海豚启发的非线性处理增强时延-多普勒分辨率技术

83 浏览量更新于2024-08-27 收藏 1.19MB PDF 举报

"这篇研究论文探讨了基于海豚启发的实线性调频脉冲信号非线性处理方法，以增强时延-多普勒分辨率。通过应用海豚回声定位系统的抽象和修改，该研究提出了两个非线性处理策略，并使用单个实线性调频（LFM）信号生成延迟-多普勒图像。接着，通过非线性算法对这些图像进行结合，以实现超高的延迟-多普勒分辨率。同时，论文还评估了这些算法的检测性能，结果显示，提出的非线性处理在保持接近常规匹配滤波接收机的检测性能的同时，显著提高了延迟-多普勒分辨率，且系统和计算复杂度增加不明显。" 本文的核心内容集中在利用生物声纳系统，尤其是海豚的回声定位机制，来改进无线通信中的信号处理技术。海豚的声纳系统以其出色的时延和多普勒分辨能力而著称，研究者从中获得灵感，设计了两个非线性处理方法。实线性调频脉冲（LFM）信号是通信和雷达系统中常用的一种信号类型，因其在频域和时域上的优良特性而受到青睐。在这项研究中，LFM信号被用来构建模型，并经过修改以更接近海豚的生物声纳信号。使用海豚启发的模型，研究人员通过单个LFM信号应用希尔伯特变换，生成了两个延迟-多普勒图像。这一步骤揭示了信号的时延和多普勒变化，是提高分辨率的关键。接下来，引入非线性算法对这些图像进行处理和分析，旨在在组合图像时进一步提升延迟-多普勒分辨率。非线性算法的选择和优化对于整体性能至关重要，因为它们可以挖掘出LFM信号的更多潜在信息，同时不会过度增加系统的复杂性。论文还进行了检测性能的比较，以验证这些非线性处理方法的有效性。通过与传统的匹配滤波接收机的检测性能相比较，发现所提出的非线性处理方案在保持相似的检测性能的同时，显著提升了延迟-多普勒分辨率。这意味着在不牺牲系统效率的情况下，可以实现更精确的目标定位和跟踪。这项研究为无线通信领域的信号处理提供了一种创新的方法，它借鉴了自然界中的生物智能，为提高无线通信系统的性能开辟了新的途径。未来的研究可能包括对更多生物声纳系统的深入研究，以及对不同环境和干扰条件下的适应性优化。

Non-Linear Processing Based on Dolphin Inspired

Real LFM Signals for Enhanced Delay-Doppler

Resolution

Dongzhou Zhan, Hangfang Zhao, and Wen Xu*

Department of Information Science and Electronic Engineering

Zhejiang University

Hangzhou, China, 310027

*e-mail: wxu@zju.edu.cn

Abstract—In this paper, we study two non-linear processing

app

roaches, inspired by biosonars, to enhance the delay-Doppler

resolution. We use real linear frequency modulated (LFM)

signals to fulfill a model, based on an abstraction and

modification of the dolphin signals of certain species. With the

help of Hilbert transform, two delay-Doppler images can be

generated through single LFM signal. Non-linear algorithms are

then applied and analyzed for producing a super delay-Doppler

resolution when combining the images together. We also

investigate the detection performance of those algorithms.

Overall, the developed non-linear processing offers a great

improvement in delay-Doppler resolution while the detection

performance is close to that of the conventional matched filter

receiver without significant increase in system or computational

complexity.

Keywords—Non-linear processing; wideband ambiguity

function; delay-Doppler resolution; Hilbert transform

I. I

NTRODUCTION

High resolution and low level of side-lobes of delay-

Doppler are highly desired for active sonar systems. The

resolution determines the ability of separating targets that are

spatially close together and/or have similar radial velocity to

the sonar, while the level of side-lobes determines the ability of

suppressing the interferences. The delay-Doppler resolution of

the conventional matched-filter sonar is mainly assessed by the

ambiguity function of the transmitted signals. Usually, the

complexity or the time-bandwidth product of the transmitted

waveform determines the resolution of the ambiguity function.

