[Abstract] This paper briefly introduces the AT91RM9200 microprocessor and the embedded LINUX operating system, and discusses the hardware design and corresponding software design methods of the seismic data acquisition system. [Keywords] AT91RM9200, Embedded LINUX operating system, Data acquisition 0 Introduction With the rapid development of digital technology, digital instruments have become the mainstream instruments in the field of observation technology. Therefore, data acquisition technology has become a very important technical link in the field of observation technology. As we all know, earthquake prediction is a global challenge. As the foundation of earthquake prediction, the importance of earthquake and earthquake precursor observation data is self-evident. Obtaining accurate and reliable observation data depends on the seismic observation instrument (including sensors and acquisition units). With the rapid development of computers, embedded data acquisition systems have emerged, which have high reliability, small size, easy expansion, powerful functions, short development cycle, and low cost. This paper is based on the seismic acquisition system design project of BGP Inc., using an ARM9 embedded system, and therefore its research has very important practical significance. 1 Overall Design Scheme As a general industrial data acquisition system hardware platform, its basic purpose is to acquire external signals, such as analog and digital signals, and to convert digital signals into analog signals for output, so as to control external devices. Based on this, the system to be designed in this paper has the following requirements: (1) Multi-channel analog signal acquisition. Because there are a lot of analog data in the industrial control field, and the amplification factor required for various analog signals is different, a variable gain amplifier is required. (2) Support for multiple communication interfaces such as Ethernet. Modern industrial measurement and control field requires the controller to transmit data more quickly and efficiently. (3) Data acquisition has mobile storage function. Based on the actual working conditions of the field, the control platform needs to be able to extract some data through mobile storage under normal working conditions so that data analysis can be performed on other devices. The design requirements put forward the minimum requirements for the performance of the entire system, and provide guiding principles for device selection and internal system design. The overall system block diagram according to the requirements is as follows: [align=center] Figure 1 Overall System Structure Block Diagram[/align] 2 System Hardware Design Constructing an embedded seismic acquisition system requires hardware support. There is no unified standard for embedded system hardware. The embedded system is tailored according to application requirements. The microprocessor used in the system design is the AT91RM9200 microprocessor manufactured by Atmel. It is a true system-on-a-chip, integrating multiple communication interfaces such as USB, Ethernet, EBI, MCI, SSC, and SPI. Its 200MIPS processing speed and advanced power management make this chip very suitable for system control applications. The hardware block diagram based on the AT91RM9200 is shown below: [align=center] Figure 2 System Hardware Structure Diagram[/align] This system is a powerful, low-power embedded high-precision data acquisition system, using an industrial-grade processor based on the ARM9 core and an embedded Linux operating system. The system has rich peripheral control and communication interfaces, and can control peripheral components and switch between multiple analog signals through IO outputs. The acquired high-precision data can be transmitted to a remote monitoring terminal via RS232 or high-speed Ethernet. Because the system uses a powerful processor and the Linux operating system, it can not only acquire high-precision data but also allow users to perform data processing and other application-level functions. 3. System Software Design The system is implemented under an embedded Linux operating system. Embedded systems are application-centric, based on computer technology, and their hardware and software are customizable, making them suitable for dedicated computer systems with strict requirements on functionality, reliability, cost, size, and power consumption. Linux possesses many advantages, such as extensive hardware support, a highly efficient and stable kernel, open source, abundant software, excellent development tools, robust network communication and file management mechanisms, and being free, making it very suitable for embedded systems. Embedded systems are minimal simplified systems designed to meet practical application needs. Embedded data acquisition systems run numerous tasks, some with high real-time requirements, and embedded microprocessors need to manage abundant resources. All of these factors necessitate the introduction of an operating system on the embedded platform. Based on the tasks required by the system, the design of each module naturally follows. 