Performance of D2D underlay and overlay for multi-class elastic traffic

Prajwal Osti, Pasi Lassila*, Samuli Aalto

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

2 Citations (Scopus)

Abstract

We explore the performance of different resource allocation schemes for transferring elastic traffic in a cellular network that is either overlaid or underlaid with D2D traffic. To this end, we model a single cell during uplink transmissions and jointly consider the presence of a randomly varying number of D2D and cellular users in the system. Various queueing models are developed to study the performance of different resource sharing schemes for D2Doverlay and underlay schemes assuming a so-called Boolean interference model between cellular and D2Dtraffic. Multi-class models have been used in particular to differentiate near and far cellular users with different achievable transmission rates. Our numerical studies reveal that the performance of the different schemes is sensitive to the load levels in the cell, as well as the interference range of cellular users in the underlay scheme. Overall, the underlay scheme appears to provide the best performance, even for quite high interference range values. Simulations with a more realistic SINR-based interference model suggest also similar gains for the underlay scheme. On the other hand, in the overlay schemes where resources are reserved separately for cellularand D2Dusers, the so-called dynamic overlay method we propose appears to perform better than the static overlay scheme.

Original languageEnglish
Pages (from-to)147-163
Number of pages17
JournalComputer Communications
Volume117
Early online date9 Aug 2017
DOIs
Publication statusPublished - 2018
MoE publication typeA1 Journal article-refereed

Keywords

  • D2D Communications
  • Flow-level models
  • M/G/1-PS
  • Markov processes

Fingerprint

Dive into the research topics of 'Performance of D2D underlay and overlay for multi-class elastic traffic'. Together they form a unique fingerprint.

Cite this