CN101001199A - Data processing method of high speed multidigit parallel data bus - Google Patents
Data processing method of high speed multidigit parallel data bus Download PDFInfo
- Publication number
- CN101001199A CN101001199A CNA200610011164XA CN200610011164A CN101001199A CN 101001199 A CN101001199 A CN 101001199A CN A200610011164X A CNA200610011164X A CN A200610011164XA CN 200610011164 A CN200610011164 A CN 200610011164A CN 101001199 A CN101001199 A CN 101001199A
- Authority
- CN
- China
- Prior art keywords
- clock
- data
- high speed
- processing method
- parallel data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
This invention discloses a data process method for high speed multi-bit parallel data buses, which divides a multi-bit parallel data bus into multiple clock domain transmissions in the same frequency and phase with the reference clock at the transmission end, the receiving end utilizes the FIFO queue of the read clock and write clock in the same frequency and phase with the reference clock as the buffer storage to recover the multi-bit parallel data bus, the data lines in the clock domains are equal to the clock lines.
Description
Technical field
The present invention relates to data communication system, particularly a kind of method of handling the high speed multidigit parallel data bus line is effectively finished the processing of bus frequency up to the multi-bit parallel data of hundreds of megahertzes.
Background technology
Data communication system develops to high-speed broadband day by day, bandwidth is that 10,000,000,000 and tens thousand of million communication system is non-rare, be in 10,000,000,000 routers of network core position or processing and the exchange that switch is responsible for data, network processing unit carries out the classification of packet or cell, processing such as route and traffic management, switching network carries out the clog-free exchange of data, communication means between network processing unit and the switching network can have several modes, a kind of is differential lines at a high speed, but this mode needs to increase in addition string and converting unit, need encode to data, and a clock and data recovery procedure will be arranged, system complexity is higher, realizes difficult.Another kind method is directly to adopt the data/address bus of multi-bit parallel to carry out the transmission of data, and the transmission speed of data/address bus can reach hundreds of megahertzes.But transmission and reception how to finish data reliably on printed wiring board (PCB) become a problem.In the data/address bus of parallel transmission,, require each data wire consistent, to satisfy the consistency of phase place with the track lengths of clock on PCB in order to guarantee well to sample to data at receiver side.But the parallel data bit wide surpasses 32 or more the time, is difficult to accomplish longer transmission such as every data lines on the circuit board.
Summary of the invention
The object of the present invention is to provide a kind of data processing method of high speed multidigit parallel data bus line, solve the multi-bit parallel data/address bus and the problem of longer transmission such as on PCB, can't realize, and then solved in the broadband system problem of data communication between the data processing unit, improve reliability of data transmission.
To achieve these goals, the invention provides a kind of data processing method of high speed multidigit parallel data bus line, at transmitting terminal the multi-bit parallel data/address bus is divided into clock zone transmission a plurality of and reference clock frequency homophase together, the receiving terminal utilization is read clock and is write clock and all reduce described multi-bit parallel data/address bus with the fifo queue of the same frequency of reference clock homophase as buffer memory, and the data wire and the clock line of each clock zone inside are isometric.
The data processing method of above-mentioned high speed multidigit parallel data bus line specifically comprises the steps:
The multi-bit parallel data/address bus is divided into the steps of many groups, and each set of dispense one clock and of deriving is used for the frame delimiter of the original position of definition of data frame, and the data wire and the clock line of packets inner are isometric;
The step of grouped data and corresponding frame delimiter being got at the rising edge of tranmitting data register or trailing edge;
Grouped data is after data wire and clock line transmission, the described clock of deriving that receiving terminal one fifo queue utilizes described grouped data correspondence writes described grouped data as writing clock, and judges according to described frame delimiter and to utilize receive clock as reading the clock dateout after receiving all described grouped datas;
The described clock of deriving, tranmitting data register and receive clock and reference clock are with the frequency homophase.
The data processing method of above-mentioned high speed multidigit parallel data bus line, wherein, the described clock of deriving, tranmitting data register and receive clock are produced according to reference clock by the zero-lag clock driver.
The data processing method of above-mentioned high speed multidigit parallel data bus line, wherein, receiving terminal begins the grouped data of correspondence is write described fifo queue after receiving described frame delimiter.
