Engineering

Engineering

Signaloid Cloud Compute Engine

Signaloid Cloud Compute Engine

September 2025

September 2025

Benchmarking

Thermal Array Sensor Calibration with the Melexis MLX90640

Sensors convert a physical signal into a voltage that an analog-to-digital converter (ADC) digitizes, and a calibration algorithm specific to each sensor converts that digital value into a measurement such as a temperature in degrees Celsius. Rounding to the nearest ADC count adds quantization error to every reading, and the emissivity of the measured surface is rarely known exactly, yet systems today report the calibrated temperature as a single number. This example runs the unmodified calibration routines of the Melexis MLX90640 32x24 far-infrared thermal sensor array [1]. It models the ADC quantization error of each raw pixel reading and an emissivity between 0.93 and 0.97 as uniform distributions, and computes the distribution of the calibrated temperature of the center pixel.

Figure 1. ...

When running on the Signaloid Cloud Compute Engine (SCCE), the calibration routines replace the usual approach of sampling followed by evaluation of each sample with a direct computation on a representation of the probability distribution. In a single computation, the SCCE computes the same output distribution that traditional Monte Carlo simulations require millions of iterations to achieve.

Figure 2. ...

The MLX90640 calibration kernel, running on a single-threaded Signaloid C0Pro-Jupiter-S core, achieves runtimes 1640× faster than an optimized C-language Monte-Carlo-based implementation (i.e., the industry standard and status quo) of the same kernel running on an Amazon EC2 R7iz instance. With 95% confidence, a 4M-iteration Monte Carlo implementation matches the accuracy of the Signaloid UxHw-based version. If higher confidence is required for the Monte Carlo results, the speedup of the UxHw approach becomes even more pronounced.

[1] https://www.melexis.com/en/product/MLX90640/Far-Infrared-Thermal-Sensor-Array

Key Performance Indicator

Signaloid Platform Solution

Competing Solution

Signaloid Benefit

Speed for the same uncertainty quantification accuracy.

Run existing non-Monte-Carlo code and use either the Signaloid Compute Engine's automated ingestion of distribution information, or use the Signaloid UxHw API to set program variables as probability distributions.

Run existing Monte Carlo code, or, starting from non-Monte-Carlo code, modify code to implement Monte Carlo sampling, iteration, and aggregation of the results from the Monte Carlo iterations of the computation.

1640x faster execution time than 4.0M-iteration Monte Carlo, while achieving same fidelity of full distribution result.

Benchmarking Methodology

Monte Carlo simulations work by statistical sampling and therefore each multi-iteration Monte Carlo run will result in a slightly different output distribution. By contrast, Signaloid's platform is deterministic and each run produces the same distribution for a given Signaloid C0 core type.

The performance improvements are calculated by comparing Signaloid's platform with a Monte Carlo simulation of a similar quality of distribution. First, we run a large Monte Carlo simulation (about 50M iterations) on an AWS r7iz high-performance AWS instance: We use this distribution result as a baseline or ground truth reference of distribution quality. Then we calculate the performance of Signaloid's technology, and compare it with the performance of a Monte Carlo iteration count where the output distribution's Wasserstein distance (to the output distribution of the ground truth reference) is as accurate as the Signaloid-core-executed algorithm's output distribution, with 95% confidence level.

Performance data based on Fall 2025 release of Signaloid's technology.

Relevant Signaloid Solutions

Schedule a Demo Call
Request Whitepaper
Schedule a Demo Call
Request Whitepaper
Schedule a Demo Call
Request Whitepaper