Over the last few years, we have seen a revolution in the field of entertainment devices. The list that began with traditional radio and TV sets has exploded into an endless number of devices ranging from smartphones, digital cameras and camcorders, portable media players, mobile Internet devices (MIDs), netbooks, all the way up to large flat-panel displays, home theatre systems and much more.
Now, for most entertainment devices, processing audio and video is one of the most important functions, and selecting an optimal processing solution is one of the keys to success for any product that has to do any significant amount of computing. The desired product features influence the product cost, power consumption and performance as well.
Aspects to be considered
Audio processors take comparatively lower processing power than video processors, but they are both equally important. Let us take a look.
Audio processors. Audio processors serve a variety of fields, and each of these fields has its own challenges and design goals. In some fields, digital signal processing is used to produce high-fidelity sounds such as in the entertainment industry where audio quality is paramount. On the other hand, communication systems require the audio to be clear while keeping within a low data rate. While designing any audio processing system, the designers have three primary targets to achieve: good audibility, intelligibility and fidelity.
Audibility. The audibility of speech or music must be sufficient to achieve the desired effect attained without distortion or feedback.
Intelligibility. Intelligibility is determined by the signal-to-noise ratio and direct-to-reverberant ratio at the listener’s end. Whilst the system must suppress external as well as electrical noise produced within the system itself, controlling the reverberation of acoustics produces good intelligibility.
Fidelity. Fidelity of sound is the overall frequency response of the sound, and a wide and relatively uniform frequency range contributes to realistic and précise augmentation of sound. Fidelity is basically contributed by every component, and any limitation at any point can affect the fidelity of the entire system.
Video processors. Video processing applications are growing exponentially, with the new kind of video-centric products surfacing rapidly. Computationally demanding video processing has different requirements for different applications. For instance, video applications such as home theatre systems require a processor which is flexible enough to connect all components together, process the signal for a large living room, creating an ultimate home theatre experience by delivering premium sound quality.
On the other hand, vehicle-reversing cameras and other small-screen LCD applications for both automotive and non-automotive electronics require good image clarity. These applications also pose challenges such as safety and reduced power consumption, especially in automotive systems in order to minimise the power burden on the battery. This increasing range of applications poses a challenge for any designer who is required to choose from so many processors and their complexities.
Let us take a look at what the audio/video processing industry has in offering for embedded multimedia designs and development of feature-rich products.
Digital and analogue solutions
Analogue processors are used where we require the best quality output. T. Anand, co-founder, Knewron explains, ”The thumb rule is, where the quality of audio/video is of prime importance, we go for the analogue processing. The output from an analogue IC is much better than a digitally processed analogue output.”
The latest analogue I²C-controlled audio processors offer a wide range of features that are suitable for stereo and multichannel applications. These can save costs and enhance the audio signal chain of the designs. Also, these devices with integrated features such as digital volume and balance control, surround sound and tone controls, further enhance the designs.
These surround or other multichannel formats are appropriate for producing virtualised 3D sound for two-speaker systems. Also, there are 3D audio processors which can create five-speaker surround sound from a two-channel stereo source.
Although digital processing can be lossy, the advent of efficient and powerful digital processors is an alternative to noise-prone analogue processing. Praveen Ganapathy, director, Business Development, Texas Instruments, India says, “Anything in the real world is analogue, so traditionally we could do processing in analogue domain; the only challenge is analogue domain is prone to a lot of noise. So we take the inputs in analogue form and then convert them into digital, process in the digital domain and then again convert to analogue form for the output.”
Digital audio processors offer more versatile handling of audio/video streams. Echo cancellation and noise suppression DSP software technology is enhancing the audio quality of wireless products while increasing versatility in multimedia application processor families is allowing developers to design a wide range of end products with minimal incremental PCB design effort. Now you can switch sample rates without changing coefficient and have more flexible designs with the new simplified multichannel designs in digital audio processors.
Modernised system on chips (SoC) solutions
On one hand there is a wide range of fully integrated smart TV SoCs, supporting full HD through the high resolution best suited for 3D graphics and 3D gaming. On the other hand there are scalable processors with affordable ARM architecture solutions. These offer a broad range of performance, price and power consumption to meet just about every need, and also include video accelerators, advanced graphics and display capabilities and high-speed connectivity with a rich set of peripherals that are optimised for a broad spectrum of digital video end equipment.
Consumer. SoCs with ARM architectures plus built-in hardware video accelerator engines are the solutions for applications such as tablets and mobile phones. “The mobile phone budget cannot afford so many dedicated chipsets for analogue-to-digital (A/D) conversion, codecs and for transmission. When you talk of an SoC, it develops a balance between the cost, performance and power, providing the complete ecosystem with the associated software at a good price that can fit into the mobile phone or a tablet,” says Avinash Babu, senior architect, Mistral Solutions.
Automotive. for automotive multimedia designs which have been gaining popularity, the multi-tuner RFCMOS single-chip solutions with embedded AM, FM and DAB tuners offer a combined car radio and audio system fully integrated on a single IC. These can help you build the system with significantly reduced system costs via a reduced bill of material (BOM).
Time to market is also a very important factor for the designers. Gaurav Kapoor, sales manager, Intersil Corp, India, says, “There are simple low-cost controllers that help designers to kick start design and time to market, as there are no software protocols required since these devices can run by setting the simple register mapping.”
