Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications

Shu Sun, Theodore Rappaport, Timothy A. Thomas, Amitava Ghosh, Huan C. Nguyen, Istvan Z. Kovacs, Ignacio Rodriguez, Ozge Koymen, Andrzej Partyka

Research output: Contribution to journalArticle

Abstract

This paper compares three candidate large-scale propagation path loss models for use over the entire microwave and millimeter-wave (mmWave) radio spectrum: The alpha-beta-gamma (ABG) model, the close-in (CI) free-space reference distance model, and the CI model with a frequency-weighted path loss exponent (CIF). Each of these models has been recently studied for use in standards bodies such as 3rd Generation Partnership Project (3GPP) and for use in the design of fifth-generation wireless systems in urban macrocell, urban microcell, and indoor office and shopping mall scenarios. Here, we compare the accuracy and sensitivity of these models using measured data from 30 propagation measurement data sets from 2 to 73 GHz over distances ranging from 4 to 1238 m. A series of sensitivity analyses of the three models shows that the four-parameter ABG model underpredicts path loss when relatively close to the transmitter, and overpredicts path loss far from the transmitter, and that the physically based two-parameter CI model and three-parameter CIF model offer computational simplicity, have very similar goodness of fit (i.e., the shadow fading standard deviation), exhibit more stable model parameter behavior across frequencies and distances, and yield smaller prediction error in sensitivity tests across distances and frequencies, when compared to the four-parameter ABG model. Results show the CI model with a 1-m reference distance is suitable for outdoor environments, while the CIF model is more appropriate for indoor modeling. The CI and CIF models are easily implemented in existing 3GPP models by making a very subtle modification-by replacing a floating non-physically based constant with a frequency-dependent constant that represents free-space path loss in the first meter of propagation. This paper shows this subtle change does not change the mathematical form of existing ITU/3GPP models and offers much easier analysis, intuitive appeal, better model parameter stability, and better accuracy in sensitivity tests over a vast range of microwave and mmWave frequencies, scenarios, and distances, while using a simpler model with fewer parameters.

Original languageEnglish (US)
Article number7434656
Pages (from-to)2843-2860
Number of pages18
JournalIEEE Transactions on Vehicular Technology
Volume65
Issue number5
DOIs
StatePublished - May 1 2016

Fingerprint

Path Loss
Wireless Communication
Propagation
Prediction
Communication
Model
Millimeter Wave
Free Space
Transmitter
Microwave
Millimeter waves
Transmitters
Microwaves
Scenarios
Stable Models
Appeal
Shopping centers
Prediction Error

Keywords

  • 5G
  • Millimeter wave
  • path loss models
  • prediction accuracy

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Aerospace Engineering
  • Automotive Engineering
  • Computer Networks and Communications
  • Applied Mathematics

Cite this

Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications. / Sun, Shu; Rappaport, Theodore; Thomas, Timothy A.; Ghosh, Amitava; Nguyen, Huan C.; Kovacs, Istvan Z.; Rodriguez, Ignacio; Koymen, Ozge; Partyka, Andrzej.

In: IEEE Transactions on Vehicular Technology, Vol. 65, No. 5, 7434656, 01.05.2016, p. 2843-2860.

Research output: Contribution to journalArticle

Sun, Shu ; Rappaport, Theodore ; Thomas, Timothy A. ; Ghosh, Amitava ; Nguyen, Huan C. ; Kovacs, Istvan Z. ; Rodriguez, Ignacio ; Koymen, Ozge ; Partyka, Andrzej. / Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications. In: IEEE Transactions on Vehicular Technology. 2016 ; Vol. 65, No. 5. pp. 2843-2860.
@article{2e85bfac0c12413cb1f035a095098c6b,
title = "Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications",
abstract = "This paper compares three candidate large-scale propagation path loss models for use over the entire microwave and millimeter-wave (mmWave) radio spectrum: The alpha-beta-gamma (ABG) model, the close-in (CI) free-space reference distance model, and the CI model with a frequency-weighted path loss exponent (CIF). Each of these models has been recently studied for use in standards bodies such as 3rd Generation Partnership Project (3GPP) and for use in the design of fifth-generation wireless systems in urban macrocell, urban microcell, and indoor office and shopping mall scenarios. Here, we compare the accuracy and sensitivity of these models using measured data from 30 propagation measurement data sets from 2 to 73 GHz over distances ranging from 4 to 1238 m. A series of sensitivity analyses of the three models shows that the four-parameter ABG model underpredicts path loss when relatively close to the transmitter, and overpredicts path loss far from the transmitter, and that the physically based two-parameter CI model and three-parameter CIF model offer computational simplicity, have very similar goodness of fit (i.e., the shadow fading standard deviation), exhibit more stable model parameter behavior across frequencies and distances, and yield smaller prediction error in sensitivity tests across distances and frequencies, when compared to the four-parameter ABG model. Results show the CI model with a 1-m reference distance is suitable for outdoor environments, while the CIF model is more appropriate for indoor modeling. The CI and CIF models are easily implemented in existing 3GPP models by making a very subtle modification-by replacing a floating non-physically based constant with a frequency-dependent constant that represents free-space path loss in the first meter of propagation. This paper shows this subtle change does not change the mathematical form of existing ITU/3GPP models and offers much easier analysis, intuitive appeal, better model parameter stability, and better accuracy in sensitivity tests over a vast range of microwave and mmWave frequencies, scenarios, and distances, while using a simpler model with fewer parameters.",
keywords = "5G, Millimeter wave, path loss models, prediction accuracy",
author = "Shu Sun and Theodore Rappaport and Thomas, {Timothy A.} and Amitava Ghosh and Nguyen, {Huan C.} and Kovacs, {Istvan Z.} and Ignacio Rodriguez and Ozge Koymen and Andrzej Partyka",
year = "2016",
month = "5",
day = "1",
doi = "10.1109/TVT.2016.2543139",
language = "English (US)",
volume = "65",
pages = "2843--2860",
journal = "IEEE Transactions on Vehicular Technology",
issn = "0018-9545",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications

