Inferensys

Glossary

Carrier Amplifier

The primary amplifier stage in a Doherty configuration, typically biased in Class-AB, that operates continuously and handles signal amplification up to the transition point where the peaking amplifier activates.
Stylish WeWork-like workspace with hot desks and document wall, professional searching through enterprise knowledge base on a mounted ultrawide display, warm industrial pendants overhead.
DOherty Architecture

What is a Carrier Amplifier?

The carrier amplifier is the primary, continuously operating amplification stage within a Doherty power amplifier architecture, responsible for handling signal amplification up to a defined transition point.

The carrier amplifier is the main amplification device in a Doherty configuration, typically biased in Class-AB to provide linear amplification of the input signal at all power levels. It operates continuously, handling the signal envelope from zero up to the transition point where the peaking amplifier activates. This biasing ensures reasonable linearity and gain for low-level signals while maintaining acceptable efficiency, forming the foundational signal path upon which the Doherty load modulation mechanism depends.

As the input drive increases, the carrier amplifier approaches voltage saturation, and its load impedance is dynamically modulated by the current injected from the peaking amplifier through the impedance inverter network. This active load-pull effect maintains the carrier at peak efficiency over a wide output back-off (OBO) range. The carrier's inherent nonlinearities, including AM-AM distortion and AM-PM distortion, are primary targets for digital predistortion (DPD) linearization to ensure overall transmitter compliance with ACLR and EVM specifications.

DOherty Architecture Fundamentals

Key Characteristics of a Carrier Amplifier

The carrier amplifier is the continuously operating backbone of the Doherty power amplifier. Understanding its biasing, linearity profile, and interaction with the impedance inverter is essential for effective digital predistortion optimization.

01

Class-AB Biasing Operation

The carrier amplifier is biased in a shallow Class-AB mode, representing a deliberate trade-off between linearity and efficiency. Unlike a deep Class-A bias, this quiescent current allows the device to handle the lower envelope of the signal with acceptable linearity while consuming less DC power.

  • Quiescent Current: Typically set between 5-15% of the peak drain current.
  • Conduction Angle: Operates for slightly more than 180 degrees of the RF cycle.
  • Efficiency Profile: Achieves peak efficiency at the back-off point, not at saturation.
  • Linearity: Provides the baseline linearity for the entire Doherty architecture before the peaking amplifier activates.
5-15%
Typical Idq (% of Imax)
>180°
Conduction Angle
02

Continuous Signal Handling

Unlike the peaking amplifier, which remains in cutoff during low-power operation, the carrier amplifier is always active. It is solely responsible for amplifying the signal envelope up to the transition point where the peaking amplifier turns on.

  • Low-Power Region: Handles the entire signal below the transition threshold.
  • Voltage Swing: Reaches its maximum voltage swing precisely at the transition point, defining the first efficiency peak.
  • Thermal Stability: Experiences a more constant thermal profile than the peaking amplifier, though self-heating memory effects still require compensation.
03

Load Modulation Target

The carrier amplifier is the primary beneficiary of the active load-pull effect generated by the Doherty combiner. As the peaking amplifier injects current, the impedance seen by the carrier is dynamically transformed from a high value (typically 2*Ropt) down to the optimal load (Ropt).

  • Impedance Trajectory: Moves from 2*Ropt at low power to Ropt at peak power.
  • Voltage Saturation: The load modulation forces the carrier into voltage saturation early, maintaining high efficiency.
  • Current Linearity: The fundamental current of the carrier remains roughly linear with input drive, simplifying the behavioral model.
04

AM-AM and AM-PM Distortion Source

While generally more linear than a Class-C peaking stage, the carrier amplifier introduces significant soft compression and phase distortion that must be corrected by digital predistortion. The nonlinear capacitance (Cgs, Cgd) varies with input drive, causing AM-PM conversion.

  • Soft Compression: Gradual gain reduction before hard saturation, characteristic of GaN HEMT devices.
  • Input Impedance Modulation: The nonlinear input capacitance causes phase shift that varies with envelope power.
  • Memory Effects: Thermal trapping and bias circuit interactions introduce short-term and long-term memory, requiring a memory polynomial or Volterra-based DPD model.
05

Impedance Inverter Interaction

The carrier amplifier sees the output load through an impedance inverter, typically a quarter-wave transmission line. This network is critical for transforming the modulated load impedance and ensuring the carrier's efficiency peak aligns with the back-off power level.

  • Quarter-Wave Transformer: Converts the high impedance at the combiner node to a low impedance at the carrier's drain.
  • Phase Alignment: The electrical length of this path must be precisely calibrated to ensure constructive combining with the peaking branch.
  • Bandwidth Limitation: The frequency-dependent nature of the quarter-wave inverter is a primary constraint on the bandwidth of the Doherty amplifier.
06

Efficiency Peak at Back-Off

The carrier amplifier achieves its maximum drain efficiency not at the peak envelope power, but at the Output Back-Off (OBO) level corresponding to the transition point. This is the fundamental mechanism enabling high average efficiency for high-PAPR signals.

  • First Efficiency Peak: Occurs typically at 6 dB OBO in a symmetric Doherty design.
  • Voltage Waveform: At this point, the voltage waveform is a full half-sinusoid while the current is half of its peak, minimizing overlap loss.
  • DPD Relevance: The predistorter must account for the rapid efficiency and gain slope change around this transition region.
CARRIER AMPLIFIER ESSENTIALS

Frequently Asked Questions

Clarifying the operational principles, biasing strategies, and critical design parameters of the primary amplifier stage in a Doherty configuration.

A carrier amplifier is the primary, continuously operating amplification stage in a Doherty power amplifier architecture, typically biased in Class-AB mode. It functions as the main signal amplification path, handling the entire input signal envelope up to a specific transition point. Below this point, the carrier amplifier operates alone, delivering power to the load through an impedance inverter. As the input signal envelope increases beyond the transition threshold, the carrier amplifier begins to compress, entering saturation, while the peaking amplifier activates. The carrier's role is to maintain linear amplification for low-to-medium power levels while enabling the active load-pull effect that modulates its load impedance for high efficiency during back-off operation. Its continuous operation ensures signal integrity across the entire dynamic range, making its linearity and gain characteristics fundamental to the overall system's error vector magnitude (EVM) performance.

Prasad Kumkar

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.