Wireless Channel Feature Extraction And Its Application Based On Markov Process

Wireless channel feature extraction is very important in the development of wireless communication systems.The extracted features are mostly used in wireless channel modeling and its related applications.The wireless channel model based on Markov process is one of the wireless channel models,and it has a good development prospect in the wireless channel modeling and its application.The Markov process has the characteristics of simple calculation and real-time fastness.However,the matching degree of different forms of Markov wireless channel model is different from each other in the current real-time scene.This makes the practicality of the Markov model have been affected.Therefore,how to use the Markov model in wireless channel modeling has great research value.This paper firstly introduces the research status of wireless channel feature extraction,wireless channel modeling and its anomaly detection.And then several common methods of wireless channel modeling and anomaly detection have been described.After that,it can be concluded that the anomaly detection accuracy at low signal-to-noise ratio(SNR)can not be guaranteed with the existing modeling methods under the premise of simple and fast calculation.At the same time,the complexity of the detection method is relatively high under the premise of ensuring the accuracy in low SNR condition.And also the real-time performance is not guaranteed.In view of the above problems,this paper connects the Markov model with the existing large-scale logarithmic power loss model.This method can improve the accuracy of modeling while keeping the calculation simple and fast.In order to improve the practicability of the Markov model in wireless channel modeling applications,an optimal selection method for wireless channel models based on the Markov process has been proposed.The optimal model selected is most suitable for the current scene and most sensitive to the environment change.It can also be used in wireless channel environment anomaly detection.Finally,the simulation results show that this method can establish a random and real-time wireless channel model based on the Markov process.The optimal model has the highest adaptability to the current scene and most sensitivity to the changes of the current scene environment.The accuracy of wireless channel modeling has been improved when the calculation keeps still simple and fast.The performance in theenvironment anomaly detection of the wireless channel is better.These results explain that the method proposed in this paper has certain value in wireless channel modeling and wireless channel environment anomaly detection.