How does Class S differ from Class D?

Class D is a topological hardware designation; it refers to using transistors purely as push-pull switches. Class S is an architectural designation; it defines how the signal is prepared before it hits the switch. Specifically, Class S uses a Delta-Sigma modulator to encode the amplitude data into a dense stream of uniform digital pulses. The Class S architecture typically uses a Class D hardware switch to physically amplify that pulse stream.

What is the 'Holy Grail' of the digital transmitter?

Historically, an RF transmitter requires an analog mixer, a local oscillator, and a linear power amplifier, all of which are inefficient and difficult to integrate on a digital silicon chip. Class S promises the 'Digital Transmitter.' The baseband processor mathematically encodes the RF signal directly into a digital bitstream. That bitstream drives a switch. A passive filter reconstructs it. No analog mixers, no complex biasing, just pure digital math driving an antenna at 80%+ efficiency.

Why isn't Class S used in every cell phone?

The required switching speeds are physically impossible with current technology. To accurately use Delta-Sigma modulation for a 2 GHz RF carrier, the digital clock rate must oversample the carrier by roughly 4x. This means the transistor must switch cleanly on and off at 8 GHz. At these speeds, the parasitic capacitance of the transistor prevents it from fully turning off, destroying the digital square wave. Furthermore, the massive amount of quantization noise generated by the Delta-Sigma modulator must be aggressively filtered out, requiring bulky, lossy output filters.

Digital RF Architecture

Class S Amplifier

The telecommunications industry wants to eliminate bulky analog RF components and move toward purely digital transmitters. The proposed solution is the Class S amplifier. Instead of sending an analog sine wave into a linear amplifier, the baseband processor uses a Delta-Sigma (ΔΣ) modulator to digitally chop the complex 5G signal into a blistering, high-speed train of 1s and 0s. When the original analog wave is at its peak, the modulator outputs a dense cluster of 1s. When the wave is at a valley, it outputs mostly 0s. This pure digital bitstream is fed directly into a switch-mode transistor (operating in Class D). Because the transistor is only ever fully ON or fully OFF, theoretical efficiency is 100%. Finally, the amplified digital pulses pass through a bandpass filter, which integrates the dense clusters of 1s and 0s back into the smooth, high-power analog RF signal right before the antenna. It is the ultimate fusion of digital signal processing and RF power.

Category: Digital RF Architecture

Modulation: Delta-Sigma (ΔΣ) / Pulse Density

Primary Barrier: Requires extreme clock oversampling rates

Switch-Mode Encoding Techniques

Encoding Technique	How it represents amplitude	Clock Requirement	Output Filter Needed
PWM (Pulse Width Modulation)	Varies the width of a single pulse per cycle	Moderate	Low-pass filter
PPM (Pulse Position Modulation)	Shifts the timing of uniform pulses	High	Bandpass filter
ΔΣ (Class S / Pulse Density)	Varies the density of rapid, uniform pulses	Extreme (Oversampled)	High-Q Bandpass filter

Oversampling Requirement for Class S:
f_clock ≥ 4 · f_carrier
To push the massive quantization noise out of the operating band, a Delta-Sigma modulator must run at a clock speed significantly faster than the RF carrier frequency. For a 2 GHz cellular signal, the digital bitstream must be clocked at a minimum of 8 GHz.

Coding Efficiency Penalty:
Unlike a continuous wave, a digital bitstream contains a massive amount of broadband noise (quantization noise). Even if the transistor is 100% efficient at amplifying the bitstream, a large percentage of that amplified power is just noise that will be blocked by the output filter. This "coding efficiency" drops the overall system efficiency from 100% down to roughly 60%.

Common Questions

Frequently Asked Questions

What is Quantization Noise?

When you convert a smooth analog wave with infinite values into a digital signal with only two values (0 and 1), you introduce mathematical errors. These errors manifest as harsh broadband RF noise. The genius of Delta-Sigma modulation is that it pushes this quantization noise away from the carrier frequency and into the higher frequencies, making it easy to filter out before it reaches the antenna.

Why does the output filter need to be High-Q?

Because the amplified signal coming out of the transistor is mostly high-frequency quantization noise. If you transmit that noise, you will illegally jam every radio frequency around you. You must use a very sharp, High-Q bandpass filter to trap the noise and only let the reconstructed analog carrier through. Designing a filter that is sharp enough to block the noise but capable of handling 50 watts of power without melting is a major engineering bottleneck.

Is Class S commercially viable yet?

For audio frequencies (Class D audio amps using Class S architecture), yes, it is the global standard. For high-frequency RF, it remains largely in the research phase. The transistors cannot switch fast enough to support 4G/5G carrier frequencies, and the power lost to the output filter (coding efficiency loss) currently negates the efficiency gained by using a switch-mode transistor.

Related Terms

← Class F Amplifier Coaxial Cable →

Digital Transmitter

Delta-Sigma Modulator Simulator

Upload a clean analog RF waveform. Set your oversampling ratio and watch the Delta-Sigma engine encode it into a pulse-density bitstream. Visualize the resulting quantization noise spectrum and design the required bandpass filter.

Simulate Pulse Density