Online Service Function Chain Planning for Satellite-Ground Integrated Networks to Minimize End-to-End (E2E) Delay

Area of Science:

• Network Engineering
• Telecommunications
• Computer Science

Background:

• Software-Defined Networking (SDN) and Satellite-Ground Integrated Networks (SGINs) are crucial for next-generation mobile communications.
• Virtual Network Function (VNF) deployment challenges in SGINs can increase end-to-end (E2E) delay due to unpredictable user requests and limited Low Earth Orbit (LEO) satellite resources.

Purpose of the Study:

• To propose an online algorithm for joint VNF deployment and routing path formation in SGINs.
• To minimize E2E delay, a critical Quality of Service (QoS) parameter, by selectively deploying essential VNFs.

Main Methods:

• An event-driven algorithm that responds to user requests by deploying necessary VNFs for Service Function Chains (SFCs).
• Defining a Minimum-Hop Region (MHR) based on geographic coordinates to reduce search space for base stations and minimize propagation delays.
• Deploying VNFs along optimized paths to further reduce E2E delay.

Main Results:

• The proposed algorithm achieves high similarity to the global optimum, with up to 97% similarity in E2E delay and CPU power consumption.
• An average similarity of approximately 90% was observed for both E2E delay and CPU power consumption.

Conclusions:

• The developed online algorithm effectively minimizes E2E delay in SDN-SGIN environments.
• The approach offers a near-optimal solution for VNF deployment and routing, enhancing QoS for mobile communications.