Barnes & Noble Inc. v. LSI Corporation et al.

Filing 303

ORDER Re Claim Construction. Signed by Judge Edward M. Chen on 4/7/2014. (emcsec, COURT STAFF) (Filed on 4/7/2014)

1
2
3
4
5

UNITED STATES DISTRICT COURT

6

NORTHERN DISTRICT OF CALIFORNIA

7
8

BARNES & NOBLE, INC., et al.,

9

No. C-11-2709 EMC

Plaintiffs,
ORDER RE CLAIM CONSTRUCTION
v.

11
For the Northern District of California

United States District Court

10

LSI CORPORATION, et al.,

12

Defendants.
___________________________________/

13
14
15

I.

INTRODUCTION

Plaintiff Barnes & Noble Inc. (“Barnes & Noble”) initiated this action against Defendants

16

LSI Corporation and Agere Systems LLC (“LSI”) seeking a declaratory judgment that the Barnes &

17

Noble “Nook” line of products do not infringe a number of patents. LSI, the assignee of the patents

18

at issue in this case, answered the complaint and asserted counterclaims (and eventually, amended

19

counterclaims) against Barnes & Noble alleging patent infringement. LSI asserts that the Nook

20

product line infringes nine patents – i.e., the ‘730 patent,‘087 patent,‘663 patent,‘006 patent, ‘867

21

patent, ‘958 patent, ‘394 patent, ‘420 patent, and the ‘552 patent. The parties have presented the

22

Court with 34 terms to be construed. This order construes thirteen of those terms – the terms

23

identified as the most significant by the parties.

24
25

II.

LEGAL STANDARD

Claim construction is a question of law to be determined by the Court. See Markman v.

26

Westview Instruments, Inc., 52 F.3d 967, 979 (Fed. Cir. 1995) (“We therefore settle inconsistencies

27

in our precedent and hold that in a case tried to a jury, the court has the power and obligation to

28

construe as a matter of law the meaning of language used in the patent claim.”). “The purpose of

1

claim construction is to ‘determin[e] the meaning and scope of the patent claims asserted to be

2

infringed.’” O2 Micro Int’l Ltd. v. Beyond Innovation Tech. Co., Ltd., 521 F.3d 1351 (Fed. Cir.

3

2008) (quoting Markman, 52 F.3d at 976).

4

Claim construction “begins with the language of the claims themselves” and “claim language

5

‘generally carries the ordinary meaning of the words in their normal usage in the field of

6

invention.’” Apio, Inc. v. Mann Packing Co., Inc., No. C07-5628 JF, 2008 WL 4571558 (N.D. Cal.

7

Oct. 14, 2008) (quoting Invitrogen Corp. v. Bicrest Mfg., L.P., 327 F.3d 1364, 1367 (Fed. Cir.

8

2003)). Thus, “[i]n some cases, the ordinary meaning of claim language . . . may be readily apparent

9

even to lay judges, and claim construction in such cases involves little more than the application of
the widely accepted meaning of commonly understood words.” Phillips v. AWH Corp., 415 F.3d

11
For the Northern District of California

United States District Court

10

1303, 1314 (Fed. Cir. 2005) (en banc). However,

12

16

[b]ecause the meaning of a claim term as understood by persons of
skill in the art is often not immediately apparent, and because
patentees frequently use terms idiosyncratically, the court looks to
“those sources available to the public that show what a person of skill
in the art would have understood disputed claim language to mean.”
Those sources include “the words of the claims themselves, the
remainder of the specification, the prosecution history, and extrinsic
evidence concerning relevant scientific principles, the meaning of
technical terms, and the state of the art.”

17

Id. (quoting Innova/Pure Water, Inc. v. Safari Water Filtration Systems, Inc., 381 F.3d 1111, 1116

18

(Fed. Cir. 2004)).

13
14
15

19

“[T]he specification ‘is always highly relevant to the claim construction analysis. Usually, it

20

is dispositive; it is the single best guide to the meaning of a disputed term.’” Id. at 1315 (quoting

21

Vitrionics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)). “It is therefore entirely

22

appropriate for a court, when conducting claim construction, to rely heavily on the written

23

description for guidance as to the meaning of the claims.” Id. at 1317. On the other hand, courts

24

must avoid “importing limitations from the specification into the claim.” Id. at 1323. In particular,

25

“although the specification often describes very specific embodiments of the invention,” courts must

26

not “confin[e] the claims to those embodiments.” Id.

27

As a general matter, extrinsic evidence such as dictionaries and expert testimony is

28

considered less reliable than intrinsic evidence (i.e., the patent and its prosecution history). See id.

2

1

at 1317-19 (noting that “extrinsic evidence may be useful to the court, but it is unlikely to result in a

2

reliable interpretation of patent claim scope unless considered in the context of the intrinsic

3

evidence”).

4
5

III.
A.

6

‘730 PATENT

Background and General Description of Patent
The ‘730 Patent is entitled “Data Protocol and Method for Segmenting Memory for a Music

7

Chip.” It explains a protocol “for labeling various types of data contained in a music chip” which

8

“includes a hierarchical arrangement of headers for storing information about selections on the chip

9

and the method in which they were coded in the memory of the chip.” ‘730 Patent, Col. 1, 47-51.
The “hierarchical arrangement of headers” includes global headers, “located at the very start of

11
For the Northern District of California

United States District Court

10

memory” which provide information that is necessary to decode the content of the music chip.

12

Examples of that information include “the necessary bit rate, as well as information pertaining to the

13

specific encoding algorithm employed in recording audio on the chip.” Id., Col. 1, 54-56. In

14

addition to the global header, each chip will have a “table of contents” that will “include information

15

on play times, song titles, music category and artist” through “individual headers.” Id., Col. 1, 57-

16

62. The benefit of the header arrangement over prior art is that it permits the user to easily find and

17

select the pre-recorded music located on the music chip. Id., Col. 1, 38-44.

18

B.

19

Representative Claims
Claims 1 and 18 are representative claims for the ‘730 patent and contain the terms the

20

parties have indicated are most significant as to this patent. Claim 1 provides (with terms to be

21

construed in bold):

22
23
24
25
26
27

1.
A data format for use in an audio system wherein pre-recorded
music is digitally encoded in memory of an integrated circuit music
chip, and said music is decoded and reproduced by means of an
associated audio player, said data format for storing information
pertaining to the contents of said music chip, wherein individual tracks
of audio are stored in designated locations in said music chip, said data
format including:
first header having parameters stored therein for use by
said audio player in decoding said digitally encoded music stored
in said memory; and

28

3

1

at least one second header, said second header including
selectable categorical information relating to said individual tracks of
audio stored in said memory.

2
3

730 Patent, Col. 6:14-27. Claim 18 provides (with claims to be construed in bold):

4

18.
A data protocol for use in storing pre-recorded audio in
memory of an integrated circuit chip, said integrated circuit chip being
adapted for use with an audio player, said data protocol comprising:

5
6

global header having parameters stored therein
corresponding to an encoding technique used for storing said prerecorded audio in memory and used by said audio player in
decoding said audio; and

7
8

at least one individual header having multiple data fields, said
data fields including general description information about individual
tracks of said pre-recorded audio.

9

11
For the Northern District of California

United States District Court

10
Id., Col. 7:19-29.

12

C.

13
14

As to the term beginning “first header having parameters . . .” the parties have provided the
following competing constructions:

15
16
17
18

“First Header” / “Global Header”

Barnes & Noble

LSI

Court

“a single data structure that
includes information used by
the audio player to decode the
collection of individual tracks
of audio stored in memory.”

Plain and ordinary meaning, or
“a data structure on a music
chip which includes
information relating to the way
the music tracks were encoded
in the memory of the music
chip for use by the audio
player in decoding the stored
music.”

“a single data structure that
includes information used by
the audio player to decode the
collection of individual tracks
of audio stored in memory.”

19
20
21
22

///

23

///

24

///

25

///

26

///

27

///

28

///

4

1
2

Similarly, for the term beginning “global header having parameters . . .” the parties have
provided the following competing constructions:

3
4
5
6

Barnes & Noble

LSI

Court

“a single data structure that
includes information used by
the audio player to decode the
collection of individual tracks
of audio stored in memory.”

Plain and ordinary meaning, or
“a data structure that includes
information about how the
prerecorded audio was
encoded and is used during
decoding.”

“a single data structure that
includes information used by
the audio player to decode the
collection of individual tracks
of audio stored in memory.”

7
8

The central dispute is whether the “first” and “global” headers are limited to a “single data

9

structure,” and whether this limitation applies to both terms. Barnes & Noble argues that both terms

11
For the Northern District of California

United States District Court

10

are limited to a “single data structure,” while LSI contends neither term is so limited.

12

These terms have been construed previously on two separate occasions. First, in Agere

13

Systems, Inc. v. Sony Corp., No. 2:06-CV-079, 2008 WL 2078308 (E.D. Tex. May 15, 2008), Sony –

14

like Barnes & Noble does here – argued that the term should be construed as referring to a “single

15

data structure” in which “all pre-recorded audio tracks are encoded for storage in memory, which is

16

used by the audio player to decode all tracks for playback.” Id. at *15. By contrast, LSI asserted the

17

same construction it advances in this case. Id.. Despite the parties treating “global” and “first”

18

headers identically, the court construed them separately. As to the “global header” term, the court

19

concluded that it “includes information common to all of the music on the chip.” Id. at *16. As to

20

the “first header” term, however, the court found the term had a “broader scope.” Id. Accordingly,

21

the court construed “global header” as “a single data structure that contains information

22

corresponding to the way in which all pre-recorded audio tracks are encoded for storage in memory,

23

which is used by the audio player to decode all tracks for playback.” Id. It construed “first header”

24

to mean “a data structure on a music chip which includes information relating to the way the music

25

tracks were encoded in the memory of the music tracks were encoded in the memory of the music

26

chip for use by the audio player in decoding the stored music.” Id.

27

///

28

///

5

1

Second, another court in this district provided a tentative claim construction1 in the case

2

Sandisk Corp. v. LSI Corp., No. C 09-02737 WHA, 2010 WL 986992 (N.D. Cal. Mar. 17, 2010).

3

As Sony did in Agere and Barnes & Noble does here, Sandisk argued that the term “first header”

4

should be construed as a “single data structure” while LSI again advanced the same construction it

5

does in this case. Id. at *3. The court found this dispute – among others – all “stem from the fact

6

that the term ‘first header’ never appeared in the specification! Rather, the term ‘first header’

7

appeared solely in claims 1 through 17. In other words, after five columns of discussion and

8

disclosure by the patentee, the claims introduced – for the first time – a term never previously seen

9

in the patent.” Id. As in Agere, the court found that the term “first header” is “broader in scope than
the disclosed ‘global header.’” Id. at *4. However, as to the question the parties have raised in this

11
For the Northern District of California

United States District Court

10

case – whether the first header can only be a “single data structure” – the Sandisk court found that

12

there “can only be a single ‘first header.’” Id. Accordingly, after resolving additional disputes, the

13

Sandisk court adopted the following construction for the term “first header”: “a single data structure

14

that includes information used by the audio player to decode digitally encoded music stored in

15

memory.” Id. at *6.

16

The Court agrees with Barnes & Noble that the term “first header” and “global header” both

17

refer to a single data structure. Beginning with the language of the patent, claim 1 and claim 18

18

speak of a “first header . . . [and] at least one second header” and a “global header . . . [and] at least

19

one individual header,” respectively. ‘730 Patent, claim 1, 18 (emphases added). See also

20

Interactive Gift Exp., Inc. v. Compuserve Inc., 256 F.3d 1323, 1331 (Fed. Cir. 2001) (“In construing

21

claims, the analytical focus must begin and remain centered on the language of the claims

22

themselves . . . .”). Accordingly, whereas the patent expressly accounts for the existence of more

23

than one individual or second header, the claims – and, in fact, everywhere in the patent – speak of a

24

global or first header in the singular. Thus, the plain text of the claim strongly supports Barnes &

25

Noble’s construction. Sandisk, 2010 WL 986992, at *4.

26
27
1

28

Subsequent to the court’s tentative construction, the parties settled the case. Accordingly,
no final construction occurred in the Sandisk case.

6

1

Similarly, as the Sandisk court recognized, the purpose of the “first header” and “global

2

header” supports the limitation of both terms to a “single data structure.” The specification explains

3

that “[t]he present invention is a protocol . . . includ[ing] a hierarchical arrangement of headers

4

about selections on the chip and the method in which they were coded.” ‘730 patent, col. 1:48-50

5

(emphasis added). The hierarchy consists of two tiers: the “global header” (or “first header”),

6

which contains information to “decode the digitally encoded music stored in the memory (e.g.,

7

encoding algorithm, bitrate, etc.),” and the “individual header” (or “second header”) which contains

8

“information about individual music tracks (e.g., artist, album, genre, etc.).” Sandisk, 2010 WL

9

986992, at *4. The invention summary provides that a “global header located at the very start of
memory will specify information needed to successfully decode the content of the music chip” while

11
For the Northern District of California

United States District Court

10

the individual headers are described as having the “music category to which a track belongs . . . the

12

artist, and information for addressing each track selection.” Id., col. 1:51-65; see also C.R. Bard,

13

Inc. v. U.S. Surgical Corp., 388 F.3d 858, 864 (Fed. Cir. 2004) (“Statements that describe the

14

invention as a whole, rather than statements that describe only preferred embodiments, are more

15

likely to support a limiting definition of a claim term.”). Accordingly, the Court agrees with the

16

Sandisk court that the “weight of intrinsic evidence” demonstrate a “clear intent by the patentee to

17

limit the invention to a hierarchy of headers, where multiple ‘music-track-specific’ headers

18

corresponded to a single ‘decoding’ header.” Sandisk, 2010 WL 986992,at *5.

