【作者】 张文博；

【导师】 周锋；

【作者基本信息】 哈尔滨工程大学 ， 信息与通信工程， 2023， 博士

【摘要】 本论文主要研究基于空时处理的移动扩频声通信技术。随着水下网络化无人平台、水下潜器集群的快速发展,目前水下潜器以及水下工作平台正在由独立系统向集群化系统演变。深海载人潜器与自主水下航行器等水下潜器的通信链路对于通信速率以及通信稳健性也有了更高的要求。由于水声信道具有明显的空时分簇特性,利用水听器阵列进行接收能够更为有效的利用信号的空间与时间增益,提高通信系统的性能。因此,研究基于空时处理的移动扩频声通信技术对于未来深海载人潜器、水下航行器自主作业等领域具有重要意义。本文首先以广义M元扩频技术为基础,提出了基于Legendre序列的正交CSK扩频通信。并在此基础上,对基于空时处理的扩频通信技术进行研究,具体包括空时二维联合Rake接收算法、基于波束跟踪的移动空时扩频接收算法。首先,针对直接序列扩频通信速率较低的缺点,提出了一种基于Legendre序列的正交CSK扩频通信技术。利用Legendre序列的序列长度具有更高的自由度,提高CSK扩频通信的码元相位信息利用率,从而改善通信速率。通过仿真,比较了基于Legendre序列的正交CSK扩频通信与直接序列扩频、传统广义M元扩频水声通信技术的通信速率以及在水声信道下抗噪声的能力。通过水池试验验证了理论算法的正确性并通过丹江口水库试验对基于Legendre序列的正交CSK扩频通信技术的方案进行了可行性验证。仿真与试验结果表明,基于Legendre序列的正交CSK扩频通信与基于m序列的CSK扩频通信相比,在相近误码率的情况下,能够为通信速率带来12%的提升。在发送声源级为186dB,接收信号信噪比为1.6dB的情况下,在3.4km距离下实现了信号频带8-12kHz、通信速率为425.5bps的无误码数据传输。其次,对空时扩频接收算法进行研究,提出了一种适用于水下复杂信道低信噪比扩频通信的空时二维联合Rake接收算法。针对扩频通信通常工作在低信噪比环境下的特点,研究了基于噪声滤除的MUSIC算法,解决了低信噪比环境下DOA估计误差较大的问题,并提出了基于特征向量的低信噪比相干信号信源估计算法,使空时二维联合Rake接收算法能够在低信噪比情况下准确估计入射信号的DOA以及信源数目。并利用Rake接收算法,更为有效地利用多径信号能量,提高接收信号信噪比。通过仿真分析,验证了基于噪声滤除的MUSIC算法与基于特征向量的低信噪比相干信源数目估计算法能够在-15dB的情况下,准确估计出入射信号的DOA与信源数目。在信道较为复杂的情况下,空时二维联合Rake接收算法与传统空时处理算法相比,在相近误码率的情况下,所需信噪比减少3dB。通过水池试验,验证了理论算法与仿真的正确性。通过松花江试验,验证了空时二维联合Rake接收算法与传统的空时处理算法相比,在相同信噪比下具有更低的误码率,在-10dB信噪比下,空时二维联合Rake接收算法在通信速率为358.1bps情况下,误码率为1.25×10^-3。最后,针对通信双方相对运动的场景,研究了移动空时扩频接收算法。由于扩频通信速率较慢,信号持续时间长。因此在通信双方存在相对运动、阵列摆动的情况下,接收信号的入射角度会产生明显变化。此时若采用常规的空时处理算法对接收信号进行处理,会导致波束成形聚焦角度与信号实际入射角度存在偏差。因此,提出了一种基于波束跟踪的移动空时扩频接收算法。利用基于扩频序列的DOA估计算法,实现通信信号逐符号DOA估计。与常规宽带DOA算法相比,基于扩频序列的DOA估计算法无需将信号转换为频域信号,算法运算量小,计算时间仅为CSM-TCT的7%,能够有效提高算法实时性。并能在-15dB信噪比的情况下,准确估计入射信号DOA与信源数目。通过基于扩频序列的DOA估计算法,能够快速、实时的完成入射信号波束跟踪。针对通信双方存在多普勒效应的问题,提出一种联合多普勒补偿方法,第一步,利用基于压缩感知技术的非一致多普勒估计算法,估计各路径信号非一致多普勒因子,降低多普勒因子搜索范围。然后通过扩频序列对多普勒不敏感的特性,利用模糊度函数逐符号进行多普勒补偿。能够在非匀速运动的情况下,更为准确地消除系统的多普勒影响,提高通信系统的可靠性。利用联合多普勒补偿算法,能够将多普勒精补偿搜索次数降低87.5%,多普勒补偿所需时间减少19.75%。通过水池试验,验证了基于波束跟踪的移动空时扩频接收算法实时跟踪性能,与不进行波束跟踪的空时处理算法相比,在相近误码率的情况下,所需信噪比减少6dB。最后通过松花江试验,验证了理论算法的正确性,在-10dB信噪比下,基于波束跟踪的移动空时扩频接收算法在通信速率为358.1bps情况下,误码率为 3.8×10^-3。更多还原

