CN101916219A - Streaming media display platform of on-chip multi-core network processor - Google Patents
Streaming media display platform of on-chip multi-core network processor Download PDFInfo
- Publication number
- CN101916219A CN101916219A CN 201010217301 CN201010217301A CN101916219A CN 101916219 A CN101916219 A CN 101916219A CN 201010217301 CN201010217301 CN 201010217301 CN 201010217301 A CN201010217301 A CN 201010217301A CN 101916219 A CN101916219 A CN 101916219A
- Authority
- CN
- China
- Prior art keywords
- video
- data
- core network
- network processor
- chip multi
- 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
The invention discloses a streaming media display platform of an on-chip multi-core network processor. The streaming media display platform is used for synchronously displaying two real-time Video windows on two personal computers (PC) respectively on the basis of ffplay, wherein the two PCs are provided with a 1G Ethernet port respectively and are connected with the on-chip multi-core network processor through the Ethernet; one of the PCs serves as a video sending terminal and is used for synchronously sending video data to the other PC through the on-chip multi-core network processor while playing videos; and the other PC serves as a video receiving terminal to receive the video data from the on-chip multi-core network processor in real time, and synchronously resolve, process and play the video data so as to display and test the streaming media of the on-chip multi-core network processor. The streaming media display platform is completely synchronous, has no disorder screens, can count the packet loss ratio and adjust the play rate and effect, and provides a good test and display platform for the on-chip multi-core processor.
Description
Technical field
The present invention relates to demonstration and the performance test thereof of on-chip multi-core network processor, be specially a kind of streaming media display platform of on-chip multi-core network processor Streaming Media.
Background technology
The chip multi-core technology is the strategic technology of contemporary integrated circuit (IC) design, and it becomes the only choosing of continuous upgrading processing device performance with stronger parallel processing and excellent calculated performance.Wherein, network-on-chip NoC (Network onChip) adopts the mode of grouping route packet, has very strong advantages such as efficient interconnection, extensibility, synchronization policy, has overcome the various bottleneck problems of being brought by bus interconnection.
In the research of network-on-chip gordian technique, Performance Evaluation and effect demonstration are the importances of NoC.The researchist studies NoC from many aspects such as topological structure, route algorithm, exchanging policy and flow-control mechanisms, but how to make up the NoC verification platform, and the performance that obtains NoC fast also is the emphasis of NoC research always.
At demonstration and the Performance Evaluation of network-on-chip to Streaming Media, yet there are no relevant report or article, need a kind ofly demonstrate platform effectively, intuitively fast, get rid of the influence of other factors, realize the demonstration Performance Evaluation of network-on-chip itself Streaming Media.
Summary of the invention
The problem to be solved in the present invention is: lack at present and be used for chip multi-core processor carries out Performance Evaluation to streaming media display simulated environment or simulator, can not realize the test and the demonstration of chip multi-core processor intuitively, efficiently, easily, needs are a kind of effectively, mode realizes the Streaming Media test and the demonstration of on-chip multi-core network processor intuitively.
Technical scheme of the present invention is: a kind of network-on-chip polycaryon processor streaming media display platform, adopt the increase income source code of project of FFmpeg, on two PC, show 2 real-time Video windows respectively synchronously based on ffplay, two PC are provided with the 1G ethernet port separately, be connected with on-chip multi-core network processor by Ethernet, a PC is as video sending end, its Video window is as the transmission video window of video sending end, video sending end in displaying video by on-chip multi-core network processor to another PC synchronized transmission video data, frame of video is packaged into the video packets of data of Ether frame form, load outputs to the 1G ethernet port as Ethernet, send video packets of data in regular turn to on-chip multi-core network processor, the video packets of data zone bit is COMMON; Another PC is as video receiver, receive the Ethernet load by the 1G ethernet port from on-chip multi-core network processor in real time, it is the video packets of data that video sending end is sent, synchronous parsing, processing video data bag, Video window on the video receiver PC is as the receiver, video window, the processed video packet data is put in the video queue, plays the video data that receives; Data transmission adopts raw socket to carry out the transmission and the reception of data, carries out the streaming media display and the test of on-chip multi-core network processor.
