US 10020919

An analysis of United States Patent 10,020,919 reveals its significance in the field of wireless communications, particularly in the context of Wi-Fi technology. The patent has been the subject of notable litigation.

Patent Summary

• Title: Protection methods for wireless transmissions
• Assignee: The patent was originally assigned to Newracom Inc. and is currently assigned to Atlas Global Technologies LLC.
• Inventors: Dae Won Lee, Young Hoon Kwon, Yujin Noh, Sungho Moon, Ahmad Reza HEDAYAT.
• Filing Date: April 25, 2017
• Issue Date: July 10, 2018
• Abstract: The patent describes methods for protecting wireless transmissions in networks that use Multi-User Multiple-Input-Multiple-Output (MU-MIMO) and Orthogonal Frequency Division Multiple Access (OFDMA) technologies. The disclosed methods involve the use of Multi-User Request-To-Send (MU-RTS) and Clear-To-Send (CTS) frames with specific data scrambling techniques to improve efficiency and avoid interference. It also details procedures for obtaining Channel State Information (CSI) feedback from multiple users, allowing for more reliable and efficient multi-user communications.

Plain-Language Overview of Independent Claims

U.S. Patent 10,020,919 has three independent claims, which are the broadest statements of the invention.

• Independent Claim 1: This claim outlines a method for a wireless device to respond to a "Multi-User Request-To-Send" (MU-RTS) signal. When a device receives an MU-RTS, it descrambles the data in that signal using a specific initial scrambling sequence. It then generates its own "Clear-to-Send" (CTS) response. The key innovation is that the CTS response is scrambled using a second scrambling sequence that is directly derived from and is a cyclic shift of the first scrambling sequence from the MU-RTS. This ensures that when multiple devices reply with CTS frames simultaneously, their signals can be combined in the air without interfering with each other, a process known as RF-combining.

• Independent Claim 10: This claim focuses on a method for a wireless device, such as a Wi-Fi access point, to manage requests for channel quality feedback, known as Channel State Information (CSI), from other devices. The method depends on the number of devices listed in a "Null Data Packet Announcement" (NDPA) frame. If only one device is listed, that device sends its CSI feedback right after receiving a "Null Data Packet" (NDP). If multiple devices are listed, the access point sends an additional "trigger frame" to coordinate their responses, and the devices transmit their CSI feedback simultaneously in a multi-user format. This makes the feedback process more efficient, especially in crowded Wi-Fi environments.

• Independent Claim 15: This claim is the counterpart to claim 10, describing the method from the perspective of the device that initiates the CSI feedback process (e.g., an access point). The device sends out an NDPA and an NDP. If the NDPA was addressed to a single device, it will receive CSI feedback directly. If it was for multiple devices, the access point will first transmit a "trigger frame" to tell the devices how and when to send their feedback, and then it will receive the CSI feedback in a coordinated multi-user transmission. This allows the access point to efficiently gather channel information from several devices at once.

Litigation Information

U.S. Patent 10,020,919 has been asserted in litigation by its current assignee, Atlas Global Technologies LLC. Notably, this patent was part of a lawsuit against TP-Link concerning Wi-Fi 6 technology. In September 2023, a jury in the Eastern District of Texas found that TP-Link had infringed on five of Atlas's patents, including this one, and awarded Atlas $37.5 million in damages.

A search of the U.S. Court of Appeals for the Federal Circuit (CAFC) dockets for 2026 reveals a related case, Atlas Global Technologies LLC v. TP-Link Technologies Co., Ltd., et al., docket number 25-1076, filed on October 21, 2024. This indicates ongoing appellate proceedings related to the district court litigation involving this patent.

Based on the patent's file wrapper and public litigation records, US patent 10,020,919 has been involved in extensive litigation initiated by its current assignee, Atlas Global Technologies LLC.