19

LSI argues, however, that the patent discloses the possibility of multiple “first” or “global”

20

headers. LSI relies on the declaration of Dr. Jayant who states, in part: “By storing information

21

about the encoding algorithm in the header of a music file (as opposed to having a single file

22

corresponding to an entire memory chip), the patent aimed to make it ‘possible to encode more on a

23

single chip using different algorithms . . . [and] at different bit rates.’” Decl. of Dr. Nikil Jayant

24

(“Jayant Decl.”) ¶ 13 (Dkt. No. 268-37) (quoting ‘730 Patent, Col. 2:58-59). However, the

25

specification language upon which Dr. Jayant relies states, in its full context, that:

26
27
28

The parameter information of the global header 22 is advantageously
included because as compression technology evolves, it may be
possible to encode more on a single chip using different algorithms,
and almost certainly at different bit rates. Thus, rather than ‘freeze’
the compression algorithm to its current quality using a specific bit

7

1
2
3

rate, it will be more cost effective to generate a specific algorithm
release for each chip. This would allow an album from a specific artist
introduced today to use 128Kbps while an album released at some
future date from the same artist could utilize a different algorithm that
would play at perhaps 32 Kbps with the same quality that the 128
Kbps piece has at present.

4
5
6

‘730 Patent, col. 2:55-67.
While this language is far from a model of clarity, the Court disagrees that this language

language recognizes that as compression technology evolves, it “may be possible” to fit more music

9

onto a single chip. This is demonstrated by the “album” example at the end of this passage – the

10

specification speaks of an album released “today” using a 128Kbps bit rate while “at some future

11
For the Northern District of California

describes a music chip containing music encoded with differing algorithms or bit rates. Rather, this

8

United States District Court

7

date” a more advanced algorithm would allow encoding at a 32Kbps bit rate but at the same quality

12

as a present 128Kbps encoding, but taking up less space in the music chip. See Declaration of Dr.

13

Paris Smaragdis (“Smaragdis Decl.”) ¶ 8 (Dkt. No. 270-44) (“In other words, while an earlier music

14

chip may encode its music using a particular algorithm, a later music chip may store its music

15

encoded using a different, better algorithm that provides the same quality as the algorithm used in

16

the earlier music chip using a much lower bit rate . . . .”). This passage read in context makes clear

17

that the “different,” more advanced, algorithm alluded to in the subject passage pertains to a new

18

algorithm on a new chip, not multiple algorithms on a single existing chip. See id. (“Thus, rather

19

than ‘freeze’ the compression algorithm to its current quality using a specific bit rate, it will be more

20

cost effective to generate a specific algorithm release for each chip.” (emphasis added)).

21

Accordingly, the Court finds that nothing in the patent discloses having multiple bit rates or

22

algorithms (and thereby requiring a multiplicity of first or global headers) on a single chip.

23

Finally, although the Agere court’s holding that “global header” and “first header” should be

24

construed differently – it may be argued that the term “global” more strongly connotes a singular

25

entity with universal application – both parties agree they should be treated similarly, at least on the

26

question of whether they are limited to a “single data structure.” The Court notes that the term “first

27

header” only appears in claim 1 and associated dependent claims. It does not appear in the

28

specification, where only “global header” is used to describe the invention as a whole. Significantly,

8

1

the inventors used the terms interchangeably during the prosecution of the patent. For example, an

2

October 1996 letter sent to a patent examiner states, in part, “[t]he data protocol [described in the

3

patent] contains a global header which will specify information needed to successfully decode the

4

content of the music and at least a second header which is a table of contents that contains various

5

fields of information.” Dkt. No. 270-32, at 11. In support of this statement, the inventor cited to

6

claims 1, 18, and 31 – despite the fact that the term “global header” appears nowhere within claim 1.

7

The Court is cognizant of the general presumption that different terms were intended to have

(Fed. Cir. 2006). However, this presumption only exists in the absence of evidence to the contrary.

10

Id. Additionally, the Federal Circuit has recognized claim drafters can, and do, use different terms

11
For the Northern District of California

different meanings. See Applied Medical Resources Corp. v. U.S. Surgical Corp., 448 F.3d 1324

9

United States District Court

8

to define the same subject matter – particularly where independent claims are involved. See Curtiss-

12

Wright Flow Control Corp. v. Velan, Inc., 438 F.3d 1374, 1380 (Fed. Cir. 2006) (“Different claims

13

with different words can . . . define different subject matter within the ambit of the invention. On

14

the other hand, claim drafters can also use different terms to define the exact same subject matter.”);

15

Mycogen Plant Science v. Monsanto Co., 243 F.3d 1316, 1329 (Fed. Cir. 2001) (citation omitted)

16

(“It is not unusual that separate claims may define the invention using different terminology,

17

especially where . . . independent claims are involved.”).

18

Accordingly, the Court finds that both the “first header” and “global header” identified in the

19

‘730 patent are limited to a “single data structure.” The Court thus provides the following

20

construction for both terms: “a single data structure that contains information corresponding to the

21

way in which pre-recorded audio tracks are encoded for storage in memory, which is used by the

22

audio player to decode tracks for playback.”

23
24
25

IV.
A.

‘958 PATENT

Background and General Patent Description
The ‘958 Patent is entitled “Digital Modulation System Using Extended Code Set.” The

26

invention is targeted at minimizing the effect of multipath interference on wireless signals. As the

27

specification describes:

28

9

1

A wireless communications channel can rarely be modeled as purely
line-of-site. Therefore, one must consider the many independent paths
that are the result of scattering the reflection of a signal between the
many objects that lie between and around the transmitting stations and
the receiving station. The scattering and reflection of the signal
creates many different “copies” of the transmitted signal (“multipath
signals” arriving at the receiving station with various amounts of
delay, phase shift and attenuation.

2
3
4
5
6

‘958 Patent, Col. 1:20-28. The resulting delay between the various signals can create “intersymbol

7

interference” (“ISI”) as well as other problems. The invention seeks to reduce these negative effects

8

on the wireless signal by using a modulation system using a larger code set of M codes of N length,

9

where M > N. Id., Col. 3: 48-50. The resulting signal consists of longer codes less susceptible to
interference while balancing the need to maintain data rate. The codes employed by the modulation

11
For the Northern District of California

United States District Court

10

system are such that they provide low auto-correlation sidelobes (i.e. reduce the chance that a given

12

code will interfere with a copy of itself that has time-shifted) and reduced cross-correlation values

13

(i.e. reduce the chance the chosen codes will interfere with each other) between the codes.

14

B.

15
16

Representative Claims
Claim 1 is a representative claim for the term the parties have identified as being most

significant as to this patent. Claim 1 provides (with terms to be construed in bold):

17

1.
A method for modulating information bits over a radio
frequency communication channel, comprising:

18
grouping a number of information bits,
19
based on the grouping, selecting a code having N chips from a
code set that includes M codes, wherein M > N, and wherein the code
set is derived from a complementary code that provides
autocorrelation sidelobes suitable for multipath environments, and

20
21
22

modulating the phase of at least one carrier signal in
accordance with the selected code.

23
24

‘958 Patent, Col. 10:62-11:4.

25

///

26

///

27

///

28

///

10

1
2
3
4
5

C.

“Code”
As to the term”code,” the parties have provided the following constructions:
Barnes & Noble

“a sequence of chips
representing a real value”

LSI
Plain and ordinary meaning, or
“a sequence of chips”

Court
“A sequence of chips
consisting of real values”

6
7

Accordingly, the parties only dispute whether the term “code” in the ‘958 Patent is limited to real

8

values or if the term also includes complex values (i.e. values with both a real and imaginary

9

component).
Both the Agere court and the International Trade Commission (“ITC”) have construed the

11
For the Northern District of California

United States District Court

10

term “Code” as requiring real values. In Agere, the court noted that the ‘958 patent describes two

12

modes of operation, “a full data rate mode of transmitting codes, and a fallback, or half data rate

13

mode. In the full data rate mode, two separate codes are transmitted at the same time, one code on

14

each of two channels (so-called ‘I’ and ‘Q’ channels). In the fallback mode, a single code is

15

transmitted on both the I and Q channels.” Agere, 2008 WL 2078308, at *3. The court found that

16

Agere had failed to explain how Figure 4 of the patent – a depiction of the invention operating in

17

“fallback mode” – could operate using complex codes. Id. at *4. Accordingly, the court concluded

18

that “[b]ecause the fallback mode requires the transmittal of the same code over both the I and Q

19

channels, it is clear that Mr. van Nee intended for his invention to include the use of only real-valued

20

code.” Id.

21

Similarly, in In Re Certain Audiovisual Components and Products Containing the Same, Inv.

22

No. 337-TA-837, 2013 WL 4406820 (USITC July 18, 2013) (initial determination), the ITC found

23

the ‘958 patent required codes with real values. It held:

24
25
26

The intrinsic evidence requires restricting the claims to real codes
because that is all the ‘958 specification discloses and allows. The
stated purpose of the ‘958 patent, which is to overcome the limitation
of “conventional M-ary keying systems” where “the number of
possible codes M is not more than the code N in chips,” makes clear
that the claim limitation “code” encompasses only real codes.

27
28

11

1

Id. at *82. The ALJ quoted an expert who noted that if the “code length N in chips” were construed

2

to include “complex chips,” a “greater number than N orthogonal sequences of ‘complex’ length N

3

would exist; accordingly, the patent’s description of both its purported problem and its purported

4

solution would be inaccurate.” Id. Finally, the ITC noted that

5

the embodiments depicted in the ‘958 specification are designed for
real codes, and not complex codes. Specifically, the system shown in
Figure 3 of the ‘958 patent cannot accommodate complex codes,
because it cannot place the imaginary part on one channel and the real
part on the other channel. Similarly, the “fallback mode” illustrated in
Figures 4 and 7 requires the simultaneous transmission of the same
code on the I and Q channels, which can be achieved only with real
codes and not complex codes.

6
7
8
9

Id. at *83 (citation omitted).

11
For the Northern District of California

United States District Court

10

The Court agrees with the Agere court and the ITC and concludes that the codes disclosed in

12

the ‘958 patent are “real” codes insofar as they are comprised of real chips. The patent states,

13

“[c]hips are actually code bits, but they are called chips to distinguish them from data bits.” ‘958

14

Col. 7:18-20. The Court begins by noting the difference between a “real” chip and a “complex”

15

chip. A real chip takes a binary value – either 1 or 0 (or, as described in the invention 1 or -1).

16

Because complex numbers take the form “a + bj,” a complex chip takes one of four values – 0 + 0j,

17

0+1j, 1+0j, or 1+1j. See Negus Decl. ¶ 16, Bambos Decl. ¶ 7.

18

There is no dispute that signals are commonly expressed as complex numbers. For example,

19

the patent itself refers to quadrature phase shift keying (QPSK). ‘958 Patent, Col. 8: 14. Nor do the

20

parties dispute that a QPSK signal can be represented by a point in the complex plane. In other

21

words, the four possible QPSK signals can be written as four complex numbers. As noted above,

22

every complex number can be written as a + bj, where a and b are both real numbers, and j is the

23

square root of -1. The numbers “a” and “b” are called the real and imaginary parts of the complex

24

number, respectively. Again, the four values of a complex chip would be 0 + 0j, 0 + 1j, 1 + 0j or 1 +

25

1j.

26
27

Here, the invention does not disclose complex chips as part of its teaching. Instead, it
employs real chips in the processing of complex signals. The specification makes this clear.

28

12

1

Figure 3 represents a digital modulation system which employs the principles of the instant

2

invention. It describes a system in which 10 bits of data are received from a multiplexer. “Then ten

3

bit data symbol is encoded into a I/Q code pair of 11 chip codes or codewords.” ‘958 Patent, Col.

4

7:9-10. The first four bits are placed through a first modulator, which, in the Figure 3 embodiment

5

“corresponds to the I phase modulation branch of the system 30 which produces the I component of

6

the . . . signal to be transmitted.” ‘958 Patent, Col. 7:10-23. The first modulator takes the four data

7

bits and produces the length 11 code described by the invention. Id. The second set of four bits are

8

then placed through a second modulator which then produces “a corresponding one of 16 length 11

9

codes from the extended code set according to the principles of the present invention.” Id., Col.
7:24-28. “The second modulator 34 corresponds to the Q phase modulation branch of the system 30

11
For the Northern District of California

United States District Court

10

which produces the Q component of the . . . signal to be transmitted.” Id., Col. 7:28-32. The I & Q

12

phases process complex signals. That complex signal is expressed as a + bj, wherein “a” is called

13

the “I component” of the signal and “b” is called the “Q component” of the signal. There is no

14

dispute that the real part of the signal is sent through the “I branch” and the imaginary part of the

15

signal is sent through the “Q branch” of a modulation system described above. See ‘958 patent,

16

Figures 1 and 3. Thus, each code in this invention encodes either the real part “a” or the imaginary

17

part “b,” but not both. In either instance, a single codeword contains information about only “a” or

18

“b,” but not both. In Re Certain Audiovisual Components, 2013 WL 4406820, at 83 (“Moreover, the

19

embodiments depicted in the ‘958 specification are designed for real codes, and not complex codes.

