EDAF: An End-to-End Delay Analytics Framework for 5G-and-Beyond Networks

TL;DR

The paper tackles the challenge of meeting strict end-to-end delay guarantees in 5G by proposing EDAF, an end-to-end delay analytics framework that inserts measurement points across the protocol stack and aggregates per-packet timestamps for analysis. It formalizes delay as $Y_n = Y_n^{C} + Y_n^{Q} + Y_n^{L}$ with $Y_n^{L} = Y_n^{Lt} + Y_n^{Ls} + Y_n^{Lr}$ and provides methods to decompose $Y_n^{L}$ into Segmentation, Retransmissions, and Transmission components, enabling targeted optimization. Through OpenAirInterface 5G uplink experiments, EDAF demonstrates that removing segmentation and frame-alignment delays can reduce average end-to-end delay from 12 ms to 4 ms, while tail latency is dominated by retransmissions in certain regimes. The framework offers practical tooling for end-to-end latency analysis, supports URLLC-oriented research, and provides a reproducible path toward latency-aware scheduling and resource allocation in 5G and beyond.

Abstract

Supporting applications in emerging domains like cyber-physical systems and human-in-the-loop scenarios typically requires adherence to strict end-to-end delay guarantees. Contributions of many tandem processes unfolding layer by layer within the wireless network result in violations of delay constraints, thereby severely degrading application performance. Meeting the application's stringent requirements necessitates coordinated optimization of the end-to-end delay by fine-tuning all contributing processes. To achieve this task, we designed and implemented EDAF, a framework to decompose packets' end-to-end delays and determine each component's significance for 5G network. We showcase EDAF on OpenAirInterface 5G uplink, modified to report timestamps across the data plane. By applying the obtained insights, we optimized end-to-end uplink delay by eliminating segmentation and frame-alignment delays, decreasing average delay from 12ms to 4ms.

Figures (8)

• Figure 1: Packet's end-to-end delay components in EDAF
• Figure 2: Link delay components in EDAF. Dashed arrows indicate HARQ attempts.
• Figure 3: EDAF implementation setup
• Figure 4: Experiments TDD frame format, "D" indicates downlink slots and "U" is uplink.
• Figure 5: EDAF e2e CCDF and decomposition in experiments feature a fixed MCS index of 23 and 500-byte packets.