Traffic Measurement Optimization for Cross-Domain Ad Hoc Networks Based on Meta-Learning and Reinforcement Learning

Abstract

Cross-domain Ad Hoc network is a network that self-organizes nodes on different media and adapts to network topology. In cross-domain communication networks, direct measurement technology helps obtain accurate end-to-end network traffic information. However, the low computational power and low storage characteristics of some nodes in the cross-domain network hinder all nodes from running the network traffic measurement process. To address this issue, a network traffic measurement optimization method based on meta-learning and proximal policy optimization(PPO) was proposed. This method determined the set of nodes that performed network traffic measurement in the next time slot according to the network operating environment of the previous time slot, so as to perform the measurement process on as few nodes as possible to obtain as much network traffic information as possible. Three network datasets were used to verify the proposed method. The experimental results show that the traffic measurement optimization algorithm for cross-domain Ad Hoc networks based on meta-learning and reinforcement learning can effectively select the nodes with large traffic flow, with faster convergence speed and higher measurement efficiency.

Keywords

• cross-domain ad hoc network
• network traffic measurement
• meta-learning
• proximal policy optimization
• reinforcement learning