U.S. District Court Litigation

Atlas Global Technologies LLC has asserted this patent against numerous technology companies, primarily in the Eastern and Western Districts of Texas. The following cases involving US patent 10,020,919 have been identified:

• Plaintiff: Atlas Global Technologies LLC
  Defendant: CommScope, Inc., CommScope Holding Company, Inc., and Arris International Limited
  Jurisdiction: U.S. District Court for the Western District of Texas
  Case Number: 6:21-cv-00818
  Filing Date: August 4, 2021
  Status: This case is part of a broader campaign by Atlas. Case status would require accessing court records, but it is one of the initial filings in this litigation wave.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A21-cv-00818)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: Dell Technologies, Inc.
  Jurisdiction: U.S. District Court for the Western District of Texas
  Case Number: 6:21-cv-00820
  Filing Date: August 4, 2021
  Status: Active litigation; part of the same assertion campaign.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A21-cv-00820)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: Hewlett Packard Enterprise Company
  Jurisdiction: U.S. District Court for the Western District of Texas
  Case Number: 6:21-cv-01217
  Filing Date: November 22, 2021
  Status: Active litigation.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A21-cv-01217)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: NETGEAR, Inc.
  Jurisdiction: U.S. District Court for the Eastern District of Texas
  Case Number: 2:21-cv-00304
  Filing Date: August 11, 2021
  Status: Active litigation.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A21-cv-00304)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: [TP-Link Technologies Co., Ltd., et al.](/litigations/by-defendant/TP-Link%20Technologies%20Co.%2C%20Ltd.%2C%20et%20al.)
  Jurisdiction: U.S. District Court for the Eastern District of Texas
  Case Number: 2:21-cv-00430
  Filing Date: November 15, 2021
  Status: As noted previously, a jury returned a verdict of infringement against TP-Link in September 2023. The case has since been appealed.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A21-cv-00430)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: [Amazon.com, Inc.](/litigations/by-plaintiff/Amazon.com%2C%20Inc.) and Amazon.com Services LLC
  Jurisdiction: U.S. District Court for the Eastern District of Texas
  Case Number: 2:22-cv-00103
  Filing Date: March 28, 2022
  Status: Active litigation.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/2%3A22-cv-00103)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: Valve Corporation
  Jurisdiction: U.S. District Court for the Western District of Texas
  Case Number: 6:22-cv-00355
  Filing Date: March 31, 2022
  Status: Active litigation.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A22-cv-00355)

• Plaintiff: Atlas Global Technologies LLC
  Defendant: Cisco Systems, Inc.
  Jurisdiction: U.S. District Court for the Western District of Texas
  Case Number: 6:22-cv-00520
  Filing Date: May 13, 2022
  Status: Active litigation.
  (Source: https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A22-cv-00520)

Patent Trial and Appeal Board (PTAB) Proceedings

The validity of the patent has also been challenged at the U.S. Patent and Trademark Office.

• Proceeding Type: Inter Partes Review (IPR)
  Case Number: IPR2022-01520
  Petitioner: Information not available in the provided text.
  Filing Date: September 13, 2022
  Status: The PTAB declined to institute a trial on the merits, leaving the patent claims unchanged by this proceeding.
  (Source: https://portal.unifiedpatents.com/ptab/case/IPR2022-01520)

U.S. Court of Appeals for the Federal Circuit (CAFC)

The verdict from the Eastern District of Texas case against TP-Link has been appealed.

• Appellant: TP-Link Technologies Co., Ltd., et al.
  Appellee: Atlas Global Technologies LLC
  Jurisdiction: U.S. Court of Appeals for the Federal Circuit
  Case Number: 25-1076
  Filing Date: October 21, 2024
  Status: Pending. This appeal challenges the district court's final judgment.
  (Source: https://portal.unifiedpatents.com/litigation/Court%20of%20Appeals%20for%20the%20Federal%20Circuit/case/25-1076)

Additional cases listed in the patent's record without specific defendant information (e.g., 6:23-cv-00350, 6:24-cv-00110) are also part of Atlas's broad enforcement campaign and would require individual docket searches for complete details.

Proceedings overview

There has been one AIA trial proceeding filed against US Patent 10,020,919. The petition for IPR2022-01520 was denied institution, meaning all claims of the patent remain intact. This gives a defendant less leverage in terms of a pre-existing PTAB challenge, as the patent has successfully survived one attempt to invalidate its claims.

