【作者】 程鹏；

【导师】 支琤；

【作者基本信息】 上海交通大学 ， 电子与通信工程（专业学位）， 2020， 硕士

【摘要】 随着通信技术的发展,无线通信网络不仅需要提供数据传输业务,还需要融合许多其他功能。高精度授时测距与无线通信一体化技术是未来无线网络的重要发展趋势。目前,一体化技术面临的挑战主要有:一是如何设计兼顾授时测距性能以及通信性能的波形;二是如何提升高速移动场景下的授时测距精度。因此,本文对授时测距与无线通信一体化技术进行研究,旨在提升授时测距精度并且获得高速的通信性能。首先,本文从双向授时同步流程入手,结合时钟模型分析了影响其精度的原因。介绍了基于前导序列的定时同步流程,并在此基础上给出基于内插的改进方案,通过提升时间戳测量精度提高系统精度。通过仿真以及上板实测验证了内插方案的时间戳记录精度,内插后最高可以将精度提升到采样速率的0.1倍左右。然后,本文针对低速场景给出了物理层波形体制设计方案。本文考虑最大发射功率受限的场景,同时考虑发射信号的峰均比(PAPR)与信道容量。当系统的峰值发射功率相同时,峰均比较低的信号在发射端能够获得较高的平均功率,从而在接收端能够获得较高的信噪比以及信道容量。基于上述分析,我们综合考虑延时估计的克拉美罗下界(CRLB)以及峰均比,给出了适合授时测距与无线通信一体化的成形滤波器参数以及混合二进制偏移载波调制参数。最后,本文针对高速移动场景提出了基于正交时频空间调制(OTFS)技术的延时速度估计方案。在高速场景下需要速度的估计才能进行高精度授时测距,因此引入了多普勒频移以及信道模型,将问题转化为对信道延时以及多普勒频移的估计。本文介绍了正交时频空间调制常用的基于离散导频信道估计的方法。针对其缺陷,我们首先给出单径下分数多普勒频移的解算。然后我们提出基于离散导频与多重信号分类(MUSIC)结合的改进算法。仿真表明,改进算法相比于离散导频算法,延时及速度估计精度均有提升。此外在较高信噪比下,改进算法可以获得与多重信号分类算法同样高的估计精度,并且大幅降低计算复杂度。然后,本文同样分析了正交时频空间调制的峰均比性能并对比正交频分复用(OFDM),说明其优越性并通过仿真验证了理论。更多还原

【Abstract】 With the development of communication technology,wireless communication networks not only need to provide data transmission services,but also need to integrate many other functions.Integration of high-precision timing and ranging and wireless communication technology is an important development trend of future wireless networks.The main challenges of integrated technology are two points: one is how to design a waveform that takes into account the timing and ranging performance and communication performance together.The second is how to improve the timing and ranging accuracy of existing waveforms.Therefore,this paper studies the timing,ranging and wireless communication integration technology in the direction of the physical layer to improve the accuracy of timing and ranging and obtain high-speed communication performance.First of all,the paper starts with the two-way timing message exchanges process and combines the clock model to analyze the accuracy error: the timestamp accuracy of the received packet.Therefore,timing synchronization process based on preamble sequence is introduced,and on this basis,an improved plan based on interpolation is given.On-board tests verify the accuracy of the time measurement of the interpolation scheme.At a sampling rate of 40 MHz,interpolation can improve the estimation accuracy of about 5ns,which can improve the performance by about 20 percent.Then,the article gives a physical layer waveform system design for low-speed scenarios.We consider the scenario where the maximum transmit power is limited and combine the peak-to-average ratio(PAPR)of the transmitted signal with the channel capacity.When the peak transmit power of the signal is the same,a signal with a lower peak-to-average ratio can obtain a higher average power at the transmitter,so that a higher signal-to-noise ratio and channel capacity can be obtained at the receiver.Based on the above analysis,we consider the Cramer-Rao lower bound(CRLB)for delay estimation and peak-to-average ratio,and give pulse-shaping fitler and composite binary offset carrier modulation parameters that suitable for the integration of timing,ranging and wireless communication.Finally,the article presents an improved delay and speed estimation scheme for orthogonal time-frequency space(OTFS)modulation technology in high-speed mobile scenarios.Speed estimation is required in high-speed scenarios for high-accuracy timing and ranging.On this basis,the Doppler frequency shift and the channel model are introduced,and the problem is transformed into an estimation of the channel delay and the Doppler frequency shift.Then,we studied a new modulation method:Orthogonal Time Frequency Space modulation.We introduce common channel estimation method for orthogonal time-frequency modulation:embedded pilot method.Aiming at the defect of the algorithm,We first give the solution of fractional Doppler frequency shift in a single path.Then we propose an improved algorithm based on the combination of embedded pilot and multiple signal classification(MUSIC).The simulation results show that Compared with the embedded pilot algorithm,the improved algorithm can improve delay and speed estimation accuracy.Besides,at higher signal-to-noise ratios,the improved algorithm can obtain the same high estimation accuracy as the multiple signal classification algorithm and meanwhile significantly reduce the complexity(about one order of magnitude).Then we analyze the peak-to-average ratio performance of orthogonal time frequency space modulation and compare it with orthogonal frequency division multiplexing.We explain its priority and verify the theory through simulation.更多还原