Wireless Sensor Networks (WSNs) consist of numerous battery-powered sensor nodes that operate with limited energy, computation, and communication capabilities. Designing routing strategies that are both energy-efficient and attack-resilient is essential for extending network lifetime and ensuring secure data delivery. This paper proposes Adaptive Federated Reinforcement Learning-Hunger Games Search (AFRL-HGS), a Hybrid Routing framework that integrates multiple advanced techniques. At the node level, tabular Q-learning enables each sensor node to act as a reinforcement learning agent, making next-hop decisions based on discretized state features such as residual energy, distance to sink, congestion, path quality, and security. At the network level, Federated Reinforcement Learning (FRL) allows the sink node to aggregate local Q-tables using adaptive, energy- and performance-weighted contributions, with Polyak-based blending to preserve stability. The binary Hunger Games Search (HGS) metaheuristic initializes Cluster Head (CH) selection and routing, providing a well-structured topology that accelerates convergence. Security is enforced as a constraint through a lightweight trust and anomaly detection module, which fuses reliability estimates with residual-based anomaly detection using Exponentially Weighted Moving Average (EWMA) on Round-Trip Time (RTT) and loss metrics. The framework further incorporates energy-accounted control plane operations with dual-format HELLO and hierarchical ADVERTISE/Service-ADVERTISE (SrvADVERTISE) messages to maintain the routing tables. Evaluation is performed in a hybrid testbed using the Graphical Network Simulator-3 (GNS3) for large-scale simulation and Kali Linux for live adversarial traffic injection, ensuring both reproducibility and realism. The proposed AFRL-HGS framework offers a scalable, secure, and energy-efficient routing solution for next-generation WSN deployments.
As smart cities nowadays use many Internet of Things (IoT) devices, the need for service composition methods that are both efficient and flexible has become unavoidable. Methodologies should be designed to provide flexibility and sustain high quality of service (QoS) under dynamic urban conditions. Deterministic techniques often have poor scalability and require complete, accurate QoS information. In contrast, nondeterministic approaches can handle uncertainty but tend to be computationally expensive and less reliable under dynamic conditions. Service composition is considered an NP-hard problem due to the exponential complexity involved in selecting and combining optimal services under multiple QoS constraints. This article proposes a novel hybrid framework called Bayesian network (BN)–Gray Wolf optimizer (BN-GWO), which integrates BNs with the Gray Wolf optimizer (GWO) algorithm. The BN captures conditional dependencies among QoS attributes and accurately estimates missing values, while the GWO algorithm efficiently explores the compositional space to optimize QoS metrics. Experimental evaluations conducted using QoS traces synthetically generated via the iFogSim2 simulation platform demonstrate that the proposed BN-GWO framework significantly outperforms state-of-the-art methods across multiple performance metrics.
1 January 2016 in International Journal of Advanced Computer Science and Applications
There are different methods for dealing with spams; however, since spammers continuously use tricks to defeat the proposed methods, hence, filters should be constantly updated. In this study, Stackelberg game was used to produce a dynamic filter and the relations between filter and adversary were modelled as a turn game in which there is a leader and a follower. Then, an attempt was made to solve the game as an optimization program via the evolutionary stable strategy (ESS). The dataset used in the study for evaluating and analyzing the proposed method was a real dataset including the emails of four users’ personal emails. The results of the conducted evaluations and investigations indicated that the proposed method had an 8% improvement over the three-class classification method and a 0.8% improvement over the ESS-based equilibrium point method