Clock recovery method and device for transmitting EI signals based on Ethernet

Abstract

The invention relates to a clock recovery method for transmitting EI signals based on the Ethernet. The clock recovery method includes the steps that an EI data package coming from the Ethernet is stored in an FIFO memorizer and buffered, the data package is read through a tooth missing clock CLKG, and EI data of clock tooth missing are output by the FIFO memorizer; the EI data of clock tooth missing are stored in an elastic memorizer, a phase-locked loop outputs synchronizing signals to the elastic memorizer, meanwhile output of the phase-locked loop is input as feedback, the phase-locked loop controls the output clock frequency of the elastic memorizer by monitoring difference between a reading pointer and a writing pointer of the elastic memorizer, if the difference between the reading pointer and the writing pointer is larger than a set value, the clock frequency of the output of the phase-locked loop is reduced, otherwise, the clock frequency of the output of the phase-locked loop is increased, and the elastic memorizer outputs the EI data with clock smoothed. The invention further discloses a clock recovery device for transmitting the EI signals based on the Ethernet. A clock is recovered with two-level buffering adjustment to reduce phase position diffuse movement, smooth missed teeth and recover the clock meeting the related ITU-TG. 823 requirements; a good diffuse movement performance index can be achieved without changing a frame structure.

Description

Technical field [0001] The invention relates to the technical field of communication and information system network transmission, in particular to a method and device for clock recovery based on Ethernet to transmit E1 signals. [0002] Background technique [0003] In the new generation of IP-based communication network, in order to achieve multi-service transmission, at the sending end, the real-time service (TDM) is packaged to make it into the data packet of the Ethernet packet, and then transmitted; at the receiving end, for recovery In the original TDM service, data was counted and jitter removed, so as to obtain the timing information of the code stream. Since in the TDMoIP system, the signal that needs to be sent at the sender is a standard E1 signal, in order to transmit in the Ethernet, the E1 signal is split and encapsulated to turn it into a fixed-size Ethernet packet, the original E1 signal All the timing information in the signal is lost. The signal received at the ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of the clock phase detector used in the PLL is that the resolution can only reach one bit. When the recovered clock is very close to the actual value, it takes a long time for the phase difference to accumulate to one bit, which will cause drift , it is difficult to meet the requirements of roaming specifications

In addition, due to the characteristics of Ethernet, the time when data packets arrive at the destination will vary randomly, and may even drop packets.

In addition, the packet data is a large block burst data. When a packet arrives, it is hundreds or even thousands of bytes. The change of the write pointer of the elastic memory is jumpy. In this case, the phase change detection filter cannot be changed from The detection of pointer difference is done by general low-pass filtering method

Images