【Abstract】 This paper focuses on the mobile spread spectrum acoustic communication technology based on space-time processing.With the rapid development of underwater networked unmanned platforms and underwater submersible clusters,underwater submersibles and underwater working platforms are evolving from independent systems to clustered systems.The communication links between deep-sea manned submersibles and autonomous underwater vehicles also have higher requirements for communication rate and communication robustness.Since the underwater acoustic channel has obvious space-time clustering characteristics,the use of hydrophone array for receiving can more effectively use the space and time gain of the signal and improve the robustness of the communication system.Therefore,the study of mobile spread spectrum acoustic communication technology based on space-time processing is of great significance for the future autonomous operation of deep-sea manned submersibles and underwater vehicles.Firstly,based on the generalized M-ary spread spectrum technology,an orthogonal cyclic shift keying spread spectrum communication based on Legendre sequence is proposed.On this basis,the spread spectrum communication technology based on space-time processing is studied,including space-time two-dimensional joint Rake receiving algorithm and mobile space-time spread spectrum receiving algorithm with beam tracking.Firstly,aiming at the disadvantage of low rate of direct sequence spread spectrum communication,an orthogonal cyclic shift keying spread spectrum communication technology based on Legendre sequence is proposed.The sequence length of the Legendre sequence has high freedom,which improves the utilization of the symbol phase information of the cyclic shift keying spread spectrum communication,thereby improving the communication rate.Through simulation,the communication rate and anti-noise ability of orthogonal CSK spread spectrum communication based on Legendre sequence,direct sequence spread spectrum and traditional generalized M-ary spread spectrum underwater acoustic communication technology are compared.The correctness of the theoretical algorithm is verified by pool experiments.And the feasibility of the orthogonal cyclic shift keying spread spectrum communication technology based on Legendre sequence is verified by Danjiangkou reservoir experiment.The simulation and experimental results show that the orthogonal cyclic shift keying spread spectrum communication based on Legendre sequence can bring a 12%increase in communication rate compared with the CSK spread spectrum communication based on m-sequence in the case of similar BER.When the transmitted sound source level is 186 dB and the received SNR is 1.6 dB,the error-free data transmission with frequency band of 8-12 kHz and a communication rate of 425.5 bps is realized at a distance of 3.4 km.Secondly,the space-time spread spectrum receiving algorithm is studied.A space-time two-dimensional joint Rake receiving algorithm for low SNR spread spectrum communication in underwater complex channels is proposed.Aiming at the characteristics of spread spectrum communication usually working in low SNR environment,the MUSIC algorithm based on noise filtering is studied.The problem of DOA estimation in low SNR environment is solved.And propose a low SNR coherent signal source estimation algorithm based on feature vector.The space-time two-dimensional joint Rake receiving algorithm can accurately estimate the DOA of the incident signal and the number of sources at low SNR.The Rake receiving algorithm is used to make more effective use of multipath signal energy and improve the SNR of the received signal.Through simulation,it is verified that the MUSIC algorithm based on noise filtering and the low SNR coherent source number estimation algorithm based on feature vector can accurately estimate the DOA and source number of the incident signal in the case of-15 dB.Compared with the traditional space-time processing algorithm,the space-time twodimensional joint Rake receiving algorithm reduces the signal-to-noise ratio by 3dB in the case of similar BER.The correctness of the theoretical algorithm and simulation is verified by the pool experiment.Through the Songhua River experiment,it is verified that the space-time twodimensional joint Rake receiving algorithm has lower BER than the traditional space-time processing algorithm under the same SNR.The BER of the space-time two-dimensional joint Rake receiving algorithm is 1.25 × 10^-3 at the communication rate of 358.1 bps at-10 dB.Finally,the mobile space-time spread spectrum receiving algorithm is studied for the scene of relative motion between the two sides of the communication.Due to the slow communication rate,the spread spectrum signal duration is long.Therefore,in the case of relative motion between the two sides of the communication and array swing,the DOA of the received signal will change significantly.At this time,if the space-time processing algorithm is used to process the received signal,there will be a deviation between the beamforming focusing angle and the actual incident angle of the signal.Therefore,a mobile space-time spread spectrum receiving algorithm based on beam tracking is proposed.Using the DOA estimation algorithm based on the spread spectrum sequence,the symbol-by-symbol DOA estimation of the communication signal is realized.Compared with the conventional wideband DOA algorithm,the DOA estimation algorithm based on spread spectrum sequence does not need to convert the signal into frequency domain signal.And the computational complexity of the algorithm is small.The computational time is only 7%of CSM-TCT,which can effectively improve the real-time performance of the algorithm.The DOA estimation algorithm based on the spread spectrum sequence can accurately estimate the DOA and the number of sources in the case of-15 dB.Through the DOA estimation algorithm based on spread spectrum sequence,the beam tracking of incident signal can be completed quickly.Aiming at the problem of Doppler effect between the two sides of the communication,a joint Doppler compensation method is proposed.First of all,the non-uniform Doppler estimation algorithm based on compressed sensing technology is used to estimate the non-uniform Doppler factor of each path signal and reduce the search range of Doppler factor.Then,through the insensitivity of the spread spectrum sequence to Doppler,the ambiguity function is used to perform Doppler compensation symbol by symbol.It can eliminate the Doppler effect of the system more accurately and improve the robustness of the communication system in the case of non-uniform motion.By using the joint Doppler compensation algorithm,the number of Doppler fine compensation searches can be reduced by 87.5%,and the time required for Doppler compensation can be reduced by 19.75%.Through the pool experiment,the beam tracking performance of the space-time spread spectrum receiving algorithm is verified.Compared with the space-time processing algorithm without beam tracking,the space-time spread spectrum receiving algorithm with beam tracking required SNR is reduced by 6 dB in the case of similar BER.Finally,the correctness of the theoretical algorithm is verified by the Songhua River experiment.Under the-10 dB SNR,the space-time spread spectrum receiving algorithm based on beam tracking has a BER of 3.8×10^-3 when the communication rate is 358.1 bps.更多还原