Introduction
Wireless Rechargeable Sensor Networks (WRSNs) are foundational to the development of 6G-enabled smart cities and industrial automation. These networks rely on Mobile Charging Vehicles (MCVs) to maintain node longevity. However, WRSNs are increasingly vulnerable to Denial of Charging (DoC) attacks. In these scenarios, malicious nodes overwhelm the system with fraudulent charging requests or physically obstruct MCVs, preventing legitimate nodes from recharging and leading to network failure. This paper proposes a novel security framework that integrates Software-Defined Networking (SDN) with Digital Twin (DT) technology to proactively detect and mitigate such threats.
System Architecture
The proposed framework utilizes a dual-layered architecture where an SDN controller manages both the physical WRSN and its virtual counterpart, the Digital Twin.
As shown in Figure 1, the SDN controller acts as the centralized intelligence, performing performance computations, policy decisions, and attack detection. The Digital Twin serves as a real-time virtual shadow, simulating expected network behavior. By comparing the actual data plane performance against the DT's simulations, the SDN controller can identify anomalies that signify a DoC attack. This centralized approach offloads the computational burden from energy-constrained sensor nodes to the sink node.
Detection Methodology
To accurately identify malicious actors, the system employs a multi-factor detection strategy. A Combined Maliciousness Score is calculated for each node based on four key metrics:
- Charging Request Patterns: Analyzing request frequency using a Poisson distribution to identify excessive, fraudulent requests.
- Energy Consumption: Comparing reported energy levels against a Normal distribution of expected usage to detect spoofed energy reports.
- Behavioral & Reputation Scores: Tracking node reliability over time using Markov chains and Beta distributions.
- Charging Efficiency: Measuring the ratio of energy transmitted by the MCV versus energy received by the node to detect transfer-phase disruptions.
Once a node exceeds a predefined maliciousness threshold, the SDN controller dynamically reconfigures the charging queue to isolate the attacker and prioritize legitimate nodes.
Performance Evaluation
The framework was evaluated in a custom simulation environment with 100 to 500 sensor nodes. The protocol was tested against Low (LAI), Moderate (MAI), and High Attack Intensity (HAI) scenarios.
As illustrated in Figure 4, the network maintained a high survival rate (above 90%) even under high-intensity attacks. Additional results indicated that the system significantly improved energy usage efficiency and reduced MCV travel distance by optimizing routes to avoid malicious detours. The probabilistic models ensured high detection accuracy across all tested node densities.
Conclusion
By combining the centralized control of SDN with the predictive capabilities of Digital Twins, this framework provides a proactive defense against DoC attacks in WRSNs. While currently based on theoretical models, the multi-factor authentication of node behavior offers a robust path toward securing 6G infrastructure. Future work will focus on testing the system against more complex environmental interferences and coordinated attack vectors.
Summary updated