【 Weak current College 】 for audio coding knowledge and technology parameters---Power By 【 China power house network 】

Summary: wireless network video communication, efficient video compression and network-friendly design is a very challenging task. On the wireless networks based on Terminal and a network of some of the technical programmes.

Keywords: wireless networking; video communication; QoS; video coding standards; multimedia applications


In recent years, wireless communication technology development has entered a flurry of historical period. Europe's third-generation (3G) mobile communication can not only provide existing mobile phone business, but also can provide a high rate of broadband multimedia services, support to high-quality voice and packet data service, and real-time video transmission. 3G pioneered the wireless communication and Internet, multimedia fusion of new era, the resulting IP wireless multimedia and wireless business will be the next wireless mobile communication business new growth points, delivered a QoS (quality of service) guarantee for real-time video stream is the main content of multimedia applications.

Mobile communication systems have in common is the error rate is high, therefore in wireless communication system for video business is difficult, requiring video decoding, and transmission system to overcome high bit error ratio of the channel, and package loss to provide QoS guarantee.

1 the technical solution based on Terminal

Resolve wireless network QoS of the programme, must be met:

(1) smooth quality degraded when the network state changes, service image quality change is stable.

(2) validity full use of bandwidth resources, when network bandwidth resources down to reduce the output image rate; if bandwidth increase is to increase the output image stream.

Hierarchical coding 1.1 image

Image classification coding is an effective way. The image encoding compression into several different quality stream, each stream has its corresponding QoS.

(1) has its superpositionof each level to code stream can and its corresponding low level code stream consists of higher level of image;

(2) have strong flexibility according to the status of network bandwidth at any hierarchical image into adapted to network bandwidth of the stream;

(3) have a very strong robust if packet loss or error, can only affect one layer of an image without affecting the entire frame of the image, the image quality can be accepted.

Hierarchical image is especially suitable for multicast traffic. Because of network asymmetry, each of your client's network is in an inconsistent state, so you can each receiving a stream; while on the sending end is the same code stream. Video of classification can have 3 hierarchical encoding: space classification, age rating, the SNR (signal to noise ratio for quality) rating. Space classification is the image into several resolution difference image encoding, respectively, to form several stream; age rating is relatively simple, direct in the Bitstream skip image frames to complete; the SNR rating is the image by DCT macroblocks used several different quantization parameter to quantify the difference results, I get several encoding stream for transmission.

1.2 video compression standards [1, 2]

1.2.1ITUT of H.263 and H.263 +

Multimedia communication system, in particular the application of the video, high-capacity data and require a large amount of bandwidth available. As the business continues to grow, but wireless network transfer rate is still limited, and this requires a new coding standard should have a higher compression rate. In 1996, ITUT proposed H.263 can meet this requirement. His coding structure based on the H.261, but to be able to lower the bitrate of the improved image quality. And H.263 + is the first one is specially designed for different network technologies to work on the internationalization of video coding standard. He provides 12 optional mode, for ease of mistakes and no QoS assurance network to improve image quality. Although error H.263 + mainly elastic model to a wired network, but he also for wireless networks. Due to the wireless network video conferencing business demand continues to grow, the International Organization for Standardization for optimized encoding technology.

1.2.2ISO/IEC of MPEG4 video standard

MPEG4 video standards, including a video object Vos (VideoObjects), video object-VOL (VideoObjectsLayer), and the video object plane VOP (VideoObjectPlane). These different elements to different levels, combined with MPEG4 video script (MPEG4VisualProfile) as representing the image data stream.

Mobile multimedia application technical challenge is different from desktop multimedia system. This is because the current mobile technologies inherent limitations, such as computing power limits, bandwidth limiting and unreliable transport, etc. MPEG4 as adaptive programme is ideal for mobile multimedia applications, because the MPEG4 has the following advantages:

(1) you can achieve very high compression ratio.

(2) a flexible encoding and decoding in complexity, such as spatial resolution, temporal resolution, and flexible quality, performance, and price of compromise.

