Research And Implementation Of DHCP Configuration Algorithm Based On User Mobility Prediction

DHCP is widely deployed in wireless LANs to automatically assign IP addresses and the lease time.In the case of a shortage of public IP addresses,the complex user mobility poses a great challenge to the management and assignment of IP addresses.In the situation where IP addresses are scarce,the complex user mobility behavior poses significant challenges for IP address management and allocation.Recently proposed IP configuration strategies(e.g.,lease time and IP address pool)rely on traditional models to study user mobility patterns.However,individual user behavior varies greatly,their online time distributions fluctuate significantly,and the number of crowds differs spatially and temporally.This leads to poor performance of DHCP server in these studies.We propose PredHCP,a predictive configuration framework to improve DHCP performance.PredHCP utilizes an attention based recurrent neural network(ARNN)to learn sequential patterns of individual mobility and accurately predicts user online time to ensure effective IP lease time configuration.Meanwhile,PredHCP introduces a spatiotemporal graph neural network(STGNN)to learn both spatial and temporal dependencies of crowd migration and accurately predicts crowd number in each area to ensure effective IP pool configuration.We conduct comprehensive experiments on real network traces for a month to evaluate the performance of PredHCP.Results on prediction accuracy show that PredHCP effectively models user mobility patterns by achieving lower RMSE.By accurately predicting mobility patterns,PredHCP makes effective IP configuration to ensure high DHCP performance.Experiment results show that PredHCP outperforms three state-of-the-art baselines that PredHCP can save up to 65%of the total IP addresses and improve the IP efficiency by 41%.