【作者】 徐冬冬；

【导师】 周杰； 魏鸣；

【作者基本信息】 南京信息工程大学 ， 大气遥感科学与技术， 2018， 博士

【摘要】 水汽是大气重要的组成部分,大气中水汽含量分布情况及水汽变化情况非常复杂,推动天气和气候变化,在中小尺度灾害性天气形成和演变过程中起重要作用。如何有效地、准确地获取水汽分布的信息,对天气预报、气候监测及防灾减灾具有特别重要的意义。一般大气水汽探测受到探测精度和时空分辨率影响,无法获取高精度、高时空分辨率的水汽信息。当前的水汽监测系统在观测时间和空间分辨率、设备造价、布网密度和监测精度等方面存在很多不足。针对当前水汽监测系统存在的问题,以电磁波载波信号作为介质反演大气折射指数,进行水汽探测。借助于目前广泛应用的第四代移动通信网络,采用多智能体系统接收电磁波信号,构建水汽监测网络,作为现有水汽监测网络的重要补充。本文提出了一种基于多智能体的边界层水汽探测方法,深入研究该探测方法中的核心技术,以下为本文的主要工作。论证了多智能体方法反演大气折射指数的可行性。对于采用通信基站发出的载波频率为2.45GHz的电磁波作为介质的探测方式研究,实质上是对于电磁波在边界层低层大气中传播路径上衰减的研究。很显然,其衰减与电磁波传播路径上大气温度、气压和湿度相关,即与电磁波传播路径上的水汽含量相关。根据利用电磁波折射率反演大气中水汽含量的原理,设计了应用于水汽探测的多智能体系统。该系统由采用改进的层次型网络结构,自下而上由传感器智能体、管理智能体和接口智能体三种智能体构成。定义了多智能体系统的组织结构和数据传输形式。对多智能系统中传感器智能体的数据采集算法进行了设计与分析。从多智能体系统探测网络技术要求、单频同步网络和第四代通信系统中采用的数据传输技术三个方面论证了多智能体系统观测边界层水汽的可行性。论证了基于多智能体的边界层水汽探测方法中利用接收到的电磁波载波信号反演电磁波传播路径上大气折射率N的反演方法和信号处理算法,并进行相应的反演实验来验证不同的算法对应的探测精度,分析了三种不同算法存在的问题和改进方法。(1)调制信号自相关函数信号处理算法。采用调制信号自相关函数计算电磁波信号从基站发出到接收端接收载波信号的传输延时,并依此计算电磁波载波信号传输伪距,反演传输路径上大气折射指数。(2)载波相位差信号处理算法。对比信号接收端接收到的载波信号与同步产生的载波信号之间的相位差,根据整周期模糊数和相位差计算电磁波载波信号传输伪距,反演传输路径上的大气折射指数。(3)导频信号追踪时间定位算法。导频信号追踪时间定位算法是利用第四代移动通信系统中采用的OFDM传输技术精确计算信号传输延迟,进而精确计算传输伪距,反演大气折射指数。通过对反演结果的比较分析,在相同区域,相同信号处理算法反演不同天气条件下大气折射指数。建立数学模型,分析大气中水不同形态对于电磁波传输衰减对反演结果的影响。采用导频信号追踪时间定位算法在不同天气条件下进行观测实验。实验结果验证了多智能体方法反演大气折射指数的可行性,并且反演结果精度满足水汽探测需求。本文的研究证明了基于多智能体的边界层水汽探测方法可行,反演结果精度较高,可以作为当前水汽探测的有效补充。更多还原