(3) the encoding based on the object, allows video, audio, interactive objects.

(4) facial animation can use parameter-driven manner.

MPEG4 video compression is very useful for image compression technology, mobile multimedia communication, software design should pay attention to MPEG4 compatibility module. Adaptation refers to software on the data flow processing according to the communication device or network conditions change, mobile communication is very important. Adaptation can alsoDynamically on the server side and client-side communication rate, 2.5 generation and third-generation wireless communications is very meaningful, because in their communication standard in the different streams of datasend rate to ensure different network conditions to ensure the quality of service.

1.2.3H.264/AVC video standard

In late 2001, the moving picture experts group (MPEG) and VCEG decided to set up a joint video team (JVT), ITUT upcoming H.264/AVC and MPEG4 AVC latest part of the creation of a separate technical design. H.264/AVC's main goal is to improve the coding efficiency and network adaptability. In the same image quality, the H.264/AVC of algorithms than previous standards such as ISO/IECJTC1MPEG4 ITUTRECH.263 and Bitstream are greatly reduced. Technically, the H.264 standard has multiple Flash, such as unified VLC symbol encoding, high precision, multiple modes of displacement estimation, based on 4 × 4 blocks of integer transform, layered coding syntax. These measures allow H.264 algorithm has a high coding efficiency, the reconstruction of the same image quality, the ability to save than H.263 50 percent rate. H.264 stream structure network adaptability, increased error recovery ability, well adapted to the IP and wireless networks.

1.3 quality control techniques [3]

Quality control of the technology's primary purpose is to ensure that the video transmission quality improvement. Including congestion control and error control. Congestion control aims to avoid network congestion caused by packet loss and loss of quality. For video streaming, congestion control, the main method is the rate control. Typical rate control method based on the coding of extensibility. Including:

(1) frame drop filter he can distinguish between different frame, such as MPEG encoding I frames, B frames, P frames, according to the importance of frames dropped frames (frames, then first B P frames, last I-frame).

(2) tiered drop filter.

(3) frequency filtering.

Congestion control in order to reduce the loss of a package, but was unable to avoid the loss of a package. In this case, you may need certain error control mechanism. Errorcontrol mechanisms include:

(1) FECFEC aim is to increase the redundant information enables packet loss recovery through other packages out of the correct information.

(2) error resilient coding (ErrorResilientEncoding) encoding via the appropriate control makes data loss is to minimize the impact on quality. The advantages of this method is to enable data loss of robustness and enhanced quality, but in the compresses of efficiency is affected.

(3) error concealment (Cancealment) error hiding refers to when an error has occurred, the receiver through certain methods to weaken the influence of human vision. The main method is the time and space of interpolation (Interpolation). Recent research also includes the largest smooth recovery, motion-compensated time predicting, etc.

2 the technical solution based on network

2.1 real-time Transport Protocol (RTP) and real-time Transport Control Protocol (RTCP)

In wireless video, you can use the IETF (Internet Engineering Task Force) recommended RTP and RTSP to complete real-time data transfer. RTP (RealtimeTransportProtocol) and RTCP (RealtimeControlProtocol) is an IP-based application layer protocol. RTP is a real-time audio/video data to provide end-to-end transmission services, including payload type ID, a sequence of labels, time labels and source identification, can provide time information and implement flow synchronization. Since TCP retransmission mechanism in will cause a delay, usually running in UDP over RTP. RTP itself does not provide a reliable delivery mechanism, nor does it provide flow control or congestion control, which is used in conjunction with RTCP, make the best transmission efficiency. RTCP used to monitor service quality and in the exchange of information in the course of the meeting. He provides QoS feedback, participants identify and control package zoom, media synchronization services. RTCP packet contains the packet number sent, lost packets, and other statistical information. On the basis of which the server can dynamically change the transfer rate even payload type. Figure 1 shows the IETF transport protocol stack of the RTP/RTCP in the protocol stack.



