Robust signal recognition algorithm based on machine learning in heterogeneous networks

There are various heterogeneous networks for terminals to deliver a better quality of service. Signal system recognition and classification contribute a lot to the process. However, in low signal to noise ratio（SNR） circumstances or under time-varying multipath channels, the majority of the existing algorithms for signal recognition are already facing limitations. In this series, we present a robust signal recognition method based upon the original and latest updated version of the extreme learning machine（ELM） to help users to switch between networks. The ELM utilizes signal characteristics to distinguish systems. The superiority of this algorithm lies in the random choices of hidden nodes and in the fact that it determines the output weights analytically, which result in lower complexity. Theoretically, the algorithm tends to offer a good generalization performance at an extremely fast speed of learning. Moreover, we implement the GSM/WCDMA/LTE models in the Matlab environment by using the Simulink tools. The simulations reveal that the signals can be recognized successfully to achieve a 95% accuracy in a low SNR（0 dB） environment in the time-varying multipath Rayleigh fading channel.

Research on the slow orbit feedback of BEPCⅡusing machine learning

Background:Long-term orbit stability is a key performance indicator in synchrotron radiation facilities and colliders nowa-days,in which the orbit correction and corresponding slow orbit feedback system are indispensable.Conventional method of orbit correction uses response matrix based on SVD algorithm,which becomes less effective after a long operation due to the fact that response matrix measurements cannot be taken during normal operation.Purpose:The purpose of this paper is to integrate machine learning model into the slow orbit feedback process and to automatically update the model online to better correct the orbit shifts.Methods:In this paper,we propose a method for slow orbit feedback of storage ring based on machine learning.Training the neural networks by using online data sets,which can establish the mapping relation between BPMs and correctors,and being updated automatically,without using extra time to remeasure the response matrix.Results:The experiments in this paper are all conducted and verified in the upgrading project of Beijing Electron-Positron Collider.By the way of learning automatically,the updated neutral network is closer to the real machine model,and the orbit after correction shows a smaller fluctuation relative to the golden orbit.Conclusion:Using the online data sets which reflect the response of orbit to correctors in real time to update the neural network can increase the orbit stability.