Analysis of Portable Mobile TV Receiving System

This paper briefly introduces the new generation of portable mobile TV receiving standard DVB-H and mobile TV source compression coding standards, focusing on the composition of DVB-H system, time slicing technology, MPE-FEC, 4K transmission mode and enabling adaptation Several major technical features of mobile TV reception.

This article refers to the address: http://

DVB-H,

Using portable communication tools such as mobile phones, watching TV anytime, anywhere is a dream. With the establishment of source coding technology, channel transmission and a new generation of basic communication networks, portable mobile reception subsystems are also received from a single text and picture format. Turn to a more colorful form of audio and video reception. In order to adapt to this trend, the television industry has also carried out standard formulation and technology research and development of related technologies. Now let's discuss the related technology.

To watch TV on a mobile phone, there are three technical aspects that need to be handled: the signal source, the transmission path, and the receiving terminal. In terms of signal sources, there is a need for a high compression ratio source compression coding standard; in terms of propagation paths, there are wireless microwave and network transmission. In order to achieve mobile reception, a digital modulation and channel processing technique with strong anti-interference capability is required. In terms of the receiving terminal, it is necessary to develop a chip with high integration, small size, light weight, low power consumption, and a small-sized, high-capacity rechargeable battery.

At present, there are three main ways to achieve this service:

1. Ways to do it with mobile networks

At present, mobile TV services launched by mobile operators in the United States and China are mainly implemented by relying on existing mobile networks. China Mobile's mobile TV service is based on its GPRS network, and China Unicom relies on its CDMA network. Regardless of whether it is a GPRS mobile phone or a CDMA mobile phone, it is necessary to install a corresponding playing software on a mobile phone terminal (usually a high-end product such as a PDA mobile phone) equipped with an operating system, and the corresponding television program source is provided by a mobile communication company or through a corresponding service. Provider to organize and provide.

2. Ways to achieve using satellite networks

It is a very new idea to use a mobile phone to receive satellite TV broadcasts. At present, only South Korea is pushing mobile TV broadcasting (DMB). Such a DMB receiver can provide high quality images, and the receiver module enables the user to simultaneously receive signals from terrestrial wireless television broadcasts and satellite television broadcasts.

3. How to install the digital TV receiver module in the mobile phone

At present, the most popular mobile TV technology is through the integration of digital TV and mobile phones. In this way, a microwave digital television receiving module needs to be installed on the mobile terminal, and the digital television signal can be directly obtained without the link of the mobile communication network. At present, the mobile digital TV standard is only the European DVB-H and Japan's single-band broadcast standard.

In China, only CCTV and a few mobile companies have launched mobile TV services. Take CCTV as an example. Since there is no digital broadcasting network of DVB-H in China, they use the 2.5G or 2.75G network transmission technology to play "mobile TV" programs, that is, using China Mobile GPRS/EDGE network or China Unicom. The CDMA network provides users with online live or on-demand streaming audio and video programs through the WAP portal.

The following discusses the transmission standards and coding standards for mobile TV:

First, the transmission standard of mobile TV - DVB-H

The DVB-H (early DVB-X) standard is called Digital Video Broadcasting Handheld, a transmission standard established by the DVB organization to provide multimedia services to portable/handheld terminals over terrestrial digital broadcast networks. Based on DVB-T, DVB-H is a system that transmits data (mainly digital multimedia data) using IPgrams. This standard is considered to be an extended application of the DVB-T standard, but compared with DVB-T, the DVB-H terminal has lower power consumption, and the mobile receiving and anti-interference performance is superior, so the standard is suitable for mobile phones, Small portable devices such as handheld computers receive signals over terrestrial digital television broadcast networks. In fact, because DVB-H is a standard that supports multimedia services, in addition to TV services, it can also carry out a variety of integrated services such as electronic newspapers, electronic auctions, travel guides, games, video on demand and interaction. In short, the DVB-H standard relies on the current DVB-T transmission system to enable mobile devices such as mobile phones to receive broadcast television signals stably by adding certain additional functions and improved technologies.

In order to reduce the power consumption of small handheld devices, DVB-H uses a technique called "time-slicing" to cut IP packets into data slots for a short period of time. Burst mode transmission. The front end of the receiver is only turned on during the selected service Data Burst. During this very short period of time, the data is received at high speed and can be stored in the buffer of the device. This buffer can store the downloaded content, or directly play the live data file.

