Wideband Unshielded-Twisted-Pair (UTP) Cable Measurements and Modeling for Multiple-Input-Multiple-Output (MIMO) Systems

The study of multiple-input-multiple-output (MIMO) systems over unshielded-twisted-pair (UTP) cables relies on or assumes models of the cable's transmission and crosstalk parameters. Several cable models have been proposed; however, there has been a lack of wide-scale line surveys to verify the...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on instrumentation and measurement 2007-12, Vol.56 (6), p.2515-2521
Main Authors: Nedev, N.H., McLaughlin, S., Cook, J.W.
Format: Article
Language:English
Subjects:
Cables
Crosstalk
Distortion measurement
Frequency measurement
Laboratories
Length measurement
Mathematical model
measurements
MIMO
modeling
multiple-input-multiple-output (MIMO)
Predictive models
Studies
unshielded-twisted-pair (UTP) cables
Wideband
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The study of multiple-input-multiple-output (MIMO) systems over unshielded-twisted-pair (UTP) cables relies on or assumes models of the cable's transmission and crosstalk parameters. Several cable models have been proposed; however, there has been a lack of wide-scale line surveys to verify the applicability of these models. This paper presents the results of wideband-crosstalk and transmission-parameter measurements on UTP cables conducted at the laboratories of BTExact. The measurement results are then used to verify Joffe's MIMO channel model, which is one of the two available but mathematically equivalent models. It is shown that crosstalk estimates from measurements taken on very short cable lengths fail to account for the effects from the pair and bundle twisting. Model parameters measured on longer cable pieces yield more realistic frequency-dependent model predictions; these are, however, distorted at high frequencies due to resonance effects. In general, the model tends to overestimate far-end crosstalk and underestimate near-end crosstalk.
ISSN:0018-9456
1557-9662
DOI:10.1109/TIM.2007.907951