Cellular network traffic prediction method based on three-dimensional convolutional neural network

Abstract

The invention provides a cellular network traffic prediction method based on a three-dimensional convolutional neural network. The cellular network traffic prediction method comprises the following steps: modeling network traffic data into a three-dimensional tensor input form to obtain a three-dimensional network traffic data model; obtaining training set data and test set data according to the three-dimensional network flow data; constructing a basic three-dimensional convolutional neural network; performing three-dimensional convolutional neural network training on the short-time dependence data to obtain short-time features, and performing three-dimensional convolutional neural network training on the long-time dependence data to obtain long-time features; carrying out fusion training on the short-time features and the long-time features to obtain a feature matrix, and taking the feature matrix as the output of the basic three-dimensional convolutional neural network to form a training model; and predicting the network traffic data to be predicted by using the training model to obtain a network traffic prediction result. According to the prediction method provided by the invention, the short-term correlation and the long-term tendency of the network traffic data are considered at the same time, and the characteristic correlation of the network traffic data in time sequence is captured.

Description

[0001] 【Technical field】 [0002] The invention relates to the application field of computer timing prediction, in particular to a three-dimensional convolutional neural network-based cellular network traffic prediction method. [0003] 【Background technique】 [0004] In recent years, with the popularization of mobile devices and mobile applications, wireless network technology has played a key role in people's daily life around the world. More and more people use mobile devices to access cellular networks, cellular network traffic and network The demand for traffic is growing rapidly. The latest industry forecast shows that by 2021, the global cellular network traffic of mobile devices is expected to exceed 48.3 exabytes per person, seven times the current usage, and smartphone traffic will exceed PC traffic in the same year. For cellular network service providers and infrastructure providers, in order to cope with the increasing demand and provide users with stable cellular ...

AI Technical Summary

Problems solved by technology

However, the ARIMA method only tends to focus on the average value of historical sequence data, so it cannot capture the rapid variation process of the underlying traffic load, and cannot model the nonlinear relationship in the real system; while the SVR method can processing, but key parameters need to be tuned to obtain accurate prediction results

At the same time, considering the influence of factors such as cellular network user mobility, arrival patterns, and user demand diversity, most of these methods ignore the potential correlation between traffic sequences in cellular networks.

For example, the spatial dependence in the cellular network, the movement of users will obviously drive the transfer of traffic demand, resulting in significant spatial dependence of traffic between different base stations, and the basic traffic demand of each area will also be affected by the surrounding environment. impact, the traffic demand in prosperous areas is obviously greater than that in remote areas, and these dependencies cannot be captured by traditional methods

Images