Optimization of Dolby Digital addition in the hott

  • Detail

Optimization of Dolby Digital add in SoC design

Dolby Digital add - the latest generation of Dolby digital technology is entering the next generation of home entertainment applications, which will bring people super sound quality, more efficient audio compression, and a more perfect user experience. Dolby Digital plus, or enhanced AC-3 (e-ac-3), has developed to meet the needs of two markets that are still in their infancy, namely HD DVD and Blu ray player. For the emerging HD DVD and Blu ray players, Dolby Digital plus's data compression capability enables people to achieve the perfect combination of HD video and super voice experience in the movie studio. This technology can also improve the performance of the latest digital TV set-top box, enabling broadcasters to carry out business at a lower bit rate, reduce costs and increase flexibility, thereby benefiting consumers

the power supply is 3-phase 4-wire system. Before a chip or a system can use Dolby digital identification, strict certification testing is essential. MIPS technology company, together with Dolby Laboratories, has developed an optimal beta version of Dolby Digital plus running on any of its 32-bit integrated processor cores, which will greatly shorten the development time of the audio system

emerging standards in the home entertainment market

HD DVD and Blu ray Disc: Dolby Digital is mandatory in the HD DVD specification, while it is optional in Blu ray disc. The specification also supports multiple separated voice streams, so that a movie can run on multiple independent channels at the same time or support multiple languages

for the first time in history, HD DVD and Blu ray players allow external bitstreams from the Internet to be synthesized with local voice on the disc after the disc is sold to consumers. For example, you can use the sound channel on the disc to accompany HD video, and also mix with the latest commentary from the director in the studio

next generation set-top box: for a complete experience, HD TV depends on high-quality voice. Both broadcast and network operators are seeking ways to significantly improve their businesses

Dolby Digital plus's high compression ratio supports 5.1 surround sound as low as 224kbps, enabling broadcasters to improve their business. For those broadcasting companies that use Kbps MPEG stereo, they can use essentially the same band, so as to ensure good machine performance and wide use effect to upgrade to 5.1 channel surround sound (see Figure 1)

figure 1: advantages brought by upgrading to Dolby Digital plus for broadcasters

Dolby Digital plus can be backward compatible with more than 36 million low-level Dolby Digital Audio/video receivers connected with s/pdif installed worldwide (Figure 2a)

figure 2A: the s/pdif connection from a set-top box or optical disc player to an a/v receiver requires that Dolby digital encoding stream be converted into a 640kbps Dolby digital bitstream

with the increasingly popular high-speed HDMI interface in the new set-top box, Dolby digital encoding stream (whether encoded or not) can be directly transmitted to a new generation of receiver, producing unprecedented higher quality voice effects (see Figure 2b)

figure 2B: the HDMI connector on the set-top box aligns the edge of the swing rod with the inner scribed line, so that the full Dolby Digital addition stream can be transmitted directly

implementation method of Dolby Digital addition

Dolby Digital addition reference software is written using floating-point C code, and provides authorization for use in the device or DSP, and then sends the device and DSP to Dolby Laboratory for approval and test

Dolby lab provides users with two sets of tests: one is to train the test vectors of each part of the algorithm at a low level and compare the output of the tested decoder with the output of the reference scheme; The other is more rigorous testing, that is, the use of sophisticated audio testing equipment to test the final implementation plan. These two tests should be passed internally before being submitted to Dolby Laboratories

in order to obtain MIPS and Dolby certification authorization, the MIPs optimized version of the code will reverse the process, which will cause the current to be too large or too small to be pipelined, and provide a high-quality implementation scheme. The adoption and approval of code can greatly reduce the implementation risk and help ensure the rapid approval of Dolby Laboratories (Figure 3)

Figure 3: using MIPS optimization code can save months of time for the R & D and approval process of Dolby Digital plus system chips and devices

optimize the implementation of mips32

in embedded systems, the most efficient audio algorithm uses fixed-point algorithm rather than floating-point algorithm. Therefore, the first step of Dolby reference implementation scheme is to convert floating-point reference code

high quality conversion requires selecting a fixed-point representation method that can achieve the best balance between dynamic range and accuracy. Most importantly, choosing the best representation method can eliminate the saturation check after each operation, which greatly improves the runtime performance

MIPS technology uses a wide area simulation and test process to realize this conversion process, ensuring that the implementation scheme will meet the strict voice quality standards issued by Dolby laboratory. An abstract c++ fixed-point class tracks the dynamic range and accuracy of the data required in the whole algorithm process. At the end of the simulation process, the fixed-point representation and operation will be selected and hardwired back to a pure C implementation

converting source code to fixed points has become an optimization method. MIPS software development environment (SDE) compilation tool series can generate efficient code. This tool is intended to be used in a dedicated MIPS core pipeline

soft audio interface integration

after generating a fully optimized fixed-point implementation of Dolby Digital plus code, MIPs integrates the compression decoder converter with its soft audio interface (SAI) - Sai is a standard API supported by all MIPS optimized audio codecs. By standardizing the method of using MIPS soft speech algorithm, Sai also simplifies the multiplexing of application layer code for different or changing speech processing tasks, and makes it easier to cascade multiple speech codecs using shared intermediate storage buffers (see Figure 4). All these will give the system designer great flexibility

Figure 4: the process of converting floating-point Dolby code into highly optimized fixed-point mip32 code

this scheme can also make the system scalable. A small, low-cost, entry-level 4ke kernel can be used for the implementation of low bandwidth Dolby Digital add, while the latest multithreaded 34k kernel can be used for HD implementation and a main controller (Figure 5)

Figure 5: the tested and optimized Dolby Digital plus code can be used on all series of MIPs cores

finally, for researchers, designing a multi-core system has obvious advantages. The team uses a general R & D environment to make cooperation and system debugging easier, and also simplifies load balancing

Author: Chris cavigioli, Brett Miller, MIPs technology

roger dresser, rob Hislop, Dolby Lab (end)

Copyright © 2011 JIN SHI