1. DVB-H system structure

DVB-H supports small mobile terminal devices such as mobile phones, and is the standard for mobile digital TV transmission. DVB-H is a standard built on DVB data broadcasting and DVB-T transmission, and more focused on the implementation of the protocol. The system front end is composed of DVB-H packager and DVB-H modulator. The DVB-H packager is responsible for encapsulating IP data into MPEG-2 system transport stream, DVB-H modulator is responsible for channel coding and modulation; system terminal is composed of DVB- The H demodulator and the DVB-H terminal are configured, the DVB-H demodulator is responsible for channel demodulation and decoding, and the DVB-H terminal is responsible for related service display and processing.

The DVB-H transmission system also has the following special requirements:

(1) In order to extend the battery life, the terminal periodically turns off a part of the receiving circuit to save power consumption;

(2) Being able to roam and receiving DVB-H services very smoothly while roaming;

(3) The transmission system can guarantee the smooth reception of DVB-H services at various mobile rates;

(4) The system has strong anti-interference ability;

(5) The system is quite flexible to accommodate different transmission bandwidth and channel bandwidth applications.

2. Protocol hierarchy

The DVB-H standard will implement the data link layer and the physical layer.

(1) Data link layer - uses time slicing technology to reduce average power consumption, facilitating smooth and seamless service switching; using MPE (multi-protocol encapsulation) forward error correction technology to improve mobile use C/N threshold and Doppler performance for enhanced anti-pulse interference.

(2) Physical layer - Compared with DVB-T, 4k transmission mode and deep symbol interleaving are added.

Other technical features include: adding DVB-H signaling in the Transmission Parameter Signaling (TPS) bits to improve the speed of service development; and cellular identity (in TPS) to support fast signal scanning during mobile reception and Frequency switching; increase 4k mode to adapt to mobile receiving and single-frequency cellular networks, improve network design and planning flexibility; 2k and 4k modes for deep symbol interleaving to further improve robustness in mobile environments and impact noise environments.

3. Key new technologies

(1) Power consumption: DVB-H requires radio frequency reception and channel demodulation, and the power consumption of the decoding part is less than 100mW.

(2) Network design

Since the receiving antenna is small and single when the DVB-H terminal moves within the network, it is necessary to optimize the design of the single frequency network. To this end, DVB-H has added new technical modules, including:

1 time slicing - based on time division multiplexing technology, saving receiving terminal power consumption and facilitating network switching;

2MPE-FEC——Based on RS error correction coding technology, additional forward error correction coding is added to improve the system's mobility and anti-pulse interference capability;

34k mode - used to increase the flexibility of network design;

4DVB-H TPS - Transmission parameter signaling dedicated to DVB-H for improving system synchronization and service access speed.

The following is a detailed introduction to time slicing, MPE-FEC, 4k mode and DVB-H TPS:

1 time slice

Time slicing technology is the most important new technology module in DVB-H. It transmits data in burst mode, and each burst transmits a service. In the service transmission time slice, the service will occupy the entire data bandwidth separately and indicate The time at which the next identical service time slice is generated, so that the handheld terminal can receive the selected service at a specified time and perform energy saving processing during the idle time of the service, thereby reducing the total average power consumption. During this period, the front-end radiographer is always working. Other business data will be transmitted between the two time slices of the same service. The DVB-H signal is composed of many such time slices. From the perspective of the receiver, the received service data is not a continuous input method like the traditional constant rate, and the data arrives at discrete intervals, called burst transmission. If the decoding terminal requires a lower data rate, it must be constant. The code rate, the receiver can buffer the received burst data first, and then generate a data stream of constant rate. It can not only effectively reduce the average power consumption of handheld terminals, but also achieve a smooth and seamless service exchange basis between different networks.

a, time slicing and power consumption

Time slicing technology uses bursty data transfer and has a higher instantaneous rate than traditional data stream services. In order to achieve power saving requirements, the burst bandwidth is generally about 10 times the fixed bandwidth. For example, a traffic stream with a constant rate of 350 kbit/s means that a burst bandwidth of about 4 Mbit/s is required. Burst bandwidth can save 50% in power consumption with twice the fixed bandwidth, so if the bandwidth is 10 times, you can save 90%.

b, time slicing and PSI/SI

The DVB-H standard specifies that PSI/SI (Program Specific Information, PSI/Service Information, SI) information is not time sliced, and they will be allocated a fixed bandwidth for transmission, mainly because of the current use. The PSI/SI information does not support time slice transfer, and it will be difficult to be compatible with current data tables if changes are made. The PSI information uses four tables to define the structure of the code stream: Program Association Table (PAT), Program Map Table (PMT), Network Information Table (NIT), and Conditional Access Table ( Conditional Access Table, CAT).