Send video window obtains the YUV4:2:0 form on this TV station PC video data, under the video data of YUV4:2:0 form, the Video window resolution is width*height, be width*height pixel, it is 1.5 bytes that each pixel takes up space, each frame of video length is width*height*1.5 byte, be divided into height*1.5 video packets of data, the data length of each packet is a width byte, be set in video sending end the interval time that sends between the adjacent data bag, the reproduction time of the every frame of setting video is t, and the maximum of then described per two packets sends Sleeptime interval time
MaxFor:
Interval time (0, Sleeptime
Max) between value, interval time, value was big more, packet loss is low more.
Video sending end sends after every frame of video sends end and has the packet that zone bit is SYNC, it is sync packet, when video receiver receives that zone bit is the sync packet of SYNC, the content of promptly representing this frame of video finishes whole transmission of transmitting terminal, no matter whether receiving end receives complete frame data, all the data importing in the buffering that will store is in the Video window audio/video player system of receiving end.
Further, video receiver is provided with packet loop variable i, initial value is 0, when checking a zone bit to be the video packets of data of COMMON, with variable i ++, when i equals the video packets of data number of frame of video encapsulation, if checking next one bag is the sync packet of SYNC for zone bit, expression normally receives sync packet, loop variable i zero clearing, a frame of video receives, and is that the video packets of data of COMMON is counted again from next zone bit; If checking next one bag is the video packets of data of COMMON for zone bit, then this video packets of data should be the video packets of data of next frame, the previous frame video Data Transmission finishes, with before the storage buffering in data importing in the Video window audio/video player system of receiving end, loop variable i zero clearing is that the video packets of data of COMMON is counted again from current mark bit.
The operating system that preferred two PC of the present invention adopt is ubuntu linux 2.6.27, and the development environment platform is c/c++eclipse, and compiler is GNU gcc 4.3.2.
The present invention is divided into video sending end and video receiver, connect by on-chip multi-core network processor between the two, video sending end is obtained the video data of yuv format, be packaged into Ether frame then, send by original socket, the video receiving terminal is to receive Ether frame by original socket, after unpacking video data is put in the video queue, plays back.By the demonstration of Streaming Media, and playback rate, effect etc. regulated, represent the performance of on-chip multi-core network processor effectively, intuitively, easily.
Simultaneously, because the network interface card processing power is limited, can cause the packet loss of some, though packet loss is extremely low, but the present invention is through the setting of sync packet, and the influence that can avoid the low volume data packet loss that take over party's video playback is caused effectively is in carrying out the Performance Evaluation of on-chip multi-core network processor to the synchronous playing of Streaming Media, avoid because the reason of network interface card, the Performance Evaluation of on-chip multi-core network processor is impacted.
The present invention be a kind of fully synchronously, do not have disorderly screen, and can add up packet loss, the streaming media display platform that playback rate, effect are regulated is for chip multi-core processor provides a good test and demonstration platform.The problem that can not realize the test and the demonstration of chip multi-core processor intuitively, efficiently, easily at prior art exists the present invention includes following advantage:
1) demo system intuitively: the present invention is based on ffplay, each motor synchronizing shows 2 real-time Video windows on two PC, window as transmitting terminal in displaying video to another PC synchronized transmission video data, another window receives the video data of sending from another PC in real time as receiving end and resolves synchronously, plays.Two online transmission of PC, two sections videos of synchronous playing.The demonstration effect is directly perceived, efficient.
2) mode of operation easily: the software systems of system's operation are operating system ubuntu linux 2.6.27, and development environment platform c/c++eclipse, compiler are GNU gcc 4.3.2.Behind debugging network connection, input compiling, operating instruction, can realize that it is very convenient to operate.
3) extremely strong compatibility: this system is transmitted based on Socket, and realizes by the C language, is applicable to the test and the demonstration of all polycaryon processors even other network environment.
4) at the performance demonstration of on-chip multi-core network processor itself: in the video Data Transmission process; because the network interface card processing power is limited; cause the phenomenon of data-bag lost through regular meeting; set, judge sync packet among the present invention; make the losing of low volume data bag; can not have influence on the video display function of whole platform, reduce the interdependency of whole platform, not disturbed by other factors the demonstration Performance Evaluation of on-chip multi-core network processor to hardware environment.
