Beginner’s Guide: Learning Embedded Systems from a Decade of Experience

Embedded systems are a type of “computer system designed for specific applications, fully embedded within controlled devices.” According to the definition by the U.K. Institution of Electrical Engineers, embedded systems are used for controlling, monitoring, or assisting devices, machines, or equipment used in factory operations. Unlike general-purpose computer systems such as personal computers, embedded systems typically perform predefined tasks with specific requirements. Since embedded systems are aimed at a single specialized task, designers can optimize them to reduce size and costs. Embedded systems are often mass-produced, so savings on a single unit can be multiplied by the production volume.

Embedded systems are application-centric, based on computer technology, and have customizable hardware and software, suitable for dedicated computer systems that have strict requirements for functionality, reliability, cost, size, and power consumption. They generally consist of four parts: embedded microprocessors, peripheral hardware devices, embedded operating systems, and user applications, used to control, monitor, or manage other devices.

The design techniques for embedded systems mainly include hardware design techniques and software design techniques. The hardware design field primarily includes chip-level design techniques and circuit board-level design techniques.The core of chip-level design techniques involves compilation/synthesis, libraries/IP, and testing/verification. Compilation/synthesis techniques allow designers to describe the required functions abstractly and automatically analyze and insert implementation details. Library/IP techniques use pre-designed low-level implementations for high-level tasks. Testing/verification techniques ensure the correctness of each functional level, reducing the costs of redesigning between levels.

Next, we will introduce some basic development processes.

Embedded System Development Process

Step 1: Establish Development EnvironmentThe operating system is generally Redhat Linux, choose custom installation or full installation, and download the corresponding GCC cross-compiler through the network (e.g., arm-linux-gcc, arm-uclibc-gcc), or install the relevant cross-compiler provided by the product manufacturer.Step 2: Configure Development HostConfigure MINICOM, with general parameters being baud rate 115200 Baud/s, data bits 8, stop bits 1, 9, no parity, and no software or hardware flow control. The configuration in Windows’ HyperTerminal is the same. The purpose of MINICOM software is to serve as a monitor for debugging the output of embedded development board information and a tool for keyboard input. Network configuration primarily involves setting up the NFS network file system, requiring the firewall to be disabled to simplify the embedded network debugging environment setup process.Step 3: Establish BootloaderDownload some publicly available source code bootloaders from the internet, such as U.BOOT, BLOB, VIVI, LILO, ARM-BOOT, RED-BOOT, etc., and modify them according to the specific chip. Some chips do not have a built-in bootloader, such as Samsung’s ARV17 and ARM9 series chips, so a burning program for the FLASH on the development board needs to be written. You can download the corresponding burning program online, as well as publicly available source code J-FLASH programs under Linux. If you cannot burn your development board, source code modifications must be made according to your specific circuit. This is the first step to ensure the system can operate normally. If users purchase the manufacturer’s emulator, it is easier to burn FLASH, although they may not understand the core technology, it can greatly speed up development for those who need to quickly develop their applications.Step 4: Download the Ported Linux Operating SystemSuch as MCLinux, ARM-Linux, PPC-Linux, etc. If there is a Linux operating system specifically ported for the CPU being used, that would be ideal. After downloading, add specific hardware drivers, and then debug and modify. For CPUs with MMU, module debugging can be used, while for systems like MCLinux, only kernel compilation can be used for debugging.Step 5: Establish Root File SystemUse BUSYBOX software to reduce functionality and create a basic root file system, then add other programs according to application needs. Since the default startup scripts generally do not meet application requirements, the startup scripts in the root file system must be modified. These scripts are stored in the /etc directory, including: /etc/init.d/rc.S, /etc/profile, /etc/.profile, etc., and the automatic mounting configuration file /etc/fstab. The specifics will vary depending on the system. The root file system in embedded systems is generally set to read-only, requiring tools like mkcramfs or genromfs to produce a burnable image file.Step 6: Establish Application Program FLASH Disk PartitionGenerally, JFFS2 or YAFFS file systems are used, which requires these file system drivers to be provided in the kernel. Some systems use a linear FLASH (NOR type) of 512KB to 32MB, some use a nonlinear FLASH (NAND type) of 8MB to 512MB, and some use both simultaneously, requiring planning of the FLASH partition scheme according to applications.Step 7: Develop Application ProgramsApplication programs can be placed in the root file system or in the YAFFS, JFFS2 file systems. Some applications do not use the root file system and directly integrate the application program with the kernel, similar to the μC/OS-II approach.Step 8: Burn KernelRoot file system and application programs, release the product.

Above is the development process for embedded systems. For more knowledge, please click below↓↓↓

“Embedded System Development”