Abstract
Real-time applications such as online gaming and video conferencing require low latency. High network latency, especially for geographically distributed users, negatively impacts these applications, as traditional network protocols prioritize throughput over latency. This document describes a network protocol to reduce latency and enable faster communication, especially for time-sensitive applications. The protocol prioritizes data packets, using custom headers to indicate priority. Additional aspects such as linking related data, sequencing, time-stamping, and lightweight session tokens can also be incorporated in the header. The protocol leverages adaptive mechanisms for congestion control, packet loss compensation, and machine learning for prediction and pre-emption of network congestion. UDP, time-aware packet scheduling, or peer-to-peer communication can be used to reduce latency. Hardware acceleration, network coding, and/or delay-based congestion control can be implemented to provide low latency. Application-specific optimizations can be implemented for time sensitive applications such as video conferencing, online gaming, virtual reality, etc. Additionally, network-aware application logic can be implemented in the real-time application.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Alam, Mohsin, "Low Latency Network Protocol for Time Sensitive Applications", Technical Disclosure Commons, (January 30, 2025)
https://www.tdcommons.org/dpubs_series/7787