Online Tool for Rapid ΔΣ Modulator Design

Delta-Sigma (ΔΣ) Analog-to-Digital Converters (ADCs) are widely used in signal processing applications because of their ability to achieve high resolution and dynamic range. This is accomplished through oversampling and noise shaping, which moves quantization noise out of the signal band, making them highly efficient for precision measurements. Depending on the application requirements, ΔΣ ADCs have different configurations such as Discrete-Time (DT), Continuous-Time (CT), Bandpass, and Incremental ΔΣ ADCs.

DT DSMs offer a precise control over the loop dynamics and are easier to simulate and design with predictable performance. Their switched-capacitor implementation provides high linearity and stability. DT DSMs are typically used as audio ADCs, instrumentation, and low-speed, high-resolution applications.

CT DSMs are better suited for high-speed applications due to their inherent sampling at the output of the modulator. They consume less power than DT DSMs because the loop filter operates in continuous time, reducing operational amplifier (op-amp) bandwidth requirements. CT DSMs inherently provide anti-aliasing functionality due to their continuous-time nature. CT DSMs are typically used in high-speed communications (e.g. wireless receivers) or as broadband ADCs.

CT Bandpass (BP) ΔΣ Modulators shape quantization noise away from a specific frequency band, making them ideal for modulating signals centered around non-zero frequencies. They are critical for systems where the signal of interest is a narrowband signal modulated at intermediate or radio frequencies. BP ΔΣ Modulators are typically used in radio receivers, narrowband communication systems, and software-defined radios.

Incremental ΔΣ Modulators (I-DSM) operate on a finite-duration input signal and, therefore, work as Nyquist ADCs. They are resettable, making them ideal for scenarios like multi-channel measurements where each input requires isolated conversion. Furthermore, they offer a lower latency compared to the freerunning DSMs making them more suitable for usage in control loops. They are typically used in sensor readout systems, multi-channel data acquisition, and low-frequency precision applications.

The design process for DSMs is often tedious and highly dependent on the knowledge and experience of the designer. In this project, an easy-to-use design tool is developed that allows the direct design of DSMs. It relies on a heuristic search based on a genetic algorithm. To achieve very short response times in the range of seconds for a complete optimization, the algorithms are implemented on a GPU to allow a heavily parallelized execution of the simulations. In addition, the tool incorporates several non-idealities to provide more than just a rough estimate. No advanced knowledge of loop-filters, modelling and transformations is required.

The web-based GUI for this design tool is available under
https://www.sigma-delta.de

Due to the demand for more sophisticated architectures and modulator types, the scope of the toolbox continues to grow. In addition, the flexibility of the underlying models and the effectiveness of the optimization algorithm are constantly being increased to extend the range of applications.