20

Specifically, the system shown in Figure 3 of the ‘958 patent cannot accommodate complex codes,

21

because it cannot place the imaginary part on one channel and the real part on the other channel.”).

22

Central to the question at bar, the code chips take a binary value (for example, either 0 or 1;

23

or -1 or 1). Thus, the codes themselves contain only real numbers even though collectively the

24

codes in the I & Q branches process complex values. At the hearing, Barnes & Noble clarified that

25

the construction it seeks does not mean that the codes could only “represent” real values, but rather

26

they contain only real values. This construction would be consistent with the notion that the signals

27

being encoded by the modulator in the I & Q branches could represent real and imaginary values, a

28

point that LSI emphasizes.

13

1

The Court notes that all the illustrative codes in the specification employ binary values – i.e.,

2

real numbers. See ‘958 Patent, col. 5 tbl 1, 2; col. 6, tbl. 3. While the specification does not

3

necessarily dictate claim limitations, it does enlighten claim construction. See Phillips., 415 F.3d at

4

1315 (“[T]he specification ‘is always highly relevant to the claim construction analysis. Usually, it

5

is dispositive; it is the single best guide to the meaning of a disputed term.’” (quoting Vitrionics., 90

6

F.3d at 1582)).

extrinsic evidence in the form of inventor testimony actually supports limiting the term “codes” to

9

“real codes.” See In re Omeprazole Patent Litig., 490 F. Supp. 2d 381, 418 (S.D.N.Y. 2007)

10

(“When the meaning cannot be determined by intrinsic evidence, a court may turn to extrinsic

11
For the Northern District of California

Further, to the extent that the intrinsic evidence is ambiguous as to the meaning of “code,”

8

United States District Court

7

evidence to construe the claims in a patent. Extrinsic evidence consists of all evidence external to

12

the patent and prosecution history, including expert and inventor testimony, dictionaries, and learned

13

treatises, and may be useful to show the state of the art at the time of the invention.” (citations

14

omitted)). For example, during his deposition, Mr. Richard Van Nee had the following exchange

15

regarding the Invention Disclosure Report he produced regarding the ‘958 patent:

16
17

Q.
That, I guess, on page – in the IDR, page 2256, I guess
in that first paragraph there it is talking about: “As an example, for 11
chip codes a system with 16 different codes can be used”?

18
19

A.
Q.
real codes?

Yes.
And the codes you were contemplating here, those were

20
A.

Yes. Those were real codes.

21
22

Q.
channels?

23

A.

24

You would transmit independent codes on the I and Q
Yes. That’s correct.

Van Nee Deposition at 33-34 (Dkt. No. 270-21). Later in the deposition, he was asked:

25

Mr. Qualey:

How is a complex code different from a real

code?
26
....
27
28

A.
That it has complex values. Not just a sign, but also a
[phase], a [phase] other than zero and [1]80 degrees.

14

1
Mr. Qualey: To transmit a complex code you have to encode
it on both the I and Q channels simultaneously?

2
3

A.

Exactly.

4

Id. at 38. Accordingly, the inventor’s testimony supports limiting the term code to “real” codes

5

insofar as the patent does not disclose a same code being simultaneously encoded on both the I and

6

Q branches simultaneously.

7

As a result, the Court construes the term “code” as “a sequence of chips consisting of real

8

values” as proposed by Barnes & Noble. This is consistent with the ultimate construction rendered

9

by the Agere court and the ITC.

11
For the Northern District of California

United States District Court

10

V.
A.

12

‘867 PATENT

Background and General Patent Description
The ‘867 Patent is entitled “Wireless Local Area Network Apparatus.” The patent discloses

13

a wireless network apparatus that provides for a transmitter that periodically sends transmission

14

signals to receivers on the network. ‘867 Patent, Col. 1, 39-45. These periodic signals contain data

15

which permits a timer in the transmitter and a timer in the receiver to be synchronized. Id., Col. 1,

16

47-50. The patent discloses that this arrangement has a power-management benefit. By

17

synchronizing the timers in both the transmitter and receiver, the receivers are able to “periodically

18

switch between a low power consumption state, in which their transceivers are de-energized, and a

19

high power consumption state, in which their transceivers are energized, and can thereby receive

20

periodic signals transmitted from some other station.” Id., Col. 1, 54-69. If the periodic signal

21

informs the receiver that there are data packets to be sent to it, it can remain energized and receive

22

those data packets. If not, the receiver can return to its low-power consumption state. See id., Col.

23

2:7-25.

24

B.

25

Representative Claims
Claims 1, 18, and 20 are representative claims for the ‘867patent and contain the terms the

26

parties have indicated are most significant as to this patent. Claim 1 provides (with terms to be

27

construed in bold):

28

15

1

1.
A method of synchronizing a receiver with a transmitter in a
wireless local area network, comprising:

2
periodically receiving a transmission signal from a transmitter,
the transmission signal including a timestamp field, the timestamp
field including a timestamp having a value m for synchronizing a
receiver timer with the transmitter timer, wherein the timestamp
represents a value within a count sequence of a timer in the
transmitter and wherein the timestamp accounts for delays due to
a busy signal on a medium access protocol, and

3
4
5
6

synchronizing the receiver with the transmitter based on the
timestamp.

7
8

‘867 Patent, Col. 8:19-31. Claim 18 provides:

9

18.
The method of claim 16, wherein the timestamp
accounts for a delay between a start of a process to transmit the
transmission signal and an actual time of transmitting the transmission
signal.

11
For the Northern District of California

United States District Court

10

12

Claim 20 provides:

13

20.

14

A receiver, comprising:
a receiver counter that counts up to n counts, and

15

a radio modem capable of periodically receiving a transmission
signal from a transmitter, the transmission signal including a
timestamp field, the timestamp field including a timestamp having a
value m for synchronizing the receiver counter with a transmitter
timer, wherein the timestamp represents a value m within a count
sequence of the transmitter timer, and wherein the timestamp
accounts for delays due to a busy signal on a medium access
protocol.

16
17
18
19
20

Id., Col. 9:57-67.

21

C.

“wherein the timestamp accounts for delays due to a busy signal on a medium access

22

protocol” / “accounts for delays”

23

The parties have provided the following constructions for the “wherein the timestamp

24

accounts for delays due to a busy signal on a medium access protocol” term that appear in the ‘867

25

Patent:

26

///

27

///

28

///

16

1
2
3
4

Barnes & Noble

LSI

Court

“wherein the timestamp
represents the amount of
delays due to a busy signal on
a medium access protocol”

Plain and ordinary meaning, or
“wherein the timestamp
accounts for the time deferred
for transmission of the
transmission signal due to a
sensed energy level above a
threshold value on the wireless
medium”

“wherein the timestamp
indicates the amount of delays
due to a busy signal on the
wireless medium”

5
6
7
8
9

11
For the Northern District of California

United States District Court

10

Similarly, the parties have provided the following constructions for the more general
“accounts for delays” language that appears in the ‘867 patent:
Barnes & Noble
“represents the amount of
delay”

LSI
Plain and ordinary meaning, or
“accounts for the delay in
transmission of a signal”

Court
“indicates the amount of
delay”

12
13

LSI argues that no construction is necessary for these terms or, in the alternative, that it

14

should be made explicit that the “amount of delay” refers to the delay in transmission of a signal

15

from an access point to a receiver. Barnes & Noble, by contrast seeks to replace “accounts” with the

16

word “represents.” LSI argues that Barnes & Noble’s construction implies that the timestamp would

17

have to be equal to the amount of delay due to a busy signal while excluding other delays (such as

18

the time it takes the transmitter to prepare and actually transmit the message in question). LSI

19

contends that the specification clearly demonstrates that the timestamp is taken from a counter only

20

after busy-signal delay and message processing- and transmission-related delays are taken into

21

account. Barnes & Noble disagrees, noting that such “message processing” and “transmission

22

related” delays “are present whether or not there has been a delay due to a busy signal and B&N’s

23

proposed construction does not exclude any such delay.” Dkt. No. 270-4, at 13 n.8.

24

The “accounts for delay” term has been construed by the Eastern District of Texas in the

25

Agere case. In Agere, Agere Systems offered the same construction as LSI does in this case, while

26

Sony proposed a construction of “the timestamp contains a value representing the amount of delay

27

resulting from the medium access protocol being busy.” Agere, 2008 WL 2078308, at *10. The

28

court ultimately concluded that:

17

1

4

Based on the claim language, the specification, and the prosecution
history, the court is persuaded that the inventors used the phrase
“wherein the timestamp accounts for” to mean “wherein the timestamp
indicates the amount of.” As such, the court construes the phrase as a
whole to mean “wherein the timestamp indicates the amount of time
deferred for transmission of the transmission signal due to a sensed
energy level above a threshold value on the wireless medium.”

5

Id. Thus, the court in Agere largely adopted the construction advanced by LSI in this case, but with

6

the substitution of “indicates” for “accounts for.” Similarly, the International Trade Commission

7

(“ITC”) has construed the more general “accounts for delay” terms as “indicates the amount of

8

delay.” In Re Certain Audiovisual Components and Products Containing the Same, Inv. No. 337-

9

TA-837, 2013 WL 4406820, at *116 (USITC July 18, 2013) (initial determination).

2
3

The Court begins by noting that while the term “wherein the timestamp accounts for delays

11
For the Northern District of California

United States District Court

10

due to a busy signal on a medium access protocol” does not expressly account for other delays that

12

are inherent in the transmission of a signal from a transmitter to a receiver, the specification

13

acknowledges there are other delays inherent in the process. Specifically, as Figure 4 demonstrates,

14

even after the transmitter has determined that the network is not busy, there is still an inherent delay

15

as a request to sent (RTS) signal and then a clear-to-send signal (CTS) are issued. ‘867 Patent, Fig.

16

4; see also id., Col. 5:25-31 (“If no medium busy is issued, so the medium is sensed ‘free’, the

17

WMAC control turns on the transmitter of the modem by issuing a request to send (RTS) signal.

18

The modem will then start to send a training sequence and will issue a clear-to-send signal (CTS)

19

once the training sequence is complete.”). It is not until this process is complete that the time stamp

20

is taken and the TIM packet sent. Id. Fig. 4; see also id., Col. 5:37-43.

21

The claim language here is elastic enough to account for these delays. In contrast, Barnes &

22

Noble’s proposed construction which would replace “accounts for” with “represents” is not

23

supported by the claims language or the specification. The ordinary meaning of represent, the

24

construction advanced by Barnes & Noble, is to “correspond to.” Webster’s Third New

25

International Dictionary 1926 (2002). By contrast, the ordinary meaning of the phrasal verb “to

26

account” as used in the subject claims is more open ended – to “supply” or “make up a specified

27

amount or proportion of.” Id.; cf. Tex. Digital Systems, Inc. v. Telegenix, 308 F.3d 1193 (Fed. Cir.

28

2002) (“Dictionaries are always available to the court to aid in the task of determining meanings that

18

1

would have been attributed by those of skill in the relevant art to any disputed terms used by the

2

inventor in the claims.”).

3

Adopting Barnes & Noble’s construction of the term would mean that as to those claims

4

which provide “wherein the timestamp accounts for delays due to a busy signal on a medium access

5

protocol,” the term would be construed such that the timestamp corresponds to any delay due to a

6

business signal. This would be inaccurate and inconsistent with the invention’s specification insofar

7

as the time stamp would also have the inherent RTS and CTS delays. Figure 4 and the invention’s

8

specification makes clear that the time stamp “represent[s] the state of the access point counter 22 at

9

the exact time of transmission of the TIM packet.” Id., Col. 7:44-45; see also id. Col. 5:2-4 (“A
TIME STAMP FIELD in which is loaded a so-called time stamp of the value of the modulo n

11
For the Northern District of California

United States District Court

10

counter in the transmitter 20 at the time of transmission of the TIM . . .”). See Phillips, 415 F.3d at

12

1315 (“[T]he specification is always highly relevant to the claim construction analysis. Usually, it is

13

dispositive; it is the single best guide to the meaning of a disputed term.” (citation omitted)). While

14

Barnes & Noble argues that its proposed construction does not “exclude any such delay,” dkt. No.

15

270-4, at 13 n.8, its choice of the term “represents” so implies.

16

In light of the specification and claims language, the Court agrees with the ITC and Agere

17

court to the extent they construed the “accounts for delay” terms by replacing “accounts for” with

18

“indicates.” Further, the Court construes the portion of the phrase providing “due to a busy signal

19

on a medium access protocol” as “due to a busy signal on the wireless medium.” Use of the phrase

20

“indicates the amount” in place of “accounts for” makes express that the timestamp must not only

21

indicate that there has been a delay, but informs the receiver as to the amount of that delay.

22

D.

“a timestamp having a value m for synchronizing a receiver timer with the transmitter timer,

23

wherein the timestamp represents a value within a count sequence of a timer in the

24

transmitter” / “a receiver counter that counts up to n counts”

25

As to the various terms in the ‘867 patent which relate to a timestamp having a value m, the

26

parties have provided the following construction:

27

///

28

///

19

1
2
3
4
5

Barnes & Noble

LSI

Court

“a timestamp representing a
value m within the range 0 to n
in the counter of the
transmitter, where n represents
the interval between
transmission signals”

Plain and ordinary meaning,
or “a timestamp representing a
value of a counter in the
transmitter”

“a timestamp representing a
value m within the range 0 to n
in the counter of the
transmitter, where n represents
the interval between attempted
transmission signals”

“receiver counter that counts
from 0 to n, where n represents
the interval between
transmission signals”

Plain and ordinary meaning, or
“a counter in the receiver
configured to count up to n
counts, where n is any whole
number.”

