An embedded processor is a microprocessor designed especially for handling the needs of an embedded system. The processor is a heart of an embedded system. Processors get the user data as an input, process them and produce an output based on the function. The processor fetches the instruction or program code, decodes the instruction and executes it.
This process is carried out by the control unit and execution unit of the embedded processor. The execution unit comprises of arithmetic and logical unit (ALU) which include a circuit that executes the instruction sets used to perform the operation inside processors.
As you know, embedded systems have unique characteristics such as less power consumption, smaller size, less user interface, etc. Embedded systems are classified into several types based on their functional performance, complexity and behavior. Inside each embedded system requires specific embedded processors to operate.
There are different types of embedded processors available in the market from different manufacturers. Those processors are used for different applications of embedded systems. Here, you can learn the different types of embedded processor in detail.
General Purpose Processors (GPP)
The general-purpose processors are the embedded processors used for general purpose applications that are not specific to any applications. It processes the instructions based on the predefined general-purpose instruction set. The processor inside the computer or laptop in your hand is a typical example of a general-purpose processor.
The microprocessor is a single VLSI chip that has the CPU and has several external units such as caches, registers, memory, floating-point processing units and advanced processing units like pipelining and superscalar processing for faster and efficient operation.
The CPU fetches an instruction from a set of general-purpose instructions, decodes and executes it. The instruction set consists of instructions for arithmetic and logic operations, data transfer operations, control operations, etc.
Z80 is an 8-bit microprocessor introduced by Zilog as the company’s first product. 8085 is an 8-bit microprocessor introduced to the market by Intel in March 1976. They upgraded the 8-bit to 16-bit processor in the name of the 8086 processor.
Now intel has a variety of processors in their hands being the front leader in processor manufacturing for computers and laptops. SAMA5D2, SAMA5D3, SAMA5D4 are the family of 32-bit microprocessors from Microchip. More number of processors are available for different types of applications, developed by top manufacturers like Intel, Texas, ARM, Altera, NXP, etc.
The microcontroller is a type of general-purpose processor. It has an integrated chip that has a processor, memory, registers, caches, timer units, interrupt control units and I/O ports. The system with a microcontroller is sometimes called a microcomputer. Since the microprocessor consists of all the functional blocks for its working, it occupied a huge place in the embedded system.
Intel’s 8051 is the popular microcontroller during the 1980-1990s. Most of the embedded systems during that period utilized the 8051 controllers. After that, the microcontrollers evolved a lot in the embedded market through different manufacturers.
PIC microcontroller is an embedded processor used for several new applications such as smartphones, audio accessories, video gaming peripherals and advanced medical devices. PIC16, PIC18 are the featured PIC families by Microchip.
The AVR microcontrollers offer unsurpassed performance, power efficiency and flexibility. Hence they are an excellent choice for a variety of embedded system designs. AVR DB, AVR DA, ATmega4809, ATtiny1607 are some of the featured AVR microcontroller families from Microchip.
ARM-based microcontrollers like S32K, EA microcontroller series, MAC57Dxxx Microcontrollers are products from NXP semiconductors used for automotive applications. The ARM-based MCUs, C2000 Real-time MCUs and MSP430 MCUs from Texas Instruments have real-time control capabilities and high-precision analog integration are optimized for industrial and automotive applications.
Intel, Philips, Atmel, ST Micro Electronics, Maxim are some of the other key manufacturers of microcontrollers.
Digital Signal Processor
Digital signal processor is a special-purpose embedded processor designed to meet the computational and power demands of audio, video and communication applications. They are either 8 bit / 16 bit / 32 bit microprocessors, designed for faster operation than the general-purpose embedded processor.
Digital signal processors have seen tremendous growth in the recent decade, finding use in everything from smart telephones to advanced scientific instruments. Modern DSP architecture achieves high performance and low power through increased parallelism and reduces power dissipation.
Sum of Products(SOP) calculation, convolution, Fast Fourier transform, Discrete Fourier transform, Finite/Infinite Impulse Response(FIR/IIR), etc. are some of the operations performed by the digital signal processors.
DSP56Fxx and MSC8xx are the two families of recent digital signal processors from NXP semiconductors. Their processors are targeted at audio/video (AV) receivers, home theaters, surround sound decoders, mini stereo systems, digital TV audio systems and automotive audio systems.
