Configured Grant Uplink Control Information

Description

Configured Grant Uplink Control Information (CG-UCI) is a critical mechanism within the 5G New Radio (NR) physical layer uplink control channel framework, specified in 3GPP TS 38.212. It refers to the Uplink Control Information (UCI) that a User Equipment (UE) transmits using pre-configured, semi-persistent uplink resources known as Configured Grants (CG), specifically Type 1 or Type 2 Configured Grants. Unlike dynamically scheduled UCI on the Physical Uplink Control Channel (PUCCH) or multiplexed on the Physical Uplink Shared Channel (PUSCH), CG-UCI is transmitted on a Configured Grant Physical Uplink Shared Channel (CG-PUSCH) resource. This means the UE does not require a dynamic Downlink Control Information (DCI) grant from the gNodeB for each transmission instance, enabling grant-free uplink access.

The operation of CG-UCI is intrinsically linked to the Configured Grant (CG) framework introduced for uplink data. When a CG resource is activated for a UE via Radio Resource Control (RRC) signaling (Type 1) or via a combination of RRC and an activating DCI (Type 2), the UE is provided with a periodic resource allocation on the PUSCH. The UE can use these resources to transmit uplink data without a scheduling request (SR) or grant, reducing latency. CG-UCI leverages these same pre-allocated resources to piggyback essential control information. The UCI payload for CG-UCI typically includes Hybrid Automatic Repeat Request Acknowledgement (HARQ-ACK) feedback for downlink transmissions, Channel State Information (CSI) reports, and Scheduling Request (SR) indications. This control information is multiplexed with the uplink data on the CG-PUSCH according to the physical layer multiplexing rules defined in TS 38.212.

Architecturally, CG-UCI resides within the UE's physical layer processing chain. The Medium Access Control (MAC) layer provides the HARQ-ACK and SR information, while the physical layer measurements generate CSI. This information is encoded, rate-matched, and multiplexed with the data transport blocks before being mapped to the allocated CG-PUSCH resource's time-frequency grid. The gNodeB, aware of the CG configuration, monitors these periodic resources to decode both the data and the embedded CG-UCI. Key components enabling CG-UCI include the CG configuration parameters (periodicity, time/frequency resources, modulation and coding scheme), the UCI encoding procedures, and the rules for resource sharing between data and control bits on the PUSCH.

CG-UCI's role in the network is pivotal for supporting Ultra-Reliable Low-Latency Communications (URLLC) and efficient uplink operation. By eliminating the need for a scheduling request and grant cycle for control feedback, it drastically reduces the control plane latency for uplink acknowledgements and reports. This is essential for applications like industrial automation, where timely HARQ-ACK for critical downlink commands is necessary. Furthermore, it improves spectral efficiency and UE power efficiency by allowing control and data to be transmitted together in a single, pre-arranged transmission burst, avoiding separate control channel transmissions.

Purpose & Motivation

CG-UCI was created to address the stringent latency and reliability requirements of 5G, particularly for URLLC use cases. Prior to 5G NR Release 16, uplink control information was primarily transmitted on dedicated PUCCH resources or multiplexed on dynamically scheduled PUSCH. Both approaches incur latency: PUCCH requires periodic dedicated resources which might not align with the need for immediate feedback, while dynamic scheduling on PUSCH requires a full scheduling request (SR) and grant procedure, adding several milliseconds of delay. For time-critical applications like closed-loop control in factory automation or tele-surgery, this latency was prohibitive.

The motivation for CG-UCI stems from the parallel development of the Configured Grant (grant-free) uplink data transmission. While CG allowed data to be sent without scheduling latency, there was a missing link: how to provide the necessary accompanying control feedback (like HARQ-ACK for downlink assignments) with equally low latency. Transmitting this feedback on a separate, dynamically scheduled resource would negate the latency benefits of grant-free data. CG-UCI solves this by enabling the UE to send this vital control information within the same grant-free transmission opportunity as its data, using the pre-configured resources. This creates a cohesive low-latency uplink path for both data and control.

Historically, LTE offered Semi-Persistent Scheduling (SPS) for uplink, but its control feedback mechanisms were not as tightly integrated or optimized for the microsecond-scale latencies targeted by 5G URLLC. CG-UCI, as part of the enhanced Configured Grant framework, represents a fundamental architectural shift to support deterministic, low-latency uplink communication, moving beyond the reactive, grant-based paradigm of previous generations.

Key Features

• Grant-free transmission of UCI on pre-configured PUSCH resources
• Multiplexing of HARQ-ACK, CSI, and SR with uplink data on CG-PUSCH
• Reduction of control-plane latency by eliminating SR/grant cycle for feedback
• Essential enabler for URLLC service requirements
• Improved UE power efficiency via combined data and control transmission
• Support via both Type 1 (RRC-configured) and Type 2 (DCI-activated) Configured Grants

Evolution Across Releases

Defining Specifications