The increase in wireless bandwidths over the next few years puts us on the verge of making wireless video as common as the cell phone in your pocket.
One can confidently forecast two
critical wireless developments in the next few years. The first is that
three years from now 65% of the U.S. population will use mobile phones
for voice transmission. This will help drive the second development, the
increase of wireless bandwidth speeds. In the next three years we will
see speeds go from the existing and rather pathetic 9600 bps to an
impressive 384 kbps. This will come about with the implementation of
Third Generation mobile networks or UMTS (Universal Mobile Telephony
Services). This second development will support the dramatic increase in
wireless data services to the general public.
The Holy Grail of wireless communications is ubiquitous wireless
video. Of course we've been able to wirelessly send and receive video
images for some time now. But the increase in wireless bandwidths over
the next few years puts us on the verge of making wireless video as
common as the cell phone in your pocket. In eager anticipation of this,
and with schoolboy zest and glee, phone manufacturers have started
releasing prototypes of their UMTS mobile videophones. Some 30 already
exist, from videophone watches to more conventional form factors (see
sidebar, page xx). Consumers may be able to sit and wait for UMTS to
take a foothold, but life and commercial pressures demand a more urgent
response from m-commerce and mobile professionals.
I have previously expressed my opinion that the combination of
Windows CE, video, and wireless communications is a match made in
Microsoft heaven (September/October '99 Handheld PC Magazine). It
now seems clear that Microsoft is abandoning the "all things to all
people"' high ground it tried to establish for Windows CE, aiming
instead at certain well-defined sectors. Hence, Microsoft repositions
the Palm-size PC (viewed as a PDA) as the Pocket PC (a device that slips
in your pocket, but has the power of a PC). You might not purchase a
Casio E-105 Palm-size PC simply to view your Pocket Outlook information
in color. After all, consumers seem to prefer the Palm PDA for simple
appointment and contact management. However, if you wanted to do that as
well as something requiring a little more power, mobile video
conferencing for example, the E-105 starts to make a lot more sense. You
need the color TFT screen to view the images and the 64bit 133MHz
processing power to compress video and push it down the mobile line.
Current 3COM Palms cannot possibly do this and given their success in
the niche they've chosen it may be a long time before they attempt it.
Vital to mobile video are highly efficient video "codecs"
tailored to specific applications. These compression/decompression
utilities aim at decreasing the amount of data transmitted, allowing a
corresponding increase in the speed or quality of video transmission.
Among the best in terms of quality are "wavelet transmission
systems," which can provide video suitable for evidence in legal
proceedings. The tradeoff is that the video (in frames per second) is
slower than Bitstream H263+ over mobile network.
The latest attempt to optimize this tradeoff is MPEG4,
which works well at delivering streaming content that has already been
encoded, rather than real-time video applications. This is because the
video encoding process typically takes 10 times more processing power than
the decoding process. Using MPEG4 for real-time video applications (i.e.
teleconferencing) at data rates of 9600 bps data over GSM does not produce
very satisfactory results.
These two examples emphasize the fact that there is no
universal format for mobile video. Developers must pick and choose codecs
depending on the application they are working on. Hybrid codecs also
exist, as recently demonstrated at 15 frames per second over 56kbps at the
World GSM Congress in Cannes by Lucent Technologies using multicasting
Shawley Ltd (www.shawley.com)
wireless cameras over a 3G mobile
Printer-friendly version
