MCU's technical principles, differences and development history

Microcontrol unit (MCU), also known as single chip microcomputer (Single Chip Microcomputer) or single chip microcomputer, is to properly reduce the frequency and specifications of the central processing unit (CPU), and the memory (memory ), counter (Timer), USB, A / D conversion, UART, PLC, DMA and other peripheral interfaces, and even LCD driver circuits are integrated on a single chip, forming a chip-level computer, for different combinations of different applications. Such as mobile phones, PC peripherals, remote controls, to automotive electronics, industrial stepper motors, robotic arm control, etc., can be seen in the MCU.

Technical principle

The MCU is connected to the temperature sensor via an I2C bus. The I2C bus occupies two MCU input and output lines, and the communication between the two is completely dependent on software. The address of the temperature sensor can be set via 2 address pins, which allows 8 such sensors to be connected simultaneously on an I2C bus. In this scenario, the 7-bit address of the sensor has been set to 1001000. When the MCU needs to access the sensor, it must first issue an 8-bit register pointer and then issue the address of the sensor (7-bit address, low-order is the WR signal). There are three registers in the sensor for the MCU to use. The 8-bit register pointer is used to determine which register the MCU will use. In this scenario, the main program constantly updates the sensor's configuration registers, which causes the sensor to operate in single-step mode, and the temperature is measured every time it is updated.

To read the contents of the sensor measurement register, the MCU must first send the sensor address and register pointer. The MCU sends a start signal, then sends the sensor address, then sets the RD/WR pin high to read the measured value register.

In order to read the 16-bit data in the sensor measurement register, the MCU must perform two 8-bit data communication with the sensor. When the sensor is powered up, the default measurement accuracy is 9 bits and the resolution is 0.5 C/LSB (range -128.5 C to 128.5 C). This solution uses the default measurement accuracy. The sensor can be reset as needed to increase the measurement accuracy to 12 bits. If only a general temperature indication, such as a thermostat, is required, then a resolution of 1 C will suffice. In this case, the lower 8 bits of the sensor data can be ignored, and only the high 8 bits of data can be used to achieve the resolution 1 C design requirement. Since the register is read in the order of the first 8 bits and the lower 8 bits, the lower 8 bits of data can be read or not. There are two advantages to reading only the upper 8 bits of data. The first is to shorten the working time of the MCU and the sensor, and reduce the power consumption. The second is that it does not affect the resolution index.

After the MCU reads the measured value of the sensor, it will then convert and display the result on the LCD. The whole process includes: judging the sign of the display result, converting the binary code to the BCD code, and transmitting the data to the relevant register of the LCD.

After the data is processed and the results are displayed, the MCU issues a single-step command to the sensor. A single-step command causes the sensor to initiate a temperature test and then automatically enters standby mode until the analog-to-digital conversion is complete. After the MCU issues a single-step instruction, it enters LPM3 mode. At this time, the MCU system clock continues to work, and a timer interrupt is generated to wake up the CPU. The length of the timing can be adjusted programmatically to suit the needs of the specific application.

Main difference

Among the most commendable achievements of the 20th century, there are developments in integrated circuits and electronic computers. Microcomputers that emerged in the 1970s caused far-reaching changes in the science and technology world. In the mid-1970s, a small faction, the MCU, was split in the family of microcomputers. With the emergence of 4-bit microcontrollers, 8-bit microcontrollers were introduced. The emergence of the MCS48 series, especially the MCS51 series of single-chip microcomputers, established the status of the microcontroller as a microcontroller (MCU), which caused a new revolution in the field of microcomputers. In today's world, microprocessors (MPUs) and microcontrollers (MCUs) form two distinct branches. They are different from each other, but they merge and promote each other. Unlike the rapid development of microprocessors (MPUs) characterized by computational performance and speed, microcontrollers (MCUs) are marked by the continuous improvement of their control functions.

The CPU (Central Processing Unit) has developed three branches, one is DSP (Digital Signal Processing/Processor), and the other two are MCU (Micro Control Unit) and MPU. (Micro Processor Unit, microprocessor unit).

