On Physical Layer Security in Energy-Efficient Wireless Health Monitoring Applications
Autor: | Belal Essam ElDiwany, Abdulla Al-Ali, Amr Mohamed, Alaa Awad Abdellatif, Mohsen Guizani, Xiaojiang Du |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2019 |
Předmět: |
Mathematical optimization
Optimization problem Branch and bound Computer science business.industry Quality of service 05 social sciences Transmitter Physical layer 050801 communication & media studies Eavesdropping Data_CODINGANDINFORMATIONTHEORY 0508 media and communications Transmission (telecommunications) wireless health monitoring Channel state information branch and bound 0502 economics and business Wireless 050211 marketing Physical layer security business secrecy outage probability Efficient energy use Communication channel |
Zdroj: | ICC |
Popis: | In this paper, we investigate a multi-objective optimization framework for secure wireless health monitoring applications. In particular, we consider a legitimate link for the transmission of a vital EEG signal, threatened by a passive eavesdropping attack, that aims at wiretapping these measurements. We incorporate in our framework the practical secrecy metric, namely secrecy outage probability (SOP), which requires only the knowledge of side information regarding the eavesdropper (Ev), instead of completely having its instantaneous channel state information (CSI). To that end, we formulate an optimization problem in the form of maximizing the energy efficiency of the transmitter, while minimizing the distortion encountered at the signal resulting from the compression process prior to transmission, under realistic quality of service (QoS) constraints. The problem is shown to be nonconvex and NP-complete. Towards solving the problem, a branch and bound (BnB)-based algorithm is presented where a ?-suboptimal solution, from the global optimal one, is obtained. Numerical results are conducted to verify the system performance, where it is shown that our proposed approach outperforms similar systems deploying fixed compression policies (FCPs). We successfully meet QoS requirements while optimizing the system objectives, at all channel conditions, which cannot be attained by these FCP approaches. Interestingly, we also show that a target secrecy rate can be practically achieved with nonzero probability, even when the Ev has a better channel condition, on the average, than that for the legitimate receiver. This work was made possible by NPRP grant # NPRP8-408-2-172 from the Qatar National Research Fund (a member of Qatar Foundation). Scopus |
Databáze: | OpenAIRE |
Externí odkaz: |