“[T]that material in the Tsang patent that exceeded Seagate’s Annual Report disclosure was not relevant to the anticipation challenge to claims 14 and 17 of the ‘601 patent.
In a precedential decision issued earlier this month, the US Court of Appeals for the Federal Circuit (CAFC) upheld a decision of the Patent Trial and Appeal Board (PTAB) in an appeal by LSI Corporation and Avago Technologies US Inc. (LSI) regarding the PTAB’s finding that the cited LSI reference in an inter partes review (IPR) did not qualify as prior art. The CAFC said in part that U.S. Patent No. 5,731,768 (“Tsang”) was not “from another” under 35 USC 102(e) for US Patent No. 5,859,601 (‘601 patent).
The ‘601 patent involves taking sequences of input data blocks containing error-prone patterns and then converting each input data block into a corresponding “codeword” that bypasses the error-prone pattern. The ‘601 patent works to convert blocks of input data into codewords that (1) “impose a limit on the maximum number of consecutive transitions” that are transcribed to the storage device and (2) impose a limit on the maximum number of non- transitions. These two restrictions on bit transitions are embodied in the claims with the ‘j’ restriction and the ‘k’ restriction. Claim 13 serves as the most general claim with these limitations while claim 14 narrows the limitations and claim 17 further narrows claim 14.
LSI argued that claim 14 of the ‘601 patent is preempted by US Pat. 5,392,270 (Okada) and US Patent No. 5,731,768 (Tsang). Okada detects the conversion of input data blocks using two rules that remove the occurrence of certain patterns in input data blocks for optical discs. Okada includes Tables 1-9 which include example mapping of all 8-bit input data blocks to 13-bit converted output data blocks. LSI initially argued that Okada’s disclosure of the second rule anticipated claims 14 and 17, but later argued that Tables 8 and 9 were embodiments that anticipated claims 14 and 17.
LSI’s second theory claimed that Tsang anticipated the ‘601 patent and, more specifically, the portions of Tsang disclosed in what is known as Seagate’s Annual Report. Some context regarding the ‘601 patent is necessary to understand the Tsang prediction argument. In September 1995, the inventors of the ‘601 patent filed the Seagate Annual Report regarding their Seagate coding. Material found in Seagate’s Annual Report was later embodied in the ‘601 patent.
In both the ‘601 patent and the Seagate Annual Report, the encoding process takes an input sequence of binary data and encodes it in a way that removes error-prone patterns of successive bit transitions before storing the sequence encoded on a computer storage device. There was no claim that the Seagate Annual Report could be relied upon as prior art to the ‘601 patent, but because the inventors of the ‘601 patent are the sole authors of the Seagate Annual Report, it cannot be considered “by another” according to the section. 102 even if it was publicly available before the priority date.
LSI, instead, relied on another available prior art patent, Tsang, to anticipate claims 14 and 17 of the ‘601 patent. Approximately four months after receiving Seagate’s Annual Report, but prior to the filing date of the ‘601 patent, Dr. Kinhing Tsang, an employee at Seagate, filed a patent application that later became the Tsang patent. There was no common invention between the ‘601 patent and Tsang. Tsang revised the encoding process described in Seagate’s Annual Report which resulted in a higher I/O rate allowing for more efficient storage.
Procedural History
In August 2016, the Regents of the University of Minnesota (UMN) filed suit against LSI for infringement of the ‘601 patent. LSI then filed for IPR of the ‘601 patent. When before the PTAB, UMN sought dismissal of LSI’s petition on the grounds of sovereign immunity. The PTAB denied this motion but stayed the proceedings while UMN appealed the PTAB decision.
In subsequent proceedings, UMN denied all contested claims except claims 13, 14, and 17. The PTAB issued its final decision in April 2021 finding that claim 13 was preempted by Okada. However, the PTAB held that LSI had not shown that claims 14 and 17 were unpatentable under either Okada or Tsang.
The PTAB determined that LSI’s argument that the Okada Tables preempt claims 14 and 17 were untimely because they were not raised on appeal. Moreover, if the arguments were timely, LSI still failed to explain how the tables independently constituted the complete embodiment. Regarding Tsang’s reasoning, the PTAB held that LSI “had not met its burden of proving that the parts of Tsang it relies on for anticipation represent the work of another to qualify as prior art under Sec. 102(e).”
Okada prior art
The CAFC began its analysis by considering whether the PTAB erred with respect to Okada. LSI argued on appeal that they did not need to challenge the untimeliness determination because the PTAB reached the merits. Quoting Intelligent Bio-Systems, Inc. v. Illumina Cambridge Ltd. (CAFC, 2016), where the court rejected a similar claim, the CAFC affirmed the PTAB’s finding of incompetence because LSI lost any challenge to the premature holding after failing to challenge it in their opening brief on appeal and the PTAB’s holding -‘s constitutes an independent. basis for its decision.
Earlier Tsang art
The court then turned to whether the PTAB was correct in rejecting LSI’s claim based on Tsang, determining whether the relevant passages in Tsang describe an invention “from another.” Citing in Duncan Parking Techs., Inc. v. IPS Group, Inc. (CAFC, 2019), determining whether a reference is “from another” involves three steps. The PTAB must (1) determine which parts of the reference were relied upon as prior art to anticipate the limitations in question, (2) assess the extent to which those parts were conceived “by another,” and (3) decide whether the contribution of that other person is important enough to make them a joint inventor.
LSI argued that it relied on Tsang to establish the presumption, thereby meeting the first step of the Duncan Parking test, and since Tsang is a species of the genus described in the ‘601 patent, Tsang anticipates claims 14 and 17. The CAFC determined that LSI misunderstood this test, because the inquiry is rather “whether Tsang’s invention is supported by and relevant to anticipation, or whether it was simply Tsang’s summary of Seagate’s previous Annual Report that supported and related to the forecast.” Tsang’s summary and reliance on Seagate’s Annual Report do not make Tsang an inventor of the ‘601 patent.
The CAFC continued:
“[T]it relied both on portions of Tsang that summarized the Seagate Annual Report and on additional portions of Tsang that were not merely derivative of the Seagate Annual Report—portions that described Tsang’s specific invention. But those extra parts of Tsang are not relevant to the prediction.”
Tsang’s unique features were irrelevant to the predicate of claims 14 and 17, and the features relevant to predicate were disclosed in Seagate’s Annual Report and merely repeated in the background section of Tsang. The court held that the PTAB did not err in stating that it was satisfied that the petition relied on information disclosed in both Tsang and Seagate’s Annual Report. The CAFC found the PTAB’s findings that material in the Tsang patent that exceeded Seagate’s Annual Report disclosure was not relevant to the anticipation challenge to claims 14 and 17 of the ‘601 patent and that Seagate’s Annual Report disclosure in Tsang was not relevant. make him an inventor of the material.