基于OFDM认知无线电网络的鲁棒资源分配算法研究

发布时间：2019-06-19 02:21

【摘要】：无线通信技术已经成为现代社会发展的一个关键技术,在我们的日常生活中,有很多设备的实现及使用都是基于无线通信,比如,车库开启工具,遥感电视,蜂窝电话,电子记事薄以及卫星电视的接收。随着无线设备的使用,频谱资源稀缺成为要解决的首要矛盾,认知无线电技术一时成为了大家关注和研究的焦点。认知无线电以提高频谱利用率为核心目标,强调不同用户级之间的合作共享。该网络由授权用户(主用户)和非授权用户(次用户或者认知用户)共同组成,其中主用户是该频谱的所有者,享有优先使用和支配的权利,次用户是外来用户,认知无线电允许次用户通过频谱检测,感知等调节系统参数的方式机会式地接入频谱,实现与主用户的协同合作或者频谱共享,最终使频谱得到合理的使用。目前关于认知无线电的研究主要由频谱感知和资源分配(功率控制)两部分组成。次用户通过感知技术或者概率估计等技术手段,对主用户的频谱进行检测,从而发现更多的频谱空洞,获取更多的通信机会;在感知结果的基础之上,次用户得以机会式接入频谱,但是需要首先保障主用户的正常通信,即次用户发射机对主用户接收机的干扰不能超过干扰门限(干扰温度)。本文围绕上述问题,针对不同层面,研究了两种不同的资源分配方案,具体工作如下:(1)从绿色能源的角度出发,考虑能耗问题(最小化发射功率)。为了减少用户间的信息交流负担,提出了多用户的分布式认知无线电网络。结合实际情况下的信道状态信息,在原名义问题(非鲁棒问题)基础上,考虑估计误差,提出了鲁棒资源分配方案。优化目标为最小化次用户的发射功率,外加次用户的信干噪比约束,主用户的干扰门限约束和发射功率阈值约束。采用区间不确定性模型和椭圆不确定性模型对不确定参数进行数学建模,运用Worst-case的方法将难解的半无限规划问题转化为简单的凸优化问题。与之前的研究不同的是,该算法提出了新的信干噪比约束,在保护主用户的前提下,能有效地保证次用户的通信质量。(2)为了全面考虑系统的参数误差问题,本文将频谱感知误差(不完美频谱感知)同鲁棒功率分配问题相结合,形成新的资源分配方案。整个方案分为两部分,第一部分,次用户对频谱进行检测,估计哪些频谱是空余的,决定是否接入,但是检测结果可能出现错误,由于虚警概率和误检概率的存在,干扰约束条件将由共道干扰和边带干扰共同组成;第二部分,在检测结果的基础上,考虑信道参数的不确定性,结合Worst-case方法和柯西-施瓦兹不等式,考虑最坏情况,运用拉格朗日对偶算法,实现功率的最优分配,达到最大化传输速率的目标。在经过算法总结和性能分析之后得到结论:不完美的频谱感知也能较好地评估检测结果,抑制对主用户不必要的干扰,保证主用户的基本通信质量,此外,鲁棒算法提前考虑信道不确定性,能有效对抗参数扰动,保证次用户的基本通信需求。
[Abstract]:Wireless communication technology has become a key technology of the development of modern society. In our daily life, there are many equipment implementation and use based on wireless communication, such as garage opening tool, remote sensing television, cellular telephone, electronic note book and satellite television receiving. With the use of wireless devices, the scarcity of spectrum resources becomes the primary contradiction to be solved, and the cognitive radio technology has become the focus of attention and research. The cognitive radio is used to improve the spectrum utilization rate as the core target, and the cooperation sharing among different user levels is emphasized. the network is composed of an authorized user (primary user) and a non-authorized user (secondary user or cognitive user), wherein the primary user is the owner of the spectrum, enjoys the right to use and dominate, the secondary user is an external user, the cognitive radio allows the secondary user to pass the spectrum detection, The mode of the system parameters such as sensing and the like is opportunistic to access the spectrum, and the cooperative cooperation or the spectrum sharing with the main user can be realized, and finally the spectrum is reasonably used. The current research on cognitive radio is mainly composed of two parts: spectrum sensing and resource allocation (power control). the secondary user detects the spectrum of the main user through the technical means such as the sensing technology or the probability estimation, so that more spectrum holes are found, and more communication opportunities are obtained; on the basis of the sensing result, the secondary user is able to access the spectrum, It is necessary to first guarantee the normal communication of the primary user, i. e. the interference of the secondary user transmitter to the main user receiver cannot exceed the interference threshold (interference temperature). In the light of the above-mentioned problems, two different resource allocation schemes are studied for different levels, and the specific work is as follows: (1) From the angle of green energy, consider the energy consumption problem (minimize transmission power). In order to reduce the burden of information exchange among users, a multi-user distributed cognitive radio network is proposed. Based on the original nominal problem (non-robust problem), a robust resource allocation scheme is proposed based on the original nominal problem (non-robust problem). The optimization objective is to minimize the transmit power of the secondary user, the interference threshold constraint of the secondary user, the interference threshold constraint of the primary user, and the transmit power threshold constraint. An interval uncertainty model and an elliptic uncertainty model are used to model the uncertain parameters, and the unbounded semi-infinite programming problem is transformed into a simple convex optimization problem by using the method of the Worst-case. It is different from the previous study that the new algorithm is a new constraint, and the communication quality of the secondary users can be effectively guaranteed on the premise of protecting the main users. (2) In order to fully consider the parameter error of the system, this paper combines the spectrum sensing error (imperfect spectrum sensing) with the problem of robust power distribution to form a new resource allocation scheme. the whole scheme is divided into two parts, The interference constraint conditions will be composed of the co-channel interference and the sideband interference. The second part, on the basis of the detection results, considers the uncertainty of the channel parameters, combining the Worst-case method and the Cauchy-Schwarz inequality, considering the worst case, using the Lagrange dual algorithm, The optimal allocation of power is realized, and the aim of maximizing the transmission rate is achieved. after the algorithm and the performance analysis, the conclusion is obtained that the imperfect spectrum sensing can also better evaluate the detection result, inhibit the unnecessary interference of the main user, ensure the basic communication quality of the main user, and moreover, the robust algorithm takes into account the channel uncertainty in advance, And the basic communication requirement of the secondary user can be ensured.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925

【相似文献】