DCT

4:26-cv-00047

Florida State University Research Foundation Inc v. AT&T Services Inc

Log In
Key Events
Complaint

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 4:26-cv-00047, N.D. Fla., 01/26/2026
  • Venue Allegations: The complaint alleges venue is proper because each Defendant operates wireless networks, advertises, maintains numerous brick-and-mortar retail stores, and employs staff within the Northern District of Florida, constituting regular and established places of business from which they allegedly commit acts of infringement.
  • Core Dispute: Plaintiff alleges that Defendants’ 5G wireless networks and related services infringe patents related to compressing Channel State Information (CSI) for efficient data transmission and to intelligent packet relay protocols for extending network range.
  • Technical Context: The technologies at issue address fundamental challenges in modern wireless communications by seeking to improve spectral efficiency, network capacity, and reliability in complex, high-data-rate environments like 5G.
  • Key Procedural History: The complaint notes that Plaintiff Red Hills Signal Systems is the exclusive licensee of the patents-in-suit from Plaintiff Florida State University Research Foundation, which owns the patents by assignment. The complaint also presents a detailed argument for joining the three distinct telecommunications carriers as defendants under 35 U.S.C. § 299, alleging their 5G network services are functionally identical, interoperable, and arise from the same series of transactions or occurrences related to the implementation of the 5G standard.

Case Timeline

Date Event
2012-11-20 ’693 Patent Priority Date
2014-08-26 ’693 Patent Issue Date
2016-02-16 ’104 Patent Priority Date
2017-12-05 ’104 Patent Issue Date
2018-03-01 5G Standard (Release 15) Completion Date
2025-02-27 Defendants Announce Joint Venture for 5G APIs
2026-01-26 Complaint Filing Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 9,838,104 - *"System and Method for Fast Compression of OFDM Channel State Information (CSI) Based on Constant Frequency Sinusoidal Approximation"*

  • Patent Identification: U.S. Patent No. 9,838,104, "System and Method for Fast Compression of OFDM Channel State Information (CSI) Based on Constant Frequency Sinusoidal Approximation," issued December 5, 2017.

The Invention Explained

  • Problem Addressed: In advanced wireless systems (like MIMO), receivers must report channel conditions back to the transmitter via Channel State Information (CSI) to optimize transmissions. As the number of antennas and subcarriers increases, the size of this CSI feedback grows, consuming excessive uplink bandwidth and creating a "feedback overhead" bottleneck that reduces network efficiency (Compl. ¶73; ’104 Patent, col. 1:24-33). Prior compression techniques were often too computationally intensive for mobile devices or sacrificed too much accuracy (Compl. ¶74).
  • The Patented Solution: The invention proposes a method to highly compress CSI by approximating the complex CSI vector as a linear combination of a small number of pre-defined "base sinusoids on constant frequencies." Because the base sinusoids are constant and pre-defined, complex real-time calculations can be replaced with much simpler and faster operations, such as dot products and matrix multiplications, which can be pre-computed. The small set of coefficients for these base sinusoids becomes the compressed CSI feedback, drastically reducing the amount of data that needs to be transmitted (Compl. ¶¶75-76; ’104 Patent, Abstract).
  • Technical Importance: This approach reduces the computational burden on the receiver, saving battery life, while simultaneously freeing up uplink bandwidth to improve overall network capacity and support higher data rates (Compl. ¶¶75-76).

Key Claims at a Glance

  • The complaint asserts at least independent claim 1 (Compl. ¶131).
  • The essential elements of method claim 1 include:
    • Receiving, at a receiver, an orthogonal frequency division multiplexing (OFDM) wireless signal.
    • Measuring a channel state information (CSI) vector from the signal.
    • Accessing a plurality of stored "configurations," where each configuration identifies a set of "base sinusoid vectors on constant frequencies."
    • For each configuration, calculating a projection vector via a dot product of the CSI vector and the base sinusoid vectors.
    • For each configuration, calculating a coefficient vector from the projection vector using a stored matrix.
    • For each configuration, calculating a minimum squared error (MSE) fit.
    • Selecting the configuration whose MSE fit residual is below a "predetermined threshold times the minimum fit residual among all configurations" and has the lowest order.
    • Using the selected configuration's coefficient vector as the compressed CSI and transmitting it to a transmitter.

