DFTS

Discrete Fourier Transform Spread Orthogonal Frequency Division Multiplexing

Physical Layer
Introduced in Rel-8
DFTS is a modulation and multiple access scheme used in the LTE uplink, combining DFT spreading with OFDM. It creates a single-carrier waveform with low peak-to-average power ratio (PAPR), enabling efficient power amplifier operation in user equipment. This technology is crucial for extending battery life and improving uplink coverage in mobile broadband systems.

Description

DFTS-OFDM, often referred to as Single-Carrier FDMA (SC-FDMA) in LTE contexts, is the fundamental physical layer transmission scheme for the LTE uplink. The architecture begins with the generation of complex-valued modulation symbols (QPSK, 16QAM, 64QAM) in the frequency domain. These symbols are then mapped to a set of contiguous or non-contiguous subcarriers assigned to a specific user. A key component is the M-point DFT (Discrete Fourier Transform) precoder, which spreads the energy of each modulation symbol across all allocated subcarriers, transforming the multi-carrier OFDM signal into a single-carrier-like waveform.

The technical operation involves several stages: First, the transmitter performs an M-point DFT on blocks of M modulation symbols, where M corresponds to the number of allocated subcarriers. The resulting frequency-domain samples are then mapped to the N-point IFFT (Inverse Fast Fourier Transform) inputs, where N is the total system bandwidth in subcarriers. Zero symbols are inserted for unallocated subcarriers. After IFFT transformation, a cyclic prefix is added to mitigate multipath interference, and the time-domain signal is transmitted. This process creates a signal with single-carrier properties despite using OFDM infrastructure.

The receiver performs the inverse operations: cyclic prefix removal, N-point FFT to return to frequency domain, extraction of the allocated subcarriers, and finally an M-point IDFT to recover the original modulation symbols. The single-carrier characteristic is maintained through the DFT spreading, which ensures all subcarriers carry a linear combination of all original symbols rather than independent modulation on each subcarrier. This architecture allows DFTS-OFDM to leverage OFDM's robustness to frequency-selective fading while maintaining the low PAPR advantages of single-carrier transmission.

In the LTE network architecture, DFTS-OFDM operates within the physical uplink shared channel (PUSCH) for data transmission and physical uplink control channel (PUCCH) for control signaling. The scheme supports both localized transmission (contiguous subcarriers) and distributed transmission (non-contiguous subcarriers with frequency hopping) to provide frequency diversity. The resource allocation granularity is one resource block (12 subcarriers × 0.5 ms slot), and multiple resource blocks can be allocated to a single user to support various data rates. The technology's integration with LTE's scheduling and link adaptation mechanisms enables dynamic resource allocation based on channel conditions and quality of service requirements.

Purpose & Motivation

DFTS-OFDM was developed specifically for the LTE uplink to address the critical challenge of power amplifier efficiency in user equipment. Traditional OFDM used in the downlink exhibits high peak-to-average power ratio (PAPR), requiring power amplifiers with large back-off to avoid nonlinear distortion. This results in poor power efficiency, which directly impacts battery life in mobile devices. By creating a single-carrier-like waveform through DFT spreading, DFTS-OFDM achieves significantly lower PAPR (approximately 2-3 dB lower than OFDM), enabling more efficient power amplifier operation and extended battery life.

The historical context for DFTS-OFDM development emerged from the limitations of previous 3GPP technologies. WCDMA used in UMTS had reasonable PAPR characteristics but suffered from poor spectral efficiency and limited support for frequency-domain scheduling. The 3GPP community sought a technology that could provide OFDM's advantages in multipath environments and frequency-domain scheduling while overcoming OFDM's high PAPR drawback for uplink transmission. DFTS-OFDM represented an elegant solution that maintained compatibility with OFDM receiver structures while transforming the transmit signal characteristics.

Beyond power efficiency, DFTS-OFDM addresses coverage limitations in cellular networks. The improved power amplifier efficiency translates to higher effective transmit power for the same battery consumption, extending uplink coverage particularly at cell edges. This was especially important for LTE's goal of providing consistent high-data-rate coverage. The technology also simplified receiver design at the base station, as the single-carrier property reduced sensitivity to amplifier nonlinearities in the user equipment, leading to more predictable signal quality at the receiver.

Key Features

  • Low peak-to-average power ratio (PAPR) for power amplifier efficiency
  • Single-carrier waveform properties with multi-carrier transmission benefits
  • Frequency-domain equalization capability at receiver
  • Support for both localized and distributed resource allocation
  • Compatibility with OFDM infrastructure and receiver structures
  • Scalable bandwidth support from 1.4 MHz to 20 MHz

Evolution Across Releases

Rel-8 Initial

Introduced DFTS-OFDM as the exclusive uplink multiple access scheme for LTE, supporting QPSK, 16QAM, and 64QAM modulation schemes. Initial implementation included both localized and distributed transmission modes with frequency hopping capabilities. The architecture was optimized for power amplifier efficiency with PAPR characteristics similar to single-carrier transmission while maintaining OFDM's robustness to multipath fading.

Defining Specifications

SpecificationTitle
TS 36.300 3GPP TR 36.300