Improved indoor range measurements at various signal bandwidths

Ian Sharp, Kegen Yu

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)


This paper investigates range error measurements over multiple paths in an indoor nonline-of-sight environment as a function of signal bandwidth. The raw range measurements were obtained by transmitting and receiving a signal of 4 GHz bandwidth. Then, through suitable processing of the received pulses, signals with an effective bandwidth from 60 MHz to 4 GHz were generated. The range errors had mean value of about 0.2 m for bandwidths greater than 300 MHz, but peak errors increased to 4-5 m with a 60-MHz bandwidth using a standard leading-edge threshold time-of-arrival (TOA) technique. These latter errors were observed to be related to distortions in the leading edge of the pulse due to multipath propagation. Based on the observations, a new enhanced leading-edge TOA algorithm is developed. The enhanced algorithm uses the initial estimate from the standard leading-edge threshold algorithm, but then uses back-projection to correct the initial estimate. Experimental results demonstrate that the proposed algorithm reduces the mean errors by about 50% at lower bandwidths. As a consequence, TOA positioning systems using the enhanced leading-edge algorithm and operating in the 2.4-GHz and especially the 5.8-GHz Industrial, Scientific and Medical band have position errors only about twice that of ultrawideband systems, despite the much smaller signal bandwidth.

Original languageEnglish
Article number7403977
Pages (from-to)1364-1373
Number of pages10
JournalIEEE Transactions on Instrumentation and Measurement
Issue number6
Publication statusPublished - 1 Jun 2016
Externally publishedYes


  • Analytical performance estimation
  • multipath and nonline-of-sight (NLOS) propagation
  • range error measurements
  • range errors versus signal bandwidth
  • time-of-arrival (TOA) leading-edge algorithm.


Dive into the research topics of 'Improved indoor range measurements at various signal bandwidths'. Together they form a unique fingerprint.

Cite this