U.S. Patent No. 8,817,693 - *"Intelligent Wi-Fi Packet Relay Protocol"*

  • Patent Identification: U.S. Patent No. 8,817,693, "Intelligent Wi-Fi Packet Relay Protocol," issued August 26, 2014.

The Invention Explained

  • Problem Addressed: Conventional wireless range extenders would "indiscriminate[ly]" rebroadcast every data packet they received from an access point. This "always-on" approach was inefficient, consuming valuable airtime with redundant retransmissions, especially when the destination device was close enough to have received the packet directly, thereby often degrading overall network performance (Compl. ¶¶83-84; ’693 Patent, col. 1:29-38).
  • The Patented Solution: The patent discloses an "intelligent" packet relayer that selectively relays packets only when necessary. The relayer passively monitors or "overhears" the communication channel between an access point and a node, including data packets and their corresponding acknowledgement messages. It uses this information to estimate "packet receive ratios" (PRRs) for the direct link. Based on an algorithm that compares these ratios, the relayer decides whether its intervention would provide a net benefit before relaying a packet, avoiding wasteful retransmissions (’693 Patent, col. 9:55-64; Compl. ¶85).
  • Technical Importance: This intelligent approach improves network efficiency by reducing unnecessary airtime consumption and interference, allowing for range extension without the typical performance degradation associated with simple repeaters (Compl. ¶¶86, 88).

Key Claims at a Glance

  • The complaint asserts at least independent claim 1 (Compl. ¶164).
  • The essential elements of method claim 1 (as corrected) include:
    • Configuring a relayer to monitor a communications channel between an access point and a node and overhear data packets and acknowledgement messages.
    • Estimating, by the relayer, a plurality of packet receive ratios based on the overheard data and acknowledgement messages.
    • Invoking the relayer to relay for the node based on the results of an algorithm that compares two or more of the plurality of packet receive ratios.

III. The Accused Instrumentality

Product Identification

The accused instrumentalities are the nationwide 5G wireless telecommunications networks operated by AT&T, Verizon, and T-Mobile (the "Accused Wireless Networks") (Compl. ¶93). This includes the network infrastructure (e.g., base stations), associated services, and end-user devices (e.g., smartphones and "5G Gateways") that operate according to the 3GPP 5G New Radio (NR) standard, specifically Release 15 and later (Compl. ¶¶98, 106, 114). The complaint provides a screenshot of an AT&T 5G Gateway user guide, which specifies its compliance with "5G NSA Sub 6 GHz" and lists supported "5G n2/n5/n14/n30/n66/n77" frequency bands (Compl. p. 37).

Functionality and Market Context

The Accused Wireless Networks provide high-speed mobile data and internet access to customers across the United States. The complaint alleges that for the purposes of infringement, these networks are "functionally identical" because they all implement the same standardized 5G network functions required by the 3GPP standard (Compl. ¶40). For the ’104 Patent, the relevant functionality is the use of the 5G NR standard's Type II CSI codebook for channel state feedback (Compl. ¶132). For the ’693 Patent, the relevant functionality is the use of Multi-Radio Dual Connectivity (MR-DC) and a split Centralized Unit/Distributed Unit (CU/DU) architecture for packet relay (Compl. ¶164).

