Broadcom Corporation v. Netflix, Inc., No. 22-1764 (Fed. Cir. June 22, 2023)

CHEN, Circuit Judge.

Broadcom Corporation appeals a decision by the Patent Trial and Appeal Board determining certain claims of Broadcom’s U.S. Patent No. 6,341,375 unpatentable under 35 U.S.C. § 103. Because we agree with the Board’s construction of the sole disputed claim term on appeal and hold its appealed findings supported by substantial evidence, we affirm.

We agree with the Board that “drive server” only requires storage capacity, not computing capabilities. The intrinsic evidence indicates a “drive server” is simply a collection of disks from which another element, the control server, retrieves data in response to a user’s request. The claim language itself calls for a “drive server” that is “configured to present a plurality of compressed data streams in response to one or more first control signals.” ’375 patent at claim 1 (emphasis added). This language indicates the drive server simply responds to requests without a need for computing capabilities. The specification further supports this understanding. A preferred embodiment of the ’375 patent includes a disk library 104, which “generally comprises one or more DVD drives and associated disks,” and a server 102, which “may be implemented as a personal computer.” Id. at col. 2 ll. 34–38 (referring to FIG. 2). The specification describes the same embodiment as having both “a server” and a “DVD drive server,” evidently referring to the server 102 and disk library 104, respectively. Id. at col. 3 ll. 25–26. Broadcom admits that under its proposed construction, claim 1 would exclude this preferred embodiment when the disk library contains no computer. Oral Arg. at 4:35–5:23. Such constructions are “rarely, if ever, correct and would require highly persuasive evidentiary support.” Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed. Cir. 1996). No such evidentiary support exists here.

3Broadcom does not assert that “drive server” has a well-established meaning in the relevant art. It instead asserts the specification statement “[t]he present invention proposes bulk drives on capable servers, with a minimal cost in the remote decoders 114a-114n” supports its proposed construction. See Appellant’s Br. 21–22 (quoting ’375 patent col. 4 ll. 14–16). We disagree. A more logical reading of the sentence—one consistent with the rest of the specification—is that it refers to the three components of the claimed invention: (1) drive servers (bulk drives), (2) control servers (capable servers), and (3) decoder devices (remote decoders), differentiating all three as separate entities. Because the intrinsic evidence sufficiently informs the meaning of “drive server,” we need not consider the extrinsic evidence to construe this claim term. See Vitronics, 90 F.3d at 1583.

The remaining issues are Broadcom’s challenges to the Board’s findings relating to one prior art reference—U.S. Patent No. 5,583,561 (Baker). Baker discloses a video-on-demand system comprising a video library, a video server, a network interface, and multiple decoders. The Board reasonably found that, under its construction of “drive server,” Baker discloses a drive server “configured to present a plurality of compressed data streams” by teaching a video library that presents multiple video streams to a video server. Netflix, Inc. v. Broadcom Corp., IPR2020-01423, 2022 WL 683412, at *10–12 (P.T.A.B. Mar. 4, 2022). Among other things, Broadcom argues Baker does not teach decoder devices that receive data streams from a control server, nor does its control server present “portions” of the data stream to different decoders. In finding otherwise, the Board found that even though Baker discloses an intermediary network interface between its video server and the decoders, it still teaches decoder devices that receive data streams from a control server because the network interface operates under the control of the video server. Id. at *12–13. The Board also found Baker’s disclosure of a video4 stream divided into blocks—each block potentially corresponding to different time periods in the same video and each block being sent to a different decoder—teaches the presenting “portions” of the data stream to different decoders limitation. Id. at *13-15. The Board’s reasoning on these findings is supported by substantial evidence.

We have considered Broadcom’s remaining arguments and find them unpersuasive. For the foregoing reasons, we affirm the Board’s decision.

AFFIRMED