【作者】 白岩夫；

【导师】 史小卫；

【作者基本信息】 西安电子科技大学 ， 电磁场与微波技术， 2014， 博士

【摘要】 近年来，随着人类对电磁波传播理论的深入研究，以及对深空探测活动的不断探索，天线组阵技术由于其低成本、高性能、高可靠性以及高灵活度等诸多优点，受到了中外学者的广泛关注，在高速率深空通信中具有良好的应用前景。本文在对国内外天线组阵研究现状进行总结后，分析了在高速率深空通信中实现天线组阵的关键难点；针对其通信频带内的相位特性，研究了子频带分割与接收信号重建技术；并在进一步完善了经典相关算法理论体系后，提出了新的高性能自适应相关算法；最终设计出了适用于高速率深空通信的天线组阵仿真系统；另一方面，针对可用于天线组阵原理样机的功分器与阻抗变换器进行了深入研究。本文所取得的主要研究成果为如下：1.在高速率深空通信中实现天线组阵的关键难点。经过系统的分析、推导与研究，得出了本课题中所需解决的关键问题与技术难点主要包括：在极低接收信噪比条件下快速、准确的获得各接收信号间的相位关系；消除在通信频带内上述相位值分布的2π模糊现象，及其引起的时延计算值错误；降低由时延校正残余量带来的频带内相位滑动现象，及其导致的信号间相关性下降。2.子频带分割与接收信号重建技术在天线组阵中的应用。针对天线组阵中高速率宽带信号合成面临的2π模糊和相位滑动现象，论述了子频带分割技术在天线组阵中的应用原理，通过理论分析与推导得出了不同条件下满足信号合成要求所需的子频带数目，并提出了相应子频带分割方法所能实现的时延差校正范围。然后，将多相滤波器组信道器应用于对接收信号的分割与重建中，设计了可重建的512路信道分解器与合成器，实现了对连续频谱信号进行有效的频带分割与重建，并通过仿真实验对比了不同子频带数目下对于QPSK调制信号的重建误差以及对子频带相位信息完成解卷绕后系统的时延差校正能力。3.经典相关算法理论体系的完善。首先，推导出了Matrix-Free算法的合成性能估算公式，并与文献中SUMPLE算法的合成性能估算公式做对比，得出Matrix-Free算法在相同条件下的合成损失低于SUMPLE算法。然后，通过引入天线阵接收信号的协方差矩阵，推导出了SUMPLE算法收敛速率的依赖关系，得出在相同条件下SUMPLE算法比Matrix-Free算法拥有更快的收敛速度。4.高性能相关算法的研究。首先，通过分析Matrix-Free算法和SUMPLE算法在运行过程中权值更新量方差的特性，找到了其随迭代计算过程运行的变化趋势，并给出了两种算法的权值更新量方差的估计方法。然后，通过引入Sigmoid函数，以权值更新量方差作为引导参数，提出了一种高性能的自适应相关算法。在不增加运算空间复杂度的情况下，SVS-MF算法具有与Matrix-Free算法接近的合成效率以及与SUMPLE算法接近的收敛速度，从而具有最优的综合性能。5.参数化天线组阵技术仿真系统的研究。提出了适用于高速率深空通信中天线组阵技术的信号合成方案，并基于SVS-MF相关算法以及子频带分割与重建技术设计了参数灵活可调的天线组阵仿真系统。在进行了大量仿真试验后得出，通过设置适当的相关平均间隔，仿真系统可以针对不同的组阵形式在极低的接收信噪比下实现较高的合成性能，并实时评估合成损失。6.可用于天线组阵技术演示验证系统的微波器件设计。针对天线组阵演示验证系统，对其中的关键器件多频带功分器以及阻抗变换器进行了深入的研究。首先，提出了一种紧凑型双频双路功率分配器设计方法，并对实物模型进行了性能评估。另一方面，提出了一种四频带阻抗变换理论以及闭合的设计公式，其可以在三个不相关频率以及一个受约束频率上实现完美的实阻抗变换。更多还原