3.1 Software Design of the A/D Channel Module The ADS1256 chip used in this system has 24 bits of conversion data. The effective number of bits varies depending on the conversion rate, input buffer, and amplifier settings. With the input buffer and amplifier settings unchanged, the conversion rate becomes the factor affecting the effective number of bits. The data acquisition frequency can be changed within the allowable range, but regardless of the acquisition frequency, the ADC conversion rate is always set to a maximum of 30Ksps. This is to ensure that the effective number of bits is always at the minimum when operating at the highest sampling frequency. However, this does not address the situation when operating at lower sampling frequencies, as the data volume is relatively low and the conversion rate requirement is not high. Therefore, when switching to a lower sampling frequency, the ADC operating value can be adjusted accordingly to operate at a lower conversion rate, provided that the data requirements at that sampling frequency are met. This can improve the conversion accuracy of the ADC at certain sampling frequencies. When using the ADC module, first set the measurement channel pin to AINx, then set the ADC operating mode, ADC conversion channel, and conversion channel (CLKDIV clock divider value) through the ADCR register, and start the ADC conversion. You can wait for the ADC conversion to complete by polling or interrupting; the converted data is stored in the ADDR register. The ADC conversion clock divider value is calculated as: CLKDIV = -1 (Fadclk is the desired ADC clock value, which cannot exceed 4.5MHz). When performing multi-channel ADC conversion, first switch to channel 1 and perform the first conversion, wait for the conversion to complete, start the conversion again, wait for the conversion result, and read the ADC result. Then switch to channel 2 and perform the first conversion, following the same procedure as channel 1. Continue switching to channels 3, 4, and so on, until all channels are converted. The A/D conversion task flowchart is shown in Figure 3: [align=center]Figure 3 A/D Task Conversion Flowchart[/align] 3.2 Software Design of the USB Channel Module The host is at the core of the USB topology. Any USB data transfer must be initiated and controlled by the host. All USB devices can only establish connections with the host, and currently, personal computers play a large role as the host. Therefore, the USB mobile devices we currently use are USB devices such as USB flash drives. Using a USB flash drive on an embedded platform requires that the embedded product support the USB host interface. The USB bus includes four basic data transfer types: control transfer, interrupt transfer, batch transfer, and synchronous transfer. This paper uses control transfer and batch transfer. Since USB flash drives generally belong to the mass-storage storage class, they follow the Bulk-Only transfer protocol and UFI command specification. Under this transfer mode, three types of data are transferred between the board and the USB flash drive: CBW, CSW, and ordinary data. CBW is the command sent from the board to the USB flash drive. Here, it is the SCSI transmission command set (including flag information, data level, UFI command). After completion, the USB flash drive reflects the current command execution status CSW to the board. The board decides whether to send data based on the CSW. [align=center] Figure 4 USB flash drive data writing flowchart[/align] 3.3 Software design of serial port module The controllers or intelligent instruments used in general industrial control sites need to have the function of communicating with PCs to give full play to the advantages of the resources of PCs and intelligent devices. The baud rate of communication can be set. The serial port is an 8-bit asynchronous communication interface. One frame of information is 10 bits: 1 start bit (0), 8 data bits (low bit first) and 1 stop bit (1). TXD1 is the transmitting end and RXD1 is the receiving end. These are the initialization of the USART register. After initialization, the following is the program flowchart: [align=center] Figure 5 Serial Port Flowchart[/align] 4 Conclusion As an application of embedded systems in the field of industrial control, this paper mainly discusses the hardware and software design of an earthquake data acquisition system based on the AT91 series processor AT91RM9200 and the embedded Linux operating system. Based on research into the current development of earthquake measurement technology, a complete system scheme for the functions and design methods of this data acquisition system is proposed. Through continuous updating and summarizing, the research and development and fabrication of the acquisition system were completed, and tests were conducted on the system's ADC performance and data storage capabilities. References [1] Jiang Junhui, Hardware Design of Embedded Systems Based on ARM, Microcomputer Information, No. 7-2, 2005 [2] Zhou Zhenan, Fan Lianglong, et al. Design and Practice of Data Acquisition System. Earthquake Press, July 2005 [3] Ma Jianming. Data Acquisition and Processing Technology. Xi'an Jiaotong University Press, September 2005 [4] Yang Heng. ARM Embedded System Design and Practice. Xi'an University of Electronic Science and Technology Press, October 2005 [5] Zou Siyi. Embedded Linux Design and Application. Tsinghua University Press, April 2002 [6] He Jiaming. Embedded 32-bit Microprocessor System Design and Application. Electronic Industry Press, December 2005