Receiver Test And Evaluation Methods Based On Playback System Of GNSS Signal Of Real Scenarios

After several decades of development,the Global Navigation Satellite System(GNSS)has been used in almost every area of people's life.The satellite navigation has become one of the fastest growing industries.It is also the most effective means of scientific research in the fields of cosmology,geophysics science and others.The receiver terminal is an important part of the application of satellite navigation system,and one of the core technologies in this system is the navigation terminal test technology.How to achieve the terminal test has significant influence on developing high-performance navigation receiver,and will bring economic and social benefit in the long term.The present researches of GNSS receiver performance testing are divided into two kinds,including simulator testing and road testing.However,the existing simulator technology cannot provide an accurate and effective model in the specific navigation signal environment.So there is a phenomenon that the theoretical results are inconsistent with the real test results.The data in the road test is true and reliable,but the operation is complex and non-repeatable,which is resulting in low test efficiency.In order to solve the above problems,this paper presents the receiver test and evaluation methods based on signal playback in the real scene.It collects the real scene signals and stores them in the database in the form of digital intermediate frequency signals.The data sets used for testing are determined by classification methods.The signal playback transforms to radio frequency GNSS signals through up-conversion for receiver testing.This article mainly focus on the following three aspects:1.The typical scene classification of GNSS signals based on KNN and decision tree algorithms.The GNSS receiver can output the parameters such as carrier-to-noise ratio,DOP value and velocity of the satellite signal received in each scene.The GNSS signal scenes are pre-classified according to the KNN algorithm.This paper makes decision according to the classification set and classifies the scene by decision tree algorithm for further verification.2.The test evaluation program based on the real signal playback.This part sets up a signal scenario playback system,which is realized the receiver test scenarios through the typical scene signal in the database.A test evaluation scheme based on the �parallel recursive model' is presented.In this test,the definition and theoretical analysis of some indicators are given,which include precision,integrity,continuity,robustness and adaptability.3.The test method and evaluation standard based on the test evaluation scheme.According to the test evaluation scheme proposed in the paper,the test method and evaluation standard of indicators are given.The performance of the receiver is evaluated by combining with the receiver modules of different performance in the market.In summary,this paper studies the typical scene classification of GNSS signals based on KNN and decision tree algorithms,and the test evaluation system and scheme based on the real signal playback.Then,the evaluation methods and evaluation standards are deduced and explained.Finally,the test scheme of the receiver is validated.