Resource Block Group

Description

A Resource Block Group is a fundamental unit for resource allocation in the frequency domain within LTE and 5G NR's OFDMA-based air interface. A Physical Resource Block (PRB) is the smallest unit of resource that can be allocated to a user, defined as 12 consecutive subcarriers for one slot (0.5 ms in LTE). However, signaling the allocation of individual PRBs for every user in every subframe would create excessive control channel overhead. To mitigate this, the standard defines RBGs, where a group of contiguous PRBs is treated as a single allocatable entity. The scheduler in the gNB (or eNB) allocates one or more RBGs to a UE via a bitmap in the Downlink Control Information (DCI).

The size of an RBG (P) is not fixed; it is a function of the system bandwidth. 3GPP specification 36.213 defines a table mapping the number of downlink resource blocks (system bandwidth) to the RBG size. For example, for a system bandwidth of 10 MHz (50 PRBs), the RBG size is typically 3 PRBs. This results in approximately 17 RBGs (50/3, rounded up). The DCI format for resource allocation type 0 uses a bitmap where each bit corresponds to an RBG. A bit set to '1' indicates that the corresponding RBG is allocated to the UE. This method, known as Resource Allocation Type 0, provides a compact representation. For smaller bandwidths or more granular allocations, Resource Allocation Type 1 allows allocation of a subset of PRBs within an RBG, but still uses RBGs as a reference structure.

RBG-based allocation directly impacts scheduling flexibility and efficiency. Larger RBG sizes reduce DCI overhead but decrease scheduling granularity, which can lead to inefficient resource utilization, especially for users with small data packets or at the cell edge requiring frequency-selective scheduling gain. Smaller RBG sizes offer finer granularity but increase signaling cost. The network can configure the preferred allocation type. In 5G NR, the concept is retained and enhanced. NR defines a Resource Block Group size for bandwidth part (BWP)-based scheduling, and the DCI can indicate resource allocation using a resource indication value (RIV) for contiguous allocations or a bitmap for non-contiguous, with the granularity tied to the PRB and the configured BWP.

Purpose & Motivation

The Resource Block Group mechanism was created to solve the critical problem of control channel overhead in OFDMA-based cellular systems like LTE and NR. These systems offer high flexibility by allowing dynamic allocation of frequency resources (PRBs) to multiple users every transmission time interval (TTI). Signaling the exact set of PRBs for each user would require a large number of bits in the downlink control channel (PDCCH), consuming significant resources that could otherwise be used for data transmission and limiting the number of users that can be scheduled simultaneously. RBGs provide a practical compromise between scheduling granularity and signaling efficiency.

Introduced in LTE Release 8, RBGs addressed the limitations of simpler allocation schemes used in prior technologies. 3G UMTS used code division multiple access (CDMA) with spreading factors, which had different overhead characteristics. The shift to OFDMA in LTE demanded a new method for efficient frequency-domain resource indication. The RBG concept allowed the network to trade off between fine-grained, frequency-selective scheduling (beneficial for channel-aware scheduling in frequency-selective fading channels) and the capacity of the control channel. It enabled the support of wider system bandwidths (up to 20 MHz in LTE Rel-8) without a proportional explosion in control signaling.

In 5G NR (from Release 15), the purpose of RBGs remains, but within a more flexible framework. NR supports much wider bandwidths (up to 400 MHz) and diverse use cases. The RBG concept is applied within a Bandwidth Part (BWP), which is a subset of the total carrier bandwidth configured for a UE. This allows for efficient scheduling even for UEs that do not support the full carrier bandwidth. The configurable RBG size within a BWP allows network operators to optimize the overhead/granularity trade-off specifically for the services (e.g., massive IoT with small packets vs. eMBB with large bursts) active on that BWP, which is essential for network slicing and service-specific optimization.

Key Features

• Overhead Reduction: Groups multiple PRBs into a single schedulable unit, significantly reducing the number of bits required in DCI for resource allocation.
• Configurable Size: RBG size (P) is determined by system bandwidth (LTE) or bandwidth part size (NR), allowing a trade-off between granularity and overhead.
• Bitmap-Based Allocation: In Resource Allocation Type 0, a simple bitmap in DCI, where each bit represents an RBG, indicates non-contiguous resource grants.
• Bandwidth-Dependent Definition: The number of RBGs in the system scales with the total available bandwidth, maintaining consistent overhead scaling.
• Foundation for Scheduling: Enables both contiguous and non-contiguous frequency-domain scheduling strategies within the control channel constraints.
• NR Enhancement: Applied within the context of a Bandwidth Part (BWP) in 5G NR, providing flexibility for UE-specific bandwidth configurations and network slicing.

Evolution Across Releases

Defining Specifications