The MCU integrates on-chip peripheral devices; the MPU does not have peripheral devices (such as memory arrays), is a highly integrated general-purpose processor, and is an MCU with integrated peripherals removed; DSP has strong computing power and is good at many repeated data operations. The MCU is suitable for processing diagnostics and calculations of multiple data sources of different information sources, focusing on control, and the speed is not as good as DSP. The biggest difference between an MCU and a DSP is its versatility, which is reflected in the instruction set and addressing mode. The combination of DSP and MCU is DSC, which will eventually replace these two chips.

1. Support for dense multiplication operations

GPP is not designed to do dense multiplication tasks. Even some modern GPPs require multiple instruction cycles to do a multiplication. The DSP processor uses specialized hardware to implement single-cycle multiplication. The DSP processor also adds an accumulator register to handle the sum of multiple products. The accumulator registers are typically wider than the other registers, adding extra bits called result bits to avoid overflow. At the same time, in order to fully embody the benefits of specialized multiply-accumulate hardware, almost all DSP instruction sets contain explicit MAC instructions.

2. Memory structure

Traditionally, GPP uses Feng. Neumann memory structure. In this configuration, only one memory space is connected to the processor core through a set of buses (an address bus and a data bus). Typically, doing a multiplication will result in 4 memory accesses, using at least four instruction cycles.

Most DSPs use a Harvard architecture that divides the memory space into two, storing programs and data separately. They have two sets of buses connected to the processor core, allowing them to be accessed at the same time. This arrangement doubles the bandwidth of the processor memory and, more importantly, provides data and instructions to the processor core at the same time. In this layout, the DSP is able to implement a single-cycle MAC instruction.

A typical high-performance GPP actually contains two on-chip caches, one for data and one for instructions, which are directly connected to the processor core to speed up access at runtime. Physically, the on-chip dual memory and bus architecture is almost identical to the Harvard architecture. However, logically, there are still important differences between the two.

GPP uses control logic to determine which data and instruction words are stored in the on-chip cache, which programmers do not specify (or may not know at all). In contrast, DSPs use multiple on-chip memories and multiple sets of buses to ensure multiple accesses to memory per instruction cycle. When using a DSP, the programmer has to explicitly control which data and instructions are stored in on-chip memory. When programmers write programs, they must ensure that the processor can effectively use its dual bus.

In addition, the DSP processor has almost no data cache. This is because the typical data of the DSP is the data stream. That is to say, after the DSP processor calculates each data sample, it discards it and almost no longer uses it.

3. Zero overhead loop

If you understand a common feature of DSP algorithms, that is, most of the processing time is spent on executing smaller loops, it is easy to understand why most DSPs have dedicated hardware for zero-overhead loops. The so-called zero-overhead loop means that the processor does not take the time to check the value of the loop counter when the loop is executed, the condition is shifted to the top of the loop, and the loop counter is decremented by one.

In contrast, the GPP cycle is implemented using software. Some high-performance GPPs use branch-prediction hardware to achieve almost the same effect as hardware-supported zero-overhead loops.

4. Fixed point calculation

Most DSPs use fixed-point calculations instead of floating point. Although DSP applications must pay great attention to the accuracy of the numbers, it should be much easier to do with floating point, but for DSP, cheap is also very important. Fixed-point machines are cheaper (and faster) than the corresponding floating-point machines. In order to ensure the accuracy of numbers without using floating-point machines, DSP processors support saturation calculations, rounding and shifting in both instruction set and hardware.

5. Specialized addressing methods

DSP processors often support specialized addressing modes that are useful for common signal processing operations and algorithms. For example, module (loop) addressing (useful for implementing digital filter delay lines), bit-reverse addressing (useful for FFT). These very specialized addressing modes are not commonly used in GPP and are only implemented in software.

