Abstract
A relay selection method is proposed for physical-layer security in multi-hop decode-and-forward (DF) relaying systems. In the proposed method, cooperative relays are selected to maximize the achievable secrecy rates under DF-relaying constraints by the classification method. Artificial neural networks (ANNs), which are used for machine learning, are applied to classify the set of cooperative relays based on the channel state information of all nodes. Simulation results show that the proposed method can achieve near-optimal performance for an exhaustive search method for all combinations of relay selection, while computation time are reduced significantly. Furthermore, the proposed method outperforms the best relay selection method, in which the best relay in terms of secrecy performance is selected among active ones.
Highlights
Security for wireless communication networks has become a crucial issue because of the broadcast nature of wireless channels
Maximum ratio combining (MRC), distributed selection combining, and distributed switch-and-stay combining schemes have been evaluated for opportunistic relay selection systems [9]
The system has Nr relays, and Π = 2 Nr + 1 class labels are employed, where 2 Nr class labels are for relay selection combinations and one class label is for no transmission (NT) scheme
Summary
Security for wireless communication networks has become a crucial issue because of the broadcast nature of wireless channels. For the physical-layer security of cooperative relays, a node selection method has been proposed for amplify-and-forward and decode-and-forward (DF) relaying in a two-hop system [6,7]. For physical-layer security, two machine learning methods, support vector machine (SVM) and naïve-Bayes, have been investigated for MIMO multiantenna-eavesdropper wiretap channels by transmit antenna selection [27]. A relay selection problem was considered to maximize the achievable secrecy rate in a cooperative DF multi-hop network with the presence of an eavesdropper. The proposed ANN model is trained using the training dataset, where the channel state information (CSI) of all nodes is the input, and the corresponding index for the activation of cooperative relays is the output. R Lx1 represents the vector space of all Lx1 real matrices
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
