WIAV Networks, LLC v. 3Com Corporation et al
Filing
680
TENTATIVE CLAIM-CONSTRUCTION ORDER AND REQUEST FOR CRITIQUE. Signed by Judge Alsup on June 2, 2011. (whalc1, COURT STAFF) (Filed on 6/2/2011)
1
2
3
4
5
6
IN THE UNITED STATES DISTRICT COURT
7
FOR THE NORTHERN DISTRICT OF CALIFORNIA
8
9
WIAV NETWORKS, LLC,
11
For the Northern District of California
United States District Court
10
12
13
No. C 10-03448 WHA
Plaintiff,
v.
TENTATIVE CLAIMCONSTRUCTION ORDER AND
REQUEST FOR CRITIQUE
HEWLETT-PACKARD CO.,
14
Defendant.
/
15
16
INTRODUCTION
17
In this patent infringement action involving communication-network technology, the
18
parties seek construction of six terms and phrases found in the two asserted patents. Those terms
19
and phrases are construed below. Each party has until NOON ON JUNE 9, 2011, to submit a five-
20
page critique (double-spaced, twelve-point Times New Roman font, with no footnotes and no
21
attachments) limited to points of critical concern. This is an opportunity for the parties to focus
22
solely on their most cogent critiques, not to rehash every point made in the briefs and at the
23
hearing. Any critiques must be limited to the claim-construction record and may not introduce
24
new evidence.
25
26
STATEMENT
The technology at issue relates to communication networks. A communication network is
27
a collection of devices interconnected by direct communication links between pairs of devices.
28
Communication networks were known in the prior art. The asserted patents purport to disclose
1
improvements over the known prior art of networking. Specifically, the claimed inventions are
2
directed at expanding the capabilities of communication networks and improving their
3
performance. The accused products are portable computers with wireless networking capabilities;
4
these products allegedly implement the claimed networking improvements.
5
Two patents are asserted in the second amended complaint (Dkt. No. 637). Two claims
6
from United States patent No. 5,400,338 and five claims from United States patent No. 6,480,497
7
are at issue. The parties seek construction of six terms and phrases appearing in these patents.
8
Overviews of the patents, the disputed phrases, and the associated claims are covered in detail in
9
the analysis below.
11
For the Northern District of California
United States District Court
10
ANALYSIS
Courts must determine the meaning of disputed claim terms from the perspective of one of
12
ordinary skill in the pertinent art at the time the patent was filed. Chamberlain Group, Inc. v.
13
Lear Corp., 516 F.3d 1331, 1335 (Fed. Cir. 2008). While claim terms “are generally given their
14
ordinary and customary meaning,” the “claims themselves provide substantial guidance as to the
15
meaning of particular claim terms.” Additionally, a patent’s specification “is always highly
16
relevant to the claim construction analysis.” Phillips v. AWH Corp., 415 F.3d 1303, 1312–15
17
(Fed. Cir. 2005) (en banc) (internal quotations omitted). Finally, courts also should consider the
18
patent’s prosecution history, which “can often inform the meaning of the claim language by
19
demonstrating how the inventor understood the invention and whether the inventor limited the
20
invention in the course of prosecution, making the claim scope narrower than it would otherwise
21
be.” These components of the intrinsic record are the primary resources in properly construing
22
claim terms. Courts have discretion to consider extrinsic evidence, including dictionaries,
23
scientific treatises, and testimony from experts and inventors, but such evidence is “less
24
significant than the intrinsic record in determining the legally operative meaning of claim
25
language.” Phillips, 415 F.3d at 1317–18 (internal quotations omitted).
26
While this order acknowledges that the parties have a right to the construction of all
27
disputed terms needing elaboration by the time the jury instructions are settled, the Court will
28
reserve the authority, on its own motion, to modify the constructions in this order if further
2
1
evidence — intrinsic or extrinsic — warrants such a modification. Given that claim construction
2
is not a purely legal matter, but is (as the Supreme Court describes it) a “mongrel practice” with
3
“evidentiary underpinnings,” it is entirely appropriate for the Court to adjust its construction of
4
claims prior to trial if the evidence compels an alternative construction. Markman, 517 U.S.
5
at 378, 390. Motions for reconsideration, however, may be made only in strict accordance with
6
the rules of procedure, if at all.
7
1.
8
The ’338 patent, entitled “Parasitic Adoption of Coordinate-Based Addressing by
9
THE ’338 PATENT.
Roaming Node,” was issued on March 21, 1995. Metricom, Inc. was the assignee of
the ’338 patent at the time of issue. WiAV states that it is now the owner of this patent
11
For the Northern District of California
United States District Court
10
(Sec. Amd. Compl. ¶ 15). Two claims from the ’338 patent are asserted in this litigation:
12
independent claim 1, and dependent claim 2. Five of the six terms and phrases construed by this
13
order are found in the ’338 patent. They are italicized in the claim language below.
14
15
16
17
Claim 1 covers the following method (col. 6:39–59):
1.
In a packet communication network with a plurality of
stationary nodes, a method by which a roaming node may establish
a communication link with said network comprising the steps of:
transmitting a link acquisition packet to one or more of said
stationary nodes;
18
19
receiving a response packet from each of said stationary
nodes that successfully receives said link
acquisition packet;
20
21
determining from data in said received response packets the
one of said stationary nodes that provides the best
communication link;
22
23
24
selecting the one of said stationary nodes that provides the
best communication link by transmitting to said selected
stationary node a packet informing said selected stationary
node that said selected stationary node is a current parent
node for said roaming node; and
25
26
transmitting data packets to nodes in the network using an
identifier of said parent node as part of the return identifier
for said roaming node.
27
28
3
1
2
3
Dependent claim 2 covers the following additional method (cols. 6:60–7:17):
2.
