A power distribution method and distribution device based on deep learning

Abstract

The embodiment of the invention provides a power distribution method based on deep learning and a distribution device. The method comprises the following steps: acquiring a channel matrix of a user; performing singular value decomposition processing on the channel matrix to obtain an equivalent channel characteristic value and a left unitary matrix; generating channel characteristic information ofthe user based on the equivalent channel characteristic value and a preset power limiting parameter of the user; inputting the channel feature information into a pre-trained full-connection neural network model to obtain the number of data streams of the user; constructing a power distribution covariance matrix of the user based on the data flow number of the user, the equivalent channel characteristic value and the left unitary matrix; and allocating transmission power to the user based on the power allocation covariance matrix of the user. According to the embodiment of the invention, the transmission power distribution efficiency can be improved.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, in particular to a power allocation method and allocation device based on deep learning. Background technique [0002] MIMO (Multiple-Input Multiple-Output) technology is a technology that can use multiple antennas at the transmitting end to send signals independently, and at the same time use multiple antennas at the receiving end to receive and restore the original information. With the continuous development of technology, MIMO technology has become one of the key technologies of wireless communication systems. [0003] MIMO technology can be applied to single-user MIMO channel scenarios or MIMO-MAC (MIMO-Multiple Access Channel, Multiple Access Channel) channel scenarios, where the single-user MIMO channel scenario is for a single user accessing the system, MIMO- The MAC channel scenario is for multiple users accessing the system. It can be considered that the single-u...

AI Technical Summary

Problems solved by technology

However, when using the water-filling algorithm to calculate the user's power allocation covariance matrix in the prior art, multiple iterative calculations are required. This multiple-iterative calculation method not only has a large amount of calculation, but also has high computational complexity, which is not conducive to quickly providing users with Allocate transmit power

Images