“receiver counter that counts
from 0 to n, where n represents
the interval between attempted
transmission signals”

6
7
8
9

11
For the Northern District of California

United States District Court

10
In this case, the two disputes between the parties appears to be whether: (1) the counter

12

resets to zero after each transmission; and (2) whether the time interval between all transmissions is

13

the same (a period labeled as “n”). The Court agrees with the construction provided by Barnes &

14

Noble – one that has been previously adopted by the ITC.

15

The court in Agere has previously construed this term. In that case, Agere Systems advanced

16

a construction that is materially similar to that advanced by LSI in this case – that the “timestamp

17

represents a value of a counter in the transmitter.” The court “expressly disapprov[ed] . . . of

18

Agere’s proposed constructions, because they appear to broaden the scope of the claims beyond their

19

plain language.” Agere, 2008 WL 2078308, at *8. Sony, in turn, advanced the following

20

construction of the timestamp terms: “the timestamp contains the value of a modulo n counter in the

21

transmitter at the time of transmission of the transmission signal, wherein n is the period between

22

successive TIM messages.” Id. at *7. The court rejected this construction as well, finding that one

23

claim – claim 49 – was “ambiguous on the issue whether the timestamp must be a value m in a count

24

sequence of the transmitter timer.” Id. at *8. Accordingly, the court stated that Sony’s construction

25

was improper because “the applicants did not clearly and unmistakably limit the scope of their

26

invention to transmitters that include modulo n counters as described in the preferred embodiment

27

and the cited portions of the prosecution history.” Id. Ultimately, the court provided the following

28

20

1

construction: “a timestamp which represents a value within a count sequence of the transmitter timer

2

for synchronizing the receiver with a transmitter timer that counts up to n counts.” Id.

3

By contrast, the ITC construed the “timestamp” phrases as meaning “a timestamp

4

representing a value m within the range 0 to n in the counter of the transmitter, where n represents

5

the interval between transmission signals” – the precise construction advanced by Barnes & Noble

6

in the instant case. In Re Certain Audiovisual Components, 2013 WL 4406820, at *113. The Court

7

agrees with the ITC’s construction.

8

First, the Court finds that requiring that a count sequence be between 0 and n and that the

Patent. In a brief appealing the PTO’s final rejection of an early version of the ‘867 patent, the

11
For the Northern District of California

timestamp be a “value m” where 0 < m < n is supported by the prosecution history of the ‘867

10

United States District Court

9

applicants noted that “the term ‘time stamp’ is well defined in Appellants’ specification,” “that it

12

“represents a value m within a count sequence,” and that the “count sequence ranges from 0 to n,

13

where 0 < m < n.” 7/18/1995 Brief on Appeal in App. Ser. No. 08/155,661, at 11-12 & n.* (Dkt.

14

No. 270-27). See, e.g., Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1325 (Fed. Cir. 2002)

15

(“Arguments and amendments made during the prosecution of a patent application and other aspects

16

of the prosecution history, as well as the specification and other claims, must be examined to

17

determine the meaning of terms in the claims.” (citation omitted)). It is apparent that the inventor

18

was not just describing a preferred embodiment, but rather relied upon the count sequence ranging

19

from 0 to n, and the term m being between 0 and n, in distinguishing prior art. Specifically, the

20

inventor sought to distinguish the Nagata invention. While the details of the Nagata invention are

21

not provided in the ‘867 inventor’s brief, the Nagata invention apparently employed a sync code

22

detector which employed a 31-bit shift register. Id. at 14. In distinguishing the Nagata invention,

23

the ‘867 inventor provided:

24
25
26
27
28

Nagata’s “sync code” does not represent a value m within a
count sequence of a counter of a transmitter . . . at a time of
transmission of a TIM packet . . . .
In contrast, detection of Nagata’s sync code by the sync code
detector 430 . . . causes a detection pulse to be generate which in turn
resets a timer 460 and causes a timer 470 to start counting . . . . Note
that timer 470 is not caused to start counting from a value which is
intermediate any [sic] count sequence such as a value m, where the
count sequence ranges from 0 to n, and 0 < m < n.

21

1

Thus, Nagata’s sync code is not equivalent to the “time stamp”
of Applicants’ claim 11.

2
3

Id. See also Andersen Corp. v. Fiber Composites, LLC, 474 F.3d 1361, 1374 (Fed. Cir. 2007) (“[A]s

4

we have made clear, an applicant’s argument that a prior art reference is distinguishable on a

5

particular ground can serve as a disclaimer of a claim scope even if the applicant distinguishes the

6

reference on other grounds as well.”).2

7

Second, while the claims do not define “n,” the specification repeatedly indicate that “n” is

61 (“The modulo n counter 22 functions as a timer and when the count value reaches n, a TIM

10

function generator 24 is triggered by way of an interrupt signal 25 indicating that the next TIM

11
For the Northern District of California

the interval between the transmission attempts of the TIM packet. See, e.g., ‘867 Patent, Col. 4: 57-

9

United States District Court

8

packet should be constructed, and transmitted . . . .”); id. at Col. 6:43-48 (“Once the modulo n

12

counters 22, 58 in the station 12.1 and the access point 14 are accurately synchronized, the counter

13

58 provides the station 12.1 with an accurate indication of the time at which the counter 22 in the

14

access point 14 reaches its n value and generates a TIM packet for transmission.”). Further, the

15

specification indicates that the transmitter attempts its transmissions at regular intervals. See, e.g.,

16

‘867 Patent, Col. 3:8-10 (recognizing that TIM packets are “transmitted at regular intervals from the

17

access point 14 and indicate for which stations . . . in the BSA 10 data packets are buffered in the

18

access point 14.”).

19

The Court is aware that claims “ are not necessarily restricted in scope to what is shown in a

20

preferred embodiment.” Phonometrics, Inc. v. N. Telecom Inc., 133 F.3d 1459, 1466 (Fed. Cir.

21

1998). At the same time, however, neither are the “specifics of the preferred embodiment irrelevant

22

to the correct meaning of claim limitations.” Id. As discussed above, the inventor expressly

23

distinguished the present invention from prior art by reference to a timer employing a count

24
2

25
26
27
28

LSI argues that the statements in the 1995 appellate brief before the PTO should be
disregarded because the ‘867 Patent was only approved after the “application was amended with the
‘account for delay’ limitations and related requirements.” Dkt. No. 273-4, at 16. While it may be
true that the ‘867 Patent was further amended to include additional limitations, it does not change
the fact that the applicants advanced an express definition of the term “time stamp” before the PTO
in an attempt to distinguish prior art. As cited above, “an applicant’s argument that a prior art
references is distinguishable on a particular ground can serve as a disclaimer of a claim even if the
applicant distinguishes the reference on other grounds as well.” Anderson, 474 F.3d at 1374.

22

1

sequence between 0 and n. In finding that the value n represents the interval between transmission

2

attempts was more than simply a “preferred embodiment,” the ITC noted that “[t]he ‘867 patent . . .

3

discloses and suggests no value for n that represents anything other than the interval between

4

transmission signals.” In Re Certain Audiovisual Components, 2013 WL 4406820, at *114. The

5

Court agrees and finds that the repeated references in the specification, combined with the inventor’s

6

statements during the prosecution history establishes the definition of n as the interval between

7

transmission attempts. See Phillips, 415 F.3d at 1315

8
9

Finally, LSI argues that Barnes & Noble’s construction would render the invention
inoperable and read out desired embodiments because if the “time between transmissions is fixed at
‘n,’” the “timestamp would never include a busy delay, excluding the embodiments in the

11
For the Northern District of California

United States District Court

10

specification.” Dkt. No. 267-4, at 22. At the hearing, Barnes & Noble acknowledged that it would

12

be more accurate to construe the term as stating that “n” represents the interval between the time

13

when the transmitter will begin its attempt to transmit. Adjusted in this way, the construction takes

14

into account potential delays. For example, consider a transmitter’s modulo n counter that is

15

configured for a count sequence of 10 (i.e., 0 through 9) and to begin its transmissions at count 0. If

16

at count 0, the transmitter encounters message processing and busy-signal related delays such that it

17

is not until count 3 of the following count sequence that the actual transmission occurs, the

18

timestamp would be “3.” This timestamp would indicate to the receiver to set its own counter to 3,

19

rather than resetting it to 0, thus synchronizing the respective counters. Hence, it is clear that a

20

regularized count sequence from 0 to n is central to the teaching of the ‘867 Patent.

21

Furthermore, it is clear from the language of Claims 1 and 20 and prosecution history

22

discussed above that the time stamp “represents a value within a count sequence” means

23

that 0 < m < n..

24

For the foregoing reasons, the court construes the timestamp phrases as “a timestamp

25

representing a value m within the range 0 to n in the counter of the transmitter, where n represents

26

the interval between attempted transmission signals.”

27

///

28

///

23

1
2

VI.
A.

3

‘087 PATENT

Background and General Patent Description
The ‘087 Patent is entitled “MPEG Decoder System and Method Having a Unified Memory

4

for Transport Decode and System Controller Functions.” The invention describes an MPEG

5

decoder system that uses a unified memory that is used by the MPEG decoder’s system controller,

6

the transport logic, and the MPEG decoder logic. ‘087 Patent, Col. 4:65-5:28. This patent

7

represents an advantage of prior art MPEG decoders that required separate memories for these three

8

functions. Id., Col. 4:28-35. By providing for a unified memory, the invention provides for reduced

9

memory requirements for the decoder and thus reduced cost. Id., Col. 4:44-48. The invention’s
memory savings gains are further advanced by the implementation of various “frame memory saving

11
For the Northern District of California

United States District Court

10

schemes” such as compression or dynamic allocation. Id., Col. 5:29-33.

12

B.

13

Representative Claims
Claims 10 is a representative claim for the ‘087 patent and contains the terms the parties

14

have indicated are most significant as to this patent. Claim 10 provides (with terms to be construed

15

in bold):

16
17
18
19
20

10.
A method for performing video decoding in an MPEG decoder
system which includes a single memory for use by transport, decode,
and system controller functions, the method comprising:
receiving an MPEG encoded stream;
demultipliexing one or more multimedia data streams from
the encoded stream, wherein said demultiplexing one or more
multimedia data streams from the encoded stream operates using a
first unified memory;

21
22

performing MPEG decoding on the multimedia data streams,
wherein said performing MPEG decoding operates using said first
unified memory; and

23
24

a system controller operations within the MPEG decoder
system, wherein said controlling operations accesses code and data
from said first unified memory;

25
26

wherein said demultiplexing one or more multimedia data
streams, said performing MPEG decoding, and said controlling
operations each use said first unified memory.

27
28

‘087 Patent, Col. 18:14-34.

24

1

C.

2
3
4
5

“single memory” / “first unified memory”
Barnes & Noble

“a single memory chip which
stores code and data for the
transport logic, system
controller and MPEG decoder
functions”

LSI
“memory functioning as a
unit”

Court
“a single memory device
which stores code and data for
the transport logic, system
controller and MPEG decoder
functions.”

6
7

The sole dispute regarding this term is whether “single memory” / “first unified memory”

8

should be construed as limiting the memory to a single chip or whether it expresses a more general

9

idea that the memory functions as a unit.
The patent’s abstract provides that the invention is an “MPEG decoder system” which

11
For the Northern District of California

United States District Court

10

“includes a unified memory for multiple functions.” ‘087 Patent, abstract. Specifically, the video

12

decoding system described in the invention “includes a single unified memory which stores code

13

and data for the transport, system controller and MPEG decoder functions.” Id. The specification

14

distinguishes this “single memory” or “first unified memory” from prior art MPEG video decoders

15

as follows:

16

Prior art MPEG video decoder systems have generally used a frame
store memory for the MPEG decoder motion compensation logic
which stores the reference frames or anchor frames as well as the
frame being reconstructed. Prior art MPEG video decoder systems
have also generally included a separate memory for the transport and
system controller functions. It has generally not been possible to
combine these memories, due to size limitations.

17
18
19
20

Id., Col. 4:28-36.

21

The ITC has construed this term, adopting the construction proffered by LSI in this case:

22

“memory functioning as a unit.” In Re Certain Audiovisual Components, 2013 WL 4406820, at *7.

23

In reaching this construction, the ITC noted that:

24

the specification indicates that the claimed memory is not limited to a
single chip. As seen in at [sic] FIG. 3 of the ‘087 patent, the 16-Mbit
SDRAM identified by reference number 212 is depicted as four
rectangles coupled together. This representation of memory 212 is
consistent with four ranks (i.e., chips) of memory coupled together to
form a 16-Mbit SDRAM.

25
26
27
28

Id.

25

1

The Court agrees with the ITC to the extent it rejects Barnes & Noble’s “single chip”

2

limitation. The specification describes the 16Mbit SDRAM identified in Figure 3 as 212 as “an

3

external memory 212, also referred to as the single unified memory 212.” ‘087 Patent, Col. 8, 40-

4

41. Figure 3 clearly depicts the “single unified memory” as four rectangles coupled together. As the

5

ITC recognized, this is consistent with four ranks (or chips) of memory coupled together.

6

Accordingly, adopting Barnes & Noble’s construction would exclude a preferred embodiment.

