Group Lasso-based neural network cutting method for power amplifier

Abstract

The invention discloses a Group Lasso-based neural network cutting method for a power amplifier, and belongs to the technical field of communication. The method comprises the following steps: for an original neural network structure, dividing all output weights connected to the same neuron in other layers except an output layer into one group; carrying out l2 norm constraint on each group of weights after grouping; taking the sum of L2 norms of the weights of all groups in the original neural network structure as a Group Lasso penalty term, and adding the Group Lasso penalty term to the original loss function Loss1 to obtain a new loss function Loss2; carrying out minimization training on the Loss2 through a BP algorithm; searching for weight groups with convergence close to 0, removing neurons connected with the weight groups to obtain a simplified neural network, training an original loss function Loss1 by adopting a BP algorithm to obtain a trained simplified neural network model, and modeling or pre-distorting the power amplifier by utilizing the model. Overfitting in the training process can be inhibited, the pre-distortion effect is improved, the calculated amount is reduced,and engineering application is facilitated.

Description

technical field [0001] The invention belongs to the technical field of communication, in particular to a Group Lasso-based neural network cutting method for power amplifiers. Background technique [0002] With the continuous development of communication technology, in order to use limited spectrum resources more efficiently, new modulation techniques with high spectrum efficiency are used in communication standards. However, these new modulation methods make the peak-to-average ratio of the signal higher and the envelope fluctuation larger, causing the signal to be seriously distorted after passing through the RF power amplifier; not only the EVM before and after the signal output is seriously deteriorated, but also serious out-of-band distortion . [0003] In order to correct the distortion produced by the signal passing through the power amplifier, the power amplifier needs to be linearized. Among the current linearization technologies, the predistortion technology is wi...

AI Technical Summary

Problems solved by technology

[0008] In order to overcome the disadvantages of the neural network in the prior art, such as difficulty in manual debugging, easy overfitting, weak robustness, difficulty in engineering implementation, and large hardware resource overhead, the present invention provides a neural network based on Group Lasso for power amplifiers The clipping method can clip or directly train the neural network structure of the real multi-layer neural network, thereby suppressing overfitting and easy engineering implementation

Images