Wireless sensor networks for monitoring gas turbine engines during development: 2017 IEEE Symposium on Computers and Communications (ISCC)

Research output: Contribution to conferenceOther

  • Authors:
  • H. H. Khalili
  • P. R. Green
  • D. George
  • G. Watson
  • W. Schiffers

Abstract

Several thousand sensors hard-wired to data acquisition units may be used to instrument a gas turbine engine during its development phase. This process of hard-wiring is time-consuming, expensive, and inflexible. A wireless sensor network (WSN) is proposed as a solution for selectively replacing wired sensors with wireless sensors and for adding new sensors. The WSN shall operate between the engine casing and the engine cowling where the presence of metal surfaces results in a complex radio frequency (RF) environment exhibiting severe multipath propagation. Although research has been performed into the use of WSNs for monitoring gas turbine engines, the propagation characteristics of the RF channel and a systematic analysis of the performance of digital modulation techniques within the environment have not been reported in the literature. The aim of this research is to investigate WSN technologies, and their communication challenges, when used to instrument gas turbine engines during their development phase, and to implement and experimentally evaluate a prototype WSN. A fully automated measurement system, utilising software defined radio (SDR) techniques, has been used to perform RF propagation measurements through more than 7,000 individual channels on a static Rolls-Royce Trent 900 engine. The analysis of these measurements are being used to characterise the RF propagation channel and to evaluate the performance of a number of modulation schemes including basic schemes such as phase shift keying (PSK) and more complex schemes such as orthogonal frequency division multiplexing (OFDM). This paper presents an overview of the project including the experimental methodology, the measurement system, and some preliminary results.

Bibliographical metadata

Original languageEnglish
Pages1325-1331
Number of pages7
DOIs
Publication statusPublished - 3 Jul 2017

Related information

Researchers

View all