The handheld terminal needs to access the NIT (Network Information Table) in the SI and the intermediate code INT table in the DVB-H system. The purpose of the NIT table is to provide information about the physical network whose content is fixed. When the handheld terminal joins a new network, it first receives the table and determines the network parameters. When switching between different transport streams, the handheld terminal needs to read the INT table. Unless the INT table changes in the future, the terminal will no longer receive the INT table. The INT table change information is in the PMT PMT (Program Map Table, PMT). The program map table is identified in the table. The PMT table indicates the PID numbers of the individual code streams that make up the program service, and describes each code stream.

Since the DVB standard stipulates that PSI information must be retransmitted every 100 ms, if the burst transmission service time is longer than 100 ms, the handheld terminal can access all PSI information while receiving the service; if the service transmission time is less than 100 ms, the handheld terminal It is necessary to continue to maintain a working time after the service is received to ensure the completion of the reception of the requested PSI form.

c, time slice and service exchange

The time slicing technology enables the handheld terminal to monitor adjacent cells during idle periods of service transmission, scan other frequency signals, and test signal strength without interrupting the reception of the service. When the user enters the new network, it switches to the different transport streams with the same service in the idle period according to the monitoring result, so as to achieve better seamless service exchange. If the business synchronization is precisely arranged at the front end, the same service can be appeared in different time slots of adjacent peaks in time, and the user will not perceive such changes.

d, time slice and conditional reception

DVB-H can implement conditional access in two ways, one is IP-based conditional access system (IP-CAS, IP data broadcast encryption). All CAS (Conditional Receiving System) related information is in IP data and can support time slicing technology to ensure power savings. However, the DVB-H standard does not need to support bidirectional transmission between CAS and receiver. IP-CAS only needs to support the broadcast environment.

Another way is the conditional acceptance system (DVB-CAS, TV Encryption System) using the DVB universal scrambling algorithm. At this time, transmitting CAS information in the DVB-H system will face some problems. Since DVB-CAS uses ECM (Electronic Counter Measure) to transmit the descrambling key, the ECM cannot perform time slicing. In addition, DVB-CAS also uses EMM (EMM-Entitlement Management Message) for transmitting authorization management information. Since the EMM time interval is random, the terminal must work all the time to ensure that the EMM is not lost, and the direct use of DVB-CAS will affect the network roaming service.

In order to ensure the descrambling work, the receiver must complete the ECM reception, and the system identifies the ECM minimum repetition period through the ECM repetition rate descriptor. If the handheld terminal completes at least one ECM minimum period reception before starting to receive the service data, at least one ECM can be received to obtain the descrambling key. Usually, the effective time of the descrambling key is 10s. For this reason, the receiver must ensure that the descrambling key is received 10s before the arrival of the service data.

The EMM will be transmitted in time slices. First, the EMM is encapsulated into an IP datagram format. The time-sliced ​​manner of EMM-IP data after encapsulation is the same as other IP data, and MPE-FEC is adopted to reduce data loss. From the perspective of the receiving terminal, the IP data carrying the EMM is an additional service that must be received, and the recovered EMM-IP data will be transmitted to the DVB-CAS specific module for EMM information processing.

After processing in the above manner, DVB-CAS will not have any impact on user roaming.

2MPE-FEC

The DVB-H standard adds RS (Reed-Solomon) error correction coding to the IP datagram at the data link layer. As the forward error correction coding of the MPE, the verification information will be transmitted in the specified FEC segment, which we call MPE-FEC. The goal of MPE-FEC is to improve C/N, Doppler performance, and immunity to impulse interference in mobile channels.