IV. Analysis of Infringement Allegations

'104 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
receiving, at a receiver..., an orthogonal frequency division multiplexing (OFDM) wireless signal over a wireless channel... The accused 5G networks utilize OFDM as the signal waveform. The User Equipment (UE) acts as the receiver, receiving OFDM-based signals like Channel State Information Reference Signals (CSI-RS). ¶133 col. 2:12-16
measuring, at a receiver..., a channel state information (CSI) vector of the wireless channel... The UE measures the CSI from received OFDM signals (including CSI-RS) on resource elements across OFDM subcarriers to produce complex-valued channel estimates. ¶134 col. 2:16-25
accessing a plurality of configurations stored at the receiver, wherein each... identifies a set of Pᵤ base sinusoid vectors on constant frequencies... The 5G NR standard's Type II CSI codebook defines multiple configurations based on the parameter L (number of beams). These configurations allegedly identify sets of L base DFT vectors (sinusoid vectors) from a stored oversampled DFT matrix. ¶135 col. 2:26-32
calculating, for each of the plurality of configurations, a dot product of the N by 1 CSI vector and a conjugate of each Pᵤ base sinusoid vector...to generate a Pᵤ by 1 projection vector The precoder codebook is constructed as a linear combination of L selected DFT vectors. The complaint alleges this mathematically corresponds to computing dot products between the channel and conjugate DFT vectors to obtain projection coefficients. A screenshot from 3GPP TS 38.214 shows the construction of vectors based on complex exponentials (sinusoids) (Compl. p. 55). ¶136 col. 2:33-38
calculating, for each of the plurality of configurations, a product of a constant Pᵤ by Pᵤ matrix stored at the receiver and the Pᵤ by 1 projection vector to generate a Pᵤ by 1 coefficient vector The accused systems allegedly apply quantization and normalization transformations to the projection coefficients, which the complaint frames as an operation equivalent to matrix multiplication for coefficient scaling. ¶137 col. 2:39-43
calculating, for each of the plurality of configurations, a minimum squared error (MSE) fit... The accused systems allegedly compute channel estimates at subband locations by a linear combination of DFT basis vectors and computed coefficients, which effectively performs an MSE-minimizing reconstruction at each location. ¶138 col. 2:44-51
selecting configuration u and use its Pᵤ by 1 coefficient vector as the compressed CSI, if the total fit residual... is below a predetermined threshold times the minimum fit residual... The accused systems allegedly use the Type II codebook to select among configurations based on reporting quality. The complaint alleges that the selection of the K(2) strongest coefficients implements a threshold-based selection. ¶139 col. 2:52-59
transmitting the compressed CSI to a transmitter... The UE transmits CSI reports containing Precoder Matrix Indicators (PMI), which include the compressed coefficient information, to the gNodeB (transmitter). ¶140 col. 2:60-61
  • Identified Points of Contention:
    • Scope Questions: A central question may be whether the 5G standard's use of a "Type II Codebook" with "DFT vectors" falls within the scope of the claim term "configurations" identifying "base sinusoid vectors on constant frequencies."
    • Technical Questions: What evidence supports the allegation that the 5G standard's method for selecting a configuration—based on parameters like "number of beams" (L) and "number of non-zero coefficients" (K(2))—is equivalent to the claim's specific mathematical criterion of selecting based on a "predetermined threshold times the minimum fit residual"? The complaint's mapping on this point may be a subject of dispute.

'693 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
configuring a relayer to monitor a communications channel between an access point and a node and overhear data packets sent by the access point to the node and acknowledgement messages sent by the node... In the accused 5G networks, a master node (access point) configures a secondary node (relayer) to handle split bearers via RRC signaling. This configuration allows the secondary node to receive data intended for the 5G Gateway (node) and monitor acknowledgements (e.g., HARQ/RLC reports) sent by the Gateway. A diagram from 3GPP TS 37.340 shows the secondary node addition procedure (Compl. p. 69). ¶¶166, 169 col. 17:41-48
estimating, by the relayer, a plurality of packet receive ratios based on the overheard data packets and the acknowledgement messages... The secondary node allegedly estimates packet receive ratios by calculating lost packet reports and tracking success/failure across a plurality of HARQ processes. The relayer also tracks the "Highest Delivered NR PDCP SN" based on acknowledgements. ¶¶170-171 col. 17:48-50
invoking the relayer to relay for the node based on the results of an algorithm that compares two or more of the plurality of packet receive ratios. The master node allegedly uses a flow control algorithm that compares delivery ratios to invoke the secondary node to relay data. The complaint also alleges that the detection of a "Radio Link Outage" (a high packet loss ratio) invokes the relayer to perform packet retransmission, which is a decision based on comparing packet receive ratios. ¶¶172-173 col. 17:50-52
  • Identified Points of Contention:
    • Scope Questions: Does the term "relayer," as described in a patent focused on Wi-Fi, read on a "secondary node" (SN) within the 5G standard's highly structured Multi-Radio Dual Connectivity (MR-DC) architecture? The defense may argue the patent contemplates a more autonomous device "overhearing" transmissions, whereas an SN is a pre-configured, integral part of the network architecture.
    • Technical Questions: Does the 5G standard's mechanism for activating the secondary node—such as responding to a "Radio Link Outage" or flow control feedback—constitute an "algorithm that compares two or more of the plurality of packet receive ratios" as claimed? This raises the question of whether reacting to a binary outage state is the same as performing a comparative analysis of multiple ratio metrics.