These controllers can be best suited for simple automotive safety applications such as rear-view camera display where, instead of having complex software-based controllers for the interaction with the display devices, you can use these dedicated LCD controllers for straight connection with digital LCD panel. You can thus make the solution ready within a couple of weeks and production ready within a few months because of simpler designing.
Critical applications. Talking about surveillance and portable processing needs for weapon-mounted sights, handheld range and target finders, and unmanned air or ground platforms, the module should first meet the demanding size, weight and power constraints. The combination of dedicated image-processing cores and an abundance of peripherals in a single integrated circuit has resulted in all-in-one image processors with a general-purpose computer. These can be useful for performing other critical non-vision tasks, such as flight control and system-to-system communication.
The optimised real-time vision-processing hardware performs operations such as noise reduction, image enhancement, image fusion, stabilisation and object tracking.
These transceivers have evolved offering flexibility for your designs. These allow you to separate the audio from video or vice versa. Such transceivers let the video pass through and bring out a compressed audio on the device for DSP to process. A good thing about such devices is that these are not only useful for home audio and video but also for professional AVs.
Subramanyam of Analog Devices, India says, “You can take the audio out from these transceivers and reinsert it back, which is a kind of requirement for studio equipment. It offers a kind of great flexibility, taking audio from one channel and putting audio and video on different locations.”
An amplifier is an integral part of an audio system. Digital amplifier products, where analogue quality performance, reduced size and improved energy efficiency are at premium, have been the focus of the market in recent times. As a result, the ‘smart’ amplifiers have surfaced for helping the design engineers. These smart amplifiers amplify the signal with constant monitoring of the functional device, thereby protecting it from damage and at the same time ensuring the quality of the output.
Ganapathy explains, “The smart amplifier technology allows you to have a 5W speaker with 10x the range without damaging the speakers, so with 5W you can have the 50W experience and thus have more compact designs. It continuously monitors speaker characteristics and responds accordingly; this is how you get good audio tones without damaging the speakers.”
Over the years use of smart devices has become common, but the potential interference they generate has always retarded the aim of providing an excellent sound performance. The new amplifiers that have come up for the smart devices, provide increased GSM robustness to mitigate the influence of smartphones on loudspeakers, resulting in a cost-effective solution that improves sound quality.
IDEs for better designing
In the recent times, IDE software with feature-rich GUI for the embedded processor families has been the focus of many leading chip vendors. These employ the latest generation of mature code-generation tools and provide seamless, intuitive C/C++ and Assembly language editing, code-gen and debug supports, thereby making processor selection and product design more engineer friendly.
Subramanyam says, “Historically what used to happen was, if you were to create a sound bar or an AVR, you would tend to have the audio engineer and the software engineer sit together and try to choose the product. but in this case the graphical user interface offered for these DSPs enables the audio engineer to change the characteristics of various builders or the crossovers.”
Additionally, the advanced algorithms for video processing have evolved for noise reduction as well as image formatting and conversion. The image enhancement algorithms add details to low-resolution images and adjust colour and contrast giving crisp, clear images on your display.
These advanced algorithms and encodings are reducing the computations, thereby reducing the power consumptions as well. Babu says, “On the way, people are trying to reduce power consumptions using accelerators, which are built using higher silicon technologies, and reducing the amount of data payload through advanced encodings.”
Managing the energy consumption is a major challenge for the application design technology today. Subramanyam says, “Energy consumption is a very important aspect of the consumer electronics industry today. the reason being, when you are watching the TV and you put it on standby, or if you put your audio system on standby, you would like the standby power of these devices to be very low. Most of the DSPs that we design have a full operational mode and then a standby mode. The standby mode helps in saving power.”
“We have ICs ranging from a few nanoamperes to a few milliamperes—it differs from product to product. and the market is pretty sensitive to the products which have sleep modes, hibernation modes, etc,” he adds.
We all know how important are analogue-to-digital converters (ADCs) and digital-to-analogue converters (DACs) for any signal processing system. These become essential while processing analogue signals such as voice or speech, and are used for enhancing voice processing features such as noise suppression, acoustic echo cancellation and multichannel beamforming. They also offer improved performance in voice capture processing such as voice control and recognition. other applications such as digital set-top box systems, digital video camcorders, smartphones and tablets operating with digital signals are also designed with reliable, low-power and high-performance ADCs.
We have single packages for sampling, A/D conversion and anti-alias filtering, generating 24-bit values for both left and right inputs in serial form at sample rates up to 200kHz per channel. Most of the packages employ fifth-order, multi-bit delta sigma modulator followed by digital filtering and decimation, which removes the need for an external anti-alias filter, thereby reducing the number of components required for the designs. Further, audio/video codecs which combine audio ADCs and DACs into single ICs provide maximum flexibility, features and performance in the multimedia designs.
The risk factor
Ever-growing and highly-demanding entertainment and multimedia industry poses some risks as well for your designs. The primary one being the reliability and your commitment to future evolutions of the design. Although multi-vendor architecture is a plus point for the designs but a roadmap for the next-generation architectures and compatibility with the future parts will ensure improved integrations and reduced costs for your designs.
Nate Srinath, founder and director, Inxee says, “Every designer should try to mitigate business risk by adopting a multi-vendor capable architecture. The selection process should consider every A/V processor vendor’s commitment and roadmap, coupled with technical and reference design support, along with proper software tools to mitigate business risk during developmental stages of the product.”
The author is a technical journalist at EFY, Gurgaon