Experiments have shown that even in a very bad receiving environment, proper use of MPE-FEC can still recover IP data without errors. The data overhead allocation of MPE-FEC is very flexible. If the other transmission parameters are unchanged, if the verification overhead is increased to 25%, the MPE-FEC can enable the handheld terminal to reach the same C/N as when using antenna diversity reception. In fact, we can compensate for the MPE-FEC overhead by selecting a high-configuration transmission parameter to increase the transmission rate, and it will provide much better performance than DVB-T (without MPE-FEC), for example at high speed, In the case of a single antenna, the handheld terminal using MPE-FEC can receive 8K/16-QAM or even 8K/64-QAM signals in the DVB-T environment. In addition, MPE-FEC provides very good anti-pulse interference capability.

34k mode and depth symbol interleaving

The DVB-H standard adds 4K (4096) mode to the original 2K (2048) and 8K (8192) modes of DVB-T, further enhancing the flexibility of network design by coordinating mobile reception performance and single-frequency network scale. At the same time, in order to further improve the anti-pulse performance of the 2K and 4K modes during movement, the DVB-H standard specifically introduces an in-depth interleaving technique.

In the DVB-T system, the 2K mode provides better mobile reception performance than the 8K mode, but the symbol period and guard interval of the 2K mode are very short, making the 2K mode suitable only for small single frequency networks. The newly added 4K mode symbol has a long period and guard interval, and can build a medium-sized single-frequency network. The network designer can better optimize the network and improve the spectrum efficiency. Although this optimization is not as efficient as the 8K mode, 4K is high. The mode is shorter than the symbol period of the 8K mode, allowing channel estimation to be performed more frequently, providing a better mobility than 8K. In summary, 4K mode performance is between 2K and 8K, providing an additional option for coverage, spectrum efficiency, and mobile reception performance.

The characteristics of the three modes in DVB-H regarding the single-frequency network peak size and mobile receiving performance can be summarized as follows:

The a8K mode is suitable for single transmitters and large, medium and small single frequency networks. Its Doppler performance allows for high speed mobile reception.

b, 4K mode is suitable for single radio and medium and small single frequency network, its Doppler performance allows for higher speed mobile reception.

c, 2K mode is suitable for single radio and small single frequency network, its Doppler performance allows ultra-high speed mobile reception.

Under the condition of impulse noise interference, the noise power is evenly distributed to 8192 subcarriers due to the longer symbol period of the 8K mode, so it has better anti-interference performance than 2K and 4K. To overcome this shortcoming, the DVB-H standard uses 8K symbol interleaver to perform deep symbol interleaving on 2K and 4K, so that both can have anti-pulse interference performance close to 8K mode.

4DVB-H transmission parameter signaling TPS

DVB-H's TPS provides a robust, easy-to-access signaling mechanism for the system to enable the receiver to discover DVB-H services faster. TPS is a signal with good robustness, and the demodulator can quickly lock it even at low C/N conditions. The DVB-H system uses two new TPS bits to identify the presence of the time slice and the optional MPE-FEC, and additionally uses the shared bits already present in DVB-T to represent the 4K mode, the symbol interleaving depth and the peak-to-peak identification.

The DVB-H standard is suitable for mobile communication and multimedia services, preparing for television broadcasting, so video compression technology is crucial. Traditional video compression standards such as MPEG-2 obviously cannot meet the requirements of DVB-H. For this reason, a variety of video compression formats have been examined for DVB-H, the most remarkable of which is.

Second, mobile phone TV source compression coding standard -

It is a new digital video coding standard developed by the ITU-T Video Coding Experts Group (VCEG) and the Joint Video Group (JVT) of the ISO/IEC Moving Picture Coding Experts Group (MPEG). It is both ITU-T and ISO. /IEC Part 10 of MPEG-4. Technically, there are several highlights in the standard, such as: unified VLC symbol encoding; high-precision, multi-mode displacement estimation; integer transform based on 4×4 blocks; hierarchical coding syntax. These technical highlights make it have better compression performance, and also enhance the adaptability to various channels. The "network-friendly" structure and syntax are used to facilitate the processing of bit errors and packet loss; the application range is wide. To meet the needs of different rates, different resolutions and different transmission (storage) occasions; these make the algorithm have high coding efficiency, its compression rate is 2 to 3 times higher than MPEG-2, and the image effect of 1Mb/s rate is close. The image quality of the DVD in MPEG-2, as well, the network adaptability of the code stream structure is also strong, which enhances its error recovery capability and can be well adapted to IP and wireless network applications. It is the most ideal source compression coding standard in mobile TV.