The method according to claim 1, further permitting said
roaming node to change its network communication link to a new
parent node in response to changed conditions and further
including the steps of:
4
5
monitoring each received data packet from said current
parent node to determine whether said communication link
is still good;
6
7
intermittently transmitting link acquisition packets to one
or more of said stationary nodes to determine the quality of
the possible communication link with them;
8
9
receiving a response packet from each of said stationary
nodes that successfully receives said link
acquisition packet;
11
For the Northern District of California
United States District Court
10
selecting from said received response packets one of said
stationary nodes to be a new parent node when the link
with said current parent node is no longer good;
12
13
14
15
16
transmitting to said current parent node a packet informing
said current parent node that said current parent node is no
longer the current parent node and that said new parent
node is a current parent node for said roaming node; and
transmitting to said new parent node a packet informing
said new parent node that said new parent node is a current
parent node for said roaming node.
17
The ’338 patent provides “a method for routing data packets through a packet
18
communication network” in which “some nodes can roam during network operation”
19
(col. 1:10–14). The invention addresses the following problem: “What is needed is a routing
20
method that permits roaming nodes to be addressed in a network in which stationary nodes are
21
addressed using a coordinate-based addressing method and that does not require excessive
22
processing by the network or the stationary nodes in the network to maintain contact with the
23
roaming nodes.” (col. 3:27–32). Prior art systems purportedly required stationary nodes “to
24
continuously keep track of and in touch with each roaming node and to ‘hand off’ roaming nodes
25
from one stationary node to another,” which required “tremendous processing and communication
26
overhead” (col. 3:16–20).
27
By contrast, the invention of the ’338 patent provides a method by which, “in a packet
28
communication system wherein stationary nodes are assigned an absolute coordinate-based
4
1
address, the addressing of roaming nodes is accomplished by parasitically adopting the
2
coordinate-based routing scheme used for addressing stationary nodes” (col. 3:35–40). As used
3
in the specification, an absolute coordinate-based address refers to the geographic location of a
4
stationary node. The parasitic adoption of that coordinate-based routing scheme refers to a
5
process by which a roaming node associates with a stationary node and uses the absolute
6
coordinate-based address of the stationary node as part of its own location identity. In essence,
7
the responsibility for keeping roaming nodes connected to the network is shifted from the
8
stationary nodes to the roaming nodes, which select stationary nodes to serve as their hosts.
9
A.
“Packet.”
The parties dispute the term “packet.” It appears in every claim of the ’338 patent, and it
11
For the Northern District of California
United States District Court
10
also appears in claim 1 of the ’497 patent. This order construes the term with a focus on its more
12
extensive use in the ’338 patent. The parties’ proposed constructions are shown below.
13
WIAV’S PROPOSED
CONSTRUCTION
14
15
HP’S PROPOSED
CONSTRUCTION
“electronic message”
“a unit of data sent on a network at
Layer 3”
16
17
The parties’ proposed constructions are vastly different, and neither one is supported by the
18
intrinsic evidence. In particular, the construction proposed by Hewlett-Packard Co. refers to a
19
standardized industry model that is not mentioned in the patent itself. HP explains that its
20
construction would support its non-infringement arguments, because the accused products
21
“address Layers 1 and 2” but not Layer 3, and deal only in frames as opposed to
22
packets (Opp. 29).
23
The claims of the ’338 patent refer to packets as being transmitted from node to node
24
within a packet communication network. The summary of the invention and detailed description
25
of specific embodiments describe packets being transmitted, received, forwarded, and delivered
26
by nodes. The background-of-the-invention portion of the specification describes “data packets”
27
and “packet communication” networks as well: “Packet communication is a form of data
28
communication whereby segments or packets of data are routed with error checking and
5
1
confirmation of receipt. Packets may be transmitted directly between a source and destination or
2
relayed via relay stations.” (col. 1:15–19). Thus, the intrinsic record of the ’338 patent indicates
3
that a packet is a segment of data that may be transmitted among nodes in a communication
4
network. The claim language of the ’497 patent also refers to packets as being sent, received, and
5
forwarded by nodes in a network communication system (col. 8:33–34).
6
The ’338 patent also teaches that “segments or packets of data” are structured in particular
of a packet (col. 1:22). The specification also describes a specific prior-art packet format that
9
included “four layers of header information” and a “tailer,” with the data payload sandwiched
10
between the headers and the tailer (col. 2:21–29). In order for data packets to be “routed with
11
For the Northern District of California
ways. For example, in discussing the prior art, the specification refers generally to “the header”
8
United States District Court
7
error checking and confirmation of receipt,” the information they contain must be formatted into
12
recognizable structures that conform to a shared protocol (col. 1:15–17). At the hearing, WiAV
13
agreed that packets must be formatted with structures such as headers.
14
WiAV Networks, LLC would construe the term “packet” as “electronic message.” This
15
broad construction is divorced from the context of the networking patents and is not supported by
16
the intrinsic record and would, due to its breadth, even comprehend a simple morse code message.
17
WiAV does not cite any evidence characterizing a packet either as electronic or as a message
18
(Br. 4–5). In responding to HP’s counterarguments, WiAV argues that the ’338 patent uses the
19
terms “packet” and “message” interchangeably in the following sentence from the specification:
20
“In another embodiment, the stationary node processes a packet error by discarding the packet
21
and sending a message to the source of the packet that delivery was not successful.”
22
(col. 6:12–15). This use of the term “message” to describe a particular embodiment of the
23
invention does not define the term “packet” as being synonymous with “message.” Indeed, the
24
message referenced in this embodiment may be complex, requiring multiple packets to transmit
25
all of the relevant information. The use of two different terms in this sentence suggests two
26
different meanings, not equivalence.
