| Microwave radiometer is a highly sensitive noise receiver.By acquiring the thermal radiation of the target,microwave radiometer can identify its characteristics.The L-band microwave radiometer can effectively detect the land surface soil moisture and ocean salinity.The thermal radiation of the ground object is extremely weak compared with the energy of the artificial microwave radiation source,resulting in that the microwave radiometer is vulnerable to the interference from it.The interference process of the artificial microwave radiation source is called Electromagnetic Interference(EMI).Strong EMI(> 300K)can lead to the saturation of microwave radiometer receiver output value;medium EMI(20K-300K)can be detected by the transient change of microwave radiometer output variables;weak EMI(< 20K)is similar to target thermal radiation and is not easy to be detected,which affects the retrieval accuracy of surface parameters.In the receiver system of the traditional integrated detection microwave radiometer,as long as the interference signal causes a power change,and within the detectable range,it will cause a change in the output.And no matter what form the interference signal is,it can be equivalent to a broadband input noise of the same magnitude.Based on this hypothesis,this paper studies the detection and suppression methods of electromagnetic interference in L-band microwave radiometer by analyzing the statistical characteristics of the received signal of microwave radiometer with digital gain automatic compensation,through data statistical analysis and controllable experimental simulation.The main contents and innovations of this paper are as follows:(1)An experiment of pulse radiation interference is designed,and the interference of different duration and magnitude is simulated.On this basis,the suppression algorithm of electromagnetic interference is studied.(2)The function of parameters in the Asynchronous Pulse Blanking algorithm(APB)is analyzed in detail.The statistical analysis method is introduced in the calculation of initial value.And the selection of initial value is defined by limiting the range of mean and variance of selected data segments.The limitation of standard deviation is added in the detection process,which is equivalent to setting the value range of detection threshold,ensuring the effective detection of continuous interference when measuring relatively stable targets.(3)On the basis of in-depth analysis of the transmission functions of microwave radiometer systems in noise measurement,through actual measurement statistics,it is found that the variation in coefficient of variation of output voltage can reflect the existence of impulse interference.A median comparison method for coefficient of variation is proposed,and the calculation method of threshold is given.The experimental results show that it is not suitable for detecting small interference when the frame capacity is large,while the frame capacity is too small,the coefficient of variation is too sensitive to statistic.(4)Based on the characteristics of ground-based remote sensing observation and the working parameters of microwave radiometer,the algorithm was improved:(a)When the interference occurs frequently and the magnitude difference is large,the values with coefficient of variation greater than 9 and less than 2 are excluded,and directly labeled as interference data,the detection rate increases by 2%;(b)When the interference of small magnitude lasts for a long time,and the whole frame data may be interfered,and the coefficient of variation may not change much,so the frame can not be identified by detection.Therefore,the filtering results of a statistical group are sorted,and the average brightness temperature of the frame within the appropriate range of the sequence is selected as the final result.(5)On the basis of theoretical analysis and simulation experiment,an L-band microwave radiometer with fast data acquisition ability which is able to capture millisecond time scale EMI information is developed.(6)Based on the L-band microwave radiometer,the ground-based remote sensing experiment of long time series is carried out.With the proposed median comparison algorithm,radar pulses interference coming from the direction of the airport is successfully detected.And the level of this type interference is about 5-10 K.(7)In the joint experiment of airborne-ground soil moisture observation in Shandian River Basin,various types of interference in the experimental area were identified.The results show that:(a)The interference of microwave radiometer increases sharply with the increase of observation angle in the complex environment of interference;(b)The EMI,less than 2K,can not be identified in the airborne and ground observation,which will affect the accuracy of soil moisture retrieval by approximately 2.2%-2.8%.The research work of this paper provides theoretical basis and technical scheme for the development of a highly sensitive L-band microwave radiometer with EMI detection capability,provides ground verification for the airborne calibration and remote sensing observation of surface water satellites of our country,and improves the precision of surface parameter retrieval in algorithm. |
PDF Full Text Request