Research On High-frequency Ground Wave OTH Radar Frequency Monitoring Algorithm

The work frequency of HF ground wave over the horizon radar is on theshortwave radio spectrum which is very crowded and complicatedin which theelectromagnetic environment is complex and changeablebecause the distributionof many interference,and atmospheric radio noise changes with location、seasonand weather all of which lead to a random disturbance in shortwaveband.Themain purpose of this project is to make high-frequency ground wave OTH radaravoid various interference in the work bands to make the radar be more reliableso it is necessary to monitoring the quality of the work spectrum band of thehigh-frequency radar for uninterrupted and real-time, and then to choose a"silent" band that can ensure high-frequency radar capable of reliable operation,thereby ensuring high-frequency ground wave OTH radars have the ability tocarry out normal work in a complex electromagnetic environment that is complexin the shortwave bands.This research aims to study a series of monitoring algori-thm for high frequency radar electromagnetic environment.The main contents of this paper include the following three aspects:The first is the study of the spectrum state prediction method. In this part,according to the historical data of measured spectrum we forecast the next cycleof spectrum using different methods, including a detailed introduction of theAR model prediction method, single exponential smoothing prediction method,then describing in detail the method of empirical mode decomposition(EMD) andsupport vector regression (SVR) forecasting method, after using support vectormachine method for spectral prediction to process the data that handlingby empirical mode decomposition method. And we compare the three kindsof prediction using standard of reliable performance. Results show that the predi-ction method of the empirical mode decomposition and support vector machine isbetter in reliabilitythan the AR model prediction method and single exponentialsmoothing method.Then the method of frequency selection. This part is divided into two parts:the first is the sliding window method selection: using the two factors includinga sliding window average power spectrum value and fluctuation value as slidingwindow selection criteria, firstly calculating the average power spectrumof sliding window value and fluctuation information and obtaining each frequen-cy point score according to certain criteria, then calculating each frequencypoint score considers the two factors, and finally obtaining the optimal frequencywhich has the highest score. Then we put forward a frequency selection method which is suitable for Bistatic Radar: state vector fusion method basedon Calman filter(the Calman filter then weighted fusion). Thefirst usingelectromagnetic spectrum of transmitting radar and receiving radarwith the designed filter to estimate Calman filter respective state value; in orderto improve the fusion accuracy, using two weighted method tofuse, first weighting to the single sensor, and then weighted to the system as awhole, from the respective state system state; the spectrum after fusion forsliding window selection to get the best band.The last is the study of frequency selective verification method, we makeVerification of radar can reliably work in the alternative band that is selected bysliding window. This part introduces power spectrum verification mode andtarget extraction verification method. Let the radar works in alternativeband, making target extraction treatment for the received data, because thetarget already is known, so we can get the processed signal SNR, using it asa basis for selecting quality, choosing the reliable working band. In the lastpart, we analyze the difference of target signal extraction underdifferent alternativeband.