Measurement and stochastic modeling of handover delay and interruption time of smartphone real-time applications on LTE networks

Donghyuk Han, Sungjin Shin, Hyoungjun Cho, Jong Moon Chung, Dongseok Ok, Iksoon Hwang

Research output: Contribution to journalArticlepeer-review

40 Citations (Scopus)

Abstract

For continuous services of mobile user equipment (UE), Long Term Evolution (LTE) systems conduct evolved node B (eNB) switching based on hard handover technology, which breaks a connection before the connection to the target eNB (T-eNB) is made. As handover service interruption seriously degrades network performance, precise knowledge of the handover (HO) performance is necessary in finding out defects of the current system and discovering clues for improvements. Although the performance of LTE handover and its anticipated effect on network services are important evaluation indexes, in existing literature only the theoretical performance is analyzed and very few actual measurements on practical LTE networks have been presented. In this article, the HO delay and handover interruption time (HIT) performance of LTE networks are measured for several cases in accordance with the average number of users in a cell. Based on the internal HO procedures that influence HO delay and HIT, the key parameters are analyzed. In addition, based on the estimated number of users in a cell, a reference probability density function (PDF) that can be used for HIT prediction is presented.

Original languageEnglish
Article number7060501
Pages (from-to)173-181
Number of pages9
JournalIEEE Communications Magazine
Volume53
Issue number3
DOIs
Publication statusPublished - 2015 Mar 1

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Measurement and stochastic modeling of handover delay and interruption time of smartphone real-time applications on LTE networks'. Together they form a unique fingerprint.

Cite this