Physical Hybrid-ARQ Indicator Channel

Description

The Physical Hybrid-ARQ Indicator Channel (PHICH) is a critical downlink physical channel in both LTE (Long-Term Evolution) and NR (New Radio) systems, specified from 3GPP Release 8 onwards. Its sole purpose is to carry the Hybrid Automatic Repeat reQuest (HARQ) acknowledgment indicators—specifically ACK (acknowledgment) or NACK (negative acknowledgment)—from the base station (eNodeB in LTE, gNB in NR) to the user equipment (UE). This feedback is sent in response to uplink data transmissions received on the Physical Uplink Shared Channel (PUSCH). The PHICH enables a fast, low-latency retransmission mechanism, which is essential for achieving high reliability and efficient use of the radio spectrum in the uplink direction.

Architecturally, the PHICH is mapped to specific resource elements within the downlink subframe. In LTE, it is transmitted in the control region of the subframe, typically occupying the first few Orthogonal Frequency-Division Multiplexing (OFDM) symbols. Multiple PHICHs are multiplexed together into PHICH groups to conserve control channel resources. Each PHICH within a group is distinguished by an orthogonal sequence (a Walsh code) spread across multiple resource elements. The key parameters defining a PHICH include the PHICH group number, the orthogonal sequence index within that group, and the PHICH duration (normal or extended). The eNodeB calculates these parameters based on the lowest Physical Resource Block (PRB) index of the corresponding PUSCH transmission and the DM-RS (Demodulation Reference Signal) cyclic shift used by the UE, ensuring a unique mapping.

How it works: When a UE transmits a data block on the PUSCH, it listens for a PHICH response in a predetermined downlink subframe (typically 4 ms later in LTE FDD). The eNodeB decodes the PUSCH transmission and generates an ACK if the data was decoded correctly or a NACK if it was not. This single-bit indicator is then BPSK-modulated, repeated, and spread with the orthogonal sequence before being mapped to the assigned resource elements. The UE, knowing its own PUSCH transmission parameters, can derive the exact PHICH resources to monitor. Upon receiving a NACK, the UE will retransmit the same or a redundant version of the data (incremental redundancy), following the synchronous HARQ process defined for the uplink. This closed-loop process continues until an ACK is received or a maximum number of retransmissions is reached.

Purpose & Motivation

The PHICH was created to address the fundamental challenge of ensuring reliable uplink data transmission over a noisy and fading wireless channel. Before HARQ with fast PHICH feedback, error correction relied more heavily on forward error correction (FEC) alone, which is less spectrum-efficient as it requires transmitting excessive redundancy for worst-case channel conditions. The PHICH enables a stop-and-wait HARQ protocol in the uplink, providing rapid feedback (ACK/NACK) that allows the UE to retransmit only when necessary. This dramatically improves uplink throughput and spectral efficiency by adapting to instantaneous channel conditions.

Historically, the design of PHICH in LTE Release 8 was motivated by the need for a low-latency control channel dedicated to HARQ feedback. Previous 3G systems like HSPA used dedicated channels or in-band signaling for similar purposes, but LTE's all-IP, OFDMA-based architecture required a new, efficient physical layer design. The PHICH solves the problem of providing timely and reliable feedback for multiple UEs simultaneously without consuming excessive downlink resources. Its group-based structure with orthogonal sequences allows multiplexing acknowledgments for many UEs onto a minimal set of resource elements.

The creation of the PHICH was driven by the overarching goals of LTE: higher data rates, lower latency, and improved spectral efficiency. By enabling fast physical-layer retransmissions (as opposed to slower RLC-layer retransmissions), the PHICH reduces the round-trip time for error recovery, which is critical for latency-sensitive applications. It forms the backbone of the uplink HARQ process, working in tandem with the PUSCH and uplink scheduling grants to create a robust and adaptive uplink data pipeline. In NR, the concept evolved but retained the same core purpose, adapting to more flexible numerology and slot structures.