A fpga-based powerlink slave frame buffer management system

Abstract

The invention discloses a POWERLINK slave station frame buffer management system based on an FPGA (Field Programmable Gate Array). A data frame type detection module and a data frame buffer management module are realized in the FPGA, a block RAM (Random Access Memory) in the FPGA is taken as a frame buffer region, the frame type is recognized in a data frame receiving process by the data frame type detection module, a result is transmitted to the buffer management module, and the buffer management module dynamically arranges a storage structure, so that different types of frames are stored in different regions of the RAM without interference on each other. Through dynamic management of the storage structure, a problem of data loss of the conventional linear storage structure due to frame coverage is avoided.

Description

technical field [0001] The invention belongs to the technical field of network communication, and more specifically relates to an FPGA-based POWERLINK slave station frame buffer management system. Background technique [0002] The POWERLINK communication profile contains 7 types of Ethernet data frames, namely SoC, PReq, PRes, SoA, ASnd, AMNI, and AInv. [0003] In the POWERLINK communication profile, frames containing application layer data are SDOs in Preq, Pres, and Asnd, and other types of data frames are used for network management. Among them, Preq and SDO frames belong to unicast frames, and Pres frames belong to broadcast frames. [0004] Such as figure 1 As shown, there are two working modes under the POWERLINK communication profile. In the Powerlink mode, the communication cycle is subdivided into an isochronous phase, an asynchronous phase, and an idle phase. The sending of the SoC frame is taken as the starting point of the isochronous synchronization phase, a...

AI Technical Summary

Problems solved by technology

[0010] In the first scheme, MAC receives all data frames without distinguishing them. When the number of slave stations in the network increases, one of the slave stations will receive a large number of data frames that do not belong to the slave station. If the slave station does not Process data in a timely manner. When the buffer area is full, the data will inevitably be overwritten, resulting in data loss.

[0011] In the second solution, OPENMAC can filter out data frames that do not belong to the slave station, but there are still problems with this method. As mentioned above, SDO communication is asynchronous communication, and the processing of SDO data frames does not require immediate processing, which is limited by the queue. length, SDO data frames may still be overwritten before being processed

But the former will affect the communication performance, while the latter is limited by the length of the buffer area, it is impossible to increase the length of the buffer area infinitely, and neither of the two measures can fundamentally solve the problem

Images