Multi-path transmission control protocol data packet scheduling method based on deep reinforcement learning

Abstract

The invention discloses a multi-path transmission control protocol data packet scheduling method based on deep reinforcement learning. In a multi-path transmission control protocol (MPTCP), a data packet scheduling process is converted into a Markov decision process by setting a periodic scheduling mechanism, a data packet scheduling strategy of the MPTCP is represented by a neural network through deep reinforcement learning, and an optimal data packet scheduling strategy under various network environments is learned. The problem that heuristic MPTCP data packet scheduling cannot adapt to complex and diverse dynamic network environments, and consequently MPTCP performance is reduced is fundamentally solved. According to the method, an Actor-Critic reinforcement learning framework is used, a deep reinforcement learning algorithm based on strategy gradient is adopted, and modeling and learning are directly carried out on an MPTCP data packet scheduling strategy, so that strategy optimization efficiency is improved, and training of an MPTCP data packet scheduling strategy neural network is accelerated.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to View More

AI Technical Summary

Benefits of technology

This patented technology allows companies or organizations to efficiently train their own machine learning algorithms that can handle complicated networks with varying environmental factors such as bandwidth usage patterns over different periods during operation. It also uses advanced techniques like Deep Learning (DL) and Artificial Intelligence (Al), which help improve its effectiveness even more effectively than traditional methods.

Problems solved by technology

Technologies described in this patents involve optimizing communication between devices over wireless local area networks (WLAN). These techniques aim to optimize resource usage while maximizing system efficiency without compromising any important factors like latency or jitter. One technique called MultiPath Transport Scheme (MTPS)-based Time Division Multiple Access protocol (TDMA PON) uses a combination of Random Interval Repeated Packet Sequencing (RISC)/Minirculate Network Reflective Algorithm (MINTRA) and Fast Retransformer (FRS) algorithms to efficiently allocate resources across WLAN connections. Another approach involves selecting certain paths based on their importance to ensure fair allocation of data streams within specific subnet boundaries.

Images