Open AccessExtreme learning machine for 60 GHz millimetre wave positioning
Author(s) -
Liang Xiaolin,
Zhang Hao,
Lu Tingting,
Gulliver Thomas Aaron
Publication year - 2017
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2016.0080
Subject(s) - computer science , robustness (evolution) , kurtosis , extreme learning machine , detector , standard deviation , algorithm , range (aeronautics) , channel (broadcasting) , performance metric , artificial intelligence , statistics , telecommunications , artificial neural network , mathematics , engineering , biochemistry , chemistry , management , economics , gene , aerospace engineering
Extreme learning machine (ELM) has attracted considerable attention in recent years due to its numerous applications in classification and regression. In this study, the authors investigate the performance of an ELM‐based threshold selection algorithm for 60 GHz millimetre wave time of arrival estimation using energy detector (ED). A hybrid metric based on the skewness, kurtosis, standard deviation, and slope of the ED values is employed. The optimal normalised threshold for different signal‐to‐noise ratios (SNRs) is investigated, and the effects of the integration period and channel model are examined. Performance results are presented which show that the proposed ELM‐based algorithm provides high precision and better robustness than existing techniques over a wide range of SNRs for the IEEE 802.15.3c CM1.1 and CM2.1 channel models. Further, the performance is largely independent of the integration period and channel model.