7

Barnes & Noble has not demonstrated the highly persuasive evidentiary support to establish such a

8

construction. See Rambus Inc. v. Rea, 731 F.3d 1248, 1253 (Fed. Cir. 2013) (“A claim construction

9

that excludes the preferred embodiment is rarely, if ever, correct and would require highly
persuasive evidentiary support.” (citation omitted)). Construing the invention as being capable of

11
For the Northern District of California

United States District Court

10

having multiple “chips” is further supported by Rambus Inc. v. Hynix Semiconductor Inc., 569 F.

12

Supp. 2d 946 (N.D. Cal. 2008), where the court declined to limit the phrase “memory device” in a

13

claim to encompass only a single chip because the specification did “not clearly limit the scope of

14

the invention to a single chip.” Id. at 974.

15

Barnes & Noble cites no intrinsic evidence in support of a single “chip” (as opposed to a

16

single “memory”) limitation. Nor does its reliance on its expert, Dr. Dan Schonfeld, support such a

17

limitation. Dr. Schonfeld merely states what the patent itself provides – that the invention uses a

18

“single” memory. Declaration of Dr. Dan Schonfeld ¶ 12-15 (Dkt. No. 270-41). Nowhere does Dr.

19

Schonfeld state or otherwise support the contention that the memory must be embodied in a single

20

“chip.”

21

The Court concludes that “single memory” and “first unified memory” refers to the fact that

22

the memory “stores code and data for the transport logic, system controller and MPEG decoder

23

functions.” ‘087 Patent, Col. 5:3-6. The patent distinguishes this from the prior art my noting that

24

prior art MPEG decoders have used: (1) “a frame store memory for the MPEG decoder motion

25

compensation logic which stores the reference frames . . . as well as the frame being reconstructed”;

26

and (2) “a separate memory for the transport and system controller functions.” Id., Col. 4:28-36.

27

Whereas combining these two discrete memories dedicated to their separate tasks was previously

28

impossible “due to size limitations,” id., Col. 4:35-36, the invention provides for “various frame

26

1

memory saving schemes . . . to reduce the required amount of frame store memory” such that the

2

“savings in memory allow portions of the memory to also be used for transport and system controller

3

functions. Id., Col. 5:37-40; see also See Phillips, 415 F.3d at 1315 (“[T]he specification is always

4

highly relevant to the claim construction analysis. Usually, it is dispositive; it is the single best

5

guide to the meaning of a disputed term.” (citation omitted)). Accordingly, the Court disagrees with

6

Barnes & Noble that rejecting its construction fails to distinguish the invention from prior art.

7

At the hearing, Barnes & Noble suggested a compromise construction, replacing the “single

8

chip” limitation in its original proposed construction with the phrase “single memory device.” The

9

Court agrees that this construction is consistent with the specification and ordinary meaning of the
term “memory.” Further, unlike LSI’s proposed construction, Barnes & Noble’s revised

11
For the Northern District of California

United States District Court

10

construction describes the true purpose of the “single unified memory” development – specifically,

12

that it stores code and data for all three of the identified functions (transport logic, system controller,

13

and video decoder).

14

Accordingly, the Court construes “single memory” and “first unified memory” as “a single

15

memory device which stores code and data for the transport logic, system controller and MPEG

16

decoder functions.”

17

D.

18
19
20
21
22

“controlling operations accesses code and data from said first unified memory,” “operates
using a first unified memory,” “operates using said first unified memory”
Barnes & Noble

LSI

“system controller
programmed so that operations
which read from memory read
exclusively from the unified
memory.”

Plain and ordinary meaning or
“system controller
programmed to access the first
unified memory”

23
24
25
26

“configured so that operations
which read and/or write to
memory read and/or write
exclusively to the first unified
memory”

Plain and ordinary meaning or
“operates by accessing a first
unified memory”

27
28

27

Court
Plain and ordinary meaning

Plain and ordinary meaning

1

The parties’ dispute as to these terms turns on whether the decoding and system controller

2

functions identified in the invention must exclusively use the unified memory. Barnes & Noble

3

contends that the terms must be so limited because the applicants expressly distinguished itself from

4

prior art MPEG decoders which utilized external memory in addition to internal memory. See Dkt.

5

No. 270-4 (“[T]he claim language and specification describe the purported invention as being

6

limited to exclusive use of a single or unified memory. LSI’s construction only requires access or

7

use of some code or data from the ‘first unified memory,’ while the remainder used for the same

8

operations may come from anywhere.”).

9

As with the “single” / “first unified memory” terms, the ITC has construed both of these
terms. The ITC found that both terms should be given its “plain and ordinary meaning as

11
For the Northern District of California

United States District Court

10

understood by a person of ordinary skill in the art.” In re Certain Audiovisual Components, 2013

12

WL 4406820, at *9, 10-11. There, respondents proposed an even broader “exclusivity” limitation,

13

propounding constructions which provided for a “system controller programmed to exclusively read

14

from and write to the unified memory” and the like. The ITC found that such a limitation did “not

15

make sense” because, on its face, it meant that the system controller could only read and write to the

16

unified memory and could not, for example, actually conduct system controller operations. Id. at *9.

17

Barnes & Noble’s exclusivity limitation advanced in the case at bar, by contrast, only provides that

18

to the extent the decoder or system controller utilize memory, it exclusively utilizes the “first unified

19

memory.”

20

Nonetheless, the “exclusive” requirement Barnes & Noble seeks to impose does not appear

21

in either the specification or the claims. Rather, Barnes & Noble points to the fact that the preamble

22

of claim 10 recites a “method for performing video decoding in an MPEG decoder system which

23

includes a single memory for use by transport, decode and system controller functions” which

24

results in “reduced memory requirements compared to prior art designs.” Dkt. No. 270-4, at 31.

25

From this – and because the specification distinguishes the use of a “first unified memory” from

26

prior art – Barnes & Noble argues that the invention is limited to “exclusive use’ of a single or

27

unified memory.

28

28

1

The Court does not agree. As an initial matter, the Court notes that claim 10’s preamble

2

language providing a “method of video decoding . . . which includes a single memory” for use by the

3

three enumerated functions is inherently open ended. See, e.g., Lucent Technologies, Inc. v.

4

Gateway, Inc., 525 F.3d 1200, 1214 (Fed. Cir. 2008) (noting that “includes” is an open-ended term);

5

see also Adobe Sys. Inc. v. Wowza Media Sys., No. 11-cv-02243-JST, 2014 WL 709865, at *5 (N.D.

6

Cal. Feb. 23, 2014) (“The word ‘includes’ is open-ended. Accordingly, the use of the word

7

‘includes’ in the claims at issue means that a ‘new media chunk’ must include, but is not limited to,

8

‘a plurality of new parameters in a corresponding one of said headers and uses at least one other

9

parameter from a previous header.’”). Accordingly, that the invention in question includes a single
memory that is accessible and used by the decoder and controller functions for purposes of video

11
For the Northern District of California

United States District Court

10

decoding does not mean that the decoder and controller may use only said single memory in the

12

decoding process.

13

In addition, the specification itself acknowledges that memory other than the “single unified

14

memory” may be used by the decoder. Figure 4 is a block diagram which illustrates the decoder

15

logic. ‘087 Patent, Col. 11:33-35. Inside the decoder is a “Motion Compensation” box located at

16

310, which contain two elements which are unnamed in the figure, but labeled as 316 and 314. Id.,

17

Col. 12:35. Outside the video decoder is the “Frame-store Memory” at 212 – this corresponds to the

18

“single unified memory” described in the invention. Id. The specification describes how the motion

19

compensation block sends and receives data from the frame store memory. It provides:

20

The motion compensation block 310 analyzes each motion vector from
the incoming temporally compressed data and retrieves a reference
block from the frame store memory 212 in response to each motion
vector. The motion compensation block 310 includes a local memory
or on-chip memory [3]16 which stores the retrieved reference block.
The motion compensation block [3]10 then uses this retrieved
reference block to decompress the temporally compressed data.

21
22
23
24

Id., Col. 12:48-56 (emphasis added).3 Accordingly, the specification describes an embodiment

25

where the decoder is, in fact, using two memories – the main “frame store memory” (which in

26
27
28

3

The patent text describes this on-chip memory as number 116. The number 116, however,
does not appear anywhere in the Figures or elsewhere in the specification. It appears to be a
typographical error as the motion compensation block lists 316 as one of the elements comprising it.

29

1

context of the figures represents the “single unified memory” accessible by the controller and

2

decoder) and a “local memory” in the motion compensation block – an arrangement that would be

3

precluded under Barnes & Noble’s construction as the decoder would only be permitted to read or

4

write to or from the “single unified memory.” See Rambus Inc., 731 F.3d at 1253 (“A claim

5

construction that excludes the preferred embodiment is rarely, if ever, correct and would require

6

highly persuasive evidentiary support.” (citation omitted)).

7

Nor does the specification require Barnes & Noble’s “exclusivity” limitation. Barnes &

8

Noble argues that the applicants expressly disavowed use of the prior art’s multiple memories. It is

9

true that the ‘087 patent specification describes at some length the limitations of the prior art. For

11
For the Northern District of California

United States District Court

10

example, it provides:

14

Prior art MPEG video decoder systems have generally used a frame
store memory for the MPEG decoder motion compensation logic
which stores the reference frames or anchor frames as well as the
frame being reconstructed. Prior art MPEG video decoder systems
have also generally included a separate memory for the transport and
system controller functions. It has generally not been possible to
combine these memories, due to size limitations.

15

‘087 Patent, Col. 4:28-36. The specification highlights the “amount of memory is a major cost item

16

in the production of video decoders” and it is “desired to reduce the memory requirements of the

17

decoder system as much as possible to reduce its size and cost.” Id. at Col. 4:44-47. Thus, the

18

specification provides: “a new video decoder system and method is desired which efficiently uses

19

memory and combines the memory subsystem for reduced memory requirements and hence reduced

20

cost.” Id. at Col. 4:59-62.

12
13

21

However, as the Federal Circuit has recognized in the context of prosecution disclaimer,

22

disclaimer or disavowal does not apply “if the applicant simply describes features of the prior art

23

and does not distinguish the claimed invention based on those features.” Computer Docking Station

24

Corp. v. Dell, Inc., 519 F.3d 1366, 1375 (Fed. Cir. 2008). Here, the reference to prior art being

25

limited to a separate memory for the decoder and a separate memory for the controller may describe

26

one problem with the prior art, but in so describing, the applicants did not expressly disavow all use

27

of additional memories. This can be seen by the specification’s recognition that “[a] typical MPEG

28

decoder includes motion compensation logic which includes local or on-chip memory.” ‘087 Patent,

30

1

at Col. 4:14-15. As discussed above, a preferred embodiment of the invention provides for both the

2

unified memory and a local or “on chip” memory for the motion compensation box within the

3

decoder. It is thus apparent that the applicants were not broadly disavowing all potential uses of

4

additional memories, but merely disclosing a way in which the three distinct functions could operate

5

on the same memory. Significantly, at the hearing, Barnes & Noble could point to no language in

6

the specification or claims which require “exclusive” use of the “single unified memory.”

7

Accordingly, the Court rejects Barnes & Noble’s construction. Instead, the Court concludes,

8

like the ITC, that these terms will be given plain and ordinary meaning as understood by a person of

9

ordinary skill in the art.

11
For the Northern District of California

United States District Court

10

12
13
14
15

E.

“MPEG encoded stream” / “demultiplexing one or more multimedia data streams from the
encoded stream”
Barnes & Noble

LSI

Court

“a plurality of encoded
multimedia data streams which
are combined into a single
stream”

Plain and ordinary meaning, or
“a stream encoded in a manner
permitted by one or more
Motion Picture Expert Group
(‘MPEG’) standards”

“a stream encoded in a manner
permitted by one or more
Motion Picture Expert Group
(‘MPEG’) standards”

“demultiplexing one or more
multimedia data streams each
composed of at least a video
stream from the plurality of
encoded multimedia data
streams”

Plain and ordinary meaning, or
“separate the encoded stream
into one or more individual
streams”

“separate the encoded stream
into one or more individual
streams”

16
17
18
19
20
21
22

The parties’ dispute as to these terms goes to whether claim 10 requires there to be many

23

encoded “multimedia data streams” in a single “MPEG encoded streams.” In other words, the

24

parties dispute whether an “MPEG encoded stream” can consist of only a single multimedia stream.

25

Barnes & Noble argues that “a single ‘MPEG encoded stream’ must be comprised of a plurality of

26

(i.e., more than one) encoded multimedia data streams.” Dkt. No. 270-4, at 32.

27

Barnes & Noble relies upon the specification which provides, in discussing Figure 3:

28

31

1
2
3
4
5

The transport and system controller block includes transport logic
which operates to demultiplex the received MPEG encoded stream
into a plurality of multimedia streams. In other words, the encoded
stream preferably includes a plurality of multiplexed encoded channels
or multimedia data streams which are combined into a single stream,
such as a broadcast signal provided from a broadcast network. The
transport logic 206 in the transport and system controller block 204
operates to demultiplex this multiplexed stream into one or more
programs, wherein each of the programs comprise individual
multimedia data streams including video and/or audio components.

