Decoding Of Space-time Code Based On Neural Networks

With the current development of communication equipment,people-to-people exchanges are becoming more and more extensive.MIMO communication technology has been applied in various fields,including business,entertainment,academic,and military fields.But the crucial part of MIMO systems,space-time coding systems have not developed new solutions in recent years.This makes the use of the original space-time coding system degrades the communication quality if the channel is deteriorated due to weather conditions or interferences,either in terms of transmission accuracy or in terms of transmission immediacy.Therefore,this paper attempts to decode the space-time coding by using a wide range of effective and effective neural network methods for batch data processing,aiming to improve the decoding efficiency of space-time coding,reduce the time and hardware cost,and improve the translation.The accuracy of the code.Since there are not many researches on the application of neural networks in the field of communications,the research in this paper is only an exploration and hopes to provide a reference or reference for follow-up researchers.Firstly,this paper introduces the communication model of MIMO system,which is the basis of space-time coding.Then it introduces the space-time coding model.The research of this topic is based on this model.At the same time,the encoding principle of V-BLAST is introduced,and the traditional zero-forcing detection method and the decoding principle of the symbol-by-symbol maximum likelihood method need to be used in this topic.Then,this paper introduces the principle and basic characteristics of neural network,and theoretically analyzes the possibility and rationality of neural network for space-time decoding.At the end of the chapter,the design idea of ??neural network is proposed.Finally,the designed neural network is verified by experiments,and the neural network method of decoding is compared with the traditional method.The decoding performance of this method is analyzed,and the feasibility and reliability of decoding the space-time code based on neural network is also proved.