27
28
HP would construe the term “packet” as “a unit of data sent on a network at Layer 3.” HP
explains that Layer 3 refers to one of the seven conceptual layers of the ISO/OSI standardized
6
1
model for developing network architectures and protocols (Opp. 5–6).1 The ISO/OSI model,
2
however, is not mentioned anywhere in either of the two asserted patents. HP attempts to
3
bootstrap the ISO/OSI model into the ’338 patent by cobbling together inapposite claim-
4
construction law and focusing on an unasserted prior-art patent referenced by the ’338 patent.
5
HP’s cumbersome arguments are misguided and unpersuasive.
6
HP’s description of Layer 3 of the ISO/OSI model is based entirely on declarations and
7
prior-art publications (ibid.). Because a functional understanding of packets can be gleaned from
8
the intrinsic record, it would be improper to narrow the term to a Layer 3 limitation from this
9
extrinsic evidence. HP’s observation that the ’338 patent is “consistent with the ISO/OSI model
terminology” does not change this fact (Opp. 7). The only direct references to the ISO/OSI model
11
For the Northern District of California
United States District Court
10
that HP cites come from the specification and prosecution history of unasserted United States
12
patent No. 4,939,726. HP argues that because the prior-art ’726 patent is referenced and
13
discussed at length in the asserted ’338 patent, the entire specification and prosecution history of
14
the ’726 patent is part of the intrinsic record for the ’338 patent (Opp. 4). Many of the decisions
15
HP relies on in making this argument do not support the propositions for which they are cited.
16
More importantly, they do not collectively support HP’s attempt to lasso the ISO/OSI model and
17
drag it from the extrinsic record into the ’338 patent. HP has not justified reading a “Layer 3”
18
limitation into the construction of the disputed term. At the hearing, HP admitted that the term
19
“packet” is used generically in networking contests unrelated to the ISO/OSI model. Any further
20
arguments that the asserted patents should be interpreted in light of the ISO/OSI model will have
21
to be made to the jury.
22
23
The term “packet” will be construed to mean “a formatted segment of digital data
typically composed of headers, a message, error correction, and a tailer.”
24
B.
25
26
“Link Acquisition Packet.”
The parties dispute the phrase “link acquisition packet.” It appears in all claims of
the ’338 patent. The parties’ proposed constructions are shown below.
27
28
1
HP further explains that “ISO/OSI” refers to the International Organization for Standardization sevenlayered reference model of Open Systems Interconnection.
7
1
2
3
4
WIAV’S PROPOSED
CONSTRUCTION
HP’S PROPOSED
CONSTRUCTION
“an electronic message requesting
information about establishing a direct
electronic connection”
“a packet for acquisition and
synchronization”
5
HP explains that its construction would support its non-infringement arguments, because the
6
accused instrumentalities do not perform synchronization. HP also suggests that the phrase may
7
be indefinite (Opp. 30).
8
9
Contrary to HP, the phrase “link acquisition packet” is not insolubly indefinite. Claim 1
recites “transmitting a link acquisition packet to one or more of said stationary nodes” as the first
step performed by a roaming node in order to perform the claimed method for establishing a
11
For the Northern District of California
United States District Court
10
communication link with a network. Thus, a link acquisition packet is something that a roaming
12
node transmits to one or more stationary nodes. Dependent claim 2 recites a similar step
13
performed by a roaming node as part of a method for changing its network communication link to
14
a new parent node: the roaming node “intermittently transmit[s] link acquisition packets to one or
15
more of said stationary nodes to determine the quality of the possible communication link with
16
them.” Thus, the roaming node transmits link acquisition packets to stationary nodes for the
17
purpose of identifying potential parent nodes. Independent claim 3 confirms that stationary nodes
18
“receiv[e]” link acquisition packets from roaming nodes (col. 7:34–35).
19
In discussing “the operation of a roaming node according to the invention,” the
20
specification repeatedly refers to link acquisition packets as “acquisition/synchronization
21
packets” (e.g., col. 4:58–62). Synchronization, therefore, is another aspect of link acquisition
22
packets. WiAV resists this conclusion on the grounds that the word “synchronization” does not
23
appear in the claims (Reply Br. 13–14). The claims, however, are only the starting point for
24
construction, and they “must be read in view of the specification, of which they are a part.”
25
Markman v. Westview Instruments, Inc., 52 F.3d 967, 979 (Fed. Cir. 1995) (en banc), aff’d,
26
517 U.S. 370 (1996).
27
28
WiAV’s proposed construction of “packet” as “electronic message” was rejected above.
WiAV attempts to support the rest of its proposed construction of “link acquisition packet” by
8
1
analyzing the purpose served by link acquisition packets. This order agrees with WiAV that link
2
acquisition packets “allow the roaming node to obtain link-quality information from stationary
3
nodes within its range so as to determine which stationary node gives the best link with the
4
roaming node,” and that “the purpose of determining the best link is to select the parent node”
5
(Br. 14–15). This outcome, however, does not establish that the link acquisition packets each
6
contain a request for “information about establishing a direct electronic connection.” The claims
7
provide that “a response packet” is sent to the roaming node by “each of said stationary nodes that
8
successfully receives said link acquisition packet,” but there is no indication that the link
9
acquisition packet includes a request for any specific type of information (col. 6:45–47). WiAV’s
11
For the Northern District of California
United States District Court
10
proposed construction overreaches.
The phrase “link acquisition packet” will be construed to mean “a non-message packet
12
sent by a roaming node to one or more stationary nodes, configured to establish a link and to
13
synchronize data rate and format.”
14
15
16
17
18
19
20
C.
“Stationary Node.”
The parties dispute the phrase “stationary node.” It appears in all claims of the ’338
patent. The parties’ proposed constructions are shown below.
