Carrier And Interference For Satellite Communication Signals Realtime Detection Method Design And Implementation

With the development of science and technology,communication technology has become an indispensable part of human society.Satellite communication is widely used in military and civil applications due to its large signal transmission bandwidth,wide network coverage area and low geographical and spatial restrictions.In the face of this situation,a powerful spectrum scanning and monitoring device is needed to monitor broadband satellite communication signals in real time to meet the current demand for spectrum awareness and electronic countermeasures.Traditional spectrum monitoring equipment often uses narrowband sweeping to achieve the spectrum monitoring function,which is difficult to monitor the satellite communication signals using broadband transmission with high accuracy in real time.To address the above situation,this thesis focuses on the research and implementation of a real-time carrier and interference detection method for satellite communication signals to monitor non-cooperative broadband satellite communication signals in real time.Firstly,this thesis analyses and simulates the existing carrier and interference detection methods and explains the limitations of these methods.A low-complexity carrier and interference detection algorithm for satellite communication signals is proposed to address the characteristics of a large number of signals under broadband satellite communication conditions,a large span of bandwidth coverage,and the dynamic undulation of the noise floor caused by the high-speed motion of low-orbiting satellites.The algorithm enables noise floor cancellation of broadband signals,carrier detection and interference detection of broadband signals after noise floor cancellation,and extraction of basic parameters of carrier and interference.The algorithm is then simulated and implemented in hardware.The advantages of this algorithm are low resource consumption and low complexity of the algorithm.This thesis then addresses the problem that low-complexity satellite communication signal carrier and interference detection algorithms require expert knowledge to adjust the detection parameters.Inspired by the structure of the Denoising Autoencoders network,a noise-substrate offset algorithm based on a one-dimensional convolutional Denoising Autoencoders network is proposed and combined with an improved LADACC algorithm to achieve automatic detection of satellite communication signal carriers and interference.After determining the network structure,this thesis also simulates the colour noise in an actual satellite communication scenario and feeds it and the real collected satellite signals into the network as training data for training.Finally,the algorithm is tested and the MAE at a maximum noise floor undulation of 18 dB is reduced by 1.36 dB compared to the low-complexity satellite communication signal carrier and interference detection method.Finally,an FPGA implementation of a one-dimensional convolutional Denoising Autoencoders network is designed,taking advantage of the parallel computing of FPGAs to implement the basic units of the encoder and decoder,and relying on data quantization operations to reduce the required storage resources as well as the computational effort to a quarter of the original.When compared with the resource consumption of lowcomplexity satellite communication signal carrier and interference detection algorithms,this method increases the resource consumption but improves the performance of realtime detection of satellite communication signal carriers and interference when the noise floor fluctuates significantly and does not require expert knowledge to adjust the detection parameters.