Quasi Co-Location

Description

Quasi Co-Location (QCL) is a fundamental concept in 3GPP New Radio (NR) that defines an assumed relationship between different reference signal antenna ports or between a reference signal port and a data channel port. When two antenna ports are configured as QCL, the UE is allowed to assume that certain large-scale properties of the radio channel experienced on the first port can be inferred and applied to assist in the reception of signals on the second port. These large-scale properties include parameters like Doppler shift, Doppler spread, average delay, delay spread, and spatial Rx parameters (which relate to the receiving beam). This assumption significantly reduces the complexity and time required for channel estimation, particularly for channels like the Physical Downlink Shared Channel (PDSCH).

The specification defines several QCL types (Type A, B, C, D) in 38.214, each allowing the inference of a different subset of these large-scale parameters. For example, Type A includes Doppler shift, Doppler spread, average delay, and delay spread. Type D is particularly critical for beam management as it includes spatial Rx parameters, meaning the UE can assume the same receive beam can be used for ports with a Type D QCL relationship. In practice, the gNB configures the UE with Transmission Configuration Indicator (TCI) states via RRC signaling and/or MAC CE activation. Each TCI state contains information that links a target reference signal (like a CSI-RS or SS/PBCH block) to a QCL type and a source reference signal. The UE then uses measurements from the source RS to derive channel estimates for the target RS or the PDSCH.

Architecturally, QCL is essential for enabling efficient beamformed transmission, especially in Frequency Range 2 (FR2 - mmWave). Due to high path loss at these frequencies, communication relies on narrow, high-gain beams. QCL Type D allows the gNB to indicate that the PDSCH is transmitted using the same beam (and thus similar spatial characteristics) as a previously measured CSI-RS or SSB. The UE can then apply the same receive beamforming weights, avoiding an exhaustive beam search for every transmission. This is managed through beam management procedures (P-1, P-2, P-3) and is tightly integrated with the control signaling for scheduling grants (where the TCI state is indicated in the DCI).

Purpose & Motivation

QCL was introduced in NR (Rel-15) to address the significant challenges of channel estimation and beam management in advanced MIMO and millimeter-wave systems, which were not sufficiently handled by LTE's antenna port quasi co-location framework. In LTE, QCL assumptions were simpler and implicit for many ports, but NR's use of massive beamforming, wider bandwidths, and higher frequencies created a scenario where the channel characteristics for different reference signals could be vastly different, especially if they were transmitted from different analog beams or different TRPs (Transmission Reception Points). Without explicit QCL relationships, the UE would need to perform independent, complex channel estimation for every signal, increasing latency, power consumption, and reducing reliability.

The primary problem QCL solves is enabling efficient UE receiver processing in a highly dynamic beamformed environment. It allows the network to explicitly inform the UE about which reference signals are 'alike' in terms of their channel statistics, so the UE can reuse prior measurements. This is critical for achieving low latency in beam switching and tracking, which is vital for maintaining connectivity for mobile users in mmWave bands where beams are narrow. It also facilitates advanced multi-TRP and coordinated multipoint (CoMP) operations by allowing the network to define relationships between signals from different geographical points, providing a flexible framework for managing spatial diversity and multiplexing gains in 5G networks.

Key Features

• Defines assumed relationships between antenna ports for large-scale channel properties
• Supports multiple QCL types (A, B, C, D) for different parameter subsets
• Enabled by TCI state configuration via RRC and MAC CE signaling
• Crucial for beam management, especially QCL Type D for spatial Rx parameters
• Reduces UE channel estimation complexity and latency
• Facilitates multi-TRP and carrier aggregation operations

Evolution Across Releases

Defining Specifications