The Precedent: Honeywell Int’l Inc. v. 3G Licensing, S.A.

In this edition of The Precedent, we outline the recent federal circuit decision in Honeywell Int’l Inc. v. 3G Licensing, S.A. (Fed. Cir. Jan. 2, 2025)

By: William Oldach & Graham Christian

Overview

The Federal Circuit reversed the Patent Trial and Appeal Board’s (the Board’s) finding of non-obviousness, finding that the Board’s decision was not supported by substantial evidence and was based on the wrong legal standard.

Issues

1. Whether the Board’s finding that a person of ordinary skill would not have been motivated to swap the last two bits of the Philips reference was supported by substantial evidence.
2. Whether the Board committed legal error by going beyond the invalidity theories set forth in the IPR petition.
3. Whether the Board committed legal error by looking only to the “best” or “preferred” combination described in the prior art to find a motivation to combine.

Holdings

1. The Board’s finding that a person of ordinary skill would not have swapped the last two bits was not supported by substantial evidence.
2. The Board committed legal error by going beyond the invalidity theories in the petition.
3. The Board committed legal error by not considering alternative motivations to combine found in the prior art.

Background and Reasoning

3G Licensing S.A. (3G) sued Honeywell Int’l for infringement of U.S. Patent No. 7,319,718 (’718 Patent). The ’718 Patent is directed to a method of coding a Channel Quality Indicator (“CQI”). Honeywell subsequently filed a petition for inter partes review (IPR), contending that claims 1, 2, 4, 5, and 15-23 of the ’718 Patent were obvious over the Philips reference alone, among other things.

The CQI indicates quality of a cellular connection received by a user’s equipment, such as a user’s mobile phone, and may be represented by a five-bit binary sequence. Rather than transmitting the five-bit sequence, user equipment instead encodes the message into a 16-bit or 20-bit “codeword” containing enough information to enable the original message to be decoded. Prior to February 13, 2002—the critical date of the ’718 Patent—Philips determined that a twenty-bit encoder of the five-bit CQI sequence (20, 5) would improve accuracy of the CQI by appending the first and second most significant bits (MSB) to the end of the 16-bit codeword.

LGE then proposed a 20-bit codeword that modified Philips by flipping the last two digits of the encoder table with the cited reason of “optimizing system throughput.” LGE incorporated this proposal into the ’718 Patent and later assigned its interest in the ’718 Patent to 3G.

The Board found that the IPR petition had not shown that a person of ordinary skill would have been motivated to switch the last two bits of the table of the Philips reference and further that, even if the petition had made this showing, it had not demonstrated the change would be desirable.

In reversing the Board’s finding of non-obviousness, a majority of the three-judge panel explained how KSR did not support the Board’s findings. Specifically, KSR states that “the problem motivating the patentee may be only one of many addressed by the patent’s subject matter.” Because the Board considered only the patentee’s avowed purpose of “optimizing system throughput,” the Board incorrectly found that swapping the last two digits would not be obvious. Moreover, the majority found that the Board ignored Honeywell’s unrebutted expert statement that swapping the last two digits “would have the purpose and effect of providing extra protection to the MSB” and that “providing extra protection to the MSB would be a desirable goal.” Accordingly, the Board’s findings were unsupported by substantial evidence.

The majority next targeted the Board’s legal conclusions, namely its conflation of anticipation and obviousness. The Board acknowledged that Philips recognized the benefit of giving extra protection to the MSB, yet found it significant that Philips did not propose swapping the final two digits, which would have the effect of protecting the MSB. Thus, the Board improperly gave weight to an anticipation theory which was not set out in the petition.

Finally, the majority found legal error in the Board’s failure to recognize that a claimed modification only need be desirable in light of the prior art, and not the “best” or “preferred approach.” The correct standard of obviousness is whether there is something in the prior art as a whole to suggest the desirability, and thus the obviousness, of the claimed invention. The Board failed to look at alternative motivations suggested in the Philips reference and thus committed legal error. In light of these findings, the majority reversed the Board’s decision and held that the ’718 Patent was obvious in view of the Philips reference.

Dissent

Judge Stoll dissented from the majority’s reversal of the Board’s decision, asserting that the majority had engaged in impermissible appellate factfinding. She argued that appellate courts should remand when a Board’s conclusion is unsupported by substantial evidence, rather than reversing. Judge Stoll clarified that the teachings of a prior art reference are a question of fact, and thus reweighing unchallenged evidence about the teachings on appeal is an intrusion into the Board’s fact-finding domain. However, Judge Stoll agreed with the majority that the Board had used improper legal standards. Therefore, in Judge Stoll’s view, the court should instead have remanded to the Board to reevaluate obviousness using the proper standard.