1 Introduction While instrumentation is developing rapidly, computer and network technologies are also developing rapidly. PCs have transitioned from high-speed growth to a period of stable development. The era in which PCs led the electronics industry to prosperity has become history. The emergence and widespread application of embedded systems have brought computers and networks into the post-PC era. The embedded intelligent instrument remote monitoring system is one of the important development directions of industrial control networks. It is the result of the joint development of various technologies such as industrial data communication and control network technology and Internet technology [1]. The development and maturity of this technology will have a profound impact on people's production and life. 2 Embedded Controller Hardware Design Definition of controller: An embedded controller is an electronic device or device controlled by a high-speed processor and other chips working together, with a high-speed processor as the core. It can complete various automated processing tasks such as monitoring and control [4]. The embedded controller is the core part of the system. 2.1 System Hardware Structure Diagram [align=center] Figure 1 Embedded Controller System Hardware Structure Diagram [/align] The system hardware structure diagram of the embedded controller is shown in Figure 1. It is powered by two DC power supplies: 5V, 3.3V and 1.8V. The system is mainly composed of DSP chip TMS320C5402, 256K FLASH memory, Ethernet interface chip RTL8019AS, A/D conversion chip, serial port chip, etc. From the system hardware structure diagram of the embedded controller, it can be seen that the embedded controller is the key part of the embedded remote monitoring system. The field intelligent instruments can input signals through the analog and digital interfaces of the embedded controller. The service program embedded in the controller can remotely view and control the signals of the field intelligent instruments through Ethernet or Modem, thereby realizing remote monitoring of the intelligent instruments. 2.2 Minimum system design of processor DSP 5402 (1) Reset circuit. The power-on reset circuit and the manual reset circuit are designed at the same time, so that the system can be easily manually reset when a fault occurs during operation. The reset circuit should ensure that the low level time of the reset is long enough to ensure reliable DSP reset; on the other hand, it should ensure good stability to prevent DSP false reset. The reset circuit uses MAX706R chip to form an automatic reset circuit. MAX706R is an automatic reset chip that can be matched with DSP chips with a working voltage of 3.3V. The specific connection of the chip is shown in Figure 2. (2) Clock circuit. External clock input is often used in general DSP systems because it has high accuracy, good stability and is easy to use. Since the DSP works based on the clock, if the clock quality is not high, the reliability and stability of the system will be difficult to guarantee. Therefore, this system will use an external clock source to provide the clock, as shown in Figure 3. The external clock signal is directly applied to the X2/CLKIN pin of the DSP chip, and the X1 pin is left floating. Set CLKMEI=1, CLKMD2=1, CLKMD3=1. After reset, the clock of the DSP chip is 1/2 of the external crystal oscillator frequency, that is, divided by 2. [align=center] Figure 2 System automatic reset circuit[/align] [align=center] Figure 3 Clock circuit[/align] (3) System power supply design. The TMS320VC5402 chip uses dual power supply. The core voltage and I/O interface voltage of the DSP are 1.8V and 3.3V respectively. This system requires three power supplies with voltages of 5V, 3.3V and 1.8V. The RTL8019AS network card chip and analog-to-digital converter circuit are both powered by 5V. The dual power supply solution of the DSP is implemented by TPS73HD318, with an input power supply voltage of 5V and output voltages of 3.3V and 1.8V respectively. The maximum output current of each power supply is 750mA. 2.3 Program memory expansion Compared with EPROM, FLASH memory has a higher performance-price ratio, and is small in size, low in power consumption, electrically erasable and easy to use. In addition, 3.3V FLASH can be directly connected to the DSP chip. Therefore, using FLASH as program memory to store programs and some fixed data is a better choice. The program memory of this system is an AT29LV020 FLASH memory. This chip has a storage space of 256K×8 and a maximum read speed of 100ns. Therefore, when reading the program, the main frequency should be lower than 10MHz. 2.4 Interface circuit design (1) Analog input interface. The analog input channel uses the TLV2544 chip produced by TI. When connected to the DSP chip, a frame synchronization signal FS can be used to control the start of a serial data frame. The TMS320VC5402 provides a high-speed, bidirectional, multi-channel buffered serial port MCBSP, which can be directly connected to the serial A/D converter. (2) Ethernet expansion interface. The primary issue in realizing the remote monitoring system of embedded intelligent instruments is to enable the intelligent instruments to have an Ethernet interface. The RTL8019AS Ethernet interface chip produced by Taiwan Realtek can enable the embedded controller to have a network interface of a general computer. (3) RS-232 serial interface design. RS-232-C does not define the physical characteristics of the connector, so there are various types of connectors such as DB-25, DB-15 and DB-9. In order for the DSP to connect to the serial port of the field intelligent instrument, we use MAXIM's MAX3111 to complete the conversion of the level and logic relationship between EIA and TTL circuits. (4) RS-485 serial communication interface. RS-485 is a serial communication bus standard developed to meet the needs of long-distance, distributed control systems. It supports multi-node, long-distance transmission. The RS-485 standard uses a balanced transmit and differential receive transceiver to drive the bus. In order to expand the number of terminal connected devices, this design adds an RS-485 interface. The connection between the DSP and the RS-485 serial port is completed using the MAX3140UART communication chip from MAXIM. 