A Study On Channel Information Analysis And Configuration Of Microwave Hyperspectral Radiometer

The microwave radiometer is a passive microwave remote sensing detection instrument.It has the advantages of full-time,all-day work and little influence by clouds.The use of microwave radiometers to detect atmospheric temperature is of great significance for numerical weather forecasting and climate detection.At present,the number of microwave radiometers channels is small.To obtain detailed information on atmospheric state parameters and improve the atmospheric detection accuracy of microwave radiometers,the number of microwave radiometers channels is increasing,and microwave hyperspectral radiometers will become a trend.The use of microwave hyperspectral detection technology for temperature retrieval means that the sensitivity is reduced,the calculation scale is increased,and the training of the retrieval model is more difficult.Therefore,it is of great significance to study the channel information extraction technology of microwave hyperspectral radiometer for improving the detection accuracy of temperature profiles and efficiently using hyperspectral data.This thesis focuses on channel information capacity and the method of channel selection.The main research contents include:Firstly,the calculation process of radiation brightness temperature is studied and the radiation brightness temperature data in the frequency band of 50?70GHz is simulated by using the Seebor V5.0 database.At the same time,the inversion method of atmospheric temperature profile by using radiation brightness temperature is studied,which provides a theoretical basis for subsequent channel information analysis and radiometer design.Second,the method of channel information extraction is studied.The effect of extracting information and the temperature retrieval results using the channel-bychannel selection method and principal component analysis method are analyzed and compared.Experiments show that principal component analysis method has advantages in extracting information and this method has higher inversion accuracy when extracting the same information capacity.Third,the parameters of the microwave radiometer are analyzed.Using the hyperspectral radiation brightness temperature simulation model in the 50?70GHz frequency band and the Seebor V5.0 atmospheric state parameter profile data,this thesis analyzes the influence of microwave radiometer bandwidth on the information capacity,the observed brightness temperature details and retrieval effect.The size of the bandwidth is determined.Under the condition that the channel bandwidth is determined,the influence of the four factors,including sensitivity,season,the number of atmospheric nodes and latitude,on the accuracy of information capacity extraction and temperature profile retrieval is analyzed.Fourth,according to the specific design specifications of the microwave radiometer,the relationship between information capacity and channel bandwidth under different system noise temperatures is analyzed,and the channel bandwidth under the design index is determined.Using the observation data of the microwave thermometer MWTS and the microwave hygrometer MWHS carried by the Fengyun 3D satellite,the temperature profiles retrieval effects of 13 channels of MWTS and 2000 channels in the50-70 GHz frequency band are analyzed and compared,and the design necessity of multi-channel temperature profile is verified.The research work in this thesis provides a theoretical basis for the development and application of microwave hyperspectral radiometers in the future.