The data processing method of above-mentioned high speed multidigit parallel data bus line, wherein, first active data signal alignment of described frame delimiter and corresponding grouped data.
Compare with prior art, the present invention directly utilizes parallel bus to carry out the transmission of wideband data, after packet, utilizes FIFO to reduce at receiving terminal, has improved reliability of data transmission.
Description of drawings
Fig. 1 is the schematic diagram of deriving of clock;
Fig. 2 is a packet schematic diagram of the present invention;
Fig. 3 is the phase relation schematic diagram of clock of the present invention and data;
Fig. 4 is the handling process schematic diagram of receiving terminal FIFO among the present invention.
Embodiment
The data processing method of high speed multidigit parallel data bus line of the present invention comprises the steps:
The data/address bus partiting step, data/address bus is divided into several clock zones, the part of each clock zone transmission multi-bit parallel data/address bus, guarantee longer transmission such as the data wire of each clock zone inside and clock line simultaneously, guarantee the consistency on the phase place, so just can be in the correct sampled data of receiving terminal, the bit wide of tentation data bus is the M position in the present invention, and it is divided into the N group;
The clock partiting step, at transmitting terminal, reference clock produces N with the synchronous clock of frequency through the zero-lag clock driver, is divided into N different clock zone, is used for the data in the synchronous described grouping;
Datum plane definition step, N after the above-mentioned division different clock zone is defined as datum plane, comprising clock signal, data-signal and a frame delimitation signal SOF who is used to Frame to delimit, this frame delimitation signal has been indicated the beginning of data, and the SOF in each grouping comes into force simultaneously;
The reduction of data step, at receiving terminal, clock signal and data-signal are through the transmission of PCB, all produced certain time-delay, owing to only guaranteed the signal lead length unanimity in the grouping, can not guarantee the consistency of its phase place between each grouping, clock line in any one grouping all is difficult to satisfy the requirement of the settling time and the retention time of data-signal, can not be with sample data-signals in all each grouping of a single clock, reduction of data step of the present invention utilizes following method to realize accurately sampled data signal and with its reorganization:
Utilize the clock signal of each datum plane sample the respectively data-signal and the frame delimitation signal on notebook data plane, utilize a FIFO (first in first out) formation then, the data that are input as the notebook data plane of sampling respectively through the clock signal of each datum plane of this fifo queue are write the clock signal of clock for this datum plane; The output signal of this fifo queue with the control of the clock of transmitting terminal clock homology under export, and the data/address bus that finally is reduced into original M position is handled for follow-up processing unit.
Further specify below in conjunction with accompanying drawing.
Fig. 1 is the schematic diagram of deriving of clock.
As shown in Figure 1, reference clock REFCLK produces several and reference clock with the clock of homophase frequently behind clock driver, one of them is used at receiving terminal as reference clock (RX_REFCLK), one is used at transmitting terminal as reference clock (TX_REFCLK), and other N is used for each packet (RCLK_1, RCLK_2 ... RCLK_N).
Fig. 2 is the grouping schematic diagram of data, as shown in Figure 2, data/address bus is divided into N group fifty-fifty, and clock signal of each set of dispense and frame delimiter are with the original position of definition of data frame.
Fig. 3 is the phase relation schematic diagram of clock and data, as shown in Figure 3, data-signal is got at the rising edge of reference clock, satisfy certain settling time (Tsetup) and retention time (Thold), RCLK_ is for transmission arrives the clock signal of receiving terminal through PCB, and RCLK_ compares with reference clock certain transmission delay (Tdelay).
Fig. 4 is the handling process schematic diagram of receiving terminal FIFO among the present invention, as shown in Figure 4, in a single day receiving terminal receives the write operation that frame delimitation signal SOF promptly begins (also comprising corresponding with it data-signal certainly) FIFO, data-signal is write FIFO, it writes the clock (RCLK_1 of clock for each grouping, RCLK_2 ... RCLK_N), treat all SOF that divide into groups all to have received, and corresponding data are write the read operation that FIFO then begins all FIFO (every group of corresponding FIFO) simultaneously, reading clock is RX_REFCLK, and the data of reading are reconfigured, handle for follow-up processing unit.