The more complex or the larger time-bandwidth product is, the

better the resolution would be. Alternatively, selection of non-

linear processing may lead to a better delay-Doppler resolution

than that of the matched filter.

The biosonar of dolphins outperforms man-made sonars in

a variety of aspects with their mediocre ‘equipment’ [1], for

instance, the ability to make fine distinction in the features of

targets. The ability indicates that the delay-Doppler resolution

of a biosonar differs significantly from the ambiguity function

of the transmitted waveform. There should be lessons we could

learn from. One of the main specialties for the sonar of

dolphins is the wideband signal that they use. From the work of

Capus, et al. [2], a dual down-chirp model of the bottlenose

dolphin (Tursiops Truncatus) is outlined. Our work is inspired

by that concept with some modification using non-linear

processing for wideband ambiguity function. That is why this

processing is called dolphin inspired, not with the exact

dolphin signals. A similar approach has been exploited, but

mainly based on a modified signal model of bats with dual

complex chirp signals and narrowband analysis for radar

processing [3].

In this paper, we would explore the potential of the real

linear frequency modulated (LFM) signals with Hilbert

transform and non-linear processing. The reason to use real

signals rather than complex signals is clear: real signals are

indeed the signals transmitted and received by sonars. The

comparison between the conventional and the novel processing

shall be based on the wideband ambiguity function naturally.

The bandwidth of the LFM signals against the carrier’s

frequency can be pretty large for an underwater acoustic

system, and thus the narrowband ambiguity function is not

appropriate. The rest of the paper is organized as follows. In

Section II, we describe the observed signals used by certain

species of dolphins. We then present a simplified model

inspired by those signals. The connection between the real

LFM signals and the model is developed in details. In Section

III, we investigate how single LFM signal could generate two

different delay-Doppler images. The non-linear processing,

which is applied to those two images for enhancing the delay-

Doppler resolution, is discussed. The detection performance of

the non-linear processing is evaluated in Section IV. At last, in

Section V, some conclusions from this paper are given.

II. D

OLPHIN

NSPIRED SIGNAL MODELS

Dolphins transmit a diversity of types of click signals,

which plays an important role in perceiving the environment

involving detection, discrimination and so on [4]. There are

seven types according to the click taxonomy based on the

dominant peaks in upper and lower frequency bands [5], and

those types of signals vary a lot both in the time and frequency

domains. However, all those types of click signals can be

abstracted and modeled as dual down chirp signals shown in

Fig. 1(a). Those down chirps strongly overlap in the time and

frequency domains and each down chirp can have different

amplitude. Here we introduce a modified model of Fig. 1(a).

One of the down chirp signals is replaced by an up chirp signal,

and two parts are totally overlapped with one another in time

下载后可阅读完整内容，剩余6页未读，立即下载

weixin_38563176

粉丝: 2
资源: 920

海豚启发的非线性处理增强时延-多普勒分辨率技术

空间谱时延估计与线性调频脉冲压缩方法研究

MATLAB实现线性调频连续波信号及其多普勒频移模拟

基于STFRFT的无源雷达高精度时延与多普勒联合估计

基于频域时延-多普勒二维聚焦的欠采样雷达信号参数估方法.docx

波导环境多目标时延-多普勒解卷成像

LFM信号模糊函数在二维时延-多普勒平面内，用matlab仿真出来

matlab_生成一个单脉冲调制的线性调频连续波信号，并且加入了距离时延与多普勒频移的信息

LFM_signal.rar_LFM时延_线性 时延_线性调频 时延_脉冲压缩_脉冲压缩 时延

xdpakzib.zip_空间谱时延_线性调频 时延_脉冲压缩 时延

关于FIR滤波器的群时延与脉冲压缩“暂态点”---讨论.pdf

最新资源

LFM_signal.rar_LFM时延_线性时延_线性调频时延_脉冲压缩_脉冲压缩时延

xdpakzib.zip_空间谱时延_线性调频时延_脉冲压缩时延