Re: [Patents] new MS patents on basis for digital tv standards

[James Heald <j.heald@ucl.ac.uk>]

PILCH Hartmut wrote:

> At
>
>   New Patents Could Propel Microsoft's Digital Media
>   http://www.internetnews.com/bus-news/article.php/3305461
>
> you find a detailed article starting with
>
>    Microsoft has just been awarded two United States patents for a
>    proprietary method of packing a high-definition television signal
>    onto a hard drive, internetnews.com has learned.
>
>    The development could help propel Microsoft's digital media strategy,
>    whereby the company has, since late 2002, worked to position its
>    Windows Media 9 technology as an industry standard for the delivery of
>    high-definition (1280 x 720) video streams directly to the PC.
>
> The patents are found at
>
>   
> http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s1=6,510,177.WKU.&OS=PN/6,510,177&RS=PN/6,510,177
>   
> http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s1=6,683,980.WKU.&OS=PN/6,683,980&RS=PN/6,683,980
>
> The article goes on to quote anonymous experts who praise the methods
> patented here as groundbreaking and speculate that the MS standards
> will become irresistible.


Looks to me to be of questionable importance.

The idea of splitting an image into a lower-resolution coarse-grained 
component and then one or more separate refinement components which take 
the image up to higher resolution has been in the mainstream for a 
/very/ long time.   It's the way that wavelets are very often presented, 
via pictures of quad-trees for example.

It's one of the nice tricks you can do with the COFDM standard for 
all-the-world-except-the-USA digital TV broadcast, for example.  COFDM 
uses spread-spectrum techniques to break the RF bandwith into several 
thousand different orthogonal virtual digital channels; the 
orthogonalisation is explicitly designed to separate the channels into 
ones which are more vulnerable  to particular sorts of interference and 
ones which are less vulnerable; or which are invariant or non-invariant 
to relative motion.  By using the hardiest channals for the 
bread-and-butter information, and putting into the most vulnerable 
channels only the finest-level refinement information, the CODFM people 
were able to demonstrate full HDTV reception on a fixed installation, 
with simultanous good-quality standard definition reception using Nokia 
portable tablets on a moving bus.

With a filing date of 24 March 2000, MS should have a big 
obviousness/prior art problem with Claim 1 of #6,510,177.

I'm not qualified as to whether the detailed design of MS's codec 
represents a breakthrough way of looking at the properties of an image, 
to reveal a new more efficient way of coding the sequential layers -- 
but on a quick reading of the claims, it doesn't look like it; and MS 
doesn't appear to advertise it as such in its 'description of the 
invention'.

Rather the claims, especially those of #6,683,980, looks like a patent 
on the not-very-innovative details of one particular algorithm, not as a 
new technological approach, but to protect low level details of this 
particular codec from interoperability.


Roll on, Europe's article 6a.