1, the technical characteristics:

(1) The following features are available to improve image quality.

The block size in motion compensation is variable, and the smallest brightness compensation block can be as small as 4×4.

Motion compensation with 1/4 sampling accuracy is used, which greatly reduces the complexity of interpolation processing.

The medium motion vector is no longer limited to the inside of the encoded reference image.

Advanced image selection techniques are used in which images can be predicted with images that have been coded and retained in the buffer.

The order in which the reference images are eliminated must depend on this correlation of the order in which the images are displayed.

The limitation of the reference image and image representation is eliminated, so that the B frame image can also be used as a reference frame to predict an image in many cases.

Weighted prediction is used, which allows a certain weighted compensation prediction and offset, which can greatly improve the coding efficiency in the fade.

In the previous standard, the "skip" area of ​​the predictive coded image was not restricted by motion. Speculation is used for the motion of the "skip" zone. For the bi-predicted B-frame image, an advanced motion prediction method called "direct" motion compensation is used to further improve the coding efficiency.

The direct spatial prediction using intra coding applies extrapolation of the edges of the encoded image to the prediction of the current intra-coded image.

A cyclic deblocking filter is employed, which eliminates the blockiness of block-based video coding in the image and improves the subjective and objective quality of the video.

(2) The following characteristics are used to improve coding efficiency in terms of good prediction methods:

1 The previous standard transform blocks are 8×8, mainly using 4×4 block transforms, which makes the encoder representation signal better local adaptability, more suitable for predictive coding, and reduces the “bell” effect. In addition, image boundaries require small block transforms.

2 Usually use small block transforms, but some signals contain enough correlation, which is required to be represented in large blocks, which is the hierarchical block transform. There are two ways to achieve this. The low frequency chrominance signal can be 8×8; for intra coding, a special coding type can be used, and the low frequency luminance signal can be used as a 16×16 block.

3 All previous standard conversions require 32-bit operations, and C uses only short word length conversions of 16-bit operations.

4 There is a certain tolerance error between the standard inverse transform and the transform. The output video signals of each decoder are different, resulting in small drift, which ultimately affects the quality of the image and achieves perfect matching.

5 Two entropy coding methods, CAVLC (Context Adaptive Variable Length Coding) and CABAC (Context Adaptive Binary Arithmetic Coding) are used, both of which are context-based entropy coding techniques.

(3) It has powerful error correction function and flexible operation in various network environments. The main features are as follows:

The parameter set structure of 1 has designed a powerful and effective transmission header information with strong anti-error characteristics. It adopts a very flexible and special way to process key information separately and can be reliably transmitted in various environments.

Each of the grammatical structures in 2 is placed in a unit called the NAL network abstraction layer, which changes the situation in which the mandatory standard bit stream interface is used in the previous standards, and can adapt to video transmission in different networks, and has a better network. Affinity.

3 can be used in very flexible strip sizes.

4 The image can be divided into strip groups, and each strip can be decoded independently. Flexible Macroblock Sorting (FMO) is highly resistant to data loss by managing the relationships between image regions.

5 supports arbitrary strip sorting, each strip can be decoded almost independently, so the strips can be sent and received in any order. In real-time applications, end-to-end latency characteristics can be improved.

6 To improve the ability to resist data loss, the encoder is allowed to transmit a redundant representation of the image area, which can still be decoded correctly when the main representation of the image area is lost.

7 According to the category of each stripe syntax element, the stripe syntax is divided into 3 parts and transmitted separately.

Here are a few of the important features:

1. Intra prediction

The encoding of I frames is achieved using spatial correlation rather than temporal correlation. The previous standard only used the correlation inside a macroblock, but ignored the correlation between macroblocks, so the amount of encoded data is large. In order to further utilize spatial correlation, intra prediction is introduced to improve compression efficiency. Simply put, intra prediction coding uses the neighboring pixel values ​​to predict the current pixel value and then encodes the prediction error. This prediction is block-based. For luminance components, the block size can be chosen between 16×16 and 4×4. The 16×16 block has 4 prediction modes 16×16, 16×8, 8×16 and 8 9 types of ×8, 4×4 blocks

Prediction mode; for chrominance components, the prediction is performed on the entire 8×8 block, with 4 prediction modes. In addition to DC prediction, each of the other prediction modes corresponds to predictions in different directions.

