Median filtering for power estimation in mobile communication systems

C. Tepedelenlioʇlu, N. Sidiropoulos, G. B. Giannakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations

Abstract

In mobile communications, accurate power estimation is important for power control and handoff decisions. Power estimation algorithms aim to filter out the fast variation in the received power due to multipath fading, and should ideally exhibit very low complexity. For this reason, linear filtering (local mean) is almost exclusively used for power estimation in practice. We derive the maximum likelihood estimator in an idealized setting, and motivated by the statistical properties of the multipath signal, we also propose median filtering for power estimation. Simulations show promising results, indicating gains in RMSE and robustness to increased window size.

Original languageEnglish (US)
Title of host publication2001 IEEE 3rd Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages229-231
Number of pages3
ISBN (Electronic)0780367200
DOIs
StatePublished - Jan 1 2001
Event3rd IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001 - Taoyuan, Taiwan, Province of China
Duration: Mar 20 2001Mar 23 2001

Publication series

NameIEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC
Volume2001-January

Other

Other3rd IEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2001
Country/TerritoryTaiwan, Province of China
CityTaoyuan
Period3/20/013/23/01

Keywords

  • Filtering
  • Maximum likelihood detection
  • Maximum likelihood estimation
  • Mobile communication
  • Nonlinear filters
  • Power control
  • Power filters
  • Power measurement
  • Robustness
  • Shadow mapping

Fingerprint

Dive into the research topics of 'Median filtering for power estimation in mobile communication systems'. Together they form a unique fingerprint.

Cite this