【Abstract】 Today, with in-depth studies of the propagation of electromagnetic waves andcontinuous exploration of deep space，antenna arraying techniques has been extensivelyconcerned and provide a good prospect of future application in deep spacecommunication at high data rate for its advantages such as low cost, high performance,high reliability and high flexibility. Based on the discussions of the present researchstatus, the author provides some analysis of key difficulties of applying antennaarraying in deep space communication at high data rate; carries out some research ontechniques for sub-band analysis and receiving signal reconfiguration according to thephase property of communication band; proposes a high performance adaptivecorrelation algorithm after providing some improvements for the classic correlationalgorithm system; presents a design of the simulation platform of antenna arraying fordeep space communication at high data rate; meanwhile, conducts a study of the powerdivider and impedance transformer for the prototype system for antenna arraying. Theauthor’s major contributions are outlined as follows:1. Key techniques and difficulties in applying antenna arraying into deep spacecommunication at high data rate have been systematically summarized. Major issuesneeded to be solved in this work include: getting the phase values effectively andaccurately in a low signal-to-noise ratio (SNR) condition, avoiding the fake delaycaused by the2π ambiguity error, and reducing the phase shift across the passband dueto the error in delay compensation.2. The application of sub-band analysis and receiving signal reconfiguration inantenna arraying has been studied. In order to avoid2π ambiguity error and phase shiftduring signal combination，the application of sub-band analyzing in antenna arrayinghas been studied. By theoretical analyses and derivations, both the amount of sub-bandthat meet the requirements of signal combination in different conditions and the rangeof delay that can be achieved have been derived. Then, polyphase filter bankchannelizer has been applied to the operation of analysis and synthesis. By selectingappropriate parameters, a512-channelizer has been designed，which realizes effectiveanalysis and synthesis for wideband signal. The reconfiguration error of quadraturephase-shift keying modulated signals in different amount of sub-band has beencompared and the calibration capability of the system with different amount of sub-bandhas been proved through simulation. 3. The improvements of the theory system of classical correlation algorithms havebeen realized. Firstly, the combining degradation for Matrix-Free algorithm has beenderived and it also has been compared with that for SUMPLE algorithm. It has beenproved that the combining degradation for Matrix-Free is lower than that for SUMPLEin the same condition. Then, the determination of the convergence rate of SUMPLE hasbeen derived by introducing the receivable covariance matrix. This indicates that theconvergence rate of SUMPLE algorithm is higher than that of Matrix-Free in the samecondition.4. A high performance correlation algorithm has been studied. Firstly, the methodto calculate the trend of the variance of weight updating values from every iteration hasbeen acquired through the analysis to Matrix-Free and SUMPLE. Secondly, a highperformance adaptive correlation algorithm is presented by introducing Sigmoidfunction for the low SNR condition in deep space communication at high data rate. Itshows that the proposed SVS-MF algorithm has a combining degradation close to thatof Matrix-Free and convergence rate approaches SUMPLE’s without increasing thecomputational complexity. This proves that SVS-MF algorithm has the best overallperformance.5. The simulation tool for antenna arraying for deep space communication at highdata rate has been studied. And based on SVS-MF correlation algorithm as well asreconfigurable channelizer, a simulation platform with flexible parameters has beendesigned. Simulation results show that, by setting an appropriate correlation averaginginterval, the system can realize a very high combining performance for differentarrangements of arrays under a very low SNR condition.6. The studies of multiband power divider and impedance transformer for theprototype system for antenna arraying have been carried out. Firstly, a compactdual-band power divider using a non-uniform transmission line is presented. Secondly, acompact quad-frequency impedance transformer that is composed of a two-sectioncoupled line to achieve ideal impedance matching at three arbitrary frequencies and arelated frequency.更多还原