Research On Key Technologies Of Hyperspectral Microwave Radiation Detection

Microwave remote sensing enables the rapid and efficient acquisition of parameters related to deep space,planets,planetary surfaces,and atmospheres from a macroscopic range,under all-day and all-weather conditions.This not only fosters scientific development but also significantly influences the sustainable progression of socioeconomic systems and the security and stability of nations.As the primary instrument for microwave passive remote sensing,the microwave radiometer has evolved as the principal tool for human exploration of Earth and beyond.Concurrent with advancements in space information technology and escalating application demands,there is an increased expectation for detection accuracy,resolution,and spectral line quantity in microwave radiometers.Consequently,the evolution of microwave radiometers is rapidly advancing toward hyper-spectral,wide-band,and high-sensitivity capabilities.The hyperspectral microwave radiometer is a sophisticated device capable of detecting hundreds or even thousands of spectral lines.Its key technology resides in the subdivision of channels within the intermediate frequency(IF)processing module.The rapid advancements in chip and digital signal processing technologies pave the way for the integration of all-digital IF modules into hyperspectral microwave radiometers.Nevertheless,the subdivision of channels in microwave radiometers can pose challenges such as reduced detection sensitivity,increased computational demands for inversion,and the complexities associated with real-time transmission of substantial data volumes.Consequently,the methodology for channel subdivision and configuration in hyperspectral microwave radiometers requires further investigation.Furthermore,spaceborne microwave radiometers are tasked with the challenge of achieving high sensitivity and ultra-multi-channel detection of broadband weak signals within constrained detection timeframes.Addressing the paradox between the speed of sampling,the processing velocity of chips,and the necessity for real-time channel subdivision processing is an exigent issue that must be resolved in the evolution of hyperspectral microwave radiometer systems.This dissertation centers on key technical research concerning the detection performance and channel configuration optimization of hyperspectral microwave radiometers,full-digital channel subdivision technology,and atmospheric temperature profile detection models.A simulation analysis platform was constructed for detecting atmospheric temperature profiles using a hyperspectral microwave radiometer.The information content analysis method was introduced into the hyperspectral microwave detection performance analysis,and key issues such as hyperspectral microwave retrieval performance and channel configuration optimization were analyzed.Theoretical models of channel subdivision key technology were proposed,and the processing logic circuits were designed and implemented.Key indicators and performance were tested.The main research and innovations of this dissertation are as follows:(1)The application of information capacity analysis from information theory has been introduced into hyperspectral microwave radiation detection.Consequently,two optimization strategies for channel subdivision have been proposed: one for maximizing detection information acquisition and the other for minimizing information loss during channel selection.Proved that hyperspectral microwave radiometers substantially enhance the precision and vertical resolution in detecting atmospheric temperature profiles,thus offering crucial support for the design of IF channel subdivision schemes within these devices.(2)A digital IF uniform channel subdivision algorithm architecture for hyperspectral microwave radiometers has been developed.This architecture addresses the conflict between high-speed sampling,chip processing speed,and real-time channel subdivision processing requirements of hyperspectral microwave radiometers,enabling hyperspectral radiation detection.The system employs multi-channel synthesis and mixed-radix fast Fourier transform algorithms,implemented on a pipeline architecture using a Field Programmable Gate Array(FPGA).This approach resolves the issues of low data throughput and processing efficiency associated with traditional Fourier transform serial algorithm architectures,enhancing channel subdivision speed and achieving real-time channel subdivision processing for hyperspectral microwave radiometers.Concurrently,research was conducted on topics such as channel response,digital noise,and design optimization.Furthermore,a resource optimization scheme based on parity decomposition algorithm was proposed to improve the utilization efficiency of FPGA resources.(3)Two distinct model types were introduced: a segmented non-uniform channel subdivision employing polyphase filter banks,and a comprehensive non-uniform channel subdivision utilizing the Goertzel optimization algorithm,both facilitating adaptable configurations of radiometer channels.To address the challenge of parallelizing real-time processing for high-speed sampling data within non-uniform channel division algorithms,a structural model based on polyphase filter banks was developed.This model partitions the detection frequency band into multiple sub bands.An investigation was conducted on the design of polyphase filter banks and the responses of the sub-frequency channels,leading to the establishment of an overarching framework for the digital IF non-uniform channel subdivision module.Concurrently,functional verification as well as evaluations of resource utilization and power consumption were performed on the channel subdivision algorithm implemented in digital logic circuits.The outcomes indicate that the non-uniform channel subdivision technique,in contrast to uniform channel subdivision algorithms,achieves autonomous frequency sub bands,thus significantly enhancing the flexibility of channel subdivision and configuration.(4)A testing platform was established for the IF module of a hyperspectral microwave radiometer.Experimental tests were performed to evaluate the input frequency range,spurious-free dynamic range,flatness,linearity,and stability of the two types of digital IF modules designed in this dissertation.Additionally,the sensitivity of a hyperspectral microwave radiometer system based on the IF module developed in this dissertation was tested.Experimental verification was also conducted on different channel bandwidths of non-uniform channel subdivision IF modules.The experimental results indicate that the sensitivity of the developed hyperspectral radiometer matches that of inorbit radiometers,with the number of spectral lines surpassing that of in-orbit radiometers by two orders of magnitude.Simultaneously,the channel configuration of the nonuniform channel subdivision module demonstrates a high degree of flexibility,marking a significant advancement in microwave and millimeter-wave radiometer technology.The research presented in the dissertation holds significant reference value for the advancement of hyperspectral microwave radiometer technology.Its findings are applicable to various fields,including atmospheric surveillance,meteorological prediction,and environmental monitoring.Furthermore,the study offers a theoretical foundation and technical guidance for the design and enhancement of future hyperspectral microwave radiometers.