Primary User Signal Recognition Algorithm based on AL-ABM in Cognitive Network

DOI

10.2991/emcs-16.2016.404How to use a DOI?

Keywords

Cognitive network; Active Learning; Primary User Signal Recognition; AdaBoost Machine

Abstract

The cognitive users can use the spectrum holes which are vacant at some time or some frequency. Once the authorized users need to use the spectrum again, the cognitive users should stop using it to avoid interfering primary users. In this paper, a method based on active learning (AL) and AdaBoost machine (ABM) for the primary user signal modulation type recognition is proposed in low signal to noise ratio. A set of cyclic spectrum features are first calculated, and the training samples and testing samples are formed for classification. Then, active learning algorithm is applied to obtain samples improved classification through a number of iterations, and AdaBoost is formed. Finally, the formed SVM is utilized to recognize the primary user signal modulation type. The simulation results shows that the proposed AL-ABM Algorithm can improve the recognition accuracy of signal samples obviously.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xin Wang
AU  - Kuan Huang
AU  - Na Zhang
AU  - Shi Wang
AU  - Changtao Wang
AU  - Shuai Wang
AU  - Xiuhong Wang
AU  - Wei Gong
AU  - Bin Wang
PY  - 2016/01
DA  - 2016/01
TI  - Primary User Signal Recognition Algorithm based on AL-ABM in Cognitive Network
BT  - Proceedings of the 2016 International Conference on Education, Management, Computer and Society
PB  - Atlantis Press
SP  - 1614
EP  - 1617
SN  - 2352-538X
UR  - https://doi.org/10.2991/emcs-16.2016.404
DO  - 10.2991/emcs-16.2016.404
ID  - Wang2016/01
ER  -