基于自适应差量的频谱状态信息压缩方法

发布时间：2018-11-02 09:27

【摘要】：由于无线通信技术的日益发展,用户群体和业务正变得越来越多,从而导致频谱资源变得越来越紧缺。于是,人们开始关注如何充分利用频谱以提高频谱的利用效率,而解决这个问题的一个好的办法便是运用认知无线电技术。认知无线电是通过感知周围的环境来发现频谱空洞,然后对它们加以利用,从而提高频谱的利用效率。由于频谱感知需要尽量准确地感知周围的频谱环境,因而获得的频谱状态信息数据量将会很大。如果不对这些感知到的海量数据进行一定的处理,而是直接将它们传送给管理中心或者相互交流,这会给认知通信系统带来很大的负载压力,会需要很长的传输时间、占用较大的带宽。因此,针对这个问题本文讨论了对海量感知数据进行处理的技术,从而在满足认知系统要求的基础上减小数据量。可以降低频谱状态信息数据量的方法有很多,比如判决处理方法、压缩感知方法和基于数据压缩的处理方法(如,Huffman编码)等。但是,它们都有各自的一些缺点。比如,判决处理方法的门限不好设定,这会导致信息量的损失。压缩感知要求信号是稀疏的,这个不一定对所有信号都成立。而常用的基于数据压缩的处理方法,比如Huffman编码,它的缺点就是其压缩效果与频谱的各个频率点的功率值出现的概率分布特点有关:当各个频率点的功率值出现的概率很不平均时,压缩效果最好;当各个频率点的功率值出现的概率是一样的时候,此时压缩效果最差。针对上述问题,本文提出了一种频谱状态信息压缩的处理方法,即对频谱状态数据之间的差值以自适应增量编码的方式转换成二进制序列再传输。这样每个状态增量数据仅用1bit信息来表示,压缩效果非常好。而且接收端在恢复数据时,能够根据压缩信息之间的内在联系,自适应地调整原数据之间的差值,并在前一个数据值的基础上追加这个差值从而得到此次数据的值,进而恢复出原来的所有频谱状态数据。这种方法在传输的过程中只需传输较少的信息,达到了信息压缩的目的,不但使传输时间大大减少,而且只需占用较小的带宽,降低了网络负载。最后,本文使用USRP软件无线电平台实现了频谱感知功能以及上述各个降低频谱状态信息数据量的方法,并分析和比较了它们的性能。
[Abstract]:With the development of wireless communication technology, the number of users and services is becoming more and more, resulting in a growing shortage of spectrum resources. Therefore, people begin to pay attention to how to make full use of spectrum to improve the efficiency of spectrum utilization, and a good way to solve this problem is to use cognitive radio technology. Cognitive radio detects spectrum holes by sensing the surrounding environment, and then uses them to improve the efficiency of spectrum utilization. Because spectrum sensing needs to perceive the spectrum environment as accurately as possible, the amount of spectrum state information obtained will be very large. If we don't deal with these perceived mass data, we can send them directly to the management center or communicate with each other, which will bring a lot of load pressure to the cognitive communication system, and it will take a long time to transmit. Take up large bandwidth. Therefore, this paper discusses the processing technology of mass perceptual data, so as to reduce the amount of data on the basis of satisfying the requirements of cognitive system. There are many methods to reduce the amount of spectrum state information, such as decision processing, compression sensing and data compression based on data compression (for example, Huffman coding) and so on. However, they all have their own shortcomings. For example, the threshold of decision processing is not well set, which can lead to loss of information. Compression sensing requires that the signal is sparse, which is not necessarily true for all signals. And common methods based on data compression, such as Huffman coding, Its disadvantage is that its compression effect is related to the probability distribution of the power value at each frequency point of the spectrum: when the probability of the power value of each frequency point is very uneven, the compression effect is the best; When the probability of power at each frequency point is the same, the compression effect is the worst. In order to solve the above problems, this paper proposes a method for the compression of spectrum state information, that is, the difference between spectrum state data is transformed into binary sequence retransmission by adaptive incremental coding. In this way, each state increment data is expressed only by 1bit information, and the compression effect is very good. Moreover, when the data is restored, the receiver can adaptively adjust the difference between the original data according to the internal relationship between the compressed information, and append the difference on the basis of the previous data value to obtain the value of the data. Then all the original spectrum state data are restored. In the process of transmission, this method only needs to transmit less information and achieves the purpose of information compression. It not only reduces the transmission time, but also takes up less bandwidth and reduces the network load. Finally, this paper uses the USRP software radio platform to realize spectrum sensing and the methods to reduce the data amount of spectrum state information, and analyzes and compares their performance.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN925

【相似文献】