Research On Drone-borne Passive Microwave Remote Sensing Observation System

A microwave radiometer is a high-sensitivity broadband noise receiver working in microwave band.It can extract the variation of weak microwave radiation signal from strong background noise.It mainly deals with the Gaussian white noise radiated by ground objects.The spaceborne microwave radiometer plays an irreplaceable role in large-scale resource and environment monitoring.Its resolution is low,and the observed areas are mostly mixed pixels,which cannot meet the application requirements of high spatial and temporal resolution surface parameter monitoring at small and medium scales(such as agriculture precision,regional resource and environment monitoring,etc.).The ground-based radiometer has high resolution,whereas it is limited to the terrain and transportation mode.The observation is based on sparse discrete points,so it is unable to form large-scale area data.Using the Unmanned Aerial Vehicle(UAV)as the carrying platform of radiometer can obtain gridded brightness temperature data with high spatial resolution,which has great advantages in maneuverability,convenience and cost compared with manned aircraft,and can fill the gap between ground-based experiment and flight experiment.In this paper,a drone-borne passive microwave remote sensing observation system is developed,and the data acquisition,data storage and data processing are discussed.Finally,high-resolution brightness temperature mapping of microwave radiation is obtained,and the main work is as follows:1.A light-weighted and miniaturized microwave radiometers are developed and carried on the UAV platform.The sensitivity and stability of the receiver are guaranteed,the complex hardware circuit feedback control link is eliminated,and the power consumption is reduced by adopting the digital automatic gain compensation(AGC)scheme.The micro control unit(MCU)is updated,as well as the AD and DA converters,the circuit board is redesigned more integrated and reduced in volume,and more efficient power supply scheme is updated in order to reduce the power comsumption,the weight and the volume of the receiver.The timing framework of the on-chip program is designed,and the efficient and stable framework algorithms are established to ensure the timeliness and stability of collaborative processing of multiple data.The sampling rate is increased(100 microseconds)to ensure the data quality and flight speed.2.A high-speed storage technology of on-board SD card is successfully realized.Aiming at the storage problems caused by continuous acquisition based on STM32processor,the advantages and disadvantages of the existing storage methods and the applicable conditions are studied,and a timer triggered dual polling storage method(DPSM)under the multitasking framework is proposed.Using time-sharing processing and parallel technology,the time consumption of the existing storage methods is reduced from 4000 microseconds to 200～400 microseconds,which is about 90～95%.The new method does not cause any delay in the acquisition process,solve the problem of data storage during the continuous acquisition operation of unmanned aerial vehicle microwave radiometer,so as to ensure the integrity of data storage,and provide data guarantee for research.3.A drone-borne passive microwave remote sensing observation system is successfully integrated.A spatial position and attitude acquisition system is established to match the acquired high-resolution data in time and space.A complete data link system is established,data is stored in real time and sent to ground control stations and remote cloud servers during the flight.The instantaneous field of view of the radiometer is analyzed and modelled,and real-time data processing and visualization are achieve d through programming.The observation system successfully flew,obtaining high-resolution passive microwave radiation brightness temperatures within a regional range,and achieving high-resolution microwave radiation brightness temperature mapping.4.A data processing process for passive microwave radiation brightness temperature observation of unmanned aerial vehicles is established,and a geolocation method considering terrain and angle correction is proposed.The gridding algorithm of spaceborne radiometer data is applied to the data acquired by the unmanned aerial vehicle platform.Aiming at the high-resolution microwave radiation brightness temperature data obtained from the drone-borne passive microwave remote sensing observation system,combined with the observation of terrain and real-time attitude of the radiometers,the error in the geolocation process is analyzed,and a data processing method considering topographic and angle correction is proposed to correct the instantaneous field of view of radiometers.5.The performance of the proposed data processing method is verified by experiments on planar terrain and complex terrain.The results show that the data processing method proposed in this study can achieve regional mapping of microwave radiation brightness temperature.Compared with the normalized water difference index(NDWI)classification results extracted from multispectral images,the consistency rate for the boundary range of water bodies extracted from radiant brightness temperature is 83%,which can well reflect the radiant brightness temperature characteristics of ground objects and provide pure pixel values.The calculated theoretical value and the corrected value based on the actual angle and terrain are very close to the 1:1 line and R~2=0.87,verifying the effectiveness of our proposed angle and terrain correction method.