WIAV’S PROPOSED
CONSTRUCTION
HP’S PROPOSED
CONSTRUCTION
No construction. Or, “a network
electronic device that is not moving and
can send and receive packets”
Node: “an element of a network that
sends, forwards, and
receives packets.”
Stationary Node: “a node in the
network that is assigned an absolute
geographic coordinate-based address
or a code conveying the same”
21
22
23
24
HP explains that its constructions would support its non-infringement arguments, because the
25
accused products do not forward packets and are not assigned an absolute geographic coordinate-
26
based address or a code conveying the same (Opp. 29–30).
27
HP argues for separate constructions of the phrase “stationary node” and its component
28
term “node.” Because the term “node” appears in both asserted patents, HP relies on evidence
9
1
from both patents in construing the term node; then, HP seeks to import its two-patent
2
construction of “node” into the construction of the disputed phrase “stationary node,” which
3
appears only in the ’338 patent. This approach is improper. The parties were instructed to
4
identify six terms or phrases for construction. “Stationary node” was one of them, and “node”
5
was not. Indeed, “node” was identified separately as one of the “most significant” terms that did
6
not make the cut for claim construction (Dkt. No. 650 at 2–3). HP may not sneak in a seventh
7
bonus term after the parties were limited to, and selected, six terms and phrases for construction.
8
The phrase identified for construction — “stationary node” — appears only in
is extrinsic evidence and will be treated as such. Although they share an inventor, the ’497 patent
11
For the Northern District of California
the ’338 patent. To the extent the ’497 patent is relevant at all to the construction of this phrase, it
10
United States District Court
9
was issued seven years after the ’338 patent, and it addresses a completely different problem
12
within the broad field of communication networks. The nodes disclosed in the ’497 patent may
13
bear broad similarities to the nodes of the ’338 patent, but network nodes are required to perform
14
different functions in the contexts of these two different inventions. The use of the term “node”
15
in the later-issued ’497 patent is not a proper basis for reading limitations into the phrase
16
“stationary node” within the ’338 patent.
17
Focusing on the ’338 patent, the claims require stationary nodes to “transmit” and
18
“receive” packets under certain circumstances. Dependent claim 4 additionally requires some
19
stationary nodes to “accept” and “forward” data packets, but these requirements are not found in
20
any of the independent claims. Indeed, the embodiment disclosed in independent claim 3 requires
21
only “that each stationary node can communicate directly with at least one other stationary node,”
22
meaning two-way communication (col. 7:26–28). Thus, a given stationary node in the network
23
may be directly linked with only one other stationary node, and not with any roaming node. Such
24
a stationary node would be able to exchange packets with the one node to which it is linked, but it
25
would not be able to receive a packet from one node and then “forward” the packet to another
26
node. The claims of the ’338 patent teach that a given stationary node must be able to transmit
27
and receive data packets, but need not be able to forward them.
28
10
1
HP seeks to add a forwarding limitation to the construction of “stationary node,” but the
patent indicate that a stationary node need not be able to forward packets. The extrinsic evidence
4
HP relies on from the ’497 patent, the unasserted ’726 patent, and the Finn publication do not
5
show that the stationary nodes of the ’338 patent must be capable for forwarding. The usage of
6
the terms “device” and “node” in the ’338 patent also does not support reading a forwarding
7
limitation into the phrase “stationary node,” as HP asserts (Opp. 10–13). HP points out that
8
Figure 3 of the ’338 patent, “a flow chart describing the operation of a stationary node,”
9
references a “‘forward’ table,” but Figure 3 depicts only one possible embodiment of a stationary
10
node. HP also characterizes the specification of the ’338 patent as requiring each node to have a
11
For the Northern District of California
record does not support doing so. As the foregoing analysis demonstrated, the claims of the ’338
3
United States District Court
2
means for routing and forwarding, but the cited language describes only the invention that was
12
claimed in the prior-art ’726 patent, not the invention claimed in the asserted ’338 patent
13
(col. 1:60–66). In short, HP cannot overcome the clear indication in the claim language that not
14
all of the stationary nodes disclosed in the ’338 patent must be able to forward data packets.
15
The specification of the ’338 patent further teaches that stationary nodes have coordinate-
16
based addresses. The background section explains that the invention addresses the following
17
problem: “What is needed is a routing method that permits roaming nodes to be addressed in a
18
network in which stationary nodes are addressed using a coordinate-based addressing
19
method . . . ” (col. 3:27–30) (emphasis added). Similarly, the summary of the invention explains
20
that it is implemented “in a packet communication system wherein stationary nodes are assigned
21
an absolute coordinate-based address” (col. 3:35–37) (emphasis added). These statements do not
22
describe particular embodiments of the invention (as WiAV argues), but rather the general
23
context and character of the invention itself. The specification further explains that a “coordinate-
24
based address” is an identifier based on “absolute geographic coordinates,” such as latitude and
25
longitude (col. 3:11). Thus, a stationary node has an addresses based on its geographic location.
26
The specification also discusses stationary nodes in opposition to “roaming nodes,” which
27
are nodes that “can roam during network operation” (col. 1:13–14). Thus, a stationary node is
28
one that can not roam during network operation. This construction differs in a subtle but
11
1
important way from WiAV’s proposal, which would define stationary nodes as “not moving,” but
2
not necessarily fixed or unable to move. WiAV’s construction would vitiate the patent’s
3
distinction between stationary nodes and roaming nodes by allowing a node that can roam but is
4
not now moving to qualify both as a stationary node and a roaming node. This result contradicts
5
the patent’s distinction between stationary and roaming nodes. A stationary node is immobile.
packets, not merely capable of doing so. According to HP, a node that merely can send, receive,
8
and forward packets, but that is not so engaged, is not “in a . . . network,” as required by the
9
claims, in any meaningful sense (Opp. 13). HP provides no support for its interpretation of what
10
it means for a node to be in a network, and this order disagrees. Requiring each stationary node
11
For the Northern District of California
HP argues that stationary nodes must be actually engaged in sending and receiving
7
United States District Court
6
to regularly participate in packet routing would presuppose a level and distribution of network
12
traffic that is both unrealistic and unduly restrictive. The capability of transmitting and receiving
13
packets is enough.