Description of drawings
Fig. 1 is a data transmission flow process of the present invention.
Fig. 2 is the FB(flow block) of video sending module among the present invention.
Fig. 3 is the FB(flow block) of video reception module among the present invention.
The packet loss test result of Fig. 4 for being provided with based on interval time among the present invention.
Embodiment
The present invention relies on the increase income source code of project of FFmpeg, use raw socket Raw Socket to carry out the data packet transceive of YUV4:2:0 form, carry out online transmission, the synchronous playing of video file, realization is complete synchronously, not having random the screen also can be to the streaming media display platform that packet loss is added up and playback rate, effect are regulated.
As shown in Figure 1, the present invention adopts the increase income source code of project of FFmpeg, on two PC, show 2 real-time Video windows respectively synchronously based on ffplay, two PC are provided with the 1G ethernet port separately, be connected with on-chip multi-core network processor by Ethernet, a PC is as video sending end, its Video window is as the transmission video window of video sending end, video sending end in displaying video by on-chip multi-core network processor to another PC synchronized transmission video data, frame of video is packaged into the video packets of data of Ether frame form, load outputs to the 1G ethernet port as Ethernet, send video packets of data in regular turn to on-chip multi-core network processor, the video packets of data zone bit is COMMON; Another PC is as video receiver, receive the Ethernet load by the 1G ethernet port from on-chip multi-core network processor in real time, it is the video packets of data that video sending end is sent, synchronous parsing, processing video data bag, Video window on the video receiver PC is as the receiver, video window, the processed video packet data is put in the video queue, plays the video data that receives; Data transmission adopts raw socket to carry out the transmission and the reception of data, and in the time of the transmitting-receiving video data, 2 Video window synchronous playing videos carry out the streaming media display and the test of on-chip multi-core network processor.
The YUV that uses among the present invention is belonged to PAL by a kind of colour coding method that the eurovision system is adopted, and is the color space that PAL and SECAM simulation color television system adopt.Y wherein, U, the several letters of V are not the portmanteau words of English word, and Y represents brightness, and U and V represent aberration, are to constitute two colored components.In modern vitascan, usually adopt three pipe colour cameras or colored CCD video camera to carry out capture, then the colour picture signal of obtaining through color separation, obtain RGB behind the amplification correction respectively, obtain brightness signal Y and two colour difference signal R-Y (being U), B-Y (being V) through matrixer again, last transmitting terminal brightness and three signals of aberration are encoded respectively, send with same channel.The method for expressing of this color is exactly that so-called YUV color space is represented.When adopting the storage of YUV4:2:0 form, it is 1.5 bytes that each pixel takes up space, and wherein the Y component accounts for 1 byte, and U and V component respectively account for 0.25 byte, and promptly every frame accounts for width*height*1.5 byte altogether.Just the Video window resolution on two PC is width*height, be width*height pixel, it is 1.5 bytes that each pixel takes up space, each frame of video length is width*height*1.5 byte, be divided into height*1.5 video packets of data, the data length of each packet is a width byte.
Program development among the present invention has utilized the increase income source code of project of FFmpeg.FFmpeg is one, and to collect and compile system, conversion, audio/video coding and decoding function be the complete solution of increasing income of one.The exploitation of FFmpeg is based on (SuSE) Linux OS, but can compile in most of operating systems and use.FFmpeg supports the codings of kind more than 40 such as MPEG, DivX, MPEG4, AC3, DV, FLV, the decoding of kinds more than 90 such as AVI, MPEG, OGG, Matroska, ASF.The player of increasing income such as TCPMP, VLC has all been used FFmpeg.
The ffmpeg project consists of the following components:
1) the capable instrument of ffmpeg video file conversion command supports that also video file is got and is encoded in catching through real-time television;
2) ffserver is used for the multimedia server of broadcasting in real time based on HTTP (RTSP is just under development);
3) a ffplay simple media player of SDL and the exploitation of FFmpeg storehouse;
4) mainly contain sub-directories such as libavcodec, libavformat and libavutil under the FFmpeg master catalogue.Wherein libavcodec is used to deposit each encode/decode module, has comprised the storehouse of all FFmpeg video and audio codings, and in order to guarantee optimal performance and high reusability, from the beginning most of codecs are developed; Libavformat is used to deposit the muxer/demuxer module, and libavutil is used to deposit modules commonly used such as internal memory operation.
