Survivability-Enhanced Virtual Network Embedding Strategy in Virtualized Wireless Sensor Networks

Area of Science:

• Computer Science
• Network Engineering
• Distributed Systems

Background:

• Wireless Sensor Networks (WSNs) are increasingly deployed, driving the need for WSN virtualization.
• The survivable virtual network embedding (SVNE) problem addresses mapping virtual networks onto WSNs while tolerating physical node failures.
• Existing research lacks sufficient focus on survivability against physical sensor node failures in virtualized WSNs.

Purpose of the Study:

• To address the lack of survivability research in virtual sensor networks (VSNs) concerning physical sensor node failures.
• To develop a mathematical model for the SVNE problem considering resource constraints and physical node failures.
• To propose a heuristic algorithm ensuring VSN survivability and efficient resource utilization.

Main Methods:

• Mathematically modeled the SVNE problem as a mixed integer programming problem.
• Developed a node reliability-aware backup survivable embedding algorithm (NCS).
• Implemented a fault recovery mechanism using resource reservation and backup switching for critical virtual nodes.

Main Results:

• The proposed NCS algorithm effectively guarantees VSN survivability against physical node failures.
• The strategy significantly reduces failure penalties associated with physical sensor node failures.
• Experimental results demonstrate improved long-term operating income for infrastructure providers.

Conclusions:

• The NCS algorithm provides a robust solution for survivable virtual network embedding in WSNs.
• Node reliability-aware mapping and backup mechanisms are crucial for VSN survivability.
• This approach enhances the resilience and economic viability of virtualized WSN infrastructure.