14
WiAV opposes construction of the phrase “stationary node.” This order finds, however
15
that the meaning of “stationary node” would not be readily apparent to the jury. A construction is
16
warranted. Assuming the phrase will be construed, WiAV argues that the ordinary meaning of
17
“stationary” is “not moving.” WiAV cites extrinsic evidence supporting this definition, and
18
WiAV also cites portions of the specification that contrast roaming nodes with stationary nodes,
19
concluding that the stationary nodes do not move during the same period of time when roaming
20
nodes move. WiAV also cites the prosecution history of the ’338 patent, noting that the phrase
21
“roaming node” in claim 3 originally read “current roaming node.” According to WiaV, the
22
dropped modifier “current” suggested that a roaming node could stop moving and become a
23
stationary node (Br. 9–10). As explained, the temporal aspect of WiAV’s “not moving”
24
construction is problematic. This construction was rejected above because it vitiates the patent’s
25
distinction between stationary nodes and roaming nodes. None of WiAV’s evidence redeems its
26
“not moving” construction from this failing.
27
28
12
1
The phrase “stationary node” will be construed to mean “an immobile network element
2
that has an address based on its geographic location, and that can transmit and receive packets,
3
but need not be able to forward packets.”
4
5
6
7
8
9
11
For the Northern District of California
United States District Court
10
D.
“Current Parent Node.”
The parties dispute the phrase “current parent node.” It appears in all claims of the ’338
patent. The parties’ proposed constructions are shown below.
WIAV’S PROPOSED
CONSTRUCTION
HP’S PROPOSED
CONSTRUCTION
“for a time, the exclusive
communicating intermediate device
with other devices in the network”
No construction. Or, “the stationary
node to which a roaming node has a
presently established link”
Claim 1 of the ’338 patent sets forth a method by which a roaming node can select a
12
stationary node to be “a current parent node” for the roaming node, thereby establishing “a
13
communication link” between the roaming node and the communication network to which the
14
stationary nodes belong. Dependent claim 2 sets forth further method steps by which the roaming
15
node can “change its network communication link to a new parent node.” When “the link with
16
said current parent node is no longer good,” the roaming node selects a different stationary node
17
“to be a new parent node.” To implement the change, the roaming node (1) informs the said
18
current parent node that it is no longer the current parent node, and that the said new parent node
19
is now a current parent node for the roaming node, and (2) informs the said new parent node that
20
it is a current parent node for the roaming node.
21
Thus, the claims of the ’338 patent teach that a “current parent node” is a stationary node
22
in a communication network that is providing a communication link between the network and a
23
roaming node. The specification uses the phrase “current parent node” consistently with this
24
understanding. HP opposes construing the phrase “current parent node” (Opp. 18). This order,
25
however, finds that the jury would benefit from elaboration of this phrase.
26
27
WiAV asserts that “for a time” is the ordinary meaning of “current,” but does not object to
“presently” or “at the present time” as alternative glosses (Br. 16). The term “present” is more
28
13
1
synonymous with “current” than is the phrase “for a time,” because the phrase “for a time”
2
connotes transience without any connection to the present time.
3
WiAV also asserts that the relationship between a roaming node and its current parent
4
node is exclusive (ibid.). HP agrees, as the term “the” in its proposed construction signifies
5
exclusivity. This order concurs with the parties on this point. Under the claimed methods, a
6
roaming node has only one “network communication link” with one parent node at a time.
7
WiAV further emphasizes the “intermediate” nature of a parent node, but WiAV does not
8
identify any language in the claims or specification that describes parent nodes as intermediaries
9
(ibid.). The fact that parent nodes serve as intermediaries does not necessarily distinguish them
from other stationary nodes. Independent claim 3 explains that “any stationary node can
11
For the Northern District of California
United States District Court
10
communicate with any other stationary node by relaying data through any number of stationary
12
nodes” in the network (col. 7:28–31). True, only a parent node serves as an intermediary for a
13
roaming node. A stationary node that is not a parent node, however, also may serve as an
14
intermediary by forwarding packets from one stationary node to another. The general concept of
15
intermediacy does not distinguish parent nodes from all other stationary nodes, and it need not be
16
incorporated into the construction of the disputed phrase.
17
The phrase “current parent node” will be construed to mean “the stationary node on which
18
a roaming node now relies as the exclusive communication link between the roaming node and
19
the communication network to which the stationary node belongs.”
20
21
22
23
E.
“Return Identifier for Said Roaming Node.”
The parties dispute the phrase “return identifier for said roaming node.” It appears in
claim 1 of the ’338 patent. The parties’ proposed constructions are shown below.
24
WIAV’S PROPOSED
CONSTRUCTION
HP’S PROPOSED
CONSTRUCTION
25
“address information for sending a reply
packet back to the roaming device”
“information sent in data packets
originating at the roaming node,
wherein such information is [used/for
use] by all other nodes to route
packets back to the roaming node”
26
27
28
14
1
HP explains that its construction would support its non-infringement arguments, because the
2
accused element is used only in local, “one-hop” communications as opposed to network-wide
3
routing (Opp. 30).
