Wireless Transmitters Without Digital Pre-Distortion (DPD): Is It Possible?
In wireless communication systems, maintaining signal linearity is crucial for achieving high data rates without signal distortion. Traditionally, engineers rely on Digital Pre-Distortion (DPD) techniques to correct transmitter non-linearities. However, DPD comes with a significant drawback: increased power consumption. This has long posed a challenge for battery-operated devices such as IoT sensors and mobile transmitters, where power efficiency is paramount.
Recently, researchers from the Institute of Science Tokyo have introduced an innovative approach that eliminates the need for DPD while maintaining both linearity and efficiency. Their breakthrough, presented at the 2025 IEEE International Solid-State Circuits Conference (ISSCC), revolves around a novel CORDIC-less digital polar transmitter architecture incorporating three key design techniques.
The Problem with DPD
DPD is widely used to address linearity issues in wireless transmitters. It works by applying an inverse distortion pattern to the signal before amplification, effectively canceling out non-linear effects. However, implementing DPD requires additional circuit blocks and computational power, resulting in higher energy consumption and increased design complexity. This trade-off has been a persistent obstacle, especially in applications requiring both high data rates and power efficiency.
A CORDIC-Less Solution: Three Innovative Techniques
The researchers from Science Tokyo tackled this problem by removing the COordinate Rotation DIgital Computer (CORDIC) circuit, a common source of power inefficiency, and introducing three new techniques to ensure linearity without DPD:
- Delta-Sigma Modulation (DSM) with a Nine-State Look-Up Table (LUT):
Instead of using CORDIC to compute amplitude and phase, the system employs two DSMs to re-encode input signals into 3-level outputs. These outputs produce only nine possible amplitude-phase combinations, which are mapped using a simple nine-state LUT. This approach eliminates CORDIC’s power drain and maintains linear amplitude and phase modulation. - 1-Bit Amplitude Quantization for Linear Amplitude Control:
The second technique simplifies amplitude modulation by using 1-bit quantization. This allows the transmitter to toggle between zero and peak amplitude without intermediate states, thereby achieving perfect linearity in amplitude control without the need for power-intensive calibration. - Phase Multiplexing Using Square Wave Edges:
The third technique achieves linear phase modulation by generating eight distinct phases from a square wave running at four times the carrier frequency. By leveraging the wave’s rising and falling edges, the system produces the necessary phase shifts without complex interpolation, ensuring phase linearity without additional power overhead.
Proven Results Without DPD
The researchers implemented their digital transmitter design using a 65nm CMOS process and benchmarked its performance against state-of-the-art designs. The results were impressive: the new transmitter achieved superior power efficiency and high data rates without relying on DPD. This innovation effectively resolves the classic efficiency-linearity trade-off that has long plagued wireless transmitter designs.
The Future of Wireless Transmitters
This advancement opens new doors for RF engineers and telecommunications researchers seeking power-efficient transmitter designs for next-generation networks, IoT devices, and 6G communications. By demonstrating that linearity can be achieved without the power cost of DPD, the study from Science Tokyo provides a blueprint for more sustainable and efficient wireless technology.
As the demand for faster and more efficient wireless communication grows, solutions like this could redefine transmitter design, offering a path forward without the compromises of traditional DPD techniques.
Original Story via Innovative Design Techniques for Better Performance of Wireless Transmitters | Science Tokyo
