Speaker

First, let’s consider the speaker. MPEG-4 encoding (which we cover in a later section)
includes higher-rate source coding for enhanced quality audio. Object coding also
supports stereo and surround sound. The low-cost speakers used in present handsets
may need to be substantially upgraded for future products, or higher-quality headsets,
needing higher-quality audio amplifiers, will need to be specified. New loudspeaker
technologies provide the basis for additional downlink audio bandwidth. One example
is a range of flat speakers from NXT (www.nxt.co.uk). A flat material (cardboard or a
translucent material) is actuated across its whole surface to produce a complex audio
waveform.
The diaphragm of a conventional loud speaker moves like a rigid piston. NXT’s
technique is to make a panel vibrate across its whole surface in a complex manner
across the entire frequency range of the speaker driver. When applied to a mobile
phone, the technology is called SoundVu, reusing the display so that it can double up
as a loudspeaker.
The audio panel is a transparent sheet of an optical polymer material positioned in
front of the display screen and separated by a small air gap. The gap varies in size
depending on the size of the display (it can be used, for example, for PDA LCD screens,
as well as mobile phones). There is some light transmission loss, but the panel can double
up to provide electromagnetic interference suppression and an antiglare screen.
The sound and image are locked together. They originate from the same point in space
and potentially can provide a left-hand channel, center channel, and right-hand channel
(surround stereo from your mobile phone).
The power consumption is claimed to be 1/25 of the power consumed by a magnetic
speaker, a few milliWatts to support a high-fidelity audio output. The device can also
be used to create a touch-screen display. Placing a finger on the screen causes the
device to change its vibrational behavior. Using digital signal processing, it is possible
to determine the finger’s position on the screen within 1 mm. The hardware is already
in place (to provide the audio output), so the only additional cost is the processing
overhead—the product is known as TouchSound.
At time of writing only proof-of-concept products exist. It does, however, illustrate
how changes/developments in handset hardware influence or can influence the offered
traffic mix—in this case, using flat panel speaker technology to support wideband audio
using the Adaptive Multi Rate Wideband (AMR-W) encoder on the downlink to the
handset. The AMR-W codec presently decodes 16 kHz of bandwidth, but consider how
user expectations increase over time. We can buy audio products with audio response
well above the limits of our hearing (super tweeters with response up to 50 kHz).
Evolution in hardware capability effectively determines the software requirements
in the handset. As speaker technology improves, there is a parallel need for increasingly
sophisticated audio management. A working group within the Internet Engineering
Task Force (IETF) is presently working on Extensible Music Format (XMF) to
provide a consistent, standardized way of producing enhanced polyphonic ring tones
and game sounds. The software processors are sometimes described as audio engines
(www.beatnik.com provides some examples).
Now let’s consider downlink image and video bandwidth. 123