The patent concerns the use of an error correcting code. If data is to be
transmitted down a noisy communications channel, then it is beneficial to
transmit the message with extra bits including redundant information so that
errors in transmission can be detected and corrected. Originally this might
have been accomplished using checksums. However a sophisticated mathematical
theory developed concerned with the design of efficient error-correcting codes
using techniques from the mathematical disciplines of finite fields (originating
from the work of Galois in the 19th century) and linear algebra (over finite
fields). The resulting theory is described in mathematical textbooks and taught
to mathematics students in universities, and is a research area for
The patent covers a code which apparently converts
10-bit codewords to 32-bit codewords using a Reed-Muller
code. The explicit encodings appear to be listed in Tables 1a to 1f of the
patent specification. These tables explicitly assign a 32-bit codeword to each
10-bit uncoded word. The underlying construction is described in the patent
specification. Presumably if a message is to be transferred from one mobile
device to another over a 3G network, both devices need to be able to encode,
transmit, receive and decode messages according to the same protocol. Therefore
both devices need to practice the patent. Hence a standards-essential patent
built around a mathematical algorithm.
(The above is an initial
assessment, since it would require more time to work in detail through the
[ Reply to This | Parent | # ]
Disclaimers: IANAL, in particular IANAPL!
The US patent system is a
peripheral claiming system. This means that if a patent application
contains patent-eligible subject matter, the scope of monopolies granted to the
patent owner are determined by the "metes and bounds" of the claims.
In other words, unless and until a particular claim in invalidated, any process,
composition of matter, manufacture or machine that falls within the scope of any
patent claim falls within the scope of the corresponding monopoly granted to the
patent owner. Therefore claim construction is an essential part of both patent
prosecution and patent litigation. For information see a webpage discussing
claim construction on the PTO website and also the Wikipedia article on
Now "it is improper to import claim limitations
from the specification" (PTO website). This means that you cannot (for
example) look at the patent specification, see some reference to hardware, and
then presume that the invention is a hardware device, and therefore only
hardware devices fall within the scope of the patent.
Assuming that all
the patent claims are valid, the US Government (through the PTO) has granted to
Samsung 88 inter-related monopolies, one monopoly per claim. In patent
prosecution or patent litigation, these claims have to be construed in order to
determine how they are to be interpreted. Another basic principle is that
"claims must be given their broadest reasonable interpretation in light of
the specification". Moreover "the broadest reasonable interpretation
must be consistent with the interpretation that those skilled in the art would
The list of claims will typically include both independent
and dependent claims. The independent claims are generally as broad as
the PTO will allow, and will be as broad as can be achieved without reading on
any known prior art. Subsequent dependent claims will narrow the scope
of the claims, so that, if broader claims are subsequently invalidated by prior
art, some of the narrower claims will survive. Also claim limitations in the
narrower dependent claims can help ensure that the broader claims are construed
sufficiently widely to encompass the narrower claims.
So much for basic
claim construction rules. Now to Samsung's patent. It would appear that the
independent claims are claims 1, 8, 10, 15, 18, 22, 27, 31, 34, 36, 42, 46, 52,
56, 61, 75, 79 and 82. If any of these independent claims (or any of the
dependent claims hanging off them) reads on an apparatus or method, then the
apparatus or method infringes the patent. And it looks as though Samsung's
patent lawyers have claimed every quirk and characteristic of their encoding
that distinguished their encoding from known prior art.
So what sort of
apparatus would infringe one of these claims? The claims repeatedly refer to a
Transport Format Combination Indicator encoding apparatus which comprises
(i.e. includes) a orthogonal sequence generator and a mask sequence
generator, together with an operation unit for adding such sequences.
Moreover, from a dependent claim, it seems that
"00101000011000111111000001110111" is an example of a mask sequence.
So the question is "whether the broadest reasonable interpretation of the
claim in the light of the specification" would read on a computer running
under appropriate software. If so, then the claim covers software
implementations of the claimed error-correcting code.
Next, look at a
method claim. One such is claim 17, which depends on claim 15. The method must
allow for generating, outputting and adding pluralities of
biorthogonal sequences and mark sequences,
forming a first sequence group, generating and applying a column
transposition function to convert sequences to orthogonal sequences,
inserting a column of zeros in front of the sequences, and generating
and applying a reverse transposition function to sequences. Is it
consistent with the "broadest reasonable interpretation of the claims"
to suppose that this claim could read on a digital computer programmed to carry
out all the necessary generating, outputting, adding, providing, forming,
applying, converting and inserting. If so, a standard digital computer
programmed to carry out the steps of the method would infringe the patent.
[ Reply to This | Parent | # ]