The present invention is based on ffplay, mainly be based upon on the basis of ffplay.c file.
Specify realization of the present invention below, adopt communication pattern,, adopt original socket, use sendto () and recvfrom () function to realize data transmission by the communication of Ethernet layer based on Socket.The software systems of platform operation are operating system ubuntu linux 2.6.27, and development environment platform c/c++eclipse, compiler are GNU gcc 4.3.2.
1 video data transmitting terminal
The flow process of video sending end as shown in Figure 2.As figure, video sending end has at first defined structure sock_info definition and procotol, all information that MAC Address is relevant.The structure element comprises Ether frame head pointer, source MAC, target MAC (Media Access Control) address or the like.The create_socket subfunction that definition subsequently is used to connect is called s=socket () and is connected.
Define the send_packet subfunction in the module and be used to send packet.The function local variable comprises the buffer memory buf1 that stores data to be sent, data length length that needs send or the like.Function is at first filled the frame head part of Ether frame.Heading length is 14 bytes, source address, each 6 byte of destination address, and the h_proto variable accounts for 2 bytes.Source address, destination address are filled into preceding 12 bytes of heading by the memcpy function.Adopt the h_proto variable identification data packet type in the heading simultaneously.It is 4 kinds that packet is divided into: flag COMMON represents video packets of data; Flag SYNC represents sync packet; Flag LAST represents the data statistics bag; Flag INVALID represents invalid bag, wherein the flag of COMMON, SYNC, LAST is determined when sending packet by video sending end, the flag INVALID of invalid bag is determined by video receiver, receiving end verifies that at first whether the packet of receiving is for being dealt into this address, as not being then to be defined as invalid bag.
Header begins to fill message information after filling and finishing, equally by the memcpy function with the data conversion storage among the buf1 in interim buffer memory.After the Ether frame encapsulation finishes, send packaged bag by the sendto function, parameter l ength is used to define the data volume that the sendto function need send.
In addition, definition global variable total_sent_packets whenever the sendto function successfully sends a packet, carries out the total_sent_packets++ operation, adds up the data packet number that successfully sends.
Principal function write_function thread at first obtains video data, calls the send_packet function then and sends data.Function at first calls the create_socket subfunction, sets up socket; Next calls the send_packet function, sends packet.In the process that sends every frame data, call 3 for circulations, send Y, U, the V component of a frame respectively, each data packet length is a width byte.For example, the Y component of each frame comprises height*width byte altogether, so need to send the transmission that height secondary data bag can finish a frame data Y component altogether.3 for loop ends, promptly Y, U, V data all send and finish, and represent that frame data transmission finishes.
It is the sync packet of SYNC that every frame data send the back zone bit that finishes, and enters next video requency frame data transmission circulation.
Be Control Network speed, preventing continues to give out a contract for a project causes network congestion, reduces packet loss, is sending next packet again behind the certain hour at interval behind packet of every transmission.
At last, definition sigint function is used for carrying out the operation of sending behind the interrupting information, at first sends the statistical packages that zone bit is LAST.Packet content is give out a contract for a project number, the i.e. value of total_sent_packets of transmitting terminal.In this simultaneously, close socket.
2 video data receiving ends
The flow process of video receiver such as Fig. 3.The create_socket subfunction that at first same definition is used to connect is called s=socket () and is connected.
Module defines the recv_packet subfunction subsequently and is used to receive packet and type of data packet is tentatively resolved.Function return value is the flag type of the packet received, COMMON or SYNC or LAST or INVALID.
Be specially, function at first calls the recvfrom function and receives data from socket, extracts the Ether frame packet header of receiving subsequently and carries out the heading parsing: judge by the memcmp function whether the destination address information in the packet header is identical with the MAC Address of this machine.As difference, illustrate that then this packet is invalid bag, function return value is INVALID; As identical, illustrate that then this packet is effective bag, then the zone bit information h_proto of reading of data packet head and with this rreturn value COMMON or SYNC or LAST as function.While global variable total_packets, the quantity of effective bag of promptly receiving adds 1.
