Advanced

Tobit Maximum-likelihood estimation of Censored Pathloss Data

Gustafson, Carl LU ; Abbas, Taimoor; Bolin, David and Tufvesson, Fredrik LU (2015) In Technical report, Dept. of Electrical and Information Technology, Lund University, Sweden
Abstract
Pathloss is typically modeled using a log-distance power law with a large-scale fading term that is log-normal. However, the received signal is affected by the dynamic range and noise floor of the measurement system used to sound the channel, which can cause measurement samples to be truncated or censored. If the information about the censored samples are not included in the estimation method, as in ordinary least squares estimation, it can result in biased estimation of both the pathloss exponent and the large scale fading. This is solved by applying a Tobit maximum-likelihood estimator, which provides consistent estimates for the pathloss parameters. This technical report provides information about the Tobit maximum-likelihood estimator... (More)
Pathloss is typically modeled using a log-distance power law with a large-scale fading term that is log-normal. However, the received signal is affected by the dynamic range and noise floor of the measurement system used to sound the channel, which can cause measurement samples to be truncated or censored. If the information about the censored samples are not included in the estimation method, as in ordinary least squares estimation, it can result in biased estimation of both the pathloss exponent and the large scale fading. This is solved by applying a Tobit maximum-likelihood estimator, which provides consistent estimates for the pathloss parameters. This technical report provides information about the Tobit maximum-likelihood estimator estimator and its asymptotic variance under certain conditions. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Book/Report
publication status
published
subject
keywords
maximum-likelihood estimation, Pathloss, ordinary least squares, truncated data, censored data, vehicular communication.
in
Technical report, Dept. of Electrical and Information Technology, Lund University, Sweden
pages
8 pages
publisher
[Publisher information missing]
ISSN
1402-8840
language
English
LU publication?
yes
id
13a7706b-33da-4b15-b087-7ed98ee53a0a (old id 7456326)
date added to LUP
2015-07-20 10:10:13
date last changed
2016-04-16 06:29:50
@misc{13a7706b-33da-4b15-b087-7ed98ee53a0a,
  abstract     = {Pathloss is typically modeled using a log-distance power law with a large-scale fading term that is log-normal. However, the received signal is affected by the dynamic range and noise floor of the measurement system used to sound the channel, which can cause measurement samples to be truncated or censored. If the information about the censored samples are not included in the estimation method, as in ordinary least squares estimation, it can result in biased estimation of both the pathloss exponent and the large scale fading. This is solved by applying a Tobit maximum-likelihood estimator, which provides consistent estimates for the pathloss parameters. This technical report provides information about the Tobit maximum-likelihood estimator estimator and its asymptotic variance under certain conditions.},
  author       = {Gustafson, Carl and Abbas, Taimoor and Bolin, David and Tufvesson, Fredrik},
  issn         = {1402-8840},
  keyword      = {maximum-likelihood estimation,Pathloss,ordinary least squares,truncated data,censored data,vehicular communication.},
  language     = {eng},
  pages        = {8},
  publisher    = {ARRAY(0x6967e70)},
  series       = {Technical report, Dept. of Electrical and Information Technology, Lund University, Sweden},
  title        = {Tobit Maximum-likelihood estimation of Censored Pathloss Data},
  year         = {2015},
}