Terminal equipment transmission power control method based on integrated neural network

Abstract

The invention discloses a terminal equipment transmission power control method based on an integrated neural network. The method comprises the following steps: collecting a channel power gain sample of a D2D link; inputting the optimal power distribution strategy into an SPCA algorithm to obtain an optimal power distribution strategy under a corresponding sample; constructing a deep neural networkand a convolutional neural network, and initializing the weight of the neural network; inputting the training data set into a neural network, constructing an MSE between the output of the neural network and the label as a loss function, and updating the weight of the neural network; when the loss function is smaller than a preset value or reaches the number of iterations, it is considered that neural network training is completed, and the neural network is stored; and constructing a selector, and selecting and outputting an allocation strategy with higher performance. According to the method,the defects that a deep neural network is poor in large-scale network learning capability and a convolutional neural network is limited in local feature extraction of a small-scale network are overcome, and the two networks are integrated by using the idea of integrated learning, so that the method can meet the real-time power distribution requirements of networks of different scales.

Description

technical field [0001] The invention belongs to the physical layer technology of the communication system, in particular to the resource allocation technology of the wireless communication system, in particular to a method for controlling the transmission power of terminal equipment based on an integrated neural network. Background technique [0002] The explosion in the number of IoT devices has resulted in fierce competition for bandwidth. If a satisfactory data delivery rate is to be achieved, the traditional solution is to use multiple users to share multiple subcarriers in a non-orthogonal manner, but this will also lead to mutual interference between multiple users, and each other becomes the noise source of the other . The result of this is that the data delivery rate of a single user in a subcarrier decreases as the transmission power of other users in the subcarrier increases. How to properly allocate power to balance interference and data rate has become a new re...

AI Technical Summary

Problems solved by technology

[0003] In addition, a typical power allocation algorithm is an iterative-water-filling algorithm, which can find a suboptimal solution that satisfies all users in the system and maximizes the performance of the rate. The disadvantage of this algorithm is poor convergence performance, including low convergence, and slow convergence It will lead to relatively high computational complexity. When facing a large-scale network, especially a communication network with a large number of terminal devices, it is difficult to guarantee the convergence of the iterative algorithm, which also limits its application field, that is, it can only be used in small used in large-scale networks

Images