精准的网络流量预测是实现网络精细化和智能化管理的关键,也是网络运营商、云服务提供商等实现网络智能运维及应用服务保障的重要支撑,属于当前业界研究的热点.网络流量预测问题一般可被视为一种时间序列预测问题,现有时间序列预测模型虽然能起到一定作用,但这些通用模型很少考虑流量数据集本身特点,从而无法在网络流量预测性能上取得突破.为此,本文重点研究了网络流量数据集中的自然周期特征,提出了一种能有效利时间序列周期性特点的网络流量预测通用模型——Cycle Little Linear Head(CycleLLH).该模型主干为Trans-former的编码器,其中两个关键设计在于:(1)周期整合.将流量序列按照一个特定周期划分步长划分为不同的周期块,然后将这些周期块对应相位的时间节点分别嵌入到不同输入令牌;(2)小线性层.由多个多层感知机组成,并且多层感知机单独作用于每个相位的时间特征.周期整合使得模型具有两个优点:更有利于模型提取数据集在一个周期内的特征;注意力矩阵的计算和内存复杂度可以看作是和周期划分步长二次方有关的常数,使得模型可以使用更大的回溯窗口而仅增加少量计算资源.通过在公共流量数据集上进行大量实验,本研究表明:与当前最先进的模型相比,CycleLLH在流量预测精度方面表现出显著优势,在六个数据集上的预测精度分别提升了12.3%、8.4%、29.9%、5.8%、8.3%和2.0%.代码可从https://github.com/wenjietang218/CycleLLH.git中获取.
CycleLLH:A New Network Traffic Prediction Model based on Cycle Integration
Accurate network traffic prediction is not only a key link to realize network refinement and intelligent management,but also an important support for network operators and cloud service providers in intelligent operation and maintenance and application service guarantee.This field has become a research hotspot in the current industry,and has a wide range of application prospects.Network traffic prediction technology plays an important role in many fields,including base station power management,base station overload prevention,unmanned aerial vehicle temporary base stations,new network deployment and base station construction,5G network slicing,Software Defined Network resource scheduling,and resource management in multi-access edge computing.Network traffic prediction problem is usually regarded as a time series prediction problem.With its self-attention mechanism,Transformer can efficiently compute the dependencies between each time node in parallel,and performs well in the field of time series prediction.However,although the existing time series prediction models,including the Transformer-based models,can play a certain role,they often ignore the unique periodic characteristics of traffic data sets,so it is difficult to make a breakthrough in network traffic prediction performance.Therefore,this paper focuses on the natural cycle of network traffic data sets,and proposes a general network traffic prediction model,Cycle Little Linear Head(CycleLLH),which can effectively exploit the periodicity of time series.The backbone of the model is the encoder of the Transformer,and the two key designs are as follows:1.Cycle Integration:we divide the traffic sequence into different cycle blocks according to a specific period,and then embed the time nodes corresponding to the phase of these cycle blocks into different input tokens;2.Little Linear Head:it is composed of multiple multi-layer perceptrons,and each multi-layer perceptron operates separately on each feature node.Cycle Integration makes the model have two advantages:it is more conducive to the model to extract the features of the dataset within one cycle;the computation and memory complexity of the attention matrix can be viewed as a constant related to the quadratic power of the period step of the dataset,allowing the model to use a larger look-back window with only a small increase in computational resources.The little linear layer preserves the spatial structure of the encoder output and significantly reduces the number of parameters in the linear layer,which helps the model to predict the detailed changes in the network traffic data.Through a large number of experiments on public traffic data sets,this study shows that CycleLLH shows significant advantages in traffic prediction accuracy compared with the current state-of-the-art models,and the prediction accuracy on the six data sets is improved by 12.3%,8.4%,29.9%,5.8%,8.3%and 2.0%,respectively.In addition,through comparative ablation experiments,this study verifies the effectiveness of Cycle Integration,Little Linear Head and normalization methods in improving the prediction performance of the model.Experimental results show that CycleLLH can achieve better prediction performance under longer backtracking window,and shows strong robustness and adaptability under different cycle division step sizes.In addition,CycleLLH also shows high stability in noisy data environment.Code can be obtained from https://github.com/wenjietang218/CycleLLH.git.
network traffic predictiontime series predictionperiodicitycycle integrationlittle linear head
万方数据
维普