The data processing method specific implementation of high speed multidigit parallel data bus line of the present invention may further comprise the steps:
Step 3 at receiving terminal, is utilized a fifo queue (FIFO), and a clock with the reference clock homology is used to control the output of FIFO;
Step 4, the method that data reconfigure, under the effect of frame delimitation signal, utilize FIFO read-write clock with characteristics frequently, in case valid data write FIFO, promptly start the read operation of FIFO, and read FIFO in data and be reassembled into the data/address bus consistent later on, handle for follow-up processing unit with form before the packet.
Certainly; the present invention also can have other various embodiments; under the situation that does not deviate from spirit of the present invention and essence thereof; those of ordinary skill in the art work as can make various corresponding changes and distortion according to the present invention, but these corresponding changes and distortion all should belong to the protection range of the appended claim of the present invention.
Claims (6)
1, a kind of data processing method of high speed multidigit parallel data bus line, it is characterized in that, transmitting terminal is divided into clock zone transmission a plurality of and reference clock frequency homophase together with the multi-bit parallel data/address bus, the receiving terminal utilization is read clock and is write clock and all reduce described multi-bit parallel data/address bus with the fifo queue of the same frequency of reference clock homophase as buffer memory, and the data wire and the clock line of each clock zone inside are isometric.
2, the data processing method of high speed multidigit parallel data bus line according to claim 1 is characterized in that, specifically comprises the steps:
The multi-bit parallel data/address bus is divided into many groups, and each set of dispense one clock and of deriving is used for the frame delimiter of the original position of definition of data frame, and the data wire and the clock line of packets inner are isometric;
The step of grouped data and corresponding frame delimiter being got at the rising edge of tranmitting data register or trailing edge;
Grouped data is after data wire and clock line transmission, the described clock of deriving that receiving terminal one fifo queue utilizes described grouped data correspondence writes described grouped data as writing clock, and judges according to described frame delimiter and to utilize receive clock as reading the clock dateout after receiving all described grouped datas;
The described clock of deriving, tranmitting data register and receive clock and reference clock are with the frequency homophase.
3, the data processing method of high speed multidigit parallel data bus line according to claim 2 is characterized in that, the described clock of deriving, tranmitting data register and receive clock are produced according to reference clock by the zero-lag clock driver.
According to the data processing method of claim 2 or 3 described high speed multidigit parallel data bus lines, it is characterized in that 4, receiving terminal begins the grouped data of correspondence is write described fifo queue after receiving described frame delimiter.
5, according to the data processing method of claim 2 or 3 described high speed multidigit parallel data bus lines, it is characterized in that first active data signal alignment of described frame delimiter and corresponding grouped data.
6, the data processing method of high speed multidigit parallel data bus line according to claim 5 is characterized in that, the described frame delimiter of each grouped data comes into force simultaneously.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA200610011164XA CN101001199A (en) | 2006-01-11 | 2006-01-11 | Data processing method of high speed multidigit parallel data bus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA200610011164XA CN101001199A (en) | 2006-01-11 | 2006-01-11 | Data processing method of high speed multidigit parallel data bus |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101001199A true CN101001199A (en) | 2007-07-18 |
Family
ID=38693015
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA200610011164XA Pending CN101001199A (en) | 2006-01-11 | 2006-01-11 | Data processing method of high speed multidigit parallel data bus |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101001199A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102394808A (en) * | 2011-09-23 | 2012-03-28 | 烽火通信科技股份有限公司 | Method and apparatus for phase adaption and frame alignment of serial media independent interface of ethernet network |
CN101946219B (en) * | 2008-02-20 | 2013-03-20 | 惠普开发有限公司 | Redriver with two reference clocks and method of operation thereof |
