axes are normalized and the results are as follows.Figure 4. Two-dimensional autocorrelation function of signalFigure 5. Relationship between delay and autocorrelation f
33、unctionThe two pictures above are two-dimensional picture of autocorrelation function of the signal , varied from cycle frequency and delay and = 0 cross-section picture. We can see that exists non-zero values in certain loop frequency and time delay, especially in the twice carrier existing the pea
34、k. Fig. 5 presents the relationship between delay and auto-correlating function, which help analyzing and selecting delay parameters to obtain detecting probability. Cyclostationary detection is in cycle frequency domain and the phase and frequency information relevant with time parameters of OFDM s
35、ignal is retained in signal processing, thus overlapping features of power density in the frequency domain is no longer overlap in the loop. Both signal and noise has spectral components when = 0, while the spectrum component of noise is zero when 0 , and the zero cycle frequency corresponds to the
36、smooth part of signal, only non-zero cycle frequency portray the cyclostationarity of signal.Figure 6. Cycle spectrum of simulated signalFigure 7. = 0, aspect picture of OFDM spectral correlation functionFrom the figures above, we can see that the spectral characteristics of OFDM signal can be react
37、ed truly by the method of cyclic spectrum related. OFDM signal shows cyclostationary characteristics, while the stationary noise does not show the relevance characteristics of cyclic frequency domain, and interference signals usually shows cyclostationary characteristics different from the signal of
38、 main users. Therefore, the greatest feature of cyclostationary detection has the ability to distinguish the spectrum from the primary user signal, noise and interference signal. We only need to verdict whether the spectrum lines appears when 0 as the verdict condition, which is an important way of
39、cyclostationary detection. Detected OFDM signal by cyclostationary detection can get rid of the influence of background noise, and distinguish the noise energy with the main user signal energy. Thus, detection performance is still good even in low SNR. The detection performance of cyclostationary de
40、tection is superior to energy detection by comparing the detection probability of them.Figure 8. Comparison between cyclostationary detection and energy detectionIV. CONCLUSIONClassic cognitive radio frequency spectrum detection method such as energy detection and cyclostationary detection in OFDM s
