IRFFT

Inverse Radio Frequency Fast Fourier Transform

Physical Layer
Introduced in Rel-15
The Inverse Fast Fourier Transform operation applied in the context of Radio Frequency signal generation, specifically for generating OFDM/OFDMA time-domain waveforms from frequency-domain symbols. It is a core mathematical operation in the transmitter chain of LTE and 5G NR.

Description

The Inverse Radio Frequency Fast Fourier Transform (IRFFT) is a critical digital signal processing (DSP) operation in the physical layer transmitter of Orthogonal Frequency Division Multiplexing (OFDM) and Orthogonal Frequency Division Multiple Access (OFDMA) based systems like LTE and 5G New Radio (NR). It is the inverse operation of the RFFT (Radio Frequency FFT) performed at the receiver. The IRFFT's function is to transform a block of complex-valued modulation symbols, mapped onto specific subcarriers in the frequency domain, into a corresponding time-domain waveform that can be converted to an analog signal and transmitted over the air.

Technically, the process begins after the stages of channel coding, modulation (e.g., QPSK, 16QAM, 64QAM, 256QAM), and resource element mapping in the transmitter chain. The output of resource mapping is a vector of complex symbols for a set of active subcarriers within an OFDM symbol period, with null values assigned to guard bands and the DC subcarrier. The IRFFT algorithm computes the Inverse Discrete Fourier Transform (IDFT) of this frequency-domain vector. For computational efficiency, the IDFT is implemented using the Inverse Fast Fourier Transform (IFFT) algorithm, with the size (N) of the IFFT determining the total number of subcarriers (including nulls) in the system's bandwidth. For example, LTE commonly uses a 2048-point IFFT for a 20 MHz channel.

The output of the IRFFT is a discrete-time sequence representing the composite OFDM symbol in the time domain. A critical subsequent step is the addition of a Cyclic Prefix (CP). The CP is created by copying a portion from the end of this time-domain symbol and prepending it to the beginning. This CP mitigates inter-symbol interference (ISI) caused by multipath propagation. The final time-domain signal, comprising consecutive CP-OFDM symbols, is then converted to analog, up-converted to the carrier frequency, amplified, and transmitted. The IRFFT is thus the mathematical engine that enables the parallel transmission of data on multiple orthogonal subcarriers, providing robustness against frequency-selective fading and forming the foundation of modern broadband wireless air interfaces.

Purpose & Motivation

The IRFFT exists as a fundamental enabling technology for OFDM, which was selected as the core modulation scheme for 4G LTE and 5G NR due to its superior performance in multipath environments. Prior to the widespread adoption of OFDM, systems like 3G UMTS used Wideband Code Division Multiple Access (WCDMA), which struggled with inter-symbol interference in high-delay-spread channels, requiring complex equalizers at the receiver. The purpose of the IRFFT (and the overall OFDM transmitter chain) is to transform a difficult frequency-selective fading wideband channel into a collection of many parallel, flat-fading narrowband subchannels.

This solves several key problems: It greatly simplifies channel equalization at the receiver (reducing to simple per-subcarrier amplitude/phase correction), provides inherent robustness against multipath delay spread through the use of the cyclic prefix, and enables flexible multi-user access (OFDMA) via allocation of different sets of subcarriers to different users. The motivation for specifically defining it in 3GPP specifications (e.g., for performance testing) is to ensure interoperability and consistent RF characteristics. By standardizing the expected output of the IRFFT process (including parameters like IFFT size, sampling rate, and spectral mask compliance), 3GPP guarantees that transmitters from different vendors produce waveforms that are correctly decoded by any standards-compliant receiver.

Historically, the computational complexity of the IFFT/FFT was a barrier, but advances in DSP and integrated circuit technology made it feasible for consumer devices. The formalization of IRFFT in 3GPP Release 15 for NR also accommodated new waveform parameters, such as flexible numerology (different subcarrier spacings) and support for wider bandwidths, which required precise definition of the transform sizes and windowing functions to maintain orthogonality and control out-of-band emissions.

Key Features

  • Transforms frequency-domain modulation symbols into a time-domain OFDM waveform
  • Implemented via computationally efficient Inverse Fast Fourier Transform (IFFT) algorithms
  • IFFT size (N) determines the system's total number of subcarriers
  • Processes input vectors containing data symbols, reference signals, and null subcarriers
  • Output is prepared for Cyclic Prefix (CP) insertion
  • A core, standardized operation ensuring transmitter waveform compliance

Evolution Across Releases

Rel-15 Initial

Formally defined in the context of 5G NR waveform generation. Specified the flexible numerology, allowing for multiple subcarrier spacings (15, 30, 60, 120, 240 kHz). The IRFFT operation was adapted to support these different numerologies, implying different IFFT sizes and sampling rates for a given channel bandwidth. It formed the basis for the CP-OFDM waveform in both downlink and uplink.

Enhanced support for integrated access and backhaul (IAB) and operation in unlicensed spectrum (NR-U). The IRFFT specifications ensured proper waveform generation for these new deployment scenarios, including adherence to specific spectral masks and power constraints required in unlicensed bands.

Extended to support non-terrestrial networks (NTN) and reduced capability (RedCap) devices. For NTN, considerations for very large delay spreads influenced CP length requirements, indirectly related to the time-domain structure created by the IRFFT. For RedCap, simplified implementations may be considered.

Under 5G-Advanced, studies on advanced antenna systems and potential waveform enhancements. While CP-OFDM remains core, research into complementary waveforms or windowing/filtering applied post-IRFFT for improved spectral containment is ongoing.

Continued refinement and optimization for extreme MIMO systems and higher frequency bands (e.g., upper mmWave), ensuring efficient and compliant waveform generation for massive antenna arrays and very wide bandwidths.

Defining Specifications

SpecificationTitle
TS 26.118 3GPP TS 26.118