Base station traffic prediction method and device

Abstract

The invention provides a base station traffic prediction method and device. The method comprises the following steps: inputting an intra-node traffic characteristic and an inter-node traffic characteristic corresponding to a base station to be predicted to a traffic prediction model that is established based on a spatial dependence relationship of the base station in advance, and obtaining an output traffic prediction value of the base station to be predicted, wherein the intra-node traffic characteristic is the intra-base station traffic of the base station to be predicted and the base station adjacent to the base station to be predicted, and the inter-node traffic characteristic is the inter-node traffic characteristic between the base station to be predicted and the base station adjacent to the base station to be predicted. By decomposing the traffic of the base station into the intra-node traffic characteristic and the inter-node traffic characteristic according to the mobility characteristics of the user, and performing traffic prediction by using the traffic prediction model that is established based on a spatial dependence relationship of the base station, the influence of the movement of the user to the base station traffic is fully considered, so that the accurate traffic prediction is achieved.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to a base station flow prediction method and equipment. Background technique [0002] In modern society, the mobile Internet has profoundly changed people's production and lifestyle. According to Cisco's traffic forecast report, mobile network traffic increased by 74% in 2015, and will reach 30.6 EB per month (1 EB is approximately equal to 1018 bytes) in 2020, which is more than 8 times the current network traffic. Such a huge traffic growth means greater challenges to mobile communication operators. [0003] In some developed areas, the network capacity has been overwhelmed; and they are facing problems such as further narrowing of base station spacing, excessive frequency reuse, and rising network noise floor. Therefore, real-time forecasting of base station traffic can help the base station to adjust parameters in real time and detect traffic anomalies early, th...

AI Technical Summary

Problems solved by technology

[0004] However, in the current network environment, considering different user states (such as working, sleeping, taking transportation, etc.), different types of applications (such as video, social, games, etc.), user mobility, urban land use ( Influenced by factors such as commercial areas, residential areas, traffic areas, etc.), population density, working days and holidays, fine-grained traffic prediction for a single base station is particularly difficult and complicated

[0005] Although there are analysis and prediction methods about network traffic in the prior art, they mainly focus on the analysis and modeling of the statistical characteristics and probability distribution characteristics of network traffic; although these works can help us deeply understand the laws of network traffic, they cannot Applied directly to network traffic forecasting

[0006] For traffic forecasting models, existing technologies include forecasting models based on autoregressive integral moving average (ARIMA), forecasting models based on long-term short-term memory networks (LSTM), and forecasting models based on spatial autocorrelation, but the above schemes either ignore spatial Influenced by various factors, or using approximate estimation models to model traffic, these models cannot reflect the long-distance spatial impact caused by user movement, and thus cannot achieve high-precision base station traffic prediction

Images