Point-to-multipoint UDP real-time data transmitting and confirming method based on FPGA

Abstract

The invention discloses an FPGA-based method for transmitting and confirming UDP real-time data from a point to multiple points, which comprises the following steps: (1) firstly, a real-time data collecting and controlling system, which is composed of a plurality of process-quantity collecting and controlling terminals, an FPGA-based communication controlling device and one or multiple upper-layer data processing devices, is established; (2) Secondly, a universal network interface card and a real-time operating system (RTOS) are configured in an industrial personal computer system so as to constitute the upper-layer data processing devices; (3) an ACK field is added into the header of application data of a UDP datagram which is from the communication controlling device to the upper-layer data processing devices; (4) after receiving the UDP datagram of the upper-layer data processing devices, the communication controlling device transmits and arbitrates the UDP datagram; and (5) after receiving the UDP datagram, the data processing devices check the self-related values in the ACK field; if the value is 1, then the current frame is confirmed, otherwise, the current frame is retransmitted after the timer overflow is sent.

Description

1. Technical field [0001] The present invention relates to FPGA application and Ethernet-based UDP real-time data point-to-multipoint transmission method and flow control mechanism, suitable for distributed acquisition and centralized data processing in occasions where real-time transmission of immediate sampling data is required and real-time reliable transmission is required for control or adjustment commands Power system protection field and industrial control field. 2. Background technology [0002] In the field of power system relay protection and industrial control, in order to improve the response speed and accurately analyze the timing of asynchronous events, it is necessary to perform high-speed synchronous sampling and real-time transmission of process variables. In order to reduce system complexity and cost and redundant processing needs, more and more emphasis is placed on data sharing, and the process data in production is collected once and released on the netw...

AI Technical Summary

Problems solved by technology

There may be multiple processes belonging to the same multicast group on a host, so when a host receives a multicast datagram, it must send a copy to each process belonging to that multicast group, increasing system overhead

[0014] (2). The interface of the switch must work in full-duplex mode, and has no address filtering capability for multicast packets and broadcast packets. If there are multiple real-time broadcast data sources on the same physical subnet, data from other ports may mutual influence, the system will not work properly

[0015] (3). Additional messages are required to confirm the downlink datagrams of multiple data processing devices, and the uplink bandwidth of the switching equipment cannot be used to transmit real-time sampling data. Although it can be solved by increasing the bandwidth, it will increase the system cost and transmission. at the expense of distance

[0016] (4). When several ports exchange data to one port in real time at the same time, the buffer overflows

Increasing the frame buffer is one of the solutions, but not completely, and increases the cost

Especially in the case of distributed collection, centralized data processing or system redundancy, the above situation is inevitable. The probability of redundant data processing devices sending real-time control commands to the same collection control node at the same time is very high, and the current Methods are not guaranteed to be real-time

Images