Principal function recv_thread module receives and handle packet.Function at first defines local variable, comprise be used for depositing temporarily the character string array recv_buffer of the every frame data that receive, temporarily deposit each packet that receives character string array packet, be used to store AVPacket type structure pkt, the loop variable i etc. of all data, loop variable i initial value is 0, and call the create_socket function and connect, call the recv_packet function subsequently and receive data.In receiving data procedures, call two while (1) circulation, be respectively subcycle and outer loop, as shown in Figure 3, a subcycle of every execution is equivalent to call packet of recv_packet function reception and stores among the character string array recv_buffer; Outer loop of every execution then is equivalent to receive the data of a frame.Particular content is:
Subcycle: the receiving video data bag, and under the situation that height*1.5 the video packets of data that a frame is comprised all receives, jump out self-loopa and enter outer loop, just the Data Receiving of a frame video is a sub-cyclic process.The rreturn value that the recv_packet function receives data and discriminant function is at first called in subcycle, if rreturn value is COMMON, then be indicated as the general data bag, cyclical function i++ then calls the data message that the memcpy function receives simultaneously and copies among the recv_buffer.If rreturn value is SYNC, show that then packet is a sync packet, promptly height*1.5 packet being comprised of a frame distributed, and has been filled with a content frame among the recv_buffer, then jumps out subcycle.
In addition, be the unexpected situation that prevents that sync packet is lost in the transmission course, set simultaneously in the program: when the value behind the i++ equals height*1.5+1, show that the data of a frame distribute, but sync packet loses, then jump out subcycle equally.
If rreturn value is LAST, show that then packet is a statistical packages, it is last packet, packet content is the quantity statistics that video sending end sends data, then by the data value in strncpy function and the atoi function analytic message, and compose and give global variable send_packets, jump out whole outer loop simultaneously, the corresponding broadcast that realizes one section complete video that video sending end sends of outer loop.
After jumping out subcycle, outer loop composes the content among the recv_buffer to structure pkt, shows that frame data receive, storage finishes, and begins outer loop next time simultaneously, receives the next frame data.Meanwhile, the video data among other modules meeting synchronization call structures pkt in the ffplayer.c program also plays back, thereby realizes the synchronous transmission and the broadcast of video.
After the EOP (end of program), call the sigint function and close socket and calculate the packet receiving rate.Video receiver has defined the sigint function equally and has been used for carrying out the operation of sending behind the interrupting information, at first close socket, the value of the quantity total_packets of effective bag that output is simultaneously finally received and the quantity send_packets of effective bag that transmitting terminal sends, and calculate the effectively packet receiving rate of bag.
Test of 3 packet loss and solution
In the video transmission debug process, through regular meeting packet takes place and slightly lose phenomenon.The reason that data-bag lost takes place mainly is that the packet transmission is too frequent, and the processing power of network interface card is limited.For addressing this problem, we are provided with the interval time that sends between per two packets in video sending end.
The setting of interval time need be satisfied the condition of two aspects: if interval time is too short, then can not effectively reduce packet loss; If interval time is long,, then can influence the normal play of transmitting terminal video because the broadcast and the transmission of transmitting terminal video are synchronous.The reproduction time of the every frame of setting video is t, and the maximum of then described per two packets sends Sleeptime interval time
MaxFor:
With 30 frame 640*480 resolution videos is example, and the reproduction time of every frame is 1/30s, and every frame comprises 480*1.5 packet altogether.For not influencing the correct broadcast of transmitting terminal video, the maximum transmission of packet is interval time:
So the value of interval time should be between 0 ~ 46.3 μ s.In debug process, we have chosen different time values, have tested the packet loss of system.Test result as shown in Figure 4.As can be seen from Figure, interval time, value was big more, and packet loss is low more.Do not influencing under the correct situation about playing of transmitting terminal video, effective bag packet receiving rate that receiving end is received can reach 99.73% at most, and promptly packet loss only is 0.27%.