1) The size of the block used in the prediction is variable

It is assumed that all pixels in the block-based motion model block do the same translation. When the motion is relatively intense or at the edge of the moving object, this assumption will be larger than the actual input, resulting in a large prediction error. Reducing the size of the block allows the assumption to remain true in small blocks. At the same time, the blockiness caused by small blocks is relatively small, thus improving the prediction effect.

A total of seven methods are used to segment a macroblock. The size and shape of the blocks are different in each mode. This allows the encoder to select the best prediction mode based on the content of the image to improve the prediction. Using blocks of different sizes and shapes can reduce the code rate by more than 15% compared to predictions using only 16x16 blocks.

(2) Finer prediction accuracy

In this, the motion vector of the luminance component uses 1/4 pixel precision. The motion vector of the chrominance component is derived from the luminance motion vector. Since the resolution of the chrominance component is half of the luminance component (pair 4:2:0), its motion vector accuracy will be 1/8. The motion vector of the chrominance component of one unit represents only 1/8 of the distance between the sampling points of the chrominance components. Such fine prediction accuracy can reduce the code rate by more than 20% than integer precision.

(3) Multiple reference frames

Multi-reference frame prediction is supported, ie more than one (up to 5) decoded frames preceding the current frame can be used as reference frames to generate predictions for the current frame. This applies to situations where the video sequence contains periodic motion. This technique can improve the performance of motion estimation and improve the decoder's error recovery capability; however, it also increases the buffer capacity and increases the complexity of the codec. Using 5 reference frames can reduce the code rate by 5 to 10% compared to using only one reference frame.

(4) Deblocking filter

Its function is to eliminate the blockiness caused by the prediction error in the reconstructed image after inverse quantization and inverse transform, that is, to eliminate the pixel value jump at the edge of the block, thereby improving the subjective quality of the image and reducing the prediction error. The deblocking filter can also make a judgment based on the image content, smoothing only the pixel value hopping due to the block effect, and leaving the pixel values ​​at the edge of the object in the image discontinuously, so as not to cause an edge. blurry. As with previous deblocking filters, the filtered image will be placed in the buffer for inter prediction as needed, rather than just improving the subjective quality when outputting the reconstructed image. For intra prediction, an unfiltered reconstructed image is used.

3, integer transformation

The DCT encoding is performed on the residual of intra or inter prediction. In order to avoid the problem of mismatch between the encoder and the decoder caused by the rounding error, the definition of the DCT is modified so that the transform can be realized only by integer addition and subtraction and shift operation, so that the influence of quantization is not considered. In this case, the output of the decoder can accurately restore the input of the encoder. Of course, the cost of doing so is that the compression performance is slightly reduced. Furthermore, the transformation is performed for 4x4 blocks, which also helps to reduce blockiness.

In order to further utilize the spatial correlation of the image, after performing the above-described integer DCT on the prediction residual of the chroma component and the prediction residual of the 16×16 intra prediction, the standard also converts the DC coefficient in each 4×4 transform coefficient block. A block of 2 x 2 or 4 x 4 size is formed, and a Hadamard transform is further performed.

4, entropy coding

For data above the Slice layer, the Exp-Golomb code is used, which is a VLC without adaptive capability. For the data below the Slice layer, if it is a residual, there are two entropy coding methods: context-based adaptive variable length code (CAVLC) and context-based adaptive binary arithmetic coding (CABAC); if not Poor, using Exp-Golomb code or CABAC code, depending on the settings of the encoder.

(1) CAVLC

The basic idea of ​​VLC is to use shorter codewords for symbols with higher frequency and longer codewords for symbols with lower frequency. This will minimize the average code length.

In CAVLC, several VLC code tables are used, and different code tables correspond to different probability models. The encoder can automatically select in these code tables according to the context, such as the non-zero coefficients of the surrounding blocks or the absolute value of the coefficients, and match the probability model of the current data as much as possible, thereby implementing the function of context adaptation.

(2) CABAC

Arithmetic coding is an efficient entropy coding scheme in which the code length corresponding to each symbol is considered to be a fraction. Since the encoding of each symbol is related to the result of the previous encoding, it considers the probability characteristic of the whole sequence of the source symbol, rather than the probability characteristic of a single symbol, so that the limit entropy of the source can be approximated to a greater extent. Reduce the bit rate.

