Numerical Estimation
Digital compensation systems use a grid of pre-recorded values to correct the output of a sensor based on measured temperature and pressure. This look-up table interpolation fills the gaps between the calibration points stored in the non-volatile memory. It allows for the correction of complex non-linearities that a simple equation cannot model.
The process involves identifying the surrounding data points and calculating a weighted average.
Memory Constraint
Storing thousands of individual points is not practical for small embedded microcontrollers. Engineers use look-up table interpolation to maintain high accuracy with a limited set of data. A typical table might contain only ten pressure steps and five temperature steps.
Resolution Enhancement
Higher density tables reduce the error introduced by the mathematical estimation between points. Through look-up table interpolation, the sensor achieves a smooth output even when the raw signal is highly irregular. This technique is particularly effective for correcting the unique behavior of individual silicon dies.
Computational Speed
Calculating a result from a table is much faster than solving a high order polynomial equation. Look-up table interpolation enables high frequency sampling in modern industrial transmitters. This speed is required for control loops in chemical processing and aerospace.
