Finding Optimal Model Parameters from Measurements with Severe Multipath

A model-weighted root-mean-square cost function is presented for finding slowly-varying model parameters from communication channel measurements in cases where frequent, deep multipath nulls are present. The new cost function is shown to produce an excellent estimate of the loss exponent in a log-di...

Full description

Saved in:
Bibliographic Details
Main Authors: Hawkins, Scot A., Pendergrass, Nixon A.
Format: Conference Proceeding
Language:English
Subjects:
Computational modeling
Cost function
Curve fitting
Data models
least mean square algorithms
multipath channels
parameter estimation
path loss
radiowave propagation
Receiving antennas
Transmitting antennas
urban environment
wireless communications
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A model-weighted root-mean-square cost function is presented for finding slowly-varying model parameters from communication channel measurements in cases where frequent, deep multipath nulls are present. The new cost function is shown to produce an excellent estimate of the loss exponent in a log-distance model with much less sensitivity to extreme dips in the data relative to the root-mean-square function commonly used. Two test cases are presented based on two-ray propagation and the resulting model parameters agree well with theory. Also shown are examples of application to measurement data from moderately urbanized areas, where severe multipath signal level drops are typical at short range in line-of-sight in built-up environments.
ISSN:2155-7578
2155-7586
DOI:10.1109/MILCOM.2013.136