6
Id., Col. 8:10-21 (emphases added); see also id., Col. 7:38-41 (“As discussed further below, the
7
video decoding system includes transport logic which operates to demultiplex received data into a
8
plurality of individual multimedia streams.”). Barnes & Noble also relies on its expert who asserts
9
an MPEG encoded stream is one where “an MPEG-encoded broadcast typically includes a plurality
of programs or channels in a single data stream, and each program is composed of video and/or
11
For the Northern District of California

United States District Court

10
audio streams.” Schonfeld Decl. ¶ 18.
12
The Court rejects Barnes & Noble’s proposed construction. The Court begins with the plain
13
language of claim 10, which, as stated above, provides in relevant part:
14
15
16
17
18

10.
A method for performing video decoding in an MPEG decoder
system . . . the method comprising:
receiving an MPEG encoded stream
demultiplexing one or more multimedia data streams from the
encoded stream from the encoded stream operates using a first unified
memory. . . .

19
‘087 Patent, Co. 18: 14-23. The use of the phrase “an MPEG encoded stream” and demultiplexing
20
“one or more multimedia data streams” directly conflicts with Barnes & Noble’s proposed
21
construction, in that the claim language expressly contemplate a single MPEG encoded stream that
22
is then demultiplexed into “one or more multimedia data streams. As the Federal Circuit has
23
recognized “‘[a]’ or ‘an’ in patent parlance carries the meaning of ‘one or more’ in open-ended
24
claims containing the transitional phrase ‘comprising.’ This convention is overcome only when ‘the
25
claim is specific as to the number of elements’ or ‘when the patentee evinces a clear intent to . . .
26
limit the article.’” Free Motion Fitness, Inc. v. Cybex Int’l, Inc., 423 F.3d 1343, 1350 (Fed. Cir.
27
2005) (quoting KCJ Corp. v. Kinetic Concepts, Inc., 223 F.3d 1351, 1356 (Fed. Cir. 2000)); see also
28

32

1

Baldwin Graphic Sys., Inc. v. Siebert, Inc., 512 F.3d 1338, 1343 (Fed. Cir. 2008) (“That ‘a’ or ‘an’

2

can mean ‘one or more’ is best described as a rule, rather than merely as a presumption or even a

3

convention.”).

4

Barnes & Noble’s contention that the Court should interpret claim 10 as limited by the

specification on this point states that “the encoded stream preferably includes a plurality of

7

multiplexed encoded channels or multimedia streams.” ‘087 Patent, Col. 8, 13-15. The use of the

8

term “preferably” in the specification suggests that the invention is not, in fact, limited in this way.

9

See, e.g., Candela Corp. v. Palomar Med. Techs., Inc., No. 9:06-CV-277, 2008 WL 3285255, at *5

10

(E.D. Tex. Aug. 6, 2008) (“Only where the specification uses language of requirement, rather than

11
For the Northern District of California

preferred embodiment contained in Figure 3 is unavailing. First, as its own brief recognizes, the

6

United States District Court

5

preference, will the specification describe an essential step or element of the claim rather than

12

merely a preferred embodiment.”). Adopting Barnes & Noble’s requirement that there be multiple

13

multimedia streams based on the language in the specification would improperly read a limitation

14

into the claim based on a preferred embodiment in the specification which makes clear the proffered

15

embodiment is not intended to be exhaustive of all possible embodiments. See Kara Tech. Inc. v.

16

Stamps.com Inc., 582 F.3d 1341, 1348 (Fed. Cir. 2009) (“The patentee is entitled to the full scope of

17

his claims, and we will not limit him to his preferred embodiment or import a limitation from the

18

specification into the claims.”).

19

Accordingly, the Court rejects Barnes & Noble’s proposed construction. Because LSI’s

20

proposed constructions provide a definition for both “MPEG stream” and “demultiplexing,” the

21

Court adopts LSI’s proposed construction.

22
23
24

VII.
A.

‘420 PATENT

Background and General Patent Description
The ‘420 Patent is entitled “Methods of and Devices for Enhancing Communications that

25

Use Spread Spectrum Technology by Using Variable Code Techniques.” The patent explains a

26

spread spectrum technique which provides for increased system capacity and/or signal quality

27

through the employment of unequal error protection (“UEP”). ‘420 Patent, Col. 1, 63-67.

28

Specifically, the technique provides a greater degree of error protection to portions of a signal that

33

1

are relatively sensitive to signal errors, while those portions fo the signal relatively insensitive to

2

such errors receive a lower degree of error protection. Id., Col. 2, 5-15. By eliminating “over-

3

coding” by providing a single level of error protection, the channel bandwidth is used more

4

efficiently. Id., Col. 2, 23-26.

5

B.

6

Representative Claims
Claims 1 and 17 are representative claims for the ‘420 patent and contain the terms the

7

parties have indicated are most significant as to this patent. Claim 1 provides (with terms to be

8

construed in bold):

9

11
For the Northern District of California

United States District Court

10

12

1.
In a method of processing a signal, comprising the steps of
applying a spread spectrum coding process, the signal being a function
of time, the improvement comprising:
(a)
encoding a first segment of the signal with a first
channel encoder operating at a first rate to generate a first encoded
segment having more significant bits using a first error protection
process; and

13
14
15
16
17
18
19
20
21

(b)
encoding a second segment of the signal with a second
channel encoder operating at a second rate to generate a second
encoded segment having less significant bits using a second error
protection process, the second rate being different from the first rate
and the first error protection process providing a greater amount of
error protection than the second error protection process.
‘420 Patent, Col. 14: 18-32. Claim 17 provides:
17.
In a method of processing a signal comprising the steps of
applying a spread spectrum multiple access decoding process, the
signal being a function of time, the improvement comprising:
(a)
decoding a first received segment of the signal with a
first channel decoder to generate a first decoded segment having more
significant bits using a first error protection process, the first channel
decoder operating at a first rate; and

22
23
24
25

(b)
decoding a second received segment of the signal
with a second channel decoder to generate a second decoded segment
having less significant bits using a second error protection process,
the second channel decoder operating at a second rate, the second rate
being different from the first rate and the first error protection process
providing a greater amount of error protection than the second error
protection process.

26
27

Id., Col. 15:57-16:5.

28

34

1

C.

“first segment of the signal” / “second segment of the signal” and “first received segment of

2

the signal” / “second received segment of the signal”

3

As to the terms “first segment of the signal” and “second segment of the signal,” the parties

4
5
6
7
8
9

have provided the following constructions:
Barnes & Noble

LSI

Court

“a first data stream that has
been separated out from the
signal to be transmitted”

Plain and ordinary meaning, or
“a group of bits”

“a group of bits”

“a second data stream that has
been separated out from the
signal to be transmitted”

Plain and ordinary meaning, or
“a second group of bits.”

“a second group of bits”

11
For the Northern District of California

United States District Court

10

12
13
14
15
16
17
18

As to the terms “first received segment of the signal” and “second received segment of the
signal” the parties have provided the following constructions:
Barnes & Noble

LSI

Court

“a first received data stream
that has been separated out
from the signal prior to
transmission to the receiver”

Plain and ordinary meaning, or
“a group of bits received by
the receiver”

“a group of bits received by
the receiver”

“a second received data stream
that has been separated out
from the signal prior to
transmission to the receiver”

Plain and ordinary meaning, or
“a second group of bits
received by the receiver.”

“a second group of bits
received by the receiver”

19
20

The dispute between the parties as to these terms is whether the “first segment” and “second

21

segment” of the signal must be distinct, constituent parts of the signal which must be physically

22

separated out from the signal prior to transmitting. Compare 270-4, at 18 (Barnes & Noble arguing

23

that “the claim language itself makes clear that the first and second segments are separate,

24

constituent parts of the signal”), with Dkt. 273-4, at 10 (LSI arguing that “[w]hile certainly some

25

implementations of the ‘420 Patent may ‘separate’ each segment from the signal, there is nothing in

26

the claim language that would lead one skilled in art to believe that separation prior to transmission

27

is required”).

28

35

1

The ‘420 invention discloses a spread spectrum transmission technology that provides for

2

increased system capacity and/or improved signal quality on a cellular network. ‘420 Patent, Col.

3

1:63-65. The invention does this by employing “unequal error protection” (“UEP”). Id. UEP works

4

by applying differing levels of error protection to different “segments” of a given signal – those

5

portions that are relatively sensitive to signal errors are afforded increased protection against such

6

errors, while those portions relatively insensitive to signal errors are afforded a reduced level of

7

protection. See id., Col. 2:5-14. The result is an efficient use of available channel bandwidth. See

8

id., Col. 2:15-17. The specification reveals that this process may be implemented in a “variety of

9

manners,” such as a variable time UEP method (“VT”), a variable code UEP method (“VC”), or a
variable power UEP method (“VP”) or some combination of these three. Id., Col. 5:15-20.

11
For the Northern District of California

United States District Court

10

However, the claims in the ‘420 Patent all address the variable code UEP process.

12

Barnes & Noble relies primarily on the specification – specifically Figure 3 and its

13

associated detailed description – in support of its proposed construction. Barnes & Noble points to

14

the fact that the figure shows two separate coders – a “first channel coder” (302) and a “second

15

channel coder” (304) that receive separate portions of the signal. In describing Figure 3, the

16

specification states “Interface 300 separates voice data into two data streams of unequal

17

significance.” Id., Col. 5:51-52. Then, a “first data stream 306 (e.g., a more significant data stream)

18

is input into the first channel coder 302 and a second data stream 308 (e.g., a less significant data

19

stream) is input into the second channel coder.” Id., Col. 5:54-56. Additional figures and

20

discussions in the specification refer to a splitting of a signal into separate data streams. See, e.g.,

21

id., Col. 10:42-47 (“Referring to FIG. 6, the pre-processor 200 comprises input voice data interface

22

600. The interface 600 separates the encoded voice data into two data streams, a first data stream

23

606 and a second data stream 608.”).

24

The Court finds that there is no basis for imposing a claim limitation that there be physical

25

separation of the signal before transmission. The asserted claims do speak of two channel encoders

26

– the “first channel encoder operating at a first rate” and a “second channel encoder operating at a

27

second rate.” It is also true that the specification and figures refer to the separation of the signal into

28

a first and second segment. However, there is nothing in the claims or specification requiring the

36

1

physical separation of the signal. For example, there is no reason, based on the plain text of the

2

claims, why the entire signal could not be processed serially through a single piece of hardware

3

controlled by software which provides unequal error protection based on the particular segment of

4

the signal currently being processed through the encoder. At the hearing, Barnes & Noble conceded

5

that such an arrangement is technologically possible. Serial processing of the data stream would not

6

require segments be physically separated.

7

Furthermore, the section of the cited specification discussing the illustrations is entitled

8

“Detailed Description of the Illustrative Embodiments.” ‘420 Patent, Col. 3, 50-51. This thus

9

suggests that the specification’s discussion is not meant to be exhaustive. See, e.g., Roberg v. 20th
Century Plastics, Inc., 40 F. Supp. 2d 208, 219 (D.N.J. 1999) (“In addition, references in the

11
For the Northern District of California

United States District Court

10

specification to a preferred embodiment, or an illustrative example, do not limit the scope of the

12

patent claim.” (citation omitted)). The patent then makes this explicit at the end of the specification,

13

providing that “[a]lthough a number of specific embodiments of this invention have been shown and

14

described herein, it is to be understood that these embodiments are merely illustrative . . . .”

15

Accordingly, nothing in the asserted claims requires the respective “segments” of a signal to

16

be separated out prior to transmission. See, e.g., Howmedica Osteonics Corp. v. Wright Med. Tech.,

17

Inc., 540 F.3d 1337, 1345 (Fed. Cir. 2008) (declining to impute a limitation into a disputed claim

18

term in the absence of a clear requirement in the specification, even where “every disclosure of [the

19

disputed term] in the specification shows [the alleged limitation]”); see also Baxter Healthcare

20

Corp. v. Fresenius Med. Care Holdings, Inc., No. C 07-1359 PJH, 2009 WL 330950 (N.D. Cal. Feb.

21

10, 2009) (“Even if every disclosed embodiment uses flexible membranes, Phillips squarely rejects

22

limiting the claim on that basis, unless the specification makes clear that the patentee . . . intends for

23

the claims and the embodiments in the specification to be strictly coextensive.”). The Court

24

concludes that the terms will be given meanings proposed by LSI.

25
26
27
28

VIII.
A.

‘394 PATENT

Background and General Patent Description
The ‘394 patent is entitled “Methods and Systems for Transmitting an Information Signal in

a Multiple Antenna Communication System” and addresses data communication in a wireless

37

1

system. The patent specification begins with a statement that “[t]he invention relates to a wireless

2

radiofrequency data communication system comprising” the following:

3
4
5
6
7
8

a base-station comprising multiple first sets and a signal processingunit, wherein each first set comprises a transmitter- and receiver-unit
provided with a transmitter and a receiver and at least one antenna
which is connected to the transmitter- and receiver-unit, wherein the
signal processing-unit is connected with each of the first sets for
processing signals received by the first sets and processing signals to
be transmitted by the first sets, and
multiple second sets, wherein each second set comprises a transmitterand receiver-unit provided with a transmitter and a receiver and at
least one antenna which is connected to the transmitter- and receiverunit.

