Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep Learning-driven Mobile Traffic Measurement Collection and Analysis

Yini Fang

TL;DR

This thesis designs Spider, a mobile traffic measurement collection and reconstruction framework and designs SDGNet, a handover-aware graph neural network model for long-term mobile traffic forecasting, which outperforms other benchmark graph models on a mobile traffic dataset collected by a major network operator.

Abstract

Modelling dynamic traffic patterns and especially the continuously changing dependencies between different base stations, which previous studies overlook, is challenging. Traditional algorithms struggle to process large volumes of data and to extract deep insights that help elucidate mobile traffic demands with fine granularity, as well as how these demands will evolve in the future. Therefore, in this thesis we harness the powerful hierarchical feature learning abilities of Deep Learning (DL) techniques in both spatial and temporal domains and develop solutions for precise city-scale mobile traffic analysis and forecasting. Firstly, we design Spider, a mobile traffic measurement collection and reconstruction framework with a view to reducing the cost of measurement collection and inferring traffic consumption with high accuracy, despite working with sparse information. In particular, we train a reinforcement learning agent to selectively sample subsets of target mobile coverage areas and tackle the large action space problem specific to this setting. We then introduce a lightweight neural network model to reconstruct the traffic consumption based on historical sparse measurements. Our proposed framework outperforms existing solutions on a real-world mobile traffic dataset. Secondly, we design SDGNet, a handover-aware graph neural network model for long-term mobile traffic forecasting. We model the cellular network as a graph, and leverage handover frequency to capture the dependencies between base stations across time. Handover information reflects user mobility such as daily commute, which helps in increasing the accuracy of the forecasts made. We proposed dynamic graph convolution to extract features from both traffic consumption and handover data, showing that our model outperforms other benchmark graph models on a mobile traffic dataset collected by a major network operator.

Deep Learning-driven Mobile Traffic Measurement Collection and Analysis

TL;DR

This thesis designs Spider, a mobile traffic measurement collection and reconstruction framework and designs SDGNet, a handover-aware graph neural network model for long-term mobile traffic forecasting, which outperforms other benchmark graph models on a mobile traffic dataset collected by a major network operator.

Abstract

Modelling dynamic traffic patterns and especially the continuously changing dependencies between different base stations, which previous studies overlook, is challenging. Traditional algorithms struggle to process large volumes of data and to extract deep insights that help elucidate mobile traffic demands with fine granularity, as well as how these demands will evolve in the future. Therefore, in this thesis we harness the powerful hierarchical feature learning abilities of Deep Learning (DL) techniques in both spatial and temporal domains and develop solutions for precise city-scale mobile traffic analysis and forecasting. Firstly, we design Spider, a mobile traffic measurement collection and reconstruction framework with a view to reducing the cost of measurement collection and inferring traffic consumption with high accuracy, despite working with sparse information. In particular, we train a reinforcement learning agent to selectively sample subsets of target mobile coverage areas and tackle the large action space problem specific to this setting. We then introduce a lightweight neural network model to reconstruct the traffic consumption based on historical sparse measurements. Our proposed framework outperforms existing solutions on a real-world mobile traffic dataset. Secondly, we design SDGNet, a handover-aware graph neural network model for long-term mobile traffic forecasting. We model the cellular network as a graph, and leverage handover frequency to capture the dependencies between base stations across time. Handover information reflects user mobility such as daily commute, which helps in increasing the accuracy of the forecasts made. We proposed dynamic graph convolution to extract features from both traffic consumption and handover data, showing that our model outperforms other benchmark graph models on a mobile traffic dataset collected by a major network operator.

Paper Structure

This paper contains 50 sections, 24 equations, 32 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Research Challenges
  3. Thesis Contributions
  4. Deep Learning-driven Sparse Mobile Traffic Measurement Collection and Reconstruction
  5. A Handover-aware Spatiotemporal Graph Neural Network for Mobile Traffic Forecasting
  6. Thesis Organization
  7. Background & Related Work
  8. Mobile Traffic Sampling and Reconstruction
  9. Sampling
  10. Data reconstruction
  11. Reconstruction Quality Estimation
  12. Reinforcement Learning and Its Applications
  13. Reinforcement Learning
  14. Sparse Mobile Crowdsensing
  15. Graph Neural Networks and Its Applications
  16. ...and 35 more sections

Figures (32)

  • Figure 1: Normal distribution of bootstrap resamples
  • Figure 2: How an agent interacts with the environment in RL.
  • Figure 3: An example of SMCS rl_mcs2
  • Figure 4: Modelling the cellular network (on the right) as a graph (on the left). Intuitively, graphs can represent accurately the spatial correlations in the cellular network. The weight (i.e., dependency or relationship) between two nodes (i.e., base stations) is quantified.
  • Figure 5: Spatio-Temporal Graph Convolutional Networks stgc
  • ...and 27 more figures