Though packet loss is lower, list in case can cause losing a video packets of data, just can have influence on the whole transmission process in back because all video packets of data are the order transmission, causes the phenomenon of take over party's video playback confusion, random screen occurs.
For the video that solves the packet loss initiation shows chaotic problem, we have introduced the sync packet sending link.Video sending end sends the sync packet that zone bit is SYNC after the video data transmission of each frame finishes.And when receiving end receives that zone bit is the sync packet of SYNC, the content transmitting terminal of promptly representing this frame all distributes, no matter whether receiving end receives a complete height*width*1.5 byte, the data importing during the recv_buffer that will jump out subcycle and will store data cushions is in audio/video player system.
Like this, even if slight video data packet loss occurred, also only can in video display process, have influence on the broadcast of this frame slightly, and the reproduction time of each frame only is about 1/30 second, so in playing process, iSCSI receiving end window only slight flashing can occur after packet loss occurring, and the broadcast of video not have to influence substantially.
Meanwhile, further, lose for preventing sync packet, in the program when checking a zone bit to be the video packets of data of COMMON, with loop variable i++.Because the every frame of video comprises height*1.5 packet altogether, when i equaled height*1.5, next a bag should be sync packet; If sync packet is lost, first video packets of data that bag is a next frame next then, value behind the i++ promptly equals height*1.5+1 at this moment, when being height*1.5+1, the value of setting loop variable i then jumps out subcycle equally, data importing among the recv_buffer in the video playback formation, can effectively be prevented to lose the phenomenon that influences whole transmission process because of sync packet.
Added sync packet video transmission afterwards, in the receiving end video display process, because the broadcast of next frame video is not subjected to the influence of former frame, and it is extremely short at interval between frame of video and the frame of video, even packet loss is arranged, phenomenon also only can occur slightly flashing, reception, broadcast not influence basically to whole video efficiently solve the influence of packet loss to the receiving end video playback.
In sum, the present invention carries out the transmitting-receiving of packet by using raw socket, realizes synchronous transmission, demonstration and the test function of on-chip multi-core network processor to Streaming Media.Setting through sync packet, the influence that can avoid the low volume data packet loss that take over party's video playback is caused effectively, get rid of the interference of external factor to the Performance Evaluation of on-chip multi-core network processor own, realized synchronous fully, as not have random screen streaming media display, every key technical index of testing on-chip multi-core network processor by the demonstration of Streaming Media effectively.
4 running examples
4.1 running environment:
Hardware environment: the high-resolution of 2 band 1G network interfaces shows the PC of (1600*1200);
Software environment: operating system ubuntu linux 2.6.27, development environment platform c/c++ eclipse, compiler are GNU gcc 4.3.2.
4.2 program run mode:
4.2.1 network connects debugging:
Obtain the authority instruction: sudo su, and input password root123
Connect the other side ip address, as: ping 10.0.0.2
4.2.2 operation eclipse: enter the eclipse file under the terminal, operation sudo./eclipse
4.2.3 compiling and operation:
Claims (7)
1. streaming media display platform of on-chip multi-core network processor, it is characterized in that adopting the FFmpeg source code of project of increasing income, on two PC, show 2 real-time Video windows respectively synchronously based on ffplay, two PC are provided with the 1G ethernet port separately, be connected with on-chip multi-core network processor by Ethernet, a PC is as video sending end, its Video window is as the transmission video window of video sending end, video sending end in displaying video by on-chip multi-core network processor to another PC synchronized transmission video data, frame of video is packaged into the video packets of data of Ether frame form, load outputs to the 1G ethernet port as Ethernet, send video packets of data in regular turn to on-chip multi-core network processor, the video packets of data zone bit is COMMON; Another PC is as video receiver, receive the Ethernet load by the 1G ethernet port from on-chip multi-core network processor in real time, it is the video packets of data that video sending end is sent, synchronous parsing, processing video data bag, Video window on the video receiver PC is as the receiver, video window, the processed video packet data is put in the video queue, plays the video data that receives; Data transmission adopts raw socket to carry out the transmission and the reception of data, carries out the streaming media display and the test of on-chip multi-core network processor.