【Abstract】 Water vapor in the atmosphere is an important part of the atmosphere.The distribution and change of water vapor content is very complex.It is the main driving force for weather and climate change and plays an important role in the formation and evolution of severe weather at small and medium scales.How to effectively and accurately obtain information on water vapor distribution is of special significance for weather forecasting,climate monitoring,and disaster prevention and mitigation.The conventional atmospheric vapor detection method is affected by the spatial and temporal resolution or detection accuracy,which limits the acquisition of high-temporal resolution and high-precision vapor information.Current water vapor monitoring systems have many shortcomings in terms of observation time and spatial resolution,equipment cost,network density,and monitoring accuracy.Aiming at the problems existing in the current water vapor monitoring system,a multi-agent system is used as a receiving end of an electromagnetic wave to construct a water vapor monitoring network by means of a fourth-generation mobile communication network that is currently widely used.The atmospheric refractive index is retrieved from electromagnetic carrier signals as an important supplement to the existing water vapor monitoring network.This paper proposes a multi-agent-based boundary layer vapor detection method,and conducts in-depth research on the key technologies of this detection method.The main work of the paper includes the following aspects.The feasibility of multi-agent method inversion of atmospheric refractive index was demonstrated.The study of the detection method using electromagnetic waves with a carrier frequency of 2.4 GHz emitted by a communication base station is essentially a study of the attenuation of electromagnetic waves in the propagation path of the boumdary layer in the lower atmosphere.Obviously,the attenuation is related to the atmospheric temperature,pressure,and humidity on the electromagnetic wave propagation path,that is,the water vapor content on the electromagnetic wave propagation path.According to the principle of retrieving the water vapor content in the atmosphere using the refractive:index of electromagnetic waves,a multi-agent system was designed.The system consists of an improved hierarchical network structure,and consists of three types of agents:sensor agent,management agent,and interface agent.A multi-agent system for water vapor detection was designed.The organizational structure and data transmission form of the multi-agent system are defined.The design and analysis of data acquisition algorithms for sensor agent in multi-intelligent systems were carried out.The feasibility of the multi-agent system to observe the water vapor at the boundary layer is demonstrated from three aspects:multi-agent system detection network technology requirements,single frequency synchronization network,and data transmission technology used in the fourth generation communication system.Systematic demonstration of inversion method and signal processing algorithm for retrieving the atmospheric refractive index N on the propagation path of electromagnetic waves based on the electromagnetic carrier signal received by the receiving end in the multi-agent boundary-layer water vapor detection method,and performing corresponding inversion experiments Verify the corresponding accuracy of different algorithls,analyze the problems and improvement methods of the three different algorithms.(l)Modulation signal autocorrelation function signal processing algorithm.The modulation signal autocorrelation function is used to calculate the propagation delay of the electromagnetic wave signal sent from the base station to the receiving end of the received carrier signal,and the electromagnetic signal carrier transmission pseudorange is calculated accordingly,and the atmospheric refractive index on the transmission path is inverted.(2)Carrier phase difference signal processing algorithm.Comparing the phase difference between the carrier signal received by the signal receiving end and the carrier signal generated syllchronously,the carrier pseudorange of the electromagnetic wave carrier signal is calculated according to the integer period fuzzy number and the phase difference,and the atmospheric refractive index on the transmission path is inverted.(3)Pilot signal tracking time positioning algorithm.The pilot signal tracking time positioning algorithm uses the OFDM transmission technology used in the fourth generation mobile communication system to accurately calculate the signal transmission delay,thereby accurately calculating the transmission pseudorange and inverting the atmospheric refractive index.Through the horizontal comparison and analysis of inversion results,in the same region,the same signal processing algorithm inverts the atmospheric refractive index under different weather conditions.A mathematical model was established to analyze the effect of different forms of water in the atmosphere on the inversion results of electromagnetic wave attenuation.The pilot signal tracking time positioning algorithm is used to perform observation experiments under different weather conditions.The experimental results verify the feasibility of the multi-agent method in retrieving the atmospheric refractive index,and the accuracy of the inversion result satisfies the demand for water vapor detection.The research in this paper proves that the multi-agent-based boundary layer vapor detection method is feasible and the inversion result is of high precision,which can be used as an effective supplement for the current vapor detection.更多还原