CN103019986A (en) * | 2012-12-07 | 2013-04-03 | 陕西千山航空电子有限责任公司 | Parallel communication bus and data transmission method |
CN104407809A (en) * | 2014-11-04 | 2015-03-11 | 盛科网络(苏州)有限公司 | Multi-channel FIFO (First In First Out) buffer and control method thereof |
CN105718389A (en) * | 2016-01-27 | 2016-06-29 | 中国电子科技集团公司第五十八研究所 | Input data encasement method applicable to programmable interface |
CN112821889A (en) * | 2019-11-15 | 2021-05-18 | 京东方科技集团股份有限公司 | Output control circuit, data transmission method and electronic equipment |
CN114710254A (en) * | 2017-07-04 | 2022-07-05 | 康杜实验室公司 | Method and device for measuring and correcting multi-line time offset |
-
2006
- 2006-01-11 CN CNA200610011164XA patent/CN101001199A/en active Pending
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101946219B (en) * | 2008-02-20 | 2013-03-20 | 惠普开发有限公司 | Redriver with two reference clocks and method of operation thereof |
CN102394808A (en) * | 2011-09-23 | 2012-03-28 | 烽火通信科技股份有限公司 | Method and apparatus for phase adaption and frame alignment of serial media independent interface of ethernet network |
CN102394808B (en) * | 2011-09-23 | 2014-12-31 | 烽火通信科技股份有限公司 | Method and apparatus for phase adaption and frame alignment of serial media independent interface of ethernet network |
CN103019986A (en) * | 2012-12-07 | 2013-04-03 | 陕西千山航空电子有限责任公司 | Parallel communication bus and data transmission method |
CN103019986B (en) * | 2012-12-07 | 2015-12-02 | 陕西千山航空电子有限责任公司 | A kind of parallel communication busses and data transmission method |
CN104407809A (en) * | 2014-11-04 | 2015-03-11 | 盛科网络(苏州)有限公司 | Multi-channel FIFO (First In First Out) buffer and control method thereof |
CN104407809B (en) * | 2014-11-04 | 2018-03-30 | 盛科网络(苏州)有限公司 | The control method of multichannel fifo buffer |
CN105718389A (en) * | 2016-01-27 | 2016-06-29 | 中国电子科技集团公司第五十八研究所 | Input data encasement method applicable to programmable interface |
CN114710254A (en) * | 2017-07-04 | 2022-07-05 | 康杜实验室公司 | Method and device for measuring and correcting multi-line time offset |
CN112821889A (en) * | 2019-11-15 | 2021-05-18 | 京东方科技集团股份有限公司 | Output control circuit, data transmission method and electronic equipment |
CN112821889B (en) * | 2019-11-15 | 2024-02-20 | 京东方科技集团股份有限公司 | Output control circuit, data transmission method and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7013359B1 (en) | High speed memory interface system and method | |
CN101001199A (en) | Data processing method of high speed multidigit parallel data bus | |
CN101095308B (en) | Communication device for packet exchange of system packet interface of queue cascade and logic mark | |
US7093061B2 (en) | FIFO module, deskew circuit and rate matching circuit having the same | |
US7082504B2 (en) | Method and apparatus for asynchronous read control | |
CN111131091A (en) | Inter-chip interconnection method and system for network on chip | |
CN109446132A (en) | A kind of interface bus device and data communication protocol | |
JP2005505200A (en) | Parallel data communication with data valid indicator and skew intolerant data group | |
CN108519857B (en) | Multi-source unformatted broadband data high-speed mass formatted storage and feature preservation method | |
CN103714038A (en) | Data processing method and device | |
CN108462620B (en) | Gilbert-level SpaceWire bus system | |
CN110471876A (en) | A kind of ultrahigh speed serial data channel system | |
CN105578585B (en) | Method, device and communication equipment for determining link delay | |
CN109951750A (en) | Data processing method and system based on mono- layer of cross architecture of FlexE | |
CN105446699A (en) | Data frame queue management method | |
CN112600551B (en) | Serdes interface circuit | |
CN112653638B (en) | Device for switching routes of multiple paths of intermediate frequencies and baseband at high speed and communication method thereof | |
CN110058207A (en) | A kind of synchronization of more lane data and recombination system and method for Radar Signal Transmission | |
CN109976704A (en) | A kind of design method of the cascade fifo module based on FPGA | |
CN100463443C (en) | Asynchronous FIFO realizing system and realizing method | |
US20070258478A1 (en) | Methods and/or apparatus for link optimization | |
CN108984451B (en) | Signal driving method and communication device | |
US6946873B1 (en) | Method and system for recovering and aligning synchronous data of multiple phase-misaligned groups of bits into a single synchronous wide bus | |
CN110008157A (en) | A kind of hardware structure of deserializer | |
CN103034610A (en) | Methods and devices for transmission and reception of advanced extensible interface (AXI) bus signal between split modules |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Open date: 20070718 |