High Reliability Protocol

Description

The High Reliability Protocol (HRP) is a 3GPP-specified protocol stack enhancement designed to meet the extreme reliability (e.g., 99.9999%) and low latency (e.g., 1 ms) requirements of Ultra-Reliable Low-Latency Communication (URLLC) services. It functions as a shim layer between the application and the standard IP/UDP transport layers, adding reliability mechanisms that are more aggressive and deterministic than those provided by TCP or conventional UDP-based protocols. Architecturally, HRP entities reside in the User Equipment (UE) and the User Plane Function (UPF) or a dedicated application server. It is designed to work in conjunction with 5G QoS mechanisms, leveraging QoS Flows with guaranteed bit rate (GBR) and priority marking.

HRP works by implementing a suite of reliability-enhancing techniques. A core mechanism is packet duplication, where identical data packets are sent over multiple independent paths (e.g., different PDU sessions, different radio bearers, or even different access technologies like 5G NR and LTE). The receiving HRP entity performs duplicate detection and discards redundant copies. HRP also employs advanced Forward Error Correction (FEC) schemes, adding redundant parity information to data blocks so that the receiver can reconstruct original data even if some packets are lost, without waiting for retransmissions. For latency control, HRP uses tight timers and may implement a selective repeat ARQ (Automatic Repeat reQuest) mechanism with a very short round-trip time expectation, but it primarily relies on FEC and duplication to avoid the latency penalty of retransmissions wherever possible.

The protocol is defined to be configurable based on the specific reliability and latency targets of the application. Network functions like the Session Management Function (SMF) can configure HRP parameters (e.g., duplication factor, FEC scheme) as part of PDU Session Establishment or modification, based on the subscribed QoS profile. HRP also includes sequence numbering and time-stamping for packet ordering and latency monitoring. Its operation is detailed in specifications such as TS 23.725 for architecture, TS 26.260 for media delivery, and TS 38.825 for radio aspects. By providing these functions at a higher layer, HRP adds an end-to-end reliability guarantee that complements the lower-layer reliability provided by the 5G New Radio (NR) air interface and RLC/MAC protocols, creating a multi-layered defense against packet loss for mission-critical communications.

Purpose & Motivation

HRP was created to address the gap between the reliability provided by traditional Internet protocols (TCP/IP) and the extreme requirements of emerging 5G vertical applications such as factory automation, remote surgery, and autonomous vehicle coordination. TCP, with its retransmission-based reliability and congestion control, introduces unpredictable and often excessive latency, making it unsuitable for sub-10ms applications. UDP offers low latency but no reliability. While the 5G NR air interface and QoS framework provide strong foundations for URLLC, packet loss can still occur due to congestion in the transport network (e.g., between UPF and application server) or from rare but catastrophic radio link failures.

The motivation for HRP stems from the need for an end-to-end, transport-layer solution that works seamlessly over 5G networks, including the core and transport segments. Previous industrial solutions often used proprietary Layer 2 protocols or rigid wired networks (e.g., PROFINET, EtherCAT). HRP provides a standardized, IP-based protocol that can leverage the flexibility and scalability of 5G while meeting hard real-time constraints. It solves the problem of achieving 'six-nines' reliability over wireless links and potentially lossy IP networks by proactively sending redundant information (via duplication and FEC) rather than reactively retransmitting lost data. This paradigm shift from reactive to proactive error recovery is key to achieving both high reliability and ultra-low latency simultaneously, enabling the wireless replacement of critical wired connections in industrial and vehicular settings.

Key Features

• Packet duplication over multiple independent transmission paths
• Advanced Forward Error Correction (FEC) for loss recovery without retransmission
• Configurable reliability and latency profiles aligned with 5G QoS Flows
• End-to-end operation between UE and application server/UPF
• Sequence numbering and duplicate detection for data integrity
• Tight integration with 5G PDU session and QoS management for resource reservation

Evolution Across Releases

Defining Specifications