Spatial Distribution of Base Stations in the Indian Scenario

Kosala, Prathyusha Devi and Kumar, Abhinav (2017) Spatial Distribution of Base Stations in the Indian Scenario. Masters thesis, Indian Institute of Technology Hyderabad.

[img] Text
EE15MTECH11001.pdf - Submitted Version
Restricted to Registered users only until 28 June 2020.

Download (2MB) | Request a copy

Abstract

Modeling the spatial distribution of base stations (BSs) is essential for evaluating differ- ent network performance metrics in cellular networks. In this work, we consider the spatial distribution of actual BSs data from Tier-1 cities of India for one of the leading network providers. To model this data, we consider the widely used homogeneous Poisson Point Process (PPP) as a benchmark. We consider the Thomas and Matern cluster point processes for modeling spatial distribu- tion of BSs. We have evaluate the performance of these point processes with homogeneous PPP. Various statistical measures of point process like Repley’s 퐾 -function, 퐿 -function, Nearest-neighbor hood function, and 퐽 -function are used as Goodness of Fit (GOF) tests for modeling spatial distribution of BSs. Point-wise envelopes of the statistical measures are computed for accepting or rejecting the hypothesis at 10% significance level. The spatial distribution of BSs in Tier-1 cities exhibit clustering property. However, this clustering is not characterized completely by either Thomas cluster process or Matern cluster process. Hence, we further analyze the spatial distribution of BSs using a combination of discrete and continuous distributions for modeling number of BSs and location of BSs, respectively. We evaluate the performance of the Poisson distribution with Discrete Exponential, Dis- crete Weibull, and Zipf-Mandelbrot distributions for modeling the distribution of number of BSs. Further, we model the spatial location of BSs using Uniform, Gaussian, and Laplace distributions. We compare the performance of these distributions using GOF tests. The nu- merical results indicate that the location of BSs can be accurately modeled as a Gaussian distribution, while, the number of BSs in a given area is best modeled as a Discreet Expo- nential distribution in the Indian scenario for Tier-1 cities.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Item Type: Thesis (Masters)
Uncontrolled Keywords: PPP, spatial distribution, TD837
Subjects: Electrical Engineering
Divisions: Department of Electrical Engineering
Depositing User: Team Library
Date Deposited: 29 Jun 2017 07:32
Last Modified: 04 Jun 2019 11:47
URI: http://raiith.iith.ac.in/id/eprint/3308
Publisher URL:
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 3308 Statistics for this ePrint Item