In this paper, we propose an efficient deep neural network-based relay selection (DNS) scheme to evaluate and improve the end-to-end throughput in wireless-powered cognitive Internet-of-Things (IoT) networks. In this system, multiple energy harvesting (EH) relays are deployed randomly to assist data transmission from a source node to multiple users under practical nonlinearity of the EH circuits. We first design an incremental relaying protocol, where a selected user will request the help from relays if the direct transmission is not favorable. In such a protocol, we develop a deep neural network framework for relay selection and throughput prediction with high accuracy, less channel feedback amount, and short execution time. Simulation results show that the proposed DNS scheme achieves higher throughput than the conventional relay selection methods, while it considerably reduces computational complexity, suggesting a real-time configuration for IoT systems under complex scenarios. Moreover, the proposed DNS scheme achieves the root-mean-square-error (RMSE) of 6.6×10-3 on the considered dataset, which exhibits the lowest RMSE as compared to the state-of-the-art machine learning approaches.

In this article, we study the physical-layer security (PHY-security) improvement method for a downlink nonorthogonal multiple access (NOMA) system in the presence of an active eavesdropper. To this end, we propose a full-duplex (FD)-enabled NOMA system and a promising scheme, called the minimal transmitter selection (MTS) scheme, to support secure transmission. Specifically, the cell-center and cell-edge users act simultaneously as both receivers and jammers to degrade the eavesdropper channel condition. Additionally, the proposed MTS scheme opportunistically selects the transmitter to minimize the maximum eavesdropper channel capacity. To estimate the secrecy performance of the proposed methods, we derive an approximated closed-form expression for secrecy outage probability (SOP) and build a deep neural network (DNN) model for SOP evaluation. Numerical results reveal that the proposed NOMA system and MTS scheme improve not only the SOP but also the secrecy sum throughput. Furthermore, the estimated SOP through the DNN model is shown to be tightly close to other approaches, i.e., the Monte-Carlo method and analytical expressions. The advantages and drawbacks of the proposed transmitter selection scheme are highlighted, along with insightful discussions.

In this paper, we propose a novel deep reinforcement learning-based quality-of-service (QoS) routing protocol, namely DRQR, exploiting cross-layer design to establish efficient QoS (EQS) routes in cognitive radio mobile ad hoc networks. An EQS route is a route with minimum end-to-end (E2E) queuing delay subject to QoS constraints such as link stability, residual energy, number of hops and avoiding licensed channels of primary users. Particularly, we propose an NP-complete optimization problem which has a feasible solution as an EQS route. To tackle this problem, we design a new deep reinforcement learning model which supports the DRQR protocol to establish EQS routes in real time by offline training instead of online training like most of literature studies. Moreover, the DRQR protocol guarantees to have high system performance. A mathematical analysis of the E2E queuing delay with random waypoint mobility model also provides to verify simulation results. Numerical results show that the DRQR protocol outperforms state-of-the-art routing protocols in terms of routing delay, queuing delay, control overhead, PDR and energy consumption.

Vehicle ad hoc networks (VANETs) are a vital part of intelligent transportation systems (ITS), offering a variety of advantages from reduced traffic to increased road safety. Despite their benefits, VANETs remain vulnerable to various security threats, including severe blackhole attacks. In this paper, we propose a deep-learning-based secure routing (DLSR) protocol using a deep-learning-based clustering (DLC) protocol to establish a secure route against blackhole attacks. The main features and contributions of this paper are as follows. First, the DLSR protocol utilizes deep learning (DL) at each node to choose secure routing or normal routing while establishing secure routes. Additionally, we can identify the behavior of malicious nodes to determine the best possible next hop based on its fitness function value. Second, the DLC protocol is considered an underlying structure to enhance connectivity between nodes and reduce control overhead. Third, we design a deep neural network (DNN) model to optimize the fitness function in both DLSR and DLC protocols. The DLSR protocol considers parameters such as remaining energy, distance, and hop count, while the DLC protocol considers cosine similarity, cosine distance, and the node’s remaining energy. Finally, from the performance results, we evaluate the performance of the proposed routing and clustering protocol in the viewpoints of packet delivery ratio, routing delay, control overhead, packet loss ratio, and number of packet losses. Additionally, we also exploit the impact of the mobility model such as reference point group mobility (RPGM) and random waypoint (RWP) on the network metrics.

In this paper, we propose a new deep Q-network (DQN) design for quality-of-service (QoS) multicast routing (DQMR) protocol to establish efficient QoS multicast (EQM) trees in cognitive radio mobile ad hoc networks (CR-MANETs). An EQM tree is a shortest-path multicast tree with minimum end-to-end (E2E) cost (a combination of queuing size ratio and link stability) subject to QoS constraints such as queuing size ratio, link stability, number of hops, number of time slots and avoiding the licensed channel of primary users. Particularly, we propose an NP-complete optimization problem such that its feasible solution is an EQM tree. To address this problem, we design a new DQN model and a new game-based model to form EQM trees in real-time by offline training instead of online training as done in previous papers. Moreover, the DQMR protocol is also guaranteed to have high stability, low routing delay, low control overhead, and high packet delivery ratio (PDR). Furthermore, one more new contribution of the paper is that exact closed-form expressions for the E2E queuing delay of a multicast routing tree are also derived assuming random waypoint mobility and the reference point group mobility models to compare with simulation results of routing delay. Simulation results show that the DQMR protocol outperforms multicast ad hoc on-demand distance vector routing protocol in terms of routing delay, control overhead, and PDR.

The frequent movement of nodes in cognitive radio mobile ad hoc networks (CRAHNs) causes challenges in scalability, stability, channel sensing and channel access problems that can be solved by using clustering technique. Game theory is a feasible approach to solve such problems by casting clustering problems as distributed optimization problems. The main contributions of this article are as follows. Firstly, we propose a minimum connected weighted inner edge spanning tree (MWIEST) game to find an approximate solution of a MWIEST problem in CRAHNs. In this game, a link-weight function of each link is designed based on a combination of link-stability and link-connectivity ratio functions. Secondly, we prove that the MWIEST game is an exact potential game that exists at least one Nash equilibrium ( NE ) point which is an approximate solution of the MWIEST problem. Besides, we also prove that best responses ( BR s) of the game converge to a NE in finite iterations. Thirdly, based on the MWIEST game, we propose four algorithms including the node information exchange (NIE), the best response selection (BRS), the intermediate nodes selection (INS) and the forming cluster (FC). Specifically, the algorithms NIE, BRS and INS provide a set of intermediate nodes (SetIN) which supports the FC algorithm to form clusters. Finally, we propose the game theory based clustering (GBC) protocol which is combination of the FC algorithm and the proposed cluster maintenance algorithm to construct high stable clusters supporting multicast routing in CRAHNs. Moreover, each obtained cluster includes most members having the same receiving channel which avoids the affected regions of licensed channels. For the performance evaluation, we implement the GBC protocol in OMNET++ platform to demonstrate its performance improvement over the state-of-the-art protocols in terms of network stability and control overheads.