Abstract
This paper explores the relationship between performance of different generations of MCU solutions for classical computing processing, as well as FPU/DSP processing of input information. Finally, it questions how fast software-controlled I/O ports (GPIO) are. By using the battery of standardized computer tests, and custom GPIO, GPU / DSP tests, the report gives clear and measurable results.
The results of measuring carried out in the study clearly show that it is possible to unload the network resources necessary for transmitting large quantities of information if they represent a part of a complex pre-treatment data for Big Data processing systems on the MCU platforms themselves with IoT sensor devices.
Keywords
Microcontrollers - MCU, Internet of Things - IoT, SoC development systems, Cyber physical systems, Big Data, Sensor system, Sensor networks, RISC CPU Architecture, Floating point unit – FPU, Digital signal processing DSP, ARM, PIC32, Xtensa LX106, ESP32, NodeMCU IoT development boards, Arduino development boards, GPIO.