6. Forecast of execution time

Most DSP applications, such as cellular phones and modems, are strictly real-time applications, and all processing must be done within a specified time. This requires the programmer to accurately determine how much processing time each sample requires, or, at a minimum, know how much time is needed in the worst case. If you plan to use low-cost GPP to complete the task of real-time signal processing, the prediction of execution time will probably not become a problem. It should have a relatively straightforward structure for low-cost GPP, and it is easier to predict the execution time. However, the processing power required by most real-time DSP applications is not available in low-cost GPP. At this time, the advantage of DSP for high-performance GPP is that even if a cached DSP is used, which instructions are put in is determined by the programmer (not the processor), so it is easy to judge whether the instruction is from the cache or Read from memory. DSPs generally do not use dynamic features such as branch prediction and inferential execution. Therefore, it is completely straightforward to predict the required execution time from a given piece of code. This allows the programmer to determine the performance limitations of the chip.

7. Fixed point DSP instruction set

The fixed-point DSP instruction set is designed with two goals: to enable the processor to perform multiple operations per instruction cycle, thereby increasing the computational efficiency of each instruction cycle. Minimize memory space for storing DSP programs (this is especially important in cost-sensitive DSP applications due to the high cost of memory for the overall system). To achieve these goals, the DSP processor's instruction set typically allows the programmer to specify several parallel operations within an instruction. For example, an instruction contains a MAC operation, that is, one or two data movements at the same time. In the typical example, an instruction contains all the operations needed to calculate a section of the FIR filter. The cost of this high efficiency is that its instruction set is neither intuitive nor easy to use (compared to the GPP instruction set). GPP programs usually don't care if the processor's instruction set is easy to use because they generally use high-level languages ​​like C or C++. For DSP programmers, unfortunately the main DSP applications are written in assembly language (at least in part, assembly language optimized). There are two reasons for this: First, most widely used high-level languages, such as C, are not suitable for describing typical DSP algorithms. Second, the complexity of DSP architectures, such as multiple memory spaces, multiple buses, irregular instruction sets, highly specialized hardware, etc., make it difficult to write efficient compilers for them. Even with the compiler compiling C source code into DSP assembly code, the optimization task is still heavy. Typical DSP applications have a large number of computational requirements and strict overhead constraints, making program optimization essential (at least for the most critical part of the program). Therefore, a key factor in considering the choice of DSP is whether there are enough programmers who can better adapt to the DSP processor instruction set.

8. Requirements for development tools

Because DSP applications require highly optimized code, most DSP vendors offer development tools to help programmers do their optimization work. For example, most vendors offer simulation tools for the processor to accurately simulate the activity of the processor during each instruction cycle. These are useful tools for ensuring real-time operations or code optimization. GPP vendors usually do not provide such tools, mainly because GPP programmers usually do not need to go to this level of information. The lack of simulation tools that are accurate to the instruction cycle of GPP is a big problem for DSP application developers: it is almost impossible to predict how high-performance GPPs need to cycle for a given task, thus failing to explain how to improve the performance of the code.

Development History

The history of single-chip microcomputers is not long, but the development is very rapid. Its generation and development are basically in sync with the generation and development of microprocessors. Since Intel Corporation first introduced 4-bit microprocessors in 1971, its development has so far been roughly divided into five stages. The following is a representative of Intel's microcontroller development.

1971-1976

The initial stage of the development of microcontroller. In November 1971, Intel first designed the Intel 4004, a 4-bit microprocessor with 2000 transistors/chips. It was equipped with RAM, ROM and shift registers to form the first MCS-4 microprocessor. An 8-bit microprocessor, the Intel 8008, and an 8-bit microprocessor from other companies have been introduced.

1976-1980

Low performance microcontroller stage. The MCS-48 series introduced by Intel Corporation in 1976 is a monolithic structure that integrates an 8-bit CPU, an 8-bit parallel I/O interface, an 8-bit timer/counter, RAM, and ROM on a semiconductor chip. Its addressing range is limited (not more than 4 KB), there is no serial I/O, RAM, ROM capacity is small, and the interrupt system is simpler, but the function can meet the needs of general industrial control and intelligent instruments and meters.