9
‘394 Patent, Col. 1:16-30. In other words, the patent defines the wireless communication system as
one having: (1) a base-station that has “multiple first sets,” each “first set” containing at least one
11
For the Northern District of California

United States District Court

10
antenna; and (2) multiple “second sets.” each with at least one antenna. It is this system in which the
12
invention described by the ‘394 patent operates.
13
In providing background for the patent, the specification noted the current limitation of
14
existing prior art wireless systems:
15
16
17
18
19

In the majority of the applications more than one second set
[computer, tablet, or other wireless device] wants to communicate
with the base-station. This means that the second set transmits signals
to this base-station and also receives signals from this base-station.
Since it would not be acceptable if all second sets would have to wait
for each other’s communication to be finished, there is a need for
simultaneous communication. Simultaneous communication allows
more second sets to communicate at the same time with the basestation.

20
21

‘394 Patent, Col. 1:51-50. The patent then recognizes that the “common way of realising

22

simultaneous communication is to assign different radiofrequencies to the respective second sets. In

23

this way all data signals can be separated easily by the first sets in the base-station by frequency

24

selective filters.” Id., Col. 1:59-63.

25

The object of the invention taught by the ‘394 patent is to “increase the capacity of the

26

wireless communication system per frequency or frequency-band used by the system. Id., Col. 2:5-

27

7. In essence, the patent increases the number of devices which may simultaneously communicate

28

with a base station over a wireless network by splitting the data signal from that device into a

38

1

multiple of signals which are then modulated on different frequencies according to an Inverse Fast

2

Fourier Transformation, and then sent to the base station’s antenna which reassembles the data

3

signal. Id., Col. 4:1-10. The specification also teaches a beneficial advantage of the multiple

4

antenna system arrangement is to minimize signal error or “deep fades” as a result of “the antenna

5

receiving diversity effect.” Id., Col. 4:33-34. Stated another way, because the base station in this

6

wireless system has multiple antenna and, therefore, receives multiple copies of the signal, the risk

7

of signal errors is reduced.

8

B.

9

Claims 1 is the asserted claim for the ‘394 patent. Claim 1 provides (with terms to be
construed in bold):

11
For the Northern District of California

United States District Court

10

Representative Claims

1.
A method for transmitting an information signal in a multiple
antenna communication system, comprising the steps of:

12
separating said information signal into K signals;
13
performing an Inverse Fast Fourier Transformation on
said K signals to generate a signal comprising K different
frequencies; and

14
15

transmitting said signal comprising K different frequencies on
only one antenna.

16
17

‘394 Patent, Col. 11:24-32.

18

C.

19
20
21
22
23
24

“multiple antenna communication system”
Barnes & Noble

“a communication system
having multiple transmitter
antennas and multiple receiver
antennas such that multiple
simultaneous communication
channels may be generated on
the same frequency or within
the same frequency band”

LSI
“system with at least one
transmitter or receiver with
more than one antenna”

Court
“a communication system
having multiple transmitter
antennas and multiple receiver
antennas such that multiple
simultaneous communication
channels may be generated on
the same frequency or within
the same frequency band”

25
26

The Court adopts Barnes & Noble’s proposed construction. As described supra, the

27

specification begins with a general statement that “[t]he invention relates to a wireless radio

28

frequency data communication system comprising” two distinct parts: (1) a base-station with

39

1

“multiple first sets,” each first set having at least one antenna; and (2) “multiple second sets,” each

2

one having at least one antenna. Id., Col. 1:16-30 (emphases added). Accordingly, the patent begins

3

by describing the “data communication system” in which the invention operates and expressly

4

provides that in such a system there are multiple transmitter and receiver antennas.4

5

Further this statement does more than describe a preferred embodiment of the invention;

6

rather, it describes the wireless system to which “[t]he invention” relates. Accordingly, it is proper

7

to read the claims as being limited to such a system. See, e.g., Edwards Lifesciences LLC v. Cook

8

Inc., 582 F.3d 1322, 1330 (Fed. Cir. 2009) (where specification makes descriptions of the “invention

9

itself” or the “present invention” rather than specific embodiments, the patent indicates an intent to
limit the invention to that description); see also Microsoft Corp v. Multi-Tech Systems, Inc., 357

11
For the Northern District of California

United States District Court

10

F.3d 1340, 1348 (Fed. Cir. 2004) (“Those statements, some of which are found in the ‘Summary of

12

the Invention’ portion of the specification, are not limited to describing a preferred embodiment, but

13

more broadly describe the overall inventions of all three patents.”). For example, in Honeywell

14

International, Inc. v. ITT Industries, Inc., 452 F.3d 1312 (Fed. Cir. 2006), the Federal Circuit

15

affirmed the district court’s construction of a “fuel injection system component” as being limited to

16

a fuel filter because “[o]n at least four occasions, the written description refers to the fuel filter as

17

‘this invention’ or ‘the present invention.’” Id. at 1318. One of these four references was, similar to

18

this case, a statement that “[t]his invention relates to a fuel filter for use in the fuel line that delivers

19

fuel to a motor vehicle engine.” Id.

20

While the language of the claims and the descriptions of “the invention” contained in the

21

specification are controlling, the Court also notes that Barnes & Noble’s construction is consistent

22

with the purpose of the invention. At the hearing, the parties disputed at length whether the

23

invention’s teachings would have any benefit in a situation where there was only a single “second

24
25
26
27
28

4

LSI and its expert appear to misconstrue Barnes & Noble’s construction. Dr. Negus states
in his declaration that it his opinion that the phrase “multiple antenna communication system does
“not require the [sic] every transmitter or receiver in the system have more than one antenna.”
Negus Decl. ¶ 38. Barnes & Noble does not argue that every transmitter or receiver in the system
have more than one antenna. Rather, it argues, and the Court agrees for the reasons articulated, that
the system must have multiple “second sets” and multiple “first sets” such that simultaneous
communications are possible.

40

1

set” with one antenna but a base station with multiple antenna. LSI correctly identified that in such

2

a situation the wireless system would benefit from reduced intersymbol interference as a result of the

3

multiple antennas in the “first set.” See ‘420 Patent, Col. 4:32-34 (“With the present multi-antenna

4

processing, the probability of deep fades is greatly reduced because of the antenna receiving

5

diversity effect.”). However, it cannot be disputed that the specification plainly discloses that the

6

main benefit of the invention is that it provides a way to permit simultaneous communications on the

7

same frequency, thus increasing the capacity of the wireless system. See, e.g., id., Col. 2:8-23. Such

8

a goal can only be realized if the wireless system described in the invention contains multiple

9

“second sets” each with at least one antenna. See, e.g., CVI/BETA Venturs, Inc. v. Tura LP, 112
F.3d 1146, 1160 (Fed. Cir. 1997) (“In construing claims, the problem the inventor was attempting to

11
For the Northern District of California

United States District Court

10

solve, as discerned from the specification and the prosecution history, is a relevant consideration.”).

12

Accordingly, the Court construes “multiple antenna communication system” as “a

13

communication system having multiple transmitter antennas and multiple receiver antennas such

14

that multiple simultaneous communication channels may be generated on the same frequency or

15

within the same frequency band.”

16

D.

17
18
19
20
21
22

“performing an Inverse Fast Fourier Transformation on said K signals to generate a signal
comprising K different frequencies”
Barnes & Noble

“performing an Inverse Fast
Fourier Transformation on a
number of signals that
corresponds to the number of
different frequencies in the
transmitted signal”

LSI
“performing Inverse Fast
Fourier Transformation to
generate another signal
containing at least as many
frequencies as the number of
divided signals”

Court
“performing an Inverse Fast
Fourier Transformation on
said K signals to generate a
signal with K different
frequencies”

23
24

The patent specification teaches that each “second set” in the multiple antenna system would

25

have a serial-to-parallel/parallel-to-serial unit which would serve to split the data signal into several

26

signals at which point the several signals would be put through an Inverse Fast Fourier

27

Transformation (“IFFT”) which would modulate these separate signals on different frequencies prior

28

to transmission. ‘394 Patent, Col. 4:3-7. Here, the parties dispute whether the claim requires that

41

1

the number of frequencies created as a result of the IFFT must equal the number of signals that went

2

into the IFFT. In other words, when K signals go into the IFFT, are K different frequencies created?

3

LSI argues that because the claim states that the IFFT on “said K signals” creates a signal

4

“comprising K different frequencies” – comprising being an inherently open-ended term – the claim

5

requires only that “at least” K frequencies are created.

6

LSI is correct that the Federal Circuit traditionally interprets the word “comprising” as being

7

an open-ended term. For example, in CIAS, Inc. v. Alliance Gaming Corp., 504 F.3d 1356, 1360

8

(Fed. Cir. 2007), the court noted that “[i]n the patent claim context the term ‘comprising’ is well

9

understood to mean ‘including but not limited to.’” Id. at 1360. However, in those cases where the
Federal Circuit has applied an open ended definition for “comprising,” the term was used in the

11
For the Northern District of California

United States District Court

10

preamble or as a transition. See, e.g., MagSil Corp. v. Hitachi Global Storage Tech., Inc., 687 F.3d

12

1377 (Fed. Cir. 2012) (“Open claim language, such as the word ‘comprising’ as a transition from the

13

preamble to the body of a claim, ‘signals that the entire claim is presumptively open-ended’”

14

(citation omitted)); Crystal Semiconductor Corp. v. TriTech Microelectronics Int’l, Inc., 246 F.3d

15

1336, 1348 (Fed. Cir. 2001) (“The transition ‘comprising’ creates a presumption that the recited

16

elements are only a part of the device, that the claim does not exclude additional, unrecited

17

elements.”).

18

That is not the case here. Rather, the term “comprising” is contained in the body of the

19

claim. Courts have held that when used in the body of a claim, and not as a transition, the term

20

“comprising” should be interpreted according to the normal rules of claim interpretation. See

21

Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 1272 n.8 (Fed. Cir. 1986) (“‘Comprising’ is

22

not used here as a transitional phrase and has no special legal effect as such. Hence, it should be

23

interpreted according to the normal rules of claim interpretation.”); see also Med. Res. Institute v.

24

Bio-Engineered Supplements & Nutrition, Inc., No. 605CV417, 2007 WL 128937 (E.D. Tex. Jan.

25

12, 2007) (“The Court agrees with BSN that ‘comprising’ in the body of the term should be

26

construed according to normal claim construction rules.”). The definition of “comprising,”

27

separated from the legal effect given to such term when used as a transition, is “to consist of” or “to

28

make up.” Webster’s Third New International Dictionary 467 (2002). The claim language includes

42

1

a numerical limitation “K” for both the signals upon which the IFFT is performed and the

2

frequencies which would result. It would be highly unusual for a writer to use the term “comprise”

3

as open-ended in this context given this express numerical limitation.

4

Further, in his expert declaration, Dr. Bambos stated:

5

A person of ordinary skill in the art would understanding that the
scheme used in the ‘394 patent (like OFDM) involves transmission of
signals that are comprised of a number of frequency subchannels onto
which information may be modulated. A person of ordinary skill
would also understand that “performing an Inverse Fast Fourier
Transformation on said K signals to generate a signal comprising K
different frequencies” refers to the number of frequency subchannels
in an OFDM-type system. Thus, that person of ordinary skill would
understand that the requirement of claim 1 for “a signal comprising K
different frequencies” means that the claim requires generation of a
signal with K frequency subchannels.

6
7
8
9

11
For the Northern District of California

United States District Court

10

Bambos Decl. ¶ 27.

12

Accordingly, the Court rejects an open ended reading of comprising that would only require

13

that the IFFT result in “at least K frequencies.” Instead, the Court construes “performing an Inverse

14

Fast Fourier Transformation on said K signals to generate a signal comprising K different

15

frequencies” as “performing an Inverse Fast Fourier Transformation on said K signals to generate a

16

signal with K different frequencies.”

17
18
19

IX.
A.

‘552 PATENT

Background and General Patent Description
The ‘552 Patent is entitled “Variable Rate Coding for Wireless Applications.” In general,

20

the patent provides a method for managing data traffic over a wireless data network. As the

21

specification describes, in the past, wireless communications supported only “a single

22

communication channel per user” and that this “limited the flexibility of the telecommunication

23

system with regard to the high data rate required in applications such as multimedia applications.”

24

‘552 Patent, Col. 1:11-15. The result was that multiple transmitters and receivers were needed for

25

“each communication channel,” thus resulting in an expensive an complicated system. Id., Col.

26

1:15-17.

27
28

The invention provides a mechanism for controlling data backlog in the wireless system.
The preferred embodiment describes a wireless base station receiving data from a variety of sources.

43

1

Id., Col. 1:46-48. These sources send data to the base station at different rates. Id., Col. 1:49. The

2

data is received by buffers which permit the data to be received at one rate and removed from the

3

buffer at a different rate. Id., Col. 1:53-54. A controller in the base station monitors the buffers “to

4

determine if an overflow situation is about to occur.” Id., Col. 2:1-2. If such a situation is about to

5

occur, the controller instructs the appropriate variable rate “Walsh coder” to “increase the coding

6

rate or number of channels for that particular buffer,” resulting in a data being transferred more

7

quickly to alleviate the overflow. Id., Col. 2:2-6. Similarly, if the controller determines that a buffer

8

has very little data backlogged, the controller instructs the appropriate variable rate “Walsh coder”

9

to decrease the coding rate or number of channels provided to that buffer “so as to free up additional
channels for other users.” Id., Col. 2:8-12.

11
For the Northern District of California

United States District Court

10

B.

12

Representative Claim
The parties have identified 9 terms in the ‘552 patent the construction of which they dispute.

13

In the parties’ first supplemental joint claim construction and prehearing statement, only one of these

14

terms (“when”) was included in the list of ten most significant terms as to the original patents. Dkt.