3 Embedded TCP/IP Protocol Design 3.1 Overall Data Flow Design The overall data flow diagram of embedded TCP/IP is shown in Figure 4. [align=center] Figure 4 Overall Data Flow Diagram of Embedded TCP/IP[/align] The application uses the data communication function provided by the TCP/IP protocol stack through the Socket application programming interface [3]. When sending data, if the connection-oriented TCP protocol is selected, the application will hand the user data over to the TCP protocol module for processing, and the TCP module will encapsulate its header and data into a TCP segment; if the connectionless UDP protocol is selected, the user data will be handed over to the UDP protocol module for processing, and the UDP module will encapsulate its header and data into a UDP datagram. The encapsulated TCP segment or UDP datagram is then handed over to the IP protocol module. The IP module adds an IP header to the TCP segment or UDP datagram and encapsulates it into an IP packet. It then determines the route for the IP packet based on the routing table. If a route is found, the packet is sent to the network interface layer. The network interface layer determines whether the data is being sent from the Ethernet interface or the asynchronous serial port. If it is from the Ethernet interface, it uses the ARP protocol to find the physical address corresponding to the destination IP address, then encapsulates it into an Ethernet frame, and the network card driver sends the Ethernet frame out. 3.2 Overall Data Structure TCP coordinates sending, receiving, and retransmission actions for each TCP connection through a data structure called the TCP Transmission Control Block (TCB). TCP maintains a TCB for each active connection. The TCB contains all information about the TCP connection, including the address and port number of the connection endpoints, the estimated current average round-trip time, the data sent or received, whether acknowledgment or retransmission is required, and a set of statistics about the connection's usage. Therefore, the TCB is relatively large. This design simplifies the TCB, retaining only the basic information necessary for controlling connection-oriented data transmission and reception. The UTCB (Universal Data Transmission Control Block) for the UDP protocol contains all the information of the UDP end node: external and local IP addresses, external and local port numbers, and other control information. The use and allocation of UTCB are exactly the same as TCB, except that because UDP does not require establishing a connection, UTCB is allocated immediately as soon as data is sent, and released once the data transmission is complete. 4 Embedded Controller Software Design According to the control requirements and programming needs of the system, the system software is divided into the following modules, a total of six modules: main control module, initialization module, protocol encapsulation module, protocol unpacking and reassembly module, network interface module, and application layer application module. Figure 5 shows the logical relationship between each module. [align=center] Figure 5 Relationship between software system modules[/align] (1) Main control module design. The main control module usually contains some variable definitions and function calls. In addition, some important I/O port definitions are also placed in the main control function. The main control module defines delay signal functions, port setting functions, synchronous serial port setting functions, interrupt service functions, etc. When the system is working, the program starts running from the main() function of the main control module. First, it calls the initialization program of DSP and RTL8019AS. After the initialization is completed, the main program enters the working state in an unconditional loop. (2) Initialization module design. When the system software is running, it first calls the initialization program to initialize the system resources before it can work normally. The initialization module includes four initialization functions, namely net_init(), dsp5402_init(), interrupt_init(), and nic_atl_init(). (3) Protocol encapsulation module design. The role of the protocol encapsulation part is to group the data and encapsulate it with different protocols to make it into packet data that conforms to the TCP/IP protocol specification. In the design, the encapsulation of different protocols is implemented by different functions. In this system, the monitoring terminal system needs to encapsulate the IP protocol located at the network layer and the TCP protocol located at the transport layer. The encapsulation order is to encapsulate the transport layer protocol TCP first, and then encapsulate the network layer protocol IP. Then it is sent out through the network interface module. (4) Protocol unpacking and reassembly module design. The working process of this module is as follows: After data is received from the Ethernet, it is first placed in the on-chip memory of the network card chip. The DSP, i.e., the network interface module, reads it and performs protocol decomposition. According to the protocol content, it is processed accordingly. If the data is encapsulated in UDP, it is placed in the system's receive buffer. The DSP regroups and queues the data packets according to the sequence number field value in the protocol. Finally, the data is unpacked, merged, and sent to the application layer for further processing after forming a continuous data stream. 5. Summary This paper details the hardware and software design and implementation process of the embedded controller, including the hardware design and implementation of the microprocessor system, the implementation of the interface module with the smart instrument, and the analysis and implementation of the TCP/IP protocol. The successful implementation of the embedded remote monitoring system will have great practical value and excellent application prospects. It will not only be applied in industrial control, but will also be widely used in meteorology, environmental monitoring, smart homes, and other fields. In future development, a more standardized application layer communication protocol can be proposed to achieve compatible communication among multiple products. It is also possible to try connecting multiple single systems together for coordinated operation, truly realizing the remote and networked nature of the monitoring system. The innovations of this paper are as follows: This system employs a 16-bit fixed-point high-speed DSP microprocessor (whose program is easily ported to similar 32-bit microprocessor chips), achieving a running speed of up to 100 MIPS, a significant improvement over the earlier popular 8-bit microprocessors. The design itself is also innovative, combining superior technology and high-performance hardware to create a high-performance monitoring system. It borrows the modular design principles of computers (including software and hardware), allowing the system to be reconfigured as needed. References: [1] Kang Jun, Dai Guanzhong. Design of Industrial Ethernet Remote Monitoring System Based on Internet[J]. Control Engineering, 2004.7: 16-20 [2] Yang Kejian, Zhao Wei, Pang Hui et al. Research on the Technology of Embedded System Products Accessing the Internet[J]. Microcomputer Information, 2006, 1-2: 43-44, 65 [3] Li Fengbao. Networked Measurement and Control System Technology[M]. Sichuan University Press, 2004.5 [4] Dong Xiaoguo, Wang Rong. Design and Implementation of Web-based Remote Monitoring System for Reclaimed Water[J]. Microcomputer Information, 2006, 12-1: 86-88