AU - Sun, Shu

AU - Rappaport, Theodore

AU - Thomas, Timothy A.

AU - Ghosh, Amitava

AU - Nguyen, Huan C.

AU - Kovacs, Istvan Z.

AU - Rodriguez, Ignacio

AU - Koymen, Ozge

AU - Partyka, Andrzej

PY - 2016/5/1

Y1 - 2016/5/1

N2 - This paper compares three candidate large-scale propagation path loss models for use over the entire microwave and millimeter-wave (mmWave) radio spectrum: The alpha-beta-gamma (ABG) model, the close-in (CI) free-space reference distance model, and the CI model with a frequency-weighted path loss exponent (CIF). Each of these models has been recently studied for use in standards bodies such as 3rd Generation Partnership Project (3GPP) and for use in the design of fifth-generation wireless systems in urban macrocell, urban microcell, and indoor office and shopping mall scenarios. Here, we compare the accuracy and sensitivity of these models using measured data from 30 propagation measurement data sets from 2 to 73 GHz over distances ranging from 4 to 1238 m. A series of sensitivity analyses of the three models shows that the four-parameter ABG model underpredicts path loss when relatively close to the transmitter, and overpredicts path loss far from the transmitter, and that the physically based two-parameter CI model and three-parameter CIF model offer computational simplicity, have very similar goodness of fit (i.e., the shadow fading standard deviation), exhibit more stable model parameter behavior across frequencies and distances, and yield smaller prediction error in sensitivity tests across distances and frequencies, when compared to the four-parameter ABG model. Results show the CI model with a 1-m reference distance is suitable for outdoor environments, while the CIF model is more appropriate for indoor modeling. The CI and CIF models are easily implemented in existing 3GPP models by making a very subtle modification-by replacing a floating non-physically based constant with a frequency-dependent constant that represents free-space path loss in the first meter of propagation. This paper shows this subtle change does not change the mathematical form of existing ITU/3GPP models and offers much easier analysis, intuitive appeal, better model parameter stability, and better accuracy in sensitivity tests over a vast range of microwave and mmWave frequencies, scenarios, and distances, while using a simpler model with fewer parameters.

AB - This paper compares three candidate large-scale propagation path loss models for use over the entire microwave and millimeter-wave (mmWave) radio spectrum: The alpha-beta-gamma (ABG) model, the close-in (CI) free-space reference distance model, and the CI model with a frequency-weighted path loss exponent (CIF). Each of these models has been recently studied for use in standards bodies such as 3rd Generation Partnership Project (3GPP) and for use in the design of fifth-generation wireless systems in urban macrocell, urban microcell, and indoor office and shopping mall scenarios. Here, we compare the accuracy and sensitivity of these models using measured data from 30 propagation measurement data sets from 2 to 73 GHz over distances ranging from 4 to 1238 m. A series of sensitivity analyses of the three models shows that the four-parameter ABG model underpredicts path loss when relatively close to the transmitter, and overpredicts path loss far from the transmitter, and that the physically based two-parameter CI model and three-parameter CIF model offer computational simplicity, have very similar goodness of fit (i.e., the shadow fading standard deviation), exhibit more stable model parameter behavior across frequencies and distances, and yield smaller prediction error in sensitivity tests across distances and frequencies, when compared to the four-parameter ABG model. Results show the CI model with a 1-m reference distance is suitable for outdoor environments, while the CIF model is more appropriate for indoor modeling. The CI and CIF models are easily implemented in existing 3GPP models by making a very subtle modification-by replacing a floating non-physically based constant with a frequency-dependent constant that represents free-space path loss in the first meter of propagation. This paper shows this subtle change does not change the mathematical form of existing ITU/3GPP models and offers much easier analysis, intuitive appeal, better model parameter stability, and better accuracy in sensitivity tests over a vast range of microwave and mmWave frequencies, scenarios, and distances, while using a simpler model with fewer parameters.

KW - 5G

KW - Millimeter wave

KW - path loss models

KW - prediction accuracy

UR - http://www.scopus.com/inward/record.url?scp=84970021874&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84970021874&partnerID=8YFLogxK

U2 - 10.1109/TVT.2016.2543139

DO - 10.1109/TVT.2016.2543139

M3 - Article

AN - SCOPUS:84970021874

VL - 65

SP - 2843

EP - 2860

JO - IEEE Transactions on Vehicular Technology

JF - IEEE Transactions on Vehicular Technology

SN - 0018-9545

IS - 5

M1 - 7434656

ER -