2. a kind of streaming media display platform of on-chip multi-core network processor according to claim 1, it is characterized in that sending video window obtains the YUV4:2:0 form on this TV station PC video data, under the video data of YUV4:2:0 form, the Video window resolution is width*height, be width*height pixel, it is 1.5 bytes that each pixel takes up space, each frame of video length is width*height*1.5 byte, be divided into height*1.5 video packets of data, the data length of each packet is a width byte, be set in video sending end the interval time that sends between the adjacent data bag, the reproduction time of the every frame of setting video is t, and the maximum of then described per two packets sends Sleeptime interval time
MaxFor:
Interval time (0, Sleeptime
Max) between value, interval time, value was big more, packet loss is low more.
3. a kind of streaming media display platform of on-chip multi-core network processor according to claim 1 and 2, it is characterized in that video sending end transmission after every frame of video sends end has the packet that zone bit is SYNC, it is sync packet, when video receiver receives that zone bit is the sync packet of SYNC, the content of promptly representing this frame of video finishes whole transmission of transmitting terminal, no matter whether receiving end receives complete frame data, all the data importing in the buffering that will store is in the Video window audio/video player system of receiving end.
4. a kind of streaming media display platform of on-chip multi-core network processor according to claim 3, it is characterized in that video receiver is provided with packet loop variable i, initial value is 0, when checking a zone bit to be the video packets of data of COMMON, with variable i ++, when i equals the video packets of data number of frame of video encapsulation, if checking next one bag is the sync packet of SYNC for zone bit, expression normally receives sync packet, loop variable i zero clearing, a frame of video receives, and is that the video packets of data of COMMON is counted again from next zone bit; If checking next one bag is the video packets of data of COMMON for zone bit, then this video packets of data should be the video packets of data of next frame, the previous frame video Data Transmission finishes, with before the storage buffering in data importing in the Video window audio/video player system of receiving end, loop variable i zero clearing is that the video packets of data of COMMON is counted again from current mark bit.
5. a kind of streaming media display platform of on-chip multi-core network processor according to claim 1 and 2, it is characterized in that the operating system that two PC adopt is ubuntu linux 2.6.27, the development environment platform is c/c++eclipse, and compiler is GNU gcc 4.3.2.
6. a kind of streaming media display platform of on-chip multi-core network processor according to claim 3 is characterized in that the operating system that two PC adopt is ubuntu linux 2.6.27, and the development environment platform is c/c++eclipse, and compiler is GNU gcc 4.3.2.
7. a kind of streaming media display platform of on-chip multi-core network processor according to claim 4 is characterized in that the operating system that two PC adopt is ubuntu linux 2.6.27, and the development environment platform is c/c++ eclipse, and compiler is GNU gcc 4.3.2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010217301 CN101916219A (en) | 2010-07-05 | 2010-07-05 | Streaming media display platform of on-chip multi-core network processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010217301 CN101916219A (en) | 2010-07-05 | 2010-07-05 | Streaming media display platform of on-chip multi-core network processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101916219A true CN101916219A (en) | 2010-12-15 |
Family
ID=43323736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010217301 Pending CN101916219A (en) | 2010-07-05 | 2010-07-05 | Streaming media display platform of on-chip multi-core network processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101916219A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554517A (en) * | 2015-12-03 | 2016-05-04 | 浙江大华技术股份有限公司 | Method and device for sending video stream |
| CN108712373A (en) * | 2018-03-30 | 2018-10-26 | 深圳市风云实业有限公司 | One-way communication method, apparatus, user terminal and computer readable storage medium |
| CN109981397A (en) * | 2019-02-13 | 2019-07-05 | 烽火通信科技股份有限公司 | Network performance acquisition methods and system |
