Inverse Discrete Fourier Transform

Description

The Inverse Discrete Fourier Transform (IDFT) is a critical signal processing algorithm in digital communications, particularly for multi-carrier modulation schemes like Orthogonal Frequency Division Multiplexing (OFDM). In the context of 3GPP LTE and NR, the transmitter uses the IDFT to transform a block of complex-valued modulation symbols, which are mapped to specific subcarriers in the frequency domain, into a corresponding sequence of time-domain samples. This time-domain sequence constitutes the baseband OFDM symbol ready for further processing (e.g., addition of cyclic prefix, digital-to-analog conversion, and up-conversion to radio frequency).

The operation is mathematically defined. For an input sequence of N complex frequency-domain symbols X[k], where k = 0, 1, ..., N-1, the IDFT computes the output time-domain sequence x[n] as: x[n] = (1/√N) * Σ_{k=0}^{N-1} X[k] * e^{(j2πkn/N)} for n = 0, 1, ..., N-1. Here, N corresponds to the size of the transform, which is directly related to the number of active subcarriers in an OFDM symbol. In practical implementations, this is computed efficiently using the Inverse Fast Fourier Transform (IFFT) algorithm, which reduces computational complexity from O(N²) to O(N log N). The 3GPP specifications define the necessary transform sizes (e.g., 128, 256, 512, 1024, 2048 for LTE) corresponding to different system bandwidths.

For the uplink, LTE introduced a variation known as DFT-spread OFDM (DFT-s-OFDM), also called Single-Carrier FDMA (SC-FDMA). Here, the process involves an additional step: a DFT precoding is applied to the time-domain modulation symbols first, spreading their energy across all subcarriers, before the larger IDFT (IFFT) is performed. This reduces the Peak-to-Average Power Ratio (PAPR) compared to conventional OFDM, which is beneficial for User Equipment (UE) power amplifier efficiency. In 5G NR, the IDFT remains central for both the CP-OFDM waveform (used in downlink and uplink) and the DFT-s-OFDM waveform (an uplink option). NR supports a more flexible numerology with variable subcarrier spacing, which is realized by adjusting the underlying IDFT/IFFT parameters and sampling rate.

The proper application of the IDFT ensures orthogonality between subcarriers, which is the key feature that allows densely packed, overlapping subcarriers to be demodulated without inter-carrier interference (ICI) in a time-invariant channel. The output of the IDFT forms the core of the physical layer signal generation chain. Its inverse operation, the Discrete Fourier Transform (DFT), is performed at the receiver to convert the received time-domain samples back into frequency-domain symbols for demodulation and decoding.

Purpose & Motivation

The IDFT is employed to solve the problem of efficient and robust data transmission over frequency-selective fading channels. Before the widespread adoption of OFDM, single-carrier modulation schemes struggled with high data rates because the channel's frequency selectivity would cause severe inter-symbol interference (ISI), requiring complex equalizers. Multi-carrier modulation, realized via the IDFT/DFT pair, divides a high-rate data stream into many parallel low-rate streams, each transmitted on a narrowband subcarrier. This makes each subcarrier experience a relatively flat fade, simplifying channel equalization to a per-subcarrier scaling operation.

The historical motivation stems from the need for spectral efficiency and resilience to multipath propagation. The IDFT enables the creation of orthogonal subcarriers whose spectra overlap, achieving a higher spectral efficiency than non-overlapping multi-carrier systems. This orthogonality, maintained as long as the channel does not destroy it (addressed by the cyclic prefix), allows simple frequency-domain processing at the receiver. The computational feasibility of the IFFT/FFT algorithms made this theoretically elegant approach practical for real-time communication systems.

In 3GPP standards, the adoption of OFDM (and DFT-s-OFDM for uplink) from LTE Release 8 onwards was a fundamental architectural break from the CDMA-based 3G systems. It was driven by the requirements for higher data rates, better spectral efficiency, and more flexible bandwidth allocation. The IDFT is the engine that makes this waveform possible, and its parameters (size, sampling rate) are carefully standardized to ensure interoperability between base stations and user equipment across the globe, while providing the flexibility needed for diverse deployment scenarios from narrowband IoT to wideband enhanced Mobile Broadband (eMBB).