GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN)
Eavesdropping activities have always been punitive threats to data security for every level of society. With the revolution and advance of technology, the tools to accomplish these have become more sophisticated, smaller yet cheaper. Medium of transmissions for transferring these data have also impr...
| Main Authors: | , , |
|---|---|
| Format: | Book Section |
| Published: |
Springer Science and Business Media Deutschland GmbH
2022
|
| Subjects: |
| _version_ | 1796866938749583360 |
|---|---|
| author | Burhan, Mohamad Fariq Nawawi, Sophan Wahyudi Yunus, Muhammad Hazim |
| author_facet | Burhan, Mohamad Fariq Nawawi, Sophan Wahyudi Yunus, Muhammad Hazim |
| author_sort | Burhan, Mohamad Fariq |
| collection | ePrints |
| description | Eavesdropping activities have always been punitive threats to data security for every level of society. With the revolution and advance of technology, the tools to accomplish these have become more sophisticated, smaller yet cheaper. Medium of transmissions for transferring these data have also improved tremendously, including over mobile telephone networks through GSM networks. In order to mitigate this threat, a fast and effective approach to localize the eavesdropping device is called for. Thus, this paper is to propose a reliable localization framework that coordinate and locate GSM eavesdropping devices in indoor environment based on Convolutional Neural Networks (CNN) algorithm and Received Signal Strength Indicator (RSSI). GSM device used in this experiment is planted in the conference rooms to prove the effectiveness of this system. 3D radio images are constructed based on RSSI fingerprints to locate GSM device accurately by determining its location coordinate. The different choice of optimization algorithms, parameters and architecture model was tested in our proposed method to achieve better localization performance. The simulation results proved that RMSProp optimization algorithm with kurtosis provide a better localization accuracy and computational complexity. |
| first_indexed | 2024-03-05T21:19:41Z |
| format | Book Section |
| id | utm.eprints-100735 |
| institution | Universiti Teknologi Malaysia - ePrints |
| last_indexed | 2024-03-05T21:19:41Z |
| publishDate | 2022 |
| publisher | Springer Science and Business Media Deutschland GmbH |
| record_format | dspace |
| spelling | utm.eprints-1007352023-04-30T10:22:14Z http://eprints.utm.my/100735/ GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) Burhan, Mohamad Fariq Nawawi, Sophan Wahyudi Yunus, Muhammad Hazim TK Electrical engineering. Electronics Nuclear engineering Eavesdropping activities have always been punitive threats to data security for every level of society. With the revolution and advance of technology, the tools to accomplish these have become more sophisticated, smaller yet cheaper. Medium of transmissions for transferring these data have also improved tremendously, including over mobile telephone networks through GSM networks. In order to mitigate this threat, a fast and effective approach to localize the eavesdropping device is called for. Thus, this paper is to propose a reliable localization framework that coordinate and locate GSM eavesdropping devices in indoor environment based on Convolutional Neural Networks (CNN) algorithm and Received Signal Strength Indicator (RSSI). GSM device used in this experiment is planted in the conference rooms to prove the effectiveness of this system. 3D radio images are constructed based on RSSI fingerprints to locate GSM device accurately by determining its location coordinate. The different choice of optimization algorithms, parameters and architecture model was tested in our proposed method to achieve better localization performance. The simulation results proved that RMSProp optimization algorithm with kurtosis provide a better localization accuracy and computational complexity. Springer Science and Business Media Deutschland GmbH 2022 Book Section PeerReviewed Burhan, Mohamad Fariq and Nawawi, Sophan Wahyudi and Yunus, Muhammad Hazim (2022) GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN). In: Control, Instrumentation and Mechatronics: Theory and Practice. Lecture Notes in Electrical Engineering, 921 (NA). Springer Science and Business Media Deutschland GmbH, Singapore, pp. 627-640. ISBN 978-981193922-8 http://dx.doi.org/10.1007/978-981-19-3923-5_54 DOI:10.1007/978-981-19-3923-5_54 |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Burhan, Mohamad Fariq Nawawi, Sophan Wahyudi Yunus, Muhammad Hazim GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title | GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title_full | GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title_fullStr | GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title_full_unstemmed | GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title_short | GSM device localization in indoor environment using Received Signal Strength Indicator (RSSI) and Convolutional Neural Networks (CNN) |
| title_sort | gsm device localization in indoor environment using received signal strength indicator rssi and convolutional neural networks cnn |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| work_keys_str_mv | AT burhanmohamadfariq gsmdevicelocalizationinindoorenvironmentusingreceivedsignalstrengthindicatorrssiandconvolutionalneuralnetworkscnn AT nawawisophanwahyudi gsmdevicelocalizationinindoorenvironmentusingreceivedsignalstrengthindicatorrssiandconvolutionalneuralnetworkscnn AT yunusmuhammadhazim gsmdevicelocalizationinindoorenvironmentusingreceivedsignalstrengthindicatorrssiandconvolutionalneuralnetworkscnn |