An Adaptive Link MCS Switching Control Method Based on Online Learning

Abstract

The invention belongs to the technical field of mobile communication, and discloses an adaptive link MCS switching control method for online learning, which selects parameters; the sending end cyclically sends different MCS schemes, and the receiving end processes and stores related data; stores data according to different MCS schemes Classify and screen valid data; calculate the relevant parameters of each MCS according to the valid data; the receiving end selects and feeds back the corresponding MCS according to the update criteria according to the current channel state, the sending end sends data according to the feedback MCS, and the receiving end processes and stores the data; After multiple updates, the final corresponding parameters of each MCS are obtained, and the receiving end feeds back the MCS according to the corresponding criteria; respectively, according to the switching criteria, EESM switching criteria, and KNN algorithm switching criteria, the system throughput of each method under different signal-to-noise ratios is counted. The present invention does not need to rely on the channel model in a specific environment, and does not need to obtain accurate classification data of each MCS in advance.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and in particular relates to an adaptive link MCS switching control method for online learning. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: With the development and evolution of mobile communication technology, the communication becomes mobile, broadband and IP-based. But in the wireless mobile communication system, the time-varying characteristics brought about by the path loss, fading and interference of the wireless channel seriously affect the system performance. The time-varying nature of the wireless channel makes the quality of the received signal a random variable. In this way, when the transmission mode and transmission parameters are fixed, the transmission performance of the system will also change with the quality of the received signal, which cannot meet the needs of communication services. The harsh...

AI Technical Summary

Problems solved by technology

In actual situations, these accurately classified data are often not known, and the computational complexity of such algorithms is very high.

[0006] The difficulty and significance of solving the above technical problems: For commonly used lookup table algorithms, accurate channel models are difficult to obtain, and some empirical models are often used, but the accuracy of empirical models is not high. Similarly, current machine learning methods must obtain accurate classification samples, but accurate classification samples are also difficult to obtain, and there is no good solution for how to obtain accurate channel models or accurate classification samples

Images