4
Claim 1 recites a method step in which a roaming node, having acquired a current parent
5
node, “transmit[s] data packets to nodes in the network using an identifier of said parent node as
6
part of the return identifier for said roaming node” (col. 6:57–59). The claims make no other
7
mention of the disputed phrase. The specification’s summary of the invention elaborates on this
8
aspect of the invention: “The coordinates of this parent stationary node are used in the header
9
block of each packet transmitted from the roaming node and these coordinates are seen by all
receiving nodes and then used in subsequent transmissions back to the roaming node.”
11
For the Northern District of California
United States District Court
10
(col. 3:42–47).
12
The parties agree that the return identifier is information that originates at the roaming
13
node and can be used by other nodes to send communications back to the roaming node
14
(Reply Br. 17 n.2). This interpretation is rooted firmly in the claim language and specification.
15
The parties’ disagreements cluster around certain details they seek to add to this
16
general description.
17
WiAV characterizes the return-identifier information as “address” information. HP
18
opposes this descriptor as unsupported and ambiguous (Opp. 21). The patent does not refer to the
19
return-identifier information as an “address.” A separate portion of the patent, however, discloses
20
various types of addresses, including local and ultimate source and destination addresses
21
(col. 2:34–49). The coordinates of the parent node do represent a type of address, but the term
22
“address” would not add value to the construction. A more precise description of the return-
23
identifier information is that it identifies the origin of a data packet in terms of the parent node.
24
In its brief, HP construes the return-identifier information as being “used by all other
25
nodes to route packets back to the roaming node.” WiAV objects that such a construction would
26
require every single other node in the network to route packets back to the roaming node, whereas
27
the patent claims contain no such requirement (Reply Br. 16–17). This order agrees. There is no
28
requirement that “all other nodes” use the return-identifier information. Only if a given node
15
1
happens to send or become involved in forwarding “subsequent transmissions back to the
2
roaming node” would that node have occasion to use the return-identifier information. The event
3
of a roaming node transmitting a data packet to the network does not automatically trigger a
4
reciprocal transmission from every other node in the network. Hence, the return-identifier
5
information need not be “used by all other nodes” as HP proposes.
6
At the hearing, HP amended its proposed construction to characterize the information as
7
being “for use” by all other nodes instead of “used” by all other nodes. HP explained that this
8
change was intended to address WiAV’s concern that HP’s construction required actual use of the
9
information by every other node in the network. WiAV did not agree that HP’s new language
solved that problem. In any event, this order concludes that the information must be only useable
11
For the Northern District of California
United States District Court
10
by the other nodes that receive or forward it.
12
The phrase “return identifier for said roaming node” will be construed to mean
13
“information in a data packet sent from a roaming node to a network, that indicates the origin of
14
the packet by identifying the parent node to which it was transmitted, and that can be used by any
15
other node in the network receiving or forwarding it to send reply or subsequent communications
16
to the roaming node.”
17
2.
18
The ’497 patent, entitled “Method and Apparatus for Maximizing Data Throughput in a
THE ’497 PATENT.
19
Packet Radio Mesh Network,” was issued on November 12, 2002. Ricochet Networks, Inc. was
20
the assignee of the ’497 patent at the time of issue. WiAV states that it is now the owner of this
21
patent (Sec. Amd. Compl. ¶ 10). Five claims from the ’497 patent are asserted in this litigation:
22
independent claims 1, 27, and 28, and dependent claims 6 and 8, which depend upon claim 1.
23
Claims 1, 6, and 8 cover methods; claims 27 and 28 cover apparatus. Only one of the phrases
24
construed by this order is found exclusively in the ’497 patent. It is italicized in the claim
25
language below.
26
27
28
16
1
2
3
Claim 1 covers the following method (col. 8:23–45):
1.
In a mesh network communication system capable of
dynamically establishing links between communicating nodes, a
method for optimizing net throughput on a link from a first node to
a second node, the method comprising steps of:
4
dynamically establishing the link between the first node
and the second node with a first signal, wherein:
5
6
the first and second nodes are part of a mesh
network communication system, and
7
each of the first and second nodes sends, receives,
forwards packets with the mesh network
communication system [sic];
8
9
determining at least one performance metric, at the second
node, of data-link on-air characteristics of the first signal
from the first node;
11
For the Northern District of California
United States District Court
10
relaying information relating to the at least one
performance metric from the second node to the first
node; and
12
13
dynamically modifying at least one signal characteristic of
a second signal transmitted from the first node to the
second node, wherein the dynamically modifying step is
responsive to at least one performance metric.
14
15
16
Dependent method claim 6 incorporates the method of claim 1 and adds the limitation that the
17
performance metric must be “based, in part, on a signal strength of the first signal received at the
18
second node” (col. 8:61–63). Dependent method claim 8 also incorporates the method of claim 1
19
and adds the limitation that the performance metric must be “based on statistical information
20
about the link” (col. 8:66–67). Independent apparatus claims 27 and 28 disclose a transceiver and
21
a controller, respectively, that could be used to perform methods related to that of claim 1
22
(col. 10:24–61).
23
The ’497 patent purports to optimize net throughput “by dynamically modifying signal
24
characteristics of the signals transmitted between nodes in response to performance metrics”
25
(col. 2:64–67). The ’497 patent recognizes that conditions may vary from node to node within a
26
network. For example, a given node might handle an unusually large volume of traffic, or it
27
might be located near machinery that generates a certain type of interference with signal
28
17
1
reception. If reception conditions vary from node to node, then the optimal characteristics of an
2
incoming signal also vary from node to node.
