Teletraffic Analysis of Direct Communication with Clustering

Janne LEHTOMAKI; Isameldin SULIMAN; Kenta UMEBAYASHI; Yasuo SUZUKI

doi:10.1587/transfun.E92.A.1356

Teletraffic Analysis of Direct Communication with Clustering

Janne LEHTOMAKI, Isameldin SULIMAN, Kenta UMEBAYASHI, Yasuo SUZUKI

Full Text Views

0

Cite this

Summary :

In direct communication, terminals that are close to each other can communicate directly without traffic going through centralized controller such as a base station (BS). This brings several advantages. We study direct communication with localized distribution, so that users tend to gather around some areas (clusters/hot-spots) within the cell such as buildings. Previous analysis about clustering has focused on one dimensional scenarios. Here we present theoretical analysis of direct communication with two dimensional clustering. Additional analysis is presented for direct communication with correlated clusters. With correlated clusters some pairs of source and destination clusters are more probable than other pairs. According to our best knowledge, this is the first time that theoretical analysis is presented about clustering and correlated clusters in two dimensional scenarios. Simulations confirm the validity of the analysis. In addition to the exact results, we also suggest using the point-based approximation to rapidly and easily obtain results. The numerical results show that the gains from direct communication, in terms of blocking probability and carried traffic, depend on the offered traffic. Additionally, correlation in cluster selection is shown to significantly improve performance. Point-based approximation is shown to be very useful when the number of clusters is large.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E92-A No.5 pp.1356-1362

Publication Date: 2009/05/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E92.A.1356

Type of Manuscript: PAPER

Category: Communication Theory and Signals

Cite this

Copy

Janne LEHTOMAKI, Isameldin SULIMAN, Kenta UMEBAYASHI, Yasuo SUZUKI, "Teletraffic Analysis of Direct Communication with Clustering" in IEICE TRANSACTIONS on Fundamentals, vol. E92-A, no. 5, pp. 1356-1362, May 2009, doi: 10.1587/transfun.E92.A.1356.
Abstract: In direct communication, terminals that are close to each other can communicate directly without traffic going through centralized controller such as a base station (BS). This brings several advantages. We study direct communication with localized distribution, so that users tend to gather around some areas (clusters/hot-spots) within the cell such as buildings. Previous analysis about clustering has focused on one dimensional scenarios. Here we present theoretical analysis of direct communication with two dimensional clustering. Additional analysis is presented for direct communication with correlated clusters. With correlated clusters some pairs of source and destination clusters are more probable than other pairs. According to our best knowledge, this is the first time that theoretical analysis is presented about clustering and correlated clusters in two dimensional scenarios. Simulations confirm the validity of the analysis. In addition to the exact results, we also suggest using the point-based approximation to rapidly and easily obtain results. The numerical results show that the gains from direct communication, in terms of blocking probability and carried traffic, depend on the offered traffic. Additionally, correlation in cluster selection is shown to significantly improve performance. Point-based approximation is shown to be very useful when the number of clusters is large.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E92.A.1356/_p

Copy

@ARTICLE{e92-a_5_1356,
author={Janne LEHTOMAKI, Isameldin SULIMAN, Kenta UMEBAYASHI, Yasuo SUZUKI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Teletraffic Analysis of Direct Communication with Clustering},
year={2009},
volume={E92-A},
number={5},
pages={1356-1362},
abstract={In direct communication, terminals that are close to each other can communicate directly without traffic going through centralized controller such as a base station (BS). This brings several advantages. We study direct communication with localized distribution, so that users tend to gather around some areas (clusters/hot-spots) within the cell such as buildings. Previous analysis about clustering has focused on one dimensional scenarios. Here we present theoretical analysis of direct communication with two dimensional clustering. Additional analysis is presented for direct communication with correlated clusters. With correlated clusters some pairs of source and destination clusters are more probable than other pairs. According to our best knowledge, this is the first time that theoretical analysis is presented about clustering and correlated clusters in two dimensional scenarios. Simulations confirm the validity of the analysis. In addition to the exact results, we also suggest using the point-based approximation to rapidly and easily obtain results. The numerical results show that the gains from direct communication, in terms of blocking probability and carried traffic, depend on the offered traffic. Additionally, correlation in cluster selection is shown to significantly improve performance. Point-based approximation is shown to be very useful when the number of clusters is large.},
keywords={},
doi={10.1587/transfun.E92.A.1356},
ISSN={1745-1337},
month={May},}

Copy

TY - JOUR
TI - Teletraffic Analysis of Direct Communication with Clustering
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1356
EP - 1362
AU - Janne LEHTOMAKI
AU - Isameldin SULIMAN
AU - Kenta UMEBAYASHI
AU - Yasuo SUZUKI
PY - 2009
DO - 10.1587/transfun.E92.A.1356
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E92-A
IS - 5
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - May 2009
AB - In direct communication, terminals that are close to each other can communicate directly without traffic going through centralized controller such as a base station (BS). This brings several advantages. We study direct communication with localized distribution, so that users tend to gather around some areas (clusters/hot-spots) within the cell such as buildings. Previous analysis about clustering has focused on one dimensional scenarios. Here we present theoretical analysis of direct communication with two dimensional clustering. Additional analysis is presented for direct communication with correlated clusters. With correlated clusters some pairs of source and destination clusters are more probable than other pairs. According to our best knowledge, this is the first time that theoretical analysis is presented about clustering and correlated clusters in two dimensional scenarios. Simulations confirm the validity of the analysis. In addition to the exact results, we also suggest using the point-based approximation to rapidly and easily obtain results. The numerical results show that the gains from direct communication, in terms of blocking probability and carried traffic, depend on the offered traffic. Additionally, correlation in cluster selection is shown to significantly improve performance. Point-based approximation is shown to be very useful when the number of clusters is large.
ER -