LEO卫星多普勒频移及信道参数估计与频谱感知研究

发布时间：2018-09-05 08:37

【摘要】：在低轨(LEO)卫星通信中,信道存在多普勒频移、多径衰落等多种干扰,严重影响接收性能,同时主用户在实际通信中只占用很小的带宽,在频域上呈现稀疏特性,频谱利用率较低。针对这些问题,本论文主要开展LEO卫星多普勒频移估计算法、稀疏信道估计与频谱感知研究,来实现LEO卫星信道的高效识别、估计及利用的目的。首先,简介了LEO卫星通信背景、现状和发展趋势。然后,概述了LEO卫星信道,针对其在传输中存在的问题,基于现有的算法提出改进方案。之后,建立了典型中低轨卫星信道的多普勒频移模型,并分析其特征。针对单颗LEO卫星频谱资源受限,重点研究采用压缩感知理论对卫星稀疏信道估计与频谱感知。最后总结了全文,即针对卫星信道中存在较大多普勒频移且呈现稀疏特性,改进了传统多普勒频移估计算法,并采用压缩感知重构算法,实现对卫星信道的频谱感知与信道估计,提高信道通信传输抗干扰性能。全文内容和创新点如下:1.在卫星信道中,多普勒频移是影响其通信性能的主要因素。针对中低椭圆轨道卫星信道特点,提出一种利用半长轴、偏心率和载波频率等参数来预测卫星多普勒频移的算法。该算法主要开展了多普勒频移公式推导和理论分析,得到其估计的解析解,可作为多普勒频移的预补偿协助信道估计。仿真验证了所提算法能较精确地计算不同轨道下卫星多普勒频移参数,且相比传统三维坐标算法降低了复杂度,提高了效率;2.传统基于导频的信道估计算法因导频占据额外信道带宽,降低了信道利用率。故利用压缩感知重构算法,提出了改进的OMP快速稀疏信道估计算法。该算法在OMP算法基础上,在每次迭代时采用多原子选择方式,即选出残差与测量矩阵内积间最优的M列向量,并对其估计。最后,将所得估计值优化,得到最优估计向量h。在相同估计效果下,该改进OMP算法导频数远小于LS算法,因而节省了较多带宽。仿真表明:在训练序列长度相同时,该改进算法较LS算法虽所耗时间相差无几,但其可获得更优越的信道估计性能;3.针对LEO卫星群中单颗卫星通信频谱资源有限问题,提出了利用认知无线电协作宽带频谱感知算法,来实现宽带频谱的精确感知。根据联合稀疏模型(JSM-2),提出了快速低复杂度频谱感知算法。其在分布式压缩感知正交匹配追踪算法(DCS-OMP)基础上,以前一时刻感知结果为先验条件,来感知当前时刻频谱占用情况。而且,在信号重构中,以残差门限为判决条件,极大减少了迭代次数,简化了重构复杂度。仿真表明:在频谱占用变化小时,相对原DCS-OMP算法,所提算法可获类似性能,且在认知用户数较多时,显著减少复杂度和延迟。当稀疏度为20,信号长度为500,及信噪比(SNR)为15dB时,所提算法在信号重构频谱误差上比DCS-OMP算法减小0.2032,所耗时间减少13.8%;本文主要针对LEO卫星信道存在的多普勒频移、多径衰落和信道频谱利用率不高这些问题,提出了多普勒频移快速估计算法,基于压缩感知的LEO卫星稀疏信道估计与频谱感知算法。所提算法有助于LEO卫星信道的高效识别、估计及有效利用,将在未来高速卫星多媒体通信及广播(如DVB-S2系统)等场合获得重要应用。
[Abstract]:In LEO satellite communication, there are many kinds of interference, such as Doppler shift, multipath fading and so on, which seriously affect the reception performance. Meanwhile, the main user only occupies a small bandwidth in the actual communication, and shows sparse characteristics in the frequency domain, and the spectrum utilization rate is low. Sparse Channel Estimation and Spectrum Sensing are studied to realize the efficient identification, estimation and utilization of LEO satellite channel. Firstly, the background, status quo and development trend of LEO satellite communication are briefly introduced. Then, the LEO satellite channel is summarized. Aiming at the problems in transmission, an improved scheme is proposed based on the existing algorithm. In view of the limited spectrum resources of a single LEO satellite, the sparse channel estimation and spectrum sensing based on compressed sensing theory are mainly studied. Finally, the full text is summarized, that is, the traditional Doppler frequency shift is improved because of the sparse Doppler frequency shift in the satellite channel. The main contents and innovations of this paper are as follows: 1. In satellite channel, Doppler shift is the main factor affecting the performance of communication. An algorithm for predicting Doppler shift of satellite by using parameters such as semi-major axis, eccentricity and carrier frequency is proposed. The algorithm derives Doppler shift formula and analyzes it theoretically. The analytical solution of the estimation is obtained, which can be used as pre-compensated assistant channel estimation for Doppler shift. Compared with the traditional three-dimensional coordinate algorithm, the traditional pilot-based channel estimation algorithm reduces the complexity and efficiency of satellite Doppler shift parameters in different orbits. 2. Because the pilot occupies the extra channel bandwidth, the channel utilization ratio is reduced. On the basis of the OMP algorithm, the algorithm uses the multi-atom selection method at each iteration, that is, to select the optimal M-column vector between the residual and the inner product of the measurement matrix, and to estimate it. Finally, the optimal estimation vector h is obtained by optimizing the estimated value. Multi-bandwidth. Simulation results show that the improved algorithm can obtain better channel estimation performance than LS algorithm even though the training sequence length is the same. 3. To solve the problem of limited spectrum resources of single satellite communication in LEO satellite group, a cooperative wideband spectrum sensing algorithm based on cognitive radio is proposed to realize wideband frequency. Accurate spectrum sensing. A fast low complexity spectrum sensing algorithm based on Joint Sparse Model (JSM-2) is proposed. Based on Distributed Compressed Sensing Orthogonal Matching Pursuit (DCS-OMP) algorithm, the spectrum occupancy at the current time can be sensed with the previous sensing result as a prior condition. Simulation results show that the proposed algorithm achieves similar performance compared with the original DCS-OMP algorithm when the spectrum occupancy varies little, and significantly reduces the complexity and delay when the number of cognitive users is large. When the sparsity is 20, the signal length is 500, and the signal-to-noise ratio (SNR) is 15 dB, the proposed algorithm achieves similar performance. Compared with DCS-OMP algorithm, the spectrum error of signal reconstruction is reduced by 0.2032 and the time consumed is reduced by 13.8%. In this paper, a fast estimation algorithm of Doppler frequency shift is proposed to solve the problems of Doppler frequency shift, multipath fading and low spectrum utilization of LEO satellite channel. The proposed algorithm is helpful to the efficient identification, estimation and utilization of LEO satellite channel, and will be used in future high-speed satellite multimedia communication and broadcasting (such as DVB-S2 system).
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN927.2

【相似文献】