RBF neural network channel prediction method based on phase space reconstruction

Abstract

The invention belongs to the technical field of channel prediction, and discloses a RBF (Radial basis function, radial basis function) neural network channel prediction method based on phase space reconstruction. The method comprises the following steps: obtaining a channel coefficient, and establishing a sample training data set; solving a phase space reconstruction parameter; performing phase space reconstruction on the sample training data set; selecting a neighbor domain node in a reconstruction phase space; training a RBF neural network by using the neighbor domain node; performing prediction by using the trained neural network; and converting a prediction result into an original space to obtain a prediction value. According to the RBF neural network channel prediction method provided by the invention, the prediction time is relatively long, and meanwhile the calculation complexity of the RBF neural network is reduced; and the RBF neural network channel prediction method can be applied to the channel prediction of an LTE uplink.

Description

technical field [0001] The invention belongs to the technical field of channel prediction, in particular to an RBF neural network channel prediction method based on phase space reconstruction. Background technique [0002] With the rapid growth of data communication services such as high-definition video and the Internet of Things, 4G with LTE as the mainstream and future 5G need to achieve reliable data transmission at a higher rate within a limited frequency band; and the harsh characteristics and complex nature of wireless channels The changeable interference inside and outside the system poses a serious challenge to the research on information transmission technology of high frequency band utilization and high power utilization in LTE system. In order to adapt to these technical challenges faced by current and future mobile communication systems, a lot of researches on link adaptation technology (AMC) based on channel information scheduling have been carried out in LTE s...

AI Technical Summary

Problems solved by technology

[0003] To sum up, the problems existing in the existing technology are: the current channel prediction method has a short channel prediction time and a large amount of calculation of the neural network.

Images