Automotive engine control systems, implantable medical devices, remote controls, office equipment, appliances, power tools, toys, and other embedded systems are just a few examples of the automatically controlled goods and gadgets that use microcontrollers. Microcontrollers make it affordable to digitally control even more devices and processes since they are smaller and less expensive than designs that require individual microprocessors, memories, and input/output devices. In order to control non-digital electronic equipment, mixed signal microcontrollers are frequently used. Microcontrollers are a popular and affordable method of data collection, sensing, and controlling the physical world as edge devices in the context of the internet of things. A system on a chip is similar to a microcontroller in modern parlance, but it is less complex. However, a system on a chip typically combines cutting-edge peripherals like a graphics processing unit (GPU) and a Wi-Fi interface controller as its internal microcontroller unit circuits. A system of the chip may connect external microcontroller chips as motherboard components.
At this time, the majority of microcontrollers had concurrent variations. One contained an EPROM program memory that could be wiped by ultraviolet radiation thanks to a transparent quartz glass in the package’s lid. For prototyping, these erasable chips were frequently employed. The other version was either a mask-programmed ROM or a PROM variant which was only programmable once. For the latter, the abbreviation OTP, which stands for “one-time programmable,” was occasionally used. The PROM was typically the same type as the EPROM in an OTP microcontroller, but the chip packaging lacked a quartz window, making it impossible to expose the EPROM to UV light, which prevented it from being erased.
Because RAM and non-volatile memory are integrated on the same chip as the CPU in microcontrollers, they may not incorporate an external address or data bus. The semiconductor can be housed in much smaller, less expensive packaging by using fewer pins. It is more expensive to integrate the memory and other peripherals onto a single chip and test them together, but doing so frequently lowers the net cost of the embedded system as a whole. Even though the price of a CPU with integrated peripherals may be slightly higher than the price of a CPU with external peripherals, having fewer chips usually enables a smaller, less expensive circuit board, reduces the labor needed to assemble and test the circuit board, and also tends to lower the defect rate for the finished assembly. Due to this integration, fewer chips, less wiring, and less circuit board space are required to create identical systems than if separate chips were used. Additionally, each pin on devices with a low pin count, in particular, may connect to a number of internal peripherals, with the pin function being determined by software. As opposed to if pins had specific functions, this enables a part to be used in a larger range of applications. Since their invention in the 1970s, microcontrollers have proven to be extremely popular in embedded systems.
Some microcontrollers employ the Harvard design, which separates the memory buses for instructions and data to enable parallel accesses. When a Harvard architecture is employed, the bit size of the processor’s instruction words may differ from the size of internal memory and registers; an example would be 12-bit instructions used with 8-bit data registers.
It can be challenging to choose which peripheral to integrate. The operating frequencies and system design flexibility of microcontroller vendors are frequently exchanged for customer-driven time-to-market demands and overall lower system costs. Manufacturers must strike a compromise between the need to reduce chip size and the addition of new functionality.
