多波束水声探测系统中的信道估计技术研究
文献类型：学位论文
| 作者 | 姜雪 |
| 学位类别 | 博士 |
| 答辩日期 | 2007-06-05 |
| 授予单位 | 中国科学院声学研究所 |
| 授予地点 | 声学研究所 |
| 关键词 | 多波束水声探测系统 盲信道估计 DS-CDMA 子空间方法 MUSIC算法 多径信道 |
| 学位专业 | 声学 |
| 中文摘要 | 多波束探测系统可以弥补单波束探测系统的覆盖范围小的缺陷，满足精密水下测量的要求，全面、准确、高效地获取海底细微结构，能实现探测范围的宽覆盖。参量阵探测声呐可以在小换能器孔径下实现低频窄波束探测，但是若像常规系统那样通过接收预成波束来实现参量阵多波束探测的话，接收换能器孔径必然要等同于常规低频声呐的孔径，失去了参量阵的低频、小孔径特点。因此，多波束参量阵探测声呐只能采用多波束发射、单一宽波束接收体制，这使得其与常规多波束探测系统有不同的信号处理要求。 本文创新地把多波束水声探测系统中的各个发射波束类比为多用户通信系统中的各个用户，并将直接序列码分多址(DS-CDMA: Direct Sequence Code-Division Multiple Access)通信系统模型应用于多波束水声探测系统，实现了同频同时的多波束信号发射。从而不同方向回波信号的同时检测，可以理解为通信中的多用户检测和信道估计问题，不同的波束就相当于通信系统中的用户。在水下多径传播环境中，多径信道的时延参数和幅度衰减参数的估计是多波束探测系统实现目标探测、定位和识别的基础，而抑制来自其它波束的多址干扰(MAI)的能力是信道估计算法的关键性能。 论文所做的工作主要有以下几个方面： 1. 首先分析了多波束系统的发射信号形式。线形调频信号可以以“频分多址”的方式作为多波束发射信号，但是，这是以牺牲时间分辨率为代价的。伪随机序列信号以正交的码序列区分不同的波束信号，实现信号的同频同时发射，信号形式灵活多样，但接收端采用经典的匹配滤波处理，无法实现对干扰信号的抑制。 2. 论文建立了基于伪随机序列的水下多波束探测系统的信号模型，研究了基于子空间方法的超分辨率盲信道估计算法，得到了合成冲激响应信道，再结合最小二乘拟合算法，计算出多径信道的时延和绝对幅度。 3. 在MATLAB仿真平台上给出了数值仿真实现，分析了信噪比、远近效应问题及多址干扰对信道估计算法性能的影响。仿真结果表明，该算法具有抑制多址干扰、克服远近效应问题的能力，信道参数的估计方差逼近Cramer-Rao下界。这种信道估计技术尤其适合应用于多波束参量阵系统，对水声通信网的研究也具有重要意义。 |
| 英文摘要 | Multi-beam detection system can supply a gap of single-beam detection system which is its finite covering. It meets the needs of precision underwater measurement. Parametric array detection sonar has low frequency and narrow beam using small transducer. But if we accomplish the parametric array multi-beam detection by beaming-forming just like common linear system, the size of the transducer will be the same as the common one. The parametric array loses the character of itself. Therefore, the typical pattern of parametric array multi-beam detection sonar is multi-beam transmission with single wide beam receiver. It requires different signal processing method from common multi-beam detection system. The innovative work in this thesis is that we consider the different beams in multi-beam system as different users in multi-user communication systems. We transmit multi-beam signal at the same time in the same band by applying the model of direct sequence code-division multiple access system to underwater acoustic multi-beam detection system. Consequently, the detection of signals from different directions in underwater acoustic multi-beam detection system can be understood as channel estimation in communication system. Different beams are equivalent to different users in communication system. Time delay and amplitude attenuation estimation are the fundamental of underwater detection, recognition and classification in multipath fading channel. It is a key performance that the algorithm can suppress multiple-access interference (MAI). The main works in this thesis include following aspects: 1. The transmitted signal model for multi-beam system is analyzed. LFM is a kind of multi-beam transmitted signal by FDMA. But it is at the cost of time resolution. Even the match filter can not suppress interference signals. We resort to orthogonal Pseudo-Noise(PN) sequences which can be transmitted at the same time in the same band. We also test the two kinds of multi-beam signals in the water pool experiment. 2. We introduce the signal model for multi-beam detection system based on PN sequences. Subspace-based channel estimation is discussed. Combined with a least-squares fit, the time delay and amplitude attenuation of the multipath channel can be obtained. 3. Then simulation results are given, the effects of signal-to-noise ratio (SNR), near-far problem and MAI are analyzed. It is shown that the proposed algorithm is robust against MAI and the estimated variance approaches the Cramer-Rao bound. This channel estimation technique is applicable to multi-beam parametric array detection system especially. It is also significant to the research of acoustics communication networks. |
| 语种 | 中文 |
| 公开日期 | 2011-05-07 |
| 页码 | 72 |
| 源URL | [http://159.226.59.140/handle/311008/268]  |
| 专题 | 声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文 |
| 推荐引用方式 GB/T 7714 | 姜雪. 多波束水声探测系统中的信道估计技术研究[D]. 声学研究所. 中国科学院声学研究所. 2007. |
入库方式： OAI收割
来源：声学研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。