ARM has also developed DSP processors such as ARM cortex-A, ARM Cortex-R, ARM Cortex-M with enhanced DSP instructions to support improved execution performance for arithmetic operations. Their processors provide a high level of signal processing.
The programmable DSPs from Texas Instruments operates in a variety of embedded real-time signal processing applications including audio and aerospace & defense. They range from low-power, single-core processors to high-performance multi-core DSP plus ARM SoCs. TMS320C6748, TMS320C6657, TMS320C6678 are few of their featured DSPs in their market.
ST Microelectronics, Analog Devices, Cirrus Logic, Infineon Technologies are some of the other top manufacturers of DSP processors.
Single Purpose Embedded Processor
As the name suggests, the single-purpose embedded processors are used in an embedded system only for a single unique purpose. They act independently by providing the necessary information to the host processor. It includes a co-processor for floating-point processing, network processor, graphics processor, pixel processor, accelerator, CODEC, etc.
Floating-point co-processors are used to speed up the floating-point operations. MC68881 and MC6882 are the floating-point co-processors from NXP semiconductors.
The network processor is used for the operation of establishing the network connection, sending and receiving acknowledgments, sending transmission requests and checking for any errors in the data transmission. C-5™ NP is the network processor manufactured by NXP semiconductors. WinPath network processor family from Microchip is purposely designed and optimized for access networks.
The graphics processor is used in a system for high-quality gaming functions, viewing videos and images and processing them. AMD and Intel and the top manufacturers of the graphics processor.
Similarly, CODEC is a single processor unit, which is used to encode the information or data which is to be stored or transmitted and decode the received information.
Application Specific Integrated Circuits (ASICs)
Application-specific Integrated Circuit is an integrated circuit designed for a particular use, rather than intended for general-purpose use. ASICs are typically designed and used by a single company for a specific system.
Such embedded processors need more time and more resources to develop the integrated circuit and hence more expensive but give extremely high performance. Modern embedded systems are fabricated with ASICs and contain over one billion transistors. They are also fabricated with Metal oxide semiconductor technology.
Modern ASICs often include microprocessors, memory blocks and other building blocks, which is often termed a system-on-chip(SoC). Altera, Honeywell, Intel, Texas Instruments, ST Microelectronics are some of the manufacturers of ASICs.
Application Specific Standard Products (ASSPs)
It is an integrated circuit that is used for a specific function, applicable for a wide range of markets for the long run. ASSPs are available as off-the-shelf components, which provide easy integration and interoperability with existing system components.
ASSPs are used in all industries, starting from automotive applications to communication devices. It is embedded in control units, microwave electronics, analog to digital converters to communications, ultra-violet detectors.
The major advantage of ASSPs are that they are cheap and cut down the development time to a greater extent. Hence it reduces the time to market such embedded systems. ASIC is sometimes called as ASSP in a way that, when the IC is sold to many consumers, we say it as ASSP.
A system-on-chip is an integrated circuit that consists of all or most components of a computer or other electronic system on a single chip. It consists of hardware units and a microprocessor or microcontroller which has software code for the hardware to function.
The hardware components include, microprocessor, memory, timers, interrupt controllers, i/o ports, modems, A/D and D/A converters, other processing units. A typical SoC has more than one processor core, a Digital signal processor, network processor, different communication interfaces, RAM, ROM, flash memory, etc.
88MW32X, MC34SB0800, QN908x are some of the system-on-chip devices manufactured by NXP semiconductors. Altera, Analog Devices, Infineon, MIPS technologies, Texas Instruments are some of the manufacturers of SoCs.
Field Programmable Gate Array (FPGA)
The processors that are discussed above are built on silicon cores. The architecture of early FPGA devices was relatively simple. Field Programmable Gate Arrays (FPGAs) are devices that are built around an array of configurable logic blocks linked via programmable interconnects.
The good thing about an FPGA is that we can configure its programmable logic to implement any combination of digital functions. Also, we can implement algorithms in a parallel fashion, which means it can process a huge amount of data in a fast and efficient manner.
The modern FPGA device consists of thousands of adders, multipliers, and digital signal processing (DSP) functions, on-chip memory, large numbers of high-speed serial interconnect blocks and so on.