41、ystem is still feasible to detect the spectrum. The algorithm of energy detection is simple and is easy to implement; and also for the phase synchronization not ask for much, but the accuracy is low. Inserting pilot signals can reduce the impact, which is a sub-optimal detecting method. In contrast,
42、 cyclostationary detection gets rid of the high accuracy requirements of SNR; but the phase request of the OFDM is higher, the complexity is bigger, and the detecting time will naturally increase. If it combines with other technologies, the signal feature is extracted by the cyclic spectrum to ident
43、ify the signal, which would increase efficiency greatly and reduce detecting time.外文资料译文ODFM系统中基于认知无线电的频谱检测的研究孙曾友，王钱春，车成华摘要这篇论文介绍了两种典型的基于认知无线电的频谱感知方式：能量检测和循环平稳检测。在OFDM认知无线电系统中，这两种方法是用来证实可行性和检测授权用户的。论文阐述了通过MATLAB仿真出当检测到OFDM信号时怎么去选择参数。论文对两种方法的优势和缺陷做出了比较，然后提出了另一种改进的解决方法。关键字认知无线电；OFDM；能量检测；循环平稳检测；MATLAB
44、。.介绍随着无线通信技术的发展，无线用户的数量急剧增加。为了缓和这样的矛盾，一方面，人们不断地开发新的有用的频谱，另一方面，人们开发新的无线接入技术，改进编码调制来提高频谱效率。然而，因为移动终端天线的大小和功率的限制造成了频谱资源的极大浪费。由此，认知无线电技术诞生了。这项技术从时域和频域充分使用空闲的频谱资源来有效地解决这些问题。认知无线电技术是通过只能频谱管理来解决频谱资源短缺的最有效的方法。它能够感知射频系统周围的空闲信道。频谱检测技术是认知无线电技术的关键，而且还是认知无线电技术应用的基础和前提。通过拦截主要用户的信号来决定信道条件，通常在这种情况下，认知用户的敏感性远远强于主要用户
45、接收机。频谱检测技术的基本想法是没有被占用的频带被感应到，然后认知无线电系统可以利用这些频带进行临时通信。当认知无线电系统检测到该频带的合法用户要继续通信，那么设备就会放弃这段频带，或者继续占有这段频带，但是不能干扰到授权用户。这就是说，认知无线电能够通过交互式地感应无线通信环境来动态地反复地使用未被利用的频谱。动态地感应频谱，同时感应周围的环境，那么就需要一个新的传输计划。正交频分复用技术已被确定是认知无线电中最佳的传输选择。OFDM的多载波调制技术和自适应的功率分配为认知无线电带来了很大的灵活性。OFDM使用多载波调制，在连续时间使用不同的子载波传输数据给用户，所以如果没有无线电频率陷波滤
46、波器，在特定的频段的非人为干涉可以避免。一个OFMD符号包括了N个平行的调制在间隔为f的不同载波频率的传输信号。这个OFDM符号的复杂包络可以表示为如下：OFDM符号的频谱可以表示为如下：认知无线电系统中物理层频谱检测有三种经典算法，匹配滤波器法，能量检测法和循环平稳检测法。但是匹配滤波器法需要主用户信号的先验信息，和高同步的需求，所以它可以与其他一起，比如循环平稳特性检测算法，来实现最高效的检测。.能量检测如果能量检测是用来感应CR-OFDM信号，那么对每一个子载波或整个信号的频谱检测可以通过较低复杂度的判定和分析信道状态来实现。能量检测是一种非相干检测方法，它通过感应电流信号的能量来检测信
47、号。这种对不确定信号的检测具有普遍性和时效性，不需要先验条件。这种周期图的方法是种经典的能量谱估计方法，具有高运算效率和低频谱分辨率。检测过程如图一：图1 OFDM能量监测模型信号S（t）的检测能量通过上述检测过程后如下：我们需要足够的采样点N来达到检测器的设计目标，如固定大小的FFT通常利用来满足预期决议，这里H0表示没有主用户，相应的H1表示有主用户。一个长度为L的OFDM信号经过傅立叶变换后转变为频率信号。不论子载波信道是否存在授权用户信号，都是由单子载波信号的正交和非相关的子载波决定的。H0假设，如果信道是空闲的，那检测器接收到的就仅仅是信道里的噪声；相反，H1假设，如果授权用户占用信
48、道，那么检测器接收到的就有授权用户信号和噪声信号。这样，根据上述的监测步骤设置噪声信号符合高斯零均值方差为。然后统计得到的能量S与设定的门限进行比较。如果测量的值超过了门限，我们就认为该信道符合H1，否则我们认定没有授权用户信号，信道符合H0。当子载波符合正交高斯函数，那么模平方的采样能量S服从分布。如果信道是高斯白噪声信道，S服从均值为零的中心分布；如果信道里有授权用户信号，接收到的信号是确定信号和噪声信号的复合，均值非零，符合无中心的分布，方程式如下：两种假设的检测概率和误警概率如下：SNR是信号噪声比。一串OFDM信号通过上述方程模拟，调制方式是QPSK，载波数是128，FFT长度是12
49、8，单个子带符号数是8，保护间隔为32。图2 能量检测信噪比与检测概率关系图3 门限值与误警概率关系从图2、3中我们可以看到只有确定合适的门限才能够得到精确检测的可能。门限越小，可能性越大。然而，随着信噪比的增加，错误保护甚至无保护会出现；对于误警概率，长度越长，感知错误判决越小，如果门限设置不合适，无保护还是会出现。为了有效检测，包括频谱检测时间和设定门线需要设定合适的参数。文献5提出了优化妥协标准，包括高传输效率，得出最小总共错误检测概率为Pe= 1 + Pf Pd 。.循环平稳检测循环平稳信号是统计特征随着时间循环变化的随机信号，包括一阶（均值），二阶（相关函数）和高阶（高阶积累量）循环平稳。对于随机信号X(t)如果满足循环平稳特性，均值和自相关函数都是循环的，循环周期与信号周期相
