Hardware-Based Data Structure Found Patent Eligible Under 35 U.S.C. § 101
The Federal Circuit recently addressed the issue of patent-eligible subject matter under 35 U.S.C. § 101 once again in ADASA Inc. v. Avery Dennison Corp., No. 22-1092, 2022 U.S. App. LEXIS 34765 (Fed. Cir. Dec. 16, 2022). The claim at issue in ADASA related to radio-frequency identification (RFID) transponders wherein the unique serial number encoded on the transponder is assigned in a specific manner.1 Despite Appellant Avery Dennison asserting that the claim “is directed to nothing more than mentally ascribing meaning to a pre-existing data field,” the Federal Circuit affirmed the district court’s holding that the claim is eligible under § 101 as a matter of law.2
The only claim at issue in ADASA was claim 1 of U.S. Patent No. 9,798,967, which relates to a system for commissioning RFID transponders or “tags,” which are used “to identify and track objects by encoding data electronically in a compact label.”3 RFID tags are encoded with information or data associated with an object through a process known as “commissioning.”4 To ensure accurate identification and tracking, it is critical that the encoded data “uniquely identify the tagged object.”5
The uniqueness of the encoded data “is ensured by assigning RFID tags . . . a serialized number that uniquely identifies the associated object.”6 This uniqueness is guaranteed by a central issuing authority, which assigns blocks or groups of serial numbers that “must generally be reconciled by comparison to a central database.”7 This “generally requires encoders to maintain a continuous network connection with the database so that new serial numbers can be retrieved when an RFID tag is commissioned.”8 However, “a continuous connection is not always possible, and, even when it is, may be plagued by network delays that slow down the commissioning process.”9
The systems of the ’967 patent seek to solve these issues by using systems capable of commissioning RFID tags “on demand,” without continuous connection to a central database, and without the need for external authorizations on a transponder-by-transponder basis.10 The systems of the ’967 patent, including those covered by claim 1, do this by subdividing the blocks of serial numbers assigned by the issuing authority “into sectors that are defined by a series of fixed ‘Most Significant Bits’ (MSBs), wherein the number of allowable sectors is determined by the number of MSBs.”11 The Federal Circuit provided the following example of how MSBs are used to subdivide the assigned serial numbers
For example, according to the SGTIN-96 standard, the serial number space consists of 38 bits which can encode 238 distinct serial numbers. If the first 14 of these bits are designated as MSBs, then the serial number space is correspondingly subdivided into 214 sectors or “blocks” which can be allocated to as many as 214 different encoders. The remaining 24 bits can then be used to encode a unique serial number space within a given block.12
In other words, the systems of the ’967 patent divide blocks of serial numbers into sectors, where the serial numbers in each sector share an identical portion of bits (i.e the MSB).
In October 2017, ADASA Inc. sued Avery Dennison Corp. in the District Court for the District of Oregon, asserting that its RFID tags infringed claim 1, among others, of the ’967 patent.13 Avery Dennison argued that claim 1 is invalid under 35 U.S.C. § 101 because it is “directed to the abstract idea of partitioning a number space in the content of an RFID transponder.”14 Specifically, Avery Dennison argued that claim 1 failed the first step of the two-step Alice framework for determining whether a claim falls within the exceptions to patentable subject matter, reasoning that because “the claim language itself does not require a limited number of most significant bits to be assigned to an allocated block of serial numbers in binary . . . the claimed invention is not directed to a hardware-based approach accomplished by managing assignment of serial numbers at the binary bit level.”15
The district court disagreed, and provided several reasons for its holding that Avery Dennison “failed to carry its burden to show by clear and convincing evidence that the ’967 patent claims are directed to an abstract idea.”16 First, the district court found that Avery Dennison’s focus on only the claim language itself is legally misguided, and explained that claims must be read in light of the specification.17 Further, the district court highlighted how Avery Dennison did not provide any evidence rebutting ADASA’s expert, who had explained that the claims encompass binary encoding.18 Avery Dennison appealed this holding, among others, to the Federal Circuit.
On appeal, Avery Dennison argued that “claim 1 is directed to the abstract idea of mentally assigning meaning to a subsection of a data field and does not recite any eligibility conferring inventive concepts.”19 The Federal Circuit disagreed, and “affirm[ed] the district court’s holding that claim 1 is eligible under § 101,” noting that, “[c]onsidered as a whole, and in view of the specification, claim 1 is not directed to an abstract idea.”20 The Federal Circuit stated that, “[s]etting aside the conventional RFID hardware components, claim 1 as a whole focuses on the data structure of the serial number space,” and went on to explain
The meaning of the MSB data field—and the improvements that flow therefrom—is the result of the unique correspondence between the data physically encoded on the claimed RFID tags with pre-authorized blocks of serial number. That is not a mere mental process, but a hardware-based data structure focused on improvements to the technological process by which that data is encoded.21
The Federal Circuit also found that prior decisions have found similar claims eligible.22 Specifically, the Federal Circuit referenced Uniloc USA, Inc. v. LG Elecs. USA, Inc., 957 F.3d 1303, 1307 (Fed. Cir. 2020), where it “held eligible claims appending an additional data field to a prior art data structure used for polling stations in a communication system.”23 In Uniloc, “the additional data field enabled eligibility-conferring improvements within the communication system.”24 Similarly, “[c]laim 1 of the ’967 patent adds an additional data field to the prior art serial number space, namely MSBs, which must uniquely correspond to an allocated block of serial numbers.”25 “This unique correspondence in turn permits unique serial numbers to be assigned without the need for a continuous database connection, reducing associated network delays and allowing encoders to operate on-demand.”26 The Federal Circuit further emphasized that
in Enfish, we held eligible at step one claims related to a “specific type of self-referential data structure designed to improve the way a computer stores and retrieves memory” . . . because it enabled greater flexibility for programmers, faster search times, and smaller memory requirements. So, too, the data structure of claim 1 of the ’967 patent is designed to enable greater flexibility by allowing encoders to commission tags on-demand without consulting a central database, while simultaneously expediting the commissioning process by reducing communication delays.
Accordingly, the Federal Circuit concluded that “claim 1, viewed in light of the specification and considered as a whole, is directed to patent eligible subject matter” and it “need not address step two” of the Alice analysis.
ADASA serves as the Federal Circuit’s most recent application of the controversial Alice test, and highlights a key aspect of § 101 analysis in the context of electronic devices. Specifically, ADASA suggests that an arguably abstract idea (i.e. data structure) may be patent eligible under 35 U.S.C. § 101, so long as it confers a technological improvement and is manifested in physical form, regardless of the conventionality of that physical form.