3
The ’497 patent discloses a way for different nodes within a network to receive signals
4
with different characteristics that are selected to maximize reception success based on the varying
5
conditions at each node. For example, a node might be located near a source of static interference
6
that causes single-bit errors. A coding technique called “interleaving” enables such errors to be
7
repaired by the recipient. Accordingly, this particular node might instruct other nodes to use the
8
interleaving technique when sending packets to this node (col. 7:55–65). When sending packets
9
to different nodes that are not subject to static interference, the sending nodes need not invest in
this extra coding. They might, however, make other adjustments to the signal characteristics
11
For the Northern District of California
United States District Court
10
(e.g., data rate, modulation type, etc.) based on the reception conditions at the other nodes.
12
Maximizing the performance of each node-to-node link based on the unique reception conditions
13
at each node has the net effect of maximizing the overall performance of the whole network
14
(cols. 2:64–3:5).
15
The ability to modify signal characteristics and use extra coding techniques such as
16
interleaving was known in the prior art. The advance claimed by the ’497 patent is a way to
17
dynamically customize the communication signal linking any two nodes in a network. In contrast
18
to the prior-art approach of using the same type of signal uniformly throughout a network, the
19
dynamically-varying links disclosed in the ’497 patent purportedly deliver the highest
20
performance capable of being supported by the arrangement and condition of the network nodes
21
at any given time. The essence of the supposed invention is the exploitation of signal variability
22
“on a per-link basis” throughout a communication network (col. 3:49–54).
23
A.
“Mesh Network Communication System” and
“Mesh Network Wireless Communication System.”
24
The parties dispute the phrases “mesh network communication system” and “mesh
25
network wireless communication system.” The former is used in asserted independent claims 1
26
and 27; the latter is used in asserted independent claim 28. The parties’ proposed constructions
27
are shown below.
28
18
1
2
3
4
WIAV’S PROPOSED
CONSTRUCTION
HP’S PROPOSED
CONSTRUCTION
“a network in which at least three
electronic devices can communicate on
alternative routes, such as directly or
though intermediate devices”
“a collection of nodes in a wireless
network which autonomously
connect, send, receive, forward and
analyze packetized traffic in the
network, wherein there are one or
more intermediate nodes between the
source and destination of a
communication, and excluding pointto-point, star, conventional wireline,
cellular, bus, and computer
backplane architectures”
5
6
7
8
would support its non-infringement theories, because the accused products do not forward packets
11
For the Northern District of California
The parties’ proposed constructions share very little in common. HP explains that its construction
10
United States District Court
9
and do not use a mesh network architecture for networking (Opp. 30). Indeed, at the hearing HP
12
stated that the accused wifi products use a star network architecture rather than a mesh one.
13
The claims of the ’497 patent use the disputed phrases without elaboration, so this order
14
looks to the specification to build a construction. To introduce the prior-art framework within
15
which the invention is implemented, the background-of-the-invention section states: “In a mesh
16
network, there is an [sic] collection of nodes which autonomously connect, send, receive, forward
17
and analyze packetized traffic in the network . . .” (col. 1:10–12). Similarly, the detailed
18
description of the preferred embodiments describes “a mesh network 10.” It states: “In such a
19
network 10, an interconnected mesh of data-packet sending and receiving nodes is collectively
20
collecting, routing and delivering data packets.” (col. 4:21–29).
21
The prosecution history of the ’497 patent provides further evidence of how the applicants
22
used the disputed phrase. In responding to a rejection based on prior art, the applicants
23
emphasized that in a mesh network, “a node could be an intermediate point in the mesh network
24
and not necessarily the source or destination” of a message. The response further explained:
25
“Mesh networks use routing and other algorithms to relay a message to its destination where one
26
node is the source, there are one or more intermediate nodes and a destination node.”
27
(Hwang Exh. 4 at 77).
28
19
1
The ’497 patent specification and prosecution history say that a mesh network contains
2
nodes that can relay a message to its destination by sending, receiving, analyzing, and forwarding
3
data packets. This form of operation is the essence of a mesh network for purposes of
4
the ’497 patent.
5
The patent also indicates that its purported invention is to be implemented wirelessly.
6
Claims 1 and 27 refer to the “data-link on-air characteristics” of a “signal” (cols. 8:37–38,
7
10:37–38). Claim 28 explicitly requires a “wireless” communication system (col. 10:43). Thus,
8
all of the asserted claims relate to wireless technology. The specification confirms this
9
interpretation with frequent references to “radio.” The summary and detailed description discuss
“radio hardware” such as “transmitter(s),” “receiver(s),” and an “antenna” (e.g., cols. 3:34,
11
For the Northern District of California
United States District Court
10
4:48–50). They also discuss the “on-air characteristics” of “signals,” including “bandwidth” and
12
“frequency” (e.g., cols. 3:3–5, 5:58). The background section begins with the plain statement,
13
“[t]his invention relates to packet communications in a radio-based mesh network” (col. 1:7–8)
14
(emphasis added). The mesh network of the ’497 patent is wireless.
15
WiAV observes that claim 28 explicitly requires “a mesh network wireless
16
communication system” (emphasis added), whereas claims 1 and 27 require only “a mesh
17
network communication system.” According to WiAV, this drafting choice raises “a
18
presumption” that the latter phrase does not incorporate a “wireless” limitation (Reply Br. 20).
19
Any such presumption is rebutted by the overwhelming evidence described above. In particular,
20
the statement quoted from the background section satisfies WiAV’s demand for “words or
21
expressions of manifest exclusion or restriction” (Br. 23). WiAV also advances the fallback
22
argument that even if the mesh network is construed to require some wireless connections, the
23
construction should not require an entirely wireless network (Reply Br. 20). The ’497 patent
24
contemplates wireless — and only wireless — technology. A construction crafted to
25
accommodate a partially wireless network would strain the patent.