15

No. 262, at 3. Given the magnitude of disputed terms in the present action, the Court subsequently

16

ordered the parties to provide the Court with a list of the ten absolute most significant terms. The

17

parties resulting list contained no terms as to ‘552 patent. However, in order to provide guidance to

18

the parties, the Court will, at this time, construe the term “when” as the parties originally identified

19

this as a significant term to the ‘552 patent.

20
21
22

The term when appears in claims 1, 3, and 4. Claim 1 provides (with the term to be
construed in bold):
1.
A method for providing variable rate wireless communications,
comprising the steps of:

23
monitoring a data backlog of data to be transmitted;
24
25
26

increasing the data transmission rate by increasing a coding
rate when the data backlog crosses a first threshold; and
decreasing the data transmission rate by decreasing the coding
rate when the data backlog crosses a second threshold.

27
28

‘552 Patent, Col. 5:48-57.

44

1

C.

2
3
4

“when”
Barnes & Noble

Plain and ordinary meaning, or
“at the time”

LSI
“After”

Court
Plain and ordinary meaning

5
6

LSI argues that Barnes & Noble’s construction is improper because requiring that the coding

would make the claim impossible to practice because electrical devices “nearly always require some

9

amount of processing time between a triggering event and a triggered action.” Dkt. No. 268-4, at

10

30-31. Barnes & Noble, by contrast, argues that “when” is used to connote a conditional, cause-

11
For the Northern District of California

rate be increased (or decreased) “at the time” a backlog crosses a given threshold, Barnes & Noble

8

United States District Court

7

effect relationship between the “data backlog” and an increase or decrease in the transmission rate.

12

Dkt. No. 270-4, at 19. It further argues that LSI’s construction of the term as meaning “After” is too

13

broad as it would, in theory, encompass any increase or decrease in coding rate occurring after a

14

backlog passes a threshold, whether or not that increase/decrease was caused by the threshold being

15

passed. Id. at 20.

16

At the hearing, Barnes & Noble asserted that it will not contend that coding rate change must

17

occur simultaneously with a threshold being crossed – the main dispute of the parties as to this term.

18

In light of this concession and the non-technical nature of the term “when” the Court finds no

19

construction necessary. Jurors will be capable of understanding the meaning of “when” in this

20

context without assistance from the Court.

21

X.

22
23

A.

‘006 PATENT

Background and General Patent Description
The ‘006 Patent is entitled “Two-Part Synchronization Scheme for Digital Video Decoders.”

24

The patent describes an improved apparatus and method for synchronizing a video decoder to a

25

compressed digitized video signal. ‘006 Patent, Col. 1:7-9. The patent describes that in

26

conventional prior-art video decoding systems, before being sent through the video decoder, the

27

coded video signal is sent through a synchronizer which “pre-processes the system bitstream prior to

28

its being input into a video decoder.” Id., Col. 6:52-56. The synchronizer in such systems had to

45

1

perform two complex operations – (1) aligning the symbols to determine where the various

2

structures making up an image or group of images start and (2) evaluating and parsing the bitstream

3

to extract the coded video data from the non-video data. Id., Col. 7: 4-9. This time consuming

4

process made “rapid synchronization difficult,” and could easily lead to either an overflow of

5

bitstream data in the buffer (as the synchronizer could not keep up) or underflow of data to the

6

decoder making the decoding process unreliable. Id.

7

To address this problem, the invention provides increased efficiency in video decoding by

synchronization of the bitstream, filtering unwanted digital information that is irrelevant to the video

10

signal. Id., Col. 9:49-68. The filtered video bitstream is then sent through a “post-parser” and then

11
For the Northern District of California

employing a “pre-parser” (or first synchronizer) which, for example, performs an initial

9

United States District Court

8

the video decoder – the operations of both being “significantly simplified” since the video bitstream

12

had already been synchronized and properly aligned. Id., Col. 9: 60-64.

13

B.

14

Representative Claim
The parties have identified 7 terms in the ‘006 patent the construction of which they dispute.

15

In the parties’ first supplemental joint claim construction and prehearing statement, three of these

16

terms were included in the list of ten most significant terms as to the supplemental patents. Dkt. No.

17

262, at 5. On the other hand, the parties second supplemental joint claim construction and

18

prehearing statement filed in response to this Court’s order contained no terms as to ‘006 patent.

19

However, in order to provide guidance to the parties, the Court will, at this time, construe the term

20

“further parsing the bitstream” – one of the terms identified as significant by the parties in their

21

original filing.

22
23

The term to be construed appears in claim 27 of the amended ‘006 patent. This claim
provides:

24

27.

25

receiving a multiplexed bitstream from a transmission channel,
the multiplexed bitstream including at least a video bitstream;

A process of decoding a bitstream comprising the steps of:

26
synchronizing to the received multiplexed bitstream;
27
28

parsing the video bitstream from the synchronized multiplexed
bitstream;

46

1
transferring the video bitstream to a channel buffer;
2
further parsing the video bitstream; and
3
decoding the parsed video bitstream,
4
wherein the multiplexed bitstream has a structure that
conforms to an MPEG format.

5
6

‘006 Patent C1, Col. 1:39-41.

7

C.

8
9

11
For the Northern District of California

United States District Court

10

12

“further parsing the bitstream”
Barnes & Noble

“extracting the layers of video
information from the video
bitstream and translating each
layer of video information
using variable length coding
look-up tables or dictionaries”

LSI
“extracting the various layers
comprising the video
bitstream”

Court
“extracting the various layers
comprising the video
bitstream”

13
14
15

The parties agree that the “further parsing” step in claim 27 includes extracting layers from a

16

video bitstream. They dispute, however, whether the “further parsing” step also requires that each

17

layer of video information be translated. The Court concludes that the “further parsing” step of

18

claim 27 does not require translation to occur.

19

Barnes & Noble’s argument that translation is a necessary component of “further parsing” is

20

premised entirely on the specification’s discussion of specific embodiments of the invention.

21

Specifically, the embodiments in the specification describe a structure in the video decoder called a

22

“post-parser” which performs both parsing and translating activities. See ‘006 Patent, Col. 14, 51-54

23

(discussing the “multi-bit symbol parallel post-parser 76” that appears in Figure 4 and stating it is

24

“essentially a one-event per cycle parser that performs look-ups and translations with respect to the

25

video bitstream and the VLC tables” (emphasis added)). However, the Federal Circuit has cautioned

26

against importing from the specification limitations into a patent’s claims, see Phillips, 415 F.3d at

27

1323, where there is nothing suggesting that the invention was intended to be limited to the

28

embodiments. To the contrary, the patent here expressly provides that “the scope of the present

47

1

invention should not be limited by the particular embodiments discussed above.” ‘006 Patent, Col.

2

15, 50-52.

3

Further, Barnes & Noble’s proposed construction is in tension with other portions of the

4

specification. In at least two places in the “Summary of the Invention,” the specification expressly

5

refers to “further parsing” as a separate act from “translation.” See id., Col. 7:61-64 (“Preferably,

6

the synchronized portion of the variable bit-rate coded signal stored in the channel buffer is further

7

parsed and translated by a second synchronizer when it is extracted from the channel buffer.”

8

(emphasis added)); id., Col. 8:38-41 (“The multi-bit symbol parallel post-parser further parses the

9

various layers comprising the video bitstream and translates the encoded video data to
corresponding video symbols.” (emphasis added)). While both are described as occurring in the

11
For the Northern District of California

United States District Court

10

“post-parser” element, the specification does not appear to define “further parsing” as including

12

translation. In short, while the post-parser can perform both “further parsing” and “translation”;

13

further parsing itself need not include translation.

14

In addition, Barnes & Noble’s construction would render certain dependent claims

15

superfluous, thus violating the doctrine of claim differentiation. See, e.g., InterDigital Commc’ns,

16

LLC v. Int’l Trade Comm’n, 690 F.3d 1318, 1315 (Fed. Cir. 2012); see also Nextec Applications v.

17

Brookwood Co., Inc., 703 F. Supp. 2d 390 (S.D.N.Y. 2010) (rejecting construction of a term that

18

would “render the dependent claims of the ‘792 patent redundant, a result that is to be avoided under

19

well-established claim construction principles”). For example, independent claim 16 of the ‘006

20

patent provides, in relevant part: “A signal decoding system comprising . . . a second synchronizing

21

means, coupled to said channel buffer, further parsing said video bitstream . . . .” ‘006 Patent, Col.

22

16, 61-63. Dependent claim 23, however, provides: “[a] signal decoding system as defined in claim

23

16 wherein said second synchronizing means further includes means for translating said parsed

24

video bitstream.” Id., Col. 17:19-21 (emphasis added). Similarly, independent claim 1 describes a

25

signal decoding system which includes, in part, “a second synchronizing means . . . for parsing said

26

portion of said variable bit-rate coded system.” Id., Col. 15:66-68. Dependent claim 13, however,

27

provides “[a] signal decoding system . . . wherein said second synchronizing means includes means

28

for translating said parsed portion of said variable bit-rate signal.” Id., Col. 16:39-42. Accordingly,

48

1

construing the term “further parsing” the video bitstream so as to require a translation function (as

2

well layer extraction) would have the effect of rendering dependent claims 13 and 23 superfluous.

3

Although Claim 27 is the asserted claim, claim 27, like the other independent claims, uses

4

the same “further parsing” term. As the Federal Circuit has noted, “[o]ther claims of the patent in

5

question, both asserted and unasserted, can also be valuable sources of enlightenment as to the

6

meaning of a claim term.” See Phillips, 415 F.3d at 1314; see also Southwall Techs., Inc. v.

7

Cardinal IG Co., 54 F.3d 1570, 1579 (Fed. Cir. 1995) (stating that a term “cannot be interpreted

8

differently in different claims because claim terms must be interpreted consistently”).

9

Accordingly, the Court rejects Barnes & Noble’s construction. While the inventor clearly
conceived of a post-parser that would both extract and translate layers, there is nothing in the

11
For the Northern District of California

United States District Court

10

applicable claim language requiring “further parsing” to include translation Accordingly, the Court

12

construes “further parsing the video bitstream” as “extracting the various layers comprising the

13

video bitstream.”

14
15
16

XI.
A.

‘663 PATENT

Background and General Patent Description
The ‘663 Patent is entitled “Method and System for Symbol Binarization.” The invention is

17

directed at an improved method of binarization of data in an MPEG data stream. ‘663 Patent,

18

abstract. Specifically, the invention provides for a hybrid binarization approach, where, if a code

19

symbol index value is less than a certain threshold value, one binzarization model will be used. If

20

the code symbol index value is greater than that threshold value, however, a second binarization

21

model will be used. Id., Col. 2:14-19. Such a method is beneficial because it would permit, for

22

example, small codewords to be readily distinguishable using a simple unary code while large

23

codewords could be kept at a reasonable length using a unary prefix and an exp-Golomb suffix.

24

Such an approach would “reduce the complexity and the bitrate/size for compressing and

25

decompressing video, images, and signals that are compressed using binary arithmetic encoding for

26

entropy encoding.” Id., Col. 2:6-11.

27

The patent’s claims involve using this process both for video decoding (i.e., taking a

28

codeword and determining the “index value” that codeword represents) and video encoding (doing

49

1

the reverse, i.e., taking an “index value” and generating a codeword from that index value). The

2

claim which includes the term to be construed in this order, as discussed below, is the former.

3

B.

Representative Claim

dispute. In the parties’ first supplemental joint claim construction and prehearing statement, three of

6

these terms were included in the list of ten most significant terms as to the supplemental patents.

7

Dkt. No. 262, at 5. However, the parties second supplemental joint claim construction and

8

prehearing statement filed in response to this Court’s order contained no terms as to ‘663 patent.

9

However, in order to provide guidance to the parties, the Court will, at this time, construe the term

10

“setting said index value to a threshold” – one of the terms identified as significant by the parties in

11
For the Northern District of California

The parties have identified 13 terms in the ‘663 patent, the construction of which they

5

United States District Court

4

their original filing.

12

This term is found in claim 1. Claim 1 provides:

13

1.
A method for generating an index value from a code-word for
digital video decoding, comprising the steps of:

14
(A) setting said index value to a threshold in response to a
first portion of said codeword having a first pattern;

15
16

(B) adding an offset to said index value based on a second
pattern in a second portion of said codeword following said first
portion in response to said first portion having said first pattern; and

17
18

(C) adding a value to said index value based on a third pattern
in a third portion of said codeword following said second portion in
response to said first portion having said first pattern.

19
20

‘663 Patent, Col. 7:31-43.

21

C.

22
23
24

“setting said index value to a threshold”
Barnes & Noble

“setting the index value to a
predetermined constant”

LSI

Court

Plain and ordinary meaning, or
“setting the index value to an
initial number”

“setting said index to an initial
predetermined number”

25
26
27
28

LSI argues that Barnes & Noble’s construction is incorrect because it improperly provides
that the “index value can only be set once during the process and never changes.” Dkt. No. 268-4, at

50

1

11. Barnes & Noble, apparently because of concerns regarding brief space, did not brief this term in

2

its claim construction briefs. See Dkt. No. 270-4, at 1 n.2.

3

At the hearing, the parties agreed that “setting said index value to a threshold” should be

4

construed as “setting said index to an initial predetermined number.” The Court adopts this

5

construction.

6
7

IT IS SO ORDERED.

8
9

Dated: April 7, 2014

11
For the Northern District of California

United States District Court

10
_________________________
EDWARD M. CHEN
United States District Judge

12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

51