V. Key Claim Terms for Construction

’104 Patent: "a predetermined threshold times the minimum fit residual among all configurations" (Claim 1)

  • The Term: "a predetermined threshold times the minimum fit residual among all configurations"
  • Context and Importance: This term defines the specific mathematical rule for selecting the optimal compression configuration. The infringement case for the ’104 patent may depend on whether the 5G standard's selection process, which the complaint alleges is based on "reporting quality" (Compl. ¶139), can be shown to meet this precise limitation.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The specification describes the overall goal as balancing accuracy and compression (e.g., ’104 Patent, col. 1:44-53). A party might argue the claim language is one exemplary way of expressing any algorithm that selects the lowest-order (most compressed) configuration that still meets a required accuracy level.
    • Evidence for a Narrower Interpretation: The claim language is highly specific and quantitative. The flowchart in FIG. 20 explicitly depicts step 530 with this exact language, suggesting it is a required calculation, not merely a conceptual goal. A party could argue this requires a literal multiplication and comparison, not a more abstract quality assessment.

’693 Patent: "relayer" (Claim 1)

  • The Term: "relayer"
  • Context and Importance: The applicability of the entire claim to the accused 5G networks hinges on whether the 5G standard’s "secondary node" in an MR-DC architecture is a "relayer." Practitioners may focus on this term because the patent's context is Wi-Fi, while the accused system is a cellular 5G network with a different architecture.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent abstract defines a device running the protocol as a "relayer" that "overhears the packet transmissions and retransmits a packet." A party could argue that any network node that performs this core function of retransmitting packets to assist delivery falls under the term's plain meaning.
    • Evidence for a Narrower Interpretation: The detailed description emphasizes that the relayer operates by "overhearing" packets on a channel between two other devices and making decisions based on "local observations" (’693 Patent, col. 9:57-60, col. 10:9-11). A party could argue this implies a degree of autonomy and passive monitoring that is distinct from a 5G "secondary node," which is actively configured and controlled by the master node as part of a split architecture.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges Defendants induce infringement by advertising and providing 5G devices and services that operate according to the infringing 5G standard, and by providing user manuals and marketing materials that encourage customers to use the networks in their normal, infringing manner (Compl. ¶¶148, 180).
  • Willful Infringement: The complaint alleges willfulness based on knowledge of the patents obtained, at the latest, upon service of the complaint (Compl. ¶¶149, 151, 181, 183).

VII. Analyst’s Conclusion: Key Questions for the Case

  • A core issue will be one of technical translation: can the specific method steps recited in the patent claims, which were drafted in the context of Wi-Fi and general MIMO systems, be shown to map directly onto the complex, multi-layered protocols of the 3GPP 5G standard? The dispute will likely focus on whether functionalities like the 5G Type II codebook selection and MR-DC outage handling are technically equivalent to the claimed methods of MSE thresholding and PRR comparison.
  • A key legal question will be one of definitional scope: can terms like "relayer", rooted in the patent's description of a semi-autonomous Wi-Fi device, be construed to cover a "secondary node" that is an integrated and centrally-managed component of a 5G cellular network? The outcome of claim construction for this and other key terms will be critical.
  • A significant procedural question will be the issue of joinder: have the Plaintiffs alleged sufficient facts about the interoperability and "fungible" nature of the Defendants' standardized 5G services to overcome the high bar for joining separate, competing companies in a single patent infringement action pursuant to 35 U.S.C. § 299?