Radio channel fingerprint model based on entropy weight method

Rugui Yao, Yan Gao, Juan Xu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Different radio channels have various characteristics, which is analogous to human fingerprint, and the difference between the characteristics is defined as radio channel fingerprint. In this paper, we construct a quantitative radio channel fingerprint model based on the entropy weight method. With this model, five indicators, including time delay spread, coherence bandwidth, Doppler spread, fading threshold, and level variance, are selected to characterize the channel fingerprint and generate the evaluation matrix. Then we apply entropy weight method to find the reasonable weights for the five indicators, and thus achieve quantitative radio channel fingerprint model. Finally, the proposed fingerprint model is used to correctly realize scene recognition with the practical measured data provided by Huawei Corporation. This success validates the effectiveness and correctness of the proposed model.

Original languageEnglish
Title of host publicationWOCC 2016 - 25th Wireless and Optical Communication Conference, Jointly held with Photonics Forum of Chiao-Tung Universities
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467399586
DOIs
StatePublished - 7 Jul 2016
Event25th Wireless and Optical Communication Conference, WOCC 2016 - Chengdu, China
Duration: 21 May 201623 May 2016

Publication series

NameWOCC 2016 - 25th Wireless and Optical Communication Conference, Jointly held with Photonics Forum of Chiao-Tung Universities

Conference

Conference25th Wireless and Optical Communication Conference, WOCC 2016
Country/TerritoryChina
CityChengdu
Period21/05/1623/05/16

Keywords

  • entropy weight method
  • indicator
  • Radio channel fingerprint model
  • scene recognition
  • weight

Fingerprint

Dive into the research topics of 'Radio channel fingerprint model based on entropy weight method'. Together they form a unique fingerprint.

Cite this