پديد آورنده :
نوروزي، مجتبي
عنوان :
طبقه بندي سيگنال هاي راداري با استفاده از روش هاي بازشناسي آماري الگو
مقطع تحصيلي :
كارشناسي ارشد
گرايش تحصيلي :
مخابرات﴿سيستم﴾
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر
صفحه شمار :
دوازده، 107ص.: مصور، جدول، نمودار
يادداشت :
ص.ع. به فارسي و انگليسي
استاد راهنما :
محمدرضا احمدزاده
توصيفگر ها :
شناسايي رادار ساطع كننده , نظريه ي مجموعه هاي راف , گسسته سازي , ماشين بردار پشتيبان , ماشين بردار رابط
تاريخ نمايه سازي :
2/3/94
استاد داور :
عبدالرضا ميرزايي، محمد مهدي نقش
تاريخ ورود اطلاعات :
1396/09/27
رشته تحصيلي :
برق و كامپيوتر
دانشكده :
مهندسي برق و كامپيوتر
چكيده انگليسي :
Radar Signal Classification using Statistical Pattern Recognition Mojtaba Norouzi m norouzi@ec iut ac ir Date of Submission 2015 01 13 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Farsi M R Ahmadzadeh Assist Prof supervisor Ahmadzadeh@cc iut ac irAbstract Data classification is an important task in a variety of scientific fields Radar signals are knownas significant information that are used for detecting objects in surrounding area Classification has amajor role in radar technology Radar emitter identification is the main part of the electronic supportmeasures ESM in electronic warfare After intercepting signals from an antenna few hierarchicalprocesses are needed to be applied These processes can be divided into feature extraction and classifi cation steps Data classification is performed on the structures called pulse descriptor words PDWs First step of this processing is the data clustering Many clustering methods have been proposedtill now in which almost all of them estimation of the number of clusters is an important issue In thisthesis we propose a new method in order to estimate the number of clusters This method is based onrough sets theory and data discretization Our proposed method has not only the ability to estimate thenumber of clusters but also determining initial centers of clusters We used hierarchical and k meansclustering algorithm in order to achieve the goal of putting data in proper clusters In the second stepwe also propose an algorithm based on support vector machine SVM and relevance vector machine RVM for classification of radar data We shall examine some validation and correctness criterionsto show the strength for our proposed method on clustering and classifying radar data Experimentalresults show a significant improvement in comparison with the other methods Keywords Radar Emitter Identification Rough Sets Theory Discretization Support Vector Machines RelevanceVector Machines
استاد راهنما :
محمدرضا احمدزاده
استاد داور :
عبدالرضا ميرزايي، محمد مهدي نقش