| CN111478958A (en) * | 2020-04-02 | 2020-07-31 | 桂林电子科技大学 | Network interface for synchronizing resources of network clock on chip and synchronization parameter configuration method |
| CN112637634A (en) * | 2020-12-24 | 2021-04-09 | 北京睿芯高通量科技有限公司 | High-concurrency video processing method and system for multi-process shared data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1812575A (en) * | 2006-02-24 | 2006-08-02 | 清华大学 | Parallel transmission dispatching method for stream media data |
| CN1933594A (en) * | 2005-09-14 | 2007-03-21 | 王世刚 | Multichannel audio-video frequency data network transmitting and synchronous playing method |
| US20080291279A1 (en) * | 2004-06-01 | 2008-11-27 | L-3 Communications Corporation | Method and System for Performing Video Flashlight |
-
2010
- 2010-07-05 CN CN 201010217301 patent/CN101916219A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080291279A1 (en) * | 2004-06-01 | 2008-11-27 | L-3 Communications Corporation | Method and System for Performing Video Flashlight |
| CN1933594A (en) * | 2005-09-14 | 2007-03-21 | 王世刚 | Multichannel audio-video frequency data network transmitting and synchronous playing method |
| CN1812575A (en) * | 2006-02-24 | 2006-08-02 | 清华大学 | Parallel transmission dispatching method for stream media data |
Non-Patent Citations (1)
| Title |
|---|
| 《电子测量技术》 20100630 潘鹏 等 基于H3MP-16多核处理器片上软件优化 第74页至第78页 1-7 第33卷, 第6期 2 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105554517A (en) * | 2015-12-03 | 2016-05-04 | 浙江大华技术股份有限公司 | Method and device for sending video stream |
| CN105554517B (en) * | 2015-12-03 | 2018-09-28 | 浙江大华技术股份有限公司 | A kind of video flow sending method and device |
| CN108712373A (en) * | 2018-03-30 | 2018-10-26 | 深圳市风云实业有限公司 | One-way communication method, apparatus, user terminal and computer readable storage medium |
| CN109981397A (en) * | 2019-02-13 | 2019-07-05 | 烽火通信科技股份有限公司 | Network performance acquisition methods and system |
| CN109981397B (en) * | 2019-02-13 | 2020-12-01 | 烽火通信科技股份有限公司 | Network performance acquisition method and system |
| CN111478958A (en) * | 2020-04-02 | 2020-07-31 | 桂林电子科技大学 | Network interface for synchronizing resources of network clock on chip and synchronization parameter configuration method |
| CN112637634A (en) * | 2020-12-24 | 2021-04-09 | 北京睿芯高通量科技有限公司 | High-concurrency video processing method and system for multi-process shared data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109996097B (en) | Screen projection method, system and storage device | |
| US10869106B2 (en) | Data transmission method and apparatus, and electronic device | |
| CN104618195B (en) | Bandwidth estimation method and apparatus | |
| US20110013772A1 (en) | Method and Apparatus for Fast Switching Between Source Multimedia Devices | |
| US20140187331A1 (en) | Latency reduction by sub-frame encoding and transmission | |
| US8098690B2 (en) | System and method for transferring high-definition multimedia signals over four twisted-pairs | |
| CN101916219A (en) | Streaming media display platform of on-chip multi-core network processor | |
| CN104038844A (en) | Mobile live broadcast system based on MPEG-2 standard | |
| CN110225347A (en) | Method of transmitting video data, device, electronic equipment and storage medium | |
| WO2016200520A1 (en) | Tunneling hdmi data over wireless connections | |
| CN103607587B (en) | Debugging method and system for video terminal | |
| CN103369299A (en) | Video monitoring method based on H.264 coding technology | |
| US9549015B2 (en) | Communication of multimedia data streams over multiple communication lanes | |
| WO2017015300A1 (en) | Devices and methods for facilitating video and graphics streams in remote display applications | |
| KR100631758B1 (en) | Network I / F card supporting multi streaming format and method | |
| US9686536B2 (en) | Method and apparatus for aggregation and streaming of monitoring data | |
| CN102368721A (en) | Performance testing system and method for MS (Media Server) clustered system | |
| CN103067690A (en) | Method and device of equalized audio-video data transmission based on intelligent monitoring platform | |
| CN113132552B (en) | Video stream processing method and device | |
| CN102215425B (en) | Method and equipment for realizing live video on demand | |
| KR101710011B1 (en) | Image data transmission and reception method and apparatus | |
| KR20000041813A (en) | Method for streaming multimedia through network | |
| CN105304001A (en) | Signal extension box based on SERDES | |
| Chen et al. | Study on relationship between network video packet loss and video quality | |
| CN104038816A (en) | Method and system for video synchronization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20101215 |