Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Smooth Handovers via Smoothed Online Learning

Michail Kalntis, Andra Lutu, Jesús Omaña Iglesias, Fernando A. Kuipers, George Iosifidis

TL;DR

This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning, and proposes a realistic system model for smooth and accurate HOs that extends existing approaches by incorporating device and cell features on HO optimization, and eliminating strong assumptions about requiring future signal measurements and knowledge of end-user mobility.

Abstract

With users demanding seamless connectivity, handovers (HOs) have become a fundamental element of cellular networks. However, optimizing HOs is a challenging problem, further exacerbated by the growing complexity of mobile networks. This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning (SOL). We first analyze an extensive dataset from a commercial mobile network operator (MNO) in Europe with more than 40M users, to understand and reveal important features and performance impacts on HOs. Our findings highlight a correlation between HO failures/delays, and the characteristics of radio cells and end-user devices, showcasing the impact of heterogeneity in mobile networks nowadays. We subsequently model UE-cell associations as dynamic decisions and propose a realistic system model for smooth and accurate HOs that extends existing approaches by (i) incorporating device and cell features on HO optimization, and (ii) eliminating (prior) strong assumptions about requiring future signal measurements and knowledge of end-user mobility. Our algorithm, aligned with the O-RAN paradigm, provides robust dynamic regret guarantees, even in challenging environments, and shows superior performance in multiple scenarios with real-world and synthetic data.

Smooth Handovers via Smoothed Online Learning

TL;DR

This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning, and proposes a realistic system model for smooth and accurate HOs that extends existing approaches by incorporating device and cell features on HO optimization, and eliminating strong assumptions about requiring future signal measurements and knowledge of end-user mobility.

Abstract

With users demanding seamless connectivity, handovers (HOs) have become a fundamental element of cellular networks. However, optimizing HOs is a challenging problem, further exacerbated by the growing complexity of mobile networks. This paper presents the first countrywide study of HO optimization, through the prism of Smoothed Online Learning (SOL). We first analyze an extensive dataset from a commercial mobile network operator (MNO) in Europe with more than 40M users, to understand and reveal important features and performance impacts on HOs. Our findings highlight a correlation between HO failures/delays, and the characteristics of radio cells and end-user devices, showcasing the impact of heterogeneity in mobile networks nowadays. We subsequently model UE-cell associations as dynamic decisions and propose a realistic system model for smooth and accurate HOs that extends existing approaches by (i) incorporating device and cell features on HO optimization, and (ii) eliminating (prior) strong assumptions about requiring future signal measurements and knowledge of end-user mobility. Our algorithm, aligned with the O-RAN paradigm, provides robust dynamic regret guarantees, even in challenging environments, and shows superior performance in multiple scenarios with real-world and synthetic data.
Paper Structure (8 sections, 32 equations, 9 figures, 1 table, 1 algorithm)

This paper contains 8 sections, 32 equations, 9 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background & Related Work
  3. Data collection & analysis
  4. System Model & Problem Statement
  5. Algorithm Design and Analysis
  6. Extensions
  7. Performance Evaluation
  8. Conclusions

Figures (9)

  • Figure 1: Median normalized DL packet loss (left, green y-axis) and normalized average UE throughput (right, orange y-axis) vs binned daily HOFs. Triangles and bold horizontal lines show the mean and median, respectively, in each boxplot.
  • Figure 2: Left: number (norm. by max) of different UE models and their RAT capabilities (up to 2G, 3G, 4G, 5G). Right: HOs (norm. by max) per day each of the UE model executes, and to what RAT.
  • Figure 3: Histogram (bars) and distribution (line) of the HO delays for each HO type (all same-colored bars sum to 1).
  • Figure 4: Histogram (bars) and distribution (line) of the HO delays for each UE model (all same-colored bars sum to 1).
  • Figure 5: Learning mechanism.
  • ...and 4 more figures