1980-1983

High-performance microcontroller stage. The high-performance 8-bit MCUs introduced at this stage generally have a serial port, a multi-stage interrupt processing system, and multiple 16-bit timer/counters. The on-chip RAM and ROM have an increased capacity and an addressing range of up to 64 KB. Individual slices also have an A/D conversion interface.

Late 1983-80s

16-bit microcontroller stage. In 1983, Intel introduced the high-performance 16-bit microcontroller MCS-96 series. Due to its latest manufacturing process, the chip integrates up to 120,000 transistors/chip.

1990s

The single-chip microcomputer develops to a higher level in all aspects of integration, function, speed, reliability, and application fields.

According to the characteristics of single-chip microcomputer, the application of single-chip microcomputer is divided into single-machine application and multi-machine application. In an application system, using only one microcontroller is called a stand-alone application. The scope of single-chip applications of single-chip microcomputers includes:

(1) Measurement and control system. With the single-chip microcomputer, various less complicated industrial control systems, adaptive control systems, data acquisition systems, etc. can be constructed to achieve the purpose of measurement and control.

(2) Smart meters. The original measurement and control instrument will be transformed by the single-chip microcomputer to promote the development of the instrument in the direction of digitization, intelligence, multi-function, integration and flexibility.

(3) Mechatronics products. The combination of single-chip microcomputer and traditional mechanical products makes the structure of traditional mechanical products simple and intelligent.

(4) Intelligent interface. In the computer control system, especially in the larger industrial measurement and control system, the control and management of the interface with the single-chip microcomputer, combined with the parallel operation of the single-chip microcomputer and the host computer, greatly improves the operating speed of the system.

(5) Intelligent civilian products. For example, in many products such as household appliances, toys, game consoles, audio-visual equipment, electronic scales, cash registers, office equipment, kitchen equipment, etc., the introduction of single-chip controllers not only greatly enhances the functions of the products, but also improves their performance. A good use effect.

The multi-machine application system of single-chip microcomputer can be divided into function distribution system, parallel multi-machine processing and local network system.

(1) Functional distribution system. The multi-functional distribution system is a multi-machine system set up to meet the requirements of various peripheral functions of the engineering system.

(2) Parallel multi-machine control system. Parallel multi-machine control system mainly solves the rapid problem of engineering application system, so as to constitute a large-scale real-time engineering application system.

(3) Local network system.

The single-chip microcomputer can be divided into general-purpose and special-purpose according to the application range. Dedicated models are designed for a specific product, such as a microcontroller for a thermometer, a microcontroller for a washing machine, and so on. In the general-purpose single-chip microcomputer, it can be divided into 4 bits, 8 bits, 16/32 bits according to the word length, although the computer microprocessor is now almost 32/64 bit world, 8-bit, 16-bit microprocessor. It has tended to shrink, but the situation of single-chip microcomputer is different. 8-bit single-chip microcomputer has low cost, low price and easy development. Its performance can meet most needs. Only in high-tech fields such as aerospace, automobile and robot, when high-speed processing of large amounts of data is required. Only need to select 16/32 bit, and in the general industrial field, 8-bit general-purpose single-chip microcomputer is still the most widely used single-chip microcomputer.

So far, China's single-chip application and embedded system development has gone through more than 20 years of history. As embedded systems gradually penetrate into all aspects of social life, the teaching of single-chip courses also goes from the traditional 8-bit processor platform to 32. The trend of the advanced RISC processor platform is changing, but the 8-bit machine is still difficult to replace. The national economic construction, military and household appliances and other fields, especially mobile phones, auto navigation devices, PDAs, smart toys, smart home appliances, medical equipment and other industries are all industries in which MCU talent is urgently needed. At present, there are more than 100,000 engineers engaged in the development and application of single-chip microcomputers in the industry. However, in the face of the trend of industrialization of embedded systems and the opportunity of China to promote the construction of “embedded software factories”, China’s embedded products should be integrated into the international market. Forming an industry, there will be an urgent need for a large number of MCU application talents, which provides a great opportunity for higher vocational students to engage in such high-tech industries.