CONSTELLATION DESIGNS, LLC v. LG ELECTRONICS INC.
Authored by: Jeremy J. Gustrowsky
The Federal Circuit recently issued a mixed decision in a patent infringement dispute involving the technology behind modern digital television broadcasts. The case centered on patents owned by Constellation Designs, LLC, covering improvements to the way digital communication systems map data onto analog signals using arrangements of points called “constellations.” Constellation had sued LG Electronics for willful infringement of nine claims across four patents, alleging that LG’s televisions compatible with the ATSC 3.0 broadcast standard infringed its inventions. After a jury trial in the Eastern District of Texas resulted in a $1.68 million verdict and a finding of willful infringement, LG appealed on issues of eligibility, infringement, and damages.
The technology at issue addresses a familiar problem: when digital information like a TV program is converted to analog signals and broadcast over the air, the signal weakens and picks up noise, which can distort the picture. Engineers have traditionally tried to solve this by using “uniform” constellations that maximize the distance between signal points. The inventors named on U.S. Patent Nos. 8,842,761, 10,693,700, 11,019,509, and 11,018,922 took a different approach, using non-uniform (unevenly spaced) constellations optimized for a measure called “parallel decode capacity” or PD capacity. This counterintuitive approach reportedly improved efficiency by over 25 percent compared to conventional designs.
The court split the asserted claims into two groups when analyzing patent eligibility under Section 101. The “optimization claims” from the ‘761 and ‘700 patents recited a constellation “optimized for capacity using parallel decode capacity” without specifying how that optimization was achieved. Drawing parallels to the Supreme Court’s classic decision in O’Reilly v. Morse, the Federal Circuit found these claims were directed to an abstract, result-oriented idea that would preempt all ways of optimizing constellations for PD capacity. Even though the specification described a detailed iterative process for selecting constellation points, the court emphasized that specification details cannot rescue claims that themselves recite only the broad concept. Finding no inventive concept beyond the abstract idea itself, the court vacated the summary judgment of eligibility for these claims.
The “constellation claims” from the ‘509 and ‘922 patents fared much better. Rather than reciting an abstract optimization goal, these claims covered specific non-uniform constellations with concrete structural parameters, such as unequally spaced point locations where at least two constellation points share the same location. The court held these claims represented a concrete technological solution to a technological problem (overcoming capacity constraints to improve coding gains) and therefore were not directed to an abstract idea at all. The court affirmed eligibility for these claims without needing to reach the second step of the Alice analysis.
On infringement, the court rejected LG’s argument that a patent owner cannot mix evidence from industry standards with product-specific evidence to prove infringement. Building on its earlier decision in Fujitsu Ltd. v. Netgear Inc., the court held that the same logic permitting standards-based infringement proof for entire claims also applies on a limitation-by-limitation basis, provided the standard is sufficiently specific and either mandatory or shown to be implemented by the accused product. Substantial evidence, including testimony that LG’s televisions were ATSC 3.0 compatible and incorporated the mandatory A/322 protocol, supported the jury’s infringement verdict.
On damages, LG challenged Constellation’s expert’s reliance on third-party Zenith licenses to support a built-in apportionment theory. The court found that LG’s challenge was really an admissibility argument improperly framed as a sufficiency challenge, and in any event, the expert had a sufficient basis to find the licenses comparable. Testimony established technical similarities between the Zenith and asserted patents, economic similarities in licensing structure, and that both sets of licenses tied value to the underlying technology rather than specific claims. The court affirmed the denial of JMOL on damages and the denial of LG’s motion to exclude the expert testimony, sending the case back to the district court for further proceedings consistent with the partial vacatur on eligibility.