26
HP argues that the specification also defines “mesh network” as a specific type of data
27
communication network architecture, as opposed to other types of network architecture such as a
28
“bus” or “star” arrangement of nodes (Opp. 23–25). The portion of the specification HP cites,
20
specifically addresses “[m]esh packet radio networks,” not simply mesh networks, and it does not
3
describe how the various architectures differ (col. 1:23–28). Construing mesh networks in
4
opposition to bus or star networks would be neither accurate nor informative in light of the cited
5
passage of the specification. HP’s other arguments for its proposed exclusionary limitation fail as
6
well. The prosecution-history statement distinguishing prior art that “only addresses point-to-
7
point communication where the transmitter is the source of the communication and the receiver is
8
the destination of the communication” may impact claim scope, but it does not define “mesh
9
network” for purposes of claim construction (Hwang Exh. 4 at 77). The specification statement
10
cited to show that architecture, or topology, is an important aspect of a mesh network, mentions
11
For the Northern District of California
however, does not support incorporating this comparison into the construction. The passage
2
United States District Court
1
topology only in passing (Opp. 25).2 The extrinsic evidence HP offers to illustrate other
12
architectures in contrast to the mesh network illustrated in Figure 1 do not compel exclusion of
13
star and bus arrangements. In sum, HP has not shown good cause to construe the disputed
14
phrases by listing a variety of things that a mesh network is not.
HP also argues that the nodes of a mesh network must be able to “autonomously connect.”
15
16
This phrase is found in the statement from the background section of the specification quoted
17
above. Autonomous connection is not mentioned in any of the other statements regarding mesh
18
networks. Indeed, the specification does not elaborate on what is meant by “autonomously
19
connect” or how autonomous connection might be relevant to the claimed invention. This single,
20
ambiguous statement does not justify reading an autonomous-connection limitation into the
21
construction of the “mesh network” phrases.
WiAV argues that the nodes of a mesh network need not be able to “forward” packets to
22
23
other nodes. WiAV’s evidence on this point is unconvincing. WiAV views the “forward”
24
limitation as “an improper attempt to incorporate an embodiment from the specification into the
25
claim” (Br. 24). The concept of forwarding, however, comes from a definitional statement in the
26
27
28
2
The topology of a network is the layout pattern of the nodes and links. For example, they may be
arranged in a straight line, or they may form a ring.
21
1
background section — not from a description of a specific embodiment (col. 1:10–12). WiAV
2
also views the mesh network depicted in Figure 1 as incompatible with a forwarding requirement.
3
Figure 1 from the ’497 Patent: A Mesh Network
4
5
6
7
8
9
11
For the Northern District of California
United States District Court
10
12
13
14
15
16
17
The specification explains that items 11a–11k in Figure 1 are “nodes,” each of which “has
18
capabilities for transmitting to and receiving from various other nodes” (col. 4:22–25). The
19
specification does not describe items 14a and 14b, which appear to be laptop computers. WiAV
20
interprets the laptops as being “part of the mesh network” and notes that the laptops “cannot
21
forward packets because each is linked to only one node” (Br. 24). It is true that the laptops
22
cannot forward packets, but there is no evidence that they are “nodes” in the mesh network.
23
Rather, the “nodes” are identified with the 11a–11k numbering scheme, and the laptops are
24
labeled with a separate numbering scheme. Thus, a more plausible reading of Figure 1 recognizes
25
that while the laptop devices are connected to the mesh network, they are not nodes within the
26
mesh network. Every node depicted in Figure 1 is connected to at least two other nodes, meaning
27
that it can forward packets. The inability of the laptops to forward packets does not detract from
28
the ability of the nodes to do so.
22
1
WiAV also would construe the disputed phrases to require at least three nodes arranged
2
such that alternative routes are available for transmitting a message between any two nodes. A
3
forwarding requirement necessarily would imply at least three nodes, so a three-node limitation
4
would be redundant in light of the foregoing analysis. WiAV’s proposed alternative-route
5
limitation is not supported by any evidence from the intrinsic record. The proffered dictionary
6
definition, standing alone, is not a sufficient basis for reading such a limitation into the
7
disputed phrase.
8
9
The phrases “mesh network communication system” and “mesh network wireless
communication system” both shall be construed to mean “a wireless communication system
composed of nodes that can relay a message to its destination by sending, receiving, analyzing,
11
For the Northern District of California
United States District Court
10
and forwarding data packets.”
12
For clarity, it is worth noting that the “mesh network communication system” of the ’497
13
patent has been construed to require nodes that can forward packets, while a forwarding limitation
14
has not been incorporated into the construction of the “stationary nodes” of the ’338 patent. This
15
difference in construction results from differences in the patents themselves. The ’338 patent
16
does not mention a “mesh network communication system.” Instead, it refers to “a packet
17
communication network” or “a packet communication system.” The “stationary nodes” of
18
the ’338 patent do not necessarily belong to a “mesh network communication system” as
19
disclosed in the ’497 patent. Thus, they do not need to be capable of forwarding packets simply
20
because nodes in a “mesh network communication system” must be able to do so.
21
CONCLUSION
22
The constructions set forth above will apply in this action. The Court reserves the
23
authority, on its own motion, to modify these constructions if further evidence warrants such a
24
modification. Additionally, by NOON ON JUNE 9, 2011, each side may file a five-page critique
25
(double-spaced, twelve-point Times New Roman font, with no footnotes and no attachments)
26
limited to points of critical concern. This is an opportunity for the parties to focus solely on their
27
28
23
1
most cogent critiques, not to rehash every point made in the briefs and at the hearing. Any
2
critiques must be limited to the claim-construction record and may not introduce new evidence.
3
4
IT IS SO ORDERED.
5
6
Dated: June 2, 2011.
WILLIAM ALSUP
UNITED STATES DISTRICT JUDGE
7
8
9
11
For the Northern District of California
United States District Court
10
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
24
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?