Integer scaled maths used on devices without a floating-point unit.
Without an FPU, fractions are stored as scaled integers in Qm.n format, e.g. Q15: value=int/32768. Multiplication needs shifts; watch overflow and rounding. Faster and smaller than software floating point but requires careful scaling.
A PID loop uses Q16.16 arithmetic on a Cortex-M0.
Fixed-point arithmetic lets small processors do fractional maths efficiently, and it is still used in many embedded and audio systems.
Integer scaled maths used on devices without a floating-point unit.
Fixed-point arithmetic lets small processors do fractional maths efficiently, and it is still used in many embedded and audio systems.
A notation that sets how many bits are fractional, such as Q15 where the value is an integer divided by 32768.
Overflow and loss of precision, so scaling and saturation must be designed carefully.
Typical applications include implementing a filter on a microcontroller without a floating-point unit, scaling sensor values to Q15 format for a DSP routine and converting a temperature reading to degrees with integer maths.
Study "Digital signal processors" first (topic 90 of 104 in Electronics & Embedded). It covers the groundwork this topic builds on.
Q-format notation. Saturate on overflow. Integer division is slow.
A PID loop uses Q16.16 arithmetic on a Cortex-M0.
The next topic is "Embedded security", which builds directly on this one.
It is classed as Advanced in Electronics & Embedded (topic 91 of 104). It is an advanced topic. It assumes you are comfortable with the intermediate topics in Electronics & Embedded.
Want to practise? Open the interactive course for quick checks, quizzes, simulations and progress tracking.