The CABAC implements the problem of circumventing the representation problem of infinite precision fractions in arithmetic coding and estimating the probability of source symbols. In CABAC, each time a binary symbol is encoded, the encoder automatically adjusts the estimate of the source probability model (represented by a "state"), and the subsequent binary symbols are encoded on the basis of this new probability model. Such encoders do not require a priori knowledge of the source statistical properties, but are adaptively estimated during the encoding process. This gives CABAC greater flexibility and better coding performance - a code rate reduction of approximately 10%.

5, SP Slice

The main purpose of SP Slice is to switch between different streams, and it can also be used for random access, fast forward/rewind and error recovery of code streams. The different code streams referred to herein refer to the code streams generated by encoding the same source under different bit rate constraints. It is assumed that the last frame in the transmission code stream before switching is Al, and the first frame of the target stream after switching is B2 (assumed to be a P frame). Since the reference frame of B2 does not exist, direct switching obviously causes serious distortion, and this The distortion will be passed backwards. The simple solution is to transmit the intra-coded B2, but the data size of the I-frame is generally large, and this method will cause the transmission rate to increase abruptly. According to the previous assumption, since the same source is encoded, although the bit rate is different, the two frames before and after the switching must have a considerable correlation, so the encoder can use Al as the reference frame of B2 and interframe between B2. Predict, the prediction error is the SP slice, and then the code stream is switched by passing the SP slice. Unlike conventional P frames, the predictions made by generating SP slices are performed in the transform domain of Al and B2. The SP slice requires that the image of B2 after switching is the same as when the target stream is directly transmitted. Of course, if the target of the switch is another stream that is irrelevant, SP Slice will not work.

6, flexible macro block sorting

Flexible macroblock ordering (FMO) refers to dividing macroblocks in an image into groups and independently encoding them. The macroblocks in a group are not necessarily continuous in the normal scanning order, but may be random. The ground is scattered in different positions in the image. Thus, at the time of transmission, if an error occurs and some macroblocks in a certain group cannot be correctly decoded, the decoder can still recover the pixels according to the spatial correlation of the image and rely on the correctly decoded pixels around it.

These features make it a wide range of applications, including video telephony (fixed or mobile), real-time video conferencing systems, video surveillance systems and Internet video transmission, multimedia information storage.

Third, summary

In the end, the DVB-H standard mainly solves two problems based on the convergence of DVB data broadcasting and terrestrial TV standards: it adopts a time division multiplexing based strategy to achieve seamless interaction between power saving and service; using MPE-FEC The technology provides a more robust signal, enabling normal service access for indoor low-rate mobile and outdoor high-speed mobile handheld terminals, especially mobile phones.

With its efficient coding performance, it can be applied to a variety of networks, while also meeting the needs of a variety of applications. Can be applied to a variety of channels based on cable, satellite, modem, DST, etc.; can also be used for storage of video data on optical or magnetic devices and public telephone services based on ISDN, Ethernet, DSL wireless and mobile networks, video Streaming services, MMS services, etc.

In the future mobile video reception, the coding problem is solved. The DVB-H standard solves the problem of video stream transmission, making it a natural development from listening to radio programs, watching video files to watching live TV on mobile phones. process. The impact of this experience on consumers is absolutely unimaginable. With the convergence of the three networks of telecommunication networks, computer networks and cable television networks, it has been determined that these various advantageous technologies will produce a variety of new media in the process of cross-cutting and complementary use, and mobile phones will become a variety of communications. The carrier of the media will become an emerging media-multimedia integrated service terminal. This will definitely bring people faster and more ways to get information.

Highlighting waterproof IP68 LED STRIP,with uniqute technology,can be used underwater.High quality smd 2835 led which is epistar chips packed.This led strip light is very safe since the input voltage is only 24vdc, so it can be avoid of electric shock even if it goes wrong operation.All the pcb and electronic parts are totally covered by special PU glue which is anti-uv for 5 years.The led strip is very energy saving,for example,60leds per meter is only 6w/m..High-flexibility,easy to be installed.These led strips are CE,ROHS,FCC certificated.

SMD2835 Led Strip Light

Smd2835 Led Strip Light,Led Strip Lights For Home,Pool Light Fixture,Outdoor Pool Lighting

Guangdong Kamtat Lighting Technology Joint Stock Co., Ltd. , http://www.ip68ledstrip.com

This entry was posted in on