2.2 video data refinement package [4]

Video data refinement packaging ensures channel use and error robustness of optimal relationship between. Single MPEG4 stream is usually divided into a series of independent can be decoded, the length of the video group rules, each group has a complex synchronization begins. These video MPEG4 encoders, grouped by, and is seen as part of the compression level. Therefore they should be IP, UDP or RTP layer created by grouping. Distinguish between RTP packet and MPEG4 video group is very important, the establishment of the RTP packet is related to MPEG4 encoding process isolation, and video group, on the other hand, is generated by the MPEG4 encoder. On a fixed-size header, the size of the video load is an important factor in controlling balance, adjustable packaging mechanism in error robustness and throughput of the best relationships.

A group head any part of the damage that will result in the loss of the RTP packet. As a result of an MPEG4 video packet data sensitivity, the situation is more complex. Have data segmentation MPEG4 group is divided into 2 parts, the first part contains the head and movement data, the second part containsThe real data. No to the first part, second part cannot be decoded. Therefore, the first part of the damage caused a complete video packet loss. Any analysis of the RTP package must consider these factors.

There are 2 kinds of packaging mechanism to MPEG4 data package into the RTP packet. In the first mechanism (see Figure 2 (a)), a packet is encapsulated into MPEG4 single RTP group; in the second mechanism (see Figure 2 (b)), an RTP packet contains a video frame, each video frame contains many MPEG4 grouping, each end are inserted into MPEG4 group 8b of cyclicredundancy code (CRC) quietness, to help realize the video group data in a error concealment while maintaining and standard MPEG4 decoder for backward compatibility.

2.3 mobile network priority transmission [1, 3]

Current video coding schemes mainly consider the bandwidth limit, and the high error rate and packet loss rate considered insufficient. Viable solution is to use based on the hierarchy of scalable encoding scheme, based on human visual characteristics, layered video encoding typically use unequal error protection (UEP), is about encoding video stream divided into several different error protection level, main Substream protecting the most important child flow so that high-priority base layer for a guaranteed service quality and refinement. This method is called with layered coding transport priority, in the video transmission system is designed to make the error recovery easier.

To improve the quality of video transport can also be passed to the video data as two separate data streams sent to realize video bit stream different parts of the optimization. In this case the encoder requires network through different priority channel to send data, will be more important and error sensitive data is assigned to a more reliable, more secure channel. Therefore, movement and header data stream is set to a higher level of error protection, more reliable than the texture data streams to transfer of carrier. In MPEG4 compression standard systems, data segmentation is crucial data on each video group, so when the second part of the low sensitivity of texture data errors, lost the possibility of video group. ｝

Priority transmission method of the range of applications including video layering, video data segmentation, UEP and priority of multiple vector video transmission, however in mobile wireless networks, the application tier priority mechanism will enable all network and transport layer header lose protection, high error rate will result in a significant part of the grouped attachment is damaged, for example, sensitive loads data header. Also, the application layer using the priority mechanism to run various applications of the common result limit, this means that system to improve service quality you need to modify the network protocol for all run.



Third-generation mobile multimedia application protocol is worked out, and now there are three basic scenarios: H.324 adaptation wireless protocol extensions; part of the multiplexer H.324 wireless application improvements; the use H.323 IP/UDP/RTP transport protocol. If the IP/UDP/RTP, the physical and link layer protocol must ensure that the error rate is very low, the basic no bit wrong, only the packet loss.

3 conclusion

Communication system characteristics determine the image communication can only be built on existing communication network. Therefore, image communication problems faced by two aspects: on the one hand, on a proper image information processing made him as much as possible to adapt the existing communication infrastructure and communication style; on the other hand, the renovation of existing communications systems so that he as much as possible to adapt to the characteristics of the image information. Therefore, we must realize that to truly wireless video communication in the relevant standards, network, Protocol, and so much more work to do, believe in our efforts, the wireless network for a better tomorrow, will become the mainstream networks for multimedia applications.