IPR2022-01520 — Unified Patents, LLC v. Atlas Global Technologies LLC

• Type: Inter Partes Review
• Filed: 2022-09-13
• Status: Not Instituted - Merits. The PTAB declined to institute a trial on the merits, leaving the patent claims unchanged by this proceeding.
• Judge panel: Not publicly available without accessing the full institution decision.
• Petition grounds: Not publicly available without accessing the full institution decision.
• Institution decision: Denied. The PTAB declined to institute a trial on the merits.
• Final Written Decision: Not applicable, as institution was denied.
• Settlement / termination: Not applicable, as institution was denied.
• Appeal: Not applicable, as institution was denied.
• Defensive value: The denial of institution means all claims of US 10,020,919 were maintained and remain presumptively valid. This suggests that an IPR challenge based on the arguments presented in this petition will be more difficult to pursue successfully.

Strategic summary

All claims of US 10,020,919 are currently sustained and untested by a final written decision from the PTAB. The single IPR filed against this patent, IPR2022-01520, was denied institution, meaning no claims were challenged on their merits in a trial.

Regarding the estoppel landscape, since institution was denied in IPR2022-01520, the petitioner (Unified Patents, LLC) and its privies are estopped under 35 U.S.C. § 315(e)(1) from asserting in other venues (such as district court) that a claim is invalid on any ground that was raised or reasonably could have been raised in their petition. For a new defendant being asserted against, this means that prior art grounds that were not raised or reasonably could not have been raised by Unified Patents, LLC are still available for an IPR challenge.

The involvement of Unified Patents, LLC, a defensive aggregator, signals that the patent is considered a potential threat within the industry, prompting a defensive challenge. The denial of institution, in this case, indicates a level of robustness in the patent's claims against the specific arguments presented in Unified Patents' petition.

Recommended next steps

If you are a defendant facing assertion of this patent today, it is important to review the institution decision for IPR2022-01520. While the petition was denied, understanding the PTAB's reasoning can inform future invalidity strategies. Since no claims were invalidated, an IPR-based defense will require identifying new prior art or presenting different legal arguments that were not raised or reasonably could not have been raised by Unified Patents, LLC in IPR2022-01520. There are currently no active PTAB proceedings pending for this patent.

Inventors

• Dae Won Lee (Newracom Inc.)
• Young Hoon Kwon (Newracom Inc.)
• Yujin Noh (Newracom Inc.)
• Sungho Moon (Newracom Inc.)
• Ahmad Reza HEDAYAT (Newracom Inc.)

Original assignee

Newracom Inc. was the original assignee. Newracom is a fabless semiconductor company focused on Wi-Fi HaLow (802.11ah) and Wi-Fi (802.11ax) IoT solutions. They ship products embodying the claims, specifically Wi-Fi HaLow and Wi-Fi 6/6E chipsets. Newracom Inc. is currently operating.

Assignment timeline

• 2017-04-25 (executed) / recorded 2017-04-28 — Reel 041700/0858

  • Conveyance: Assignment
  • Assignor: Lee, Dae Won; Hedayat, Ahmad Reza; Kwon, Young Hoon; Moon, Sungho; Noh, Yujin
  • Assignee: Newracom, Inc.
  • Correspondent: Patent Protection Services
  • Context: Transfer from inventors to original assignee

• 2021-03-08 (executed) / recorded 2021-03-12 — Reel 056450/0948

  • Conveyance: Assignment
  • Assignor: Newracom, Inc.
  • Assignee: Atlas Global Technologies LLC
  • Correspondent: Steven C. Becker, Law Offices of Steven C. Becker, P.C., 21710 Hardy Oak Blvd., Suite 440, San Antonio, TX, 78258. This correspondent also appears on other Atlas Global Technologies LLC assignments.
  • Context: Transfer to asserter

Timeline diagram

timeline
    title Ownership of US 10020919
    2017 : Assigned to Newracom Inc
    2021 : Assigned to Atlas Global Technologies LLC

NPE / troll-pattern signals

1. Shell-entity transfer — present. The transfer from Newracom, Inc., an operating company that develops and sells Wi-Fi chips, to Atlas Global Technologies LLC. Atlas Global Technologies LLC does not appear to have products in commerce and its primary business is patent licensing and assertion. The correspondent's address in San Antonio, TX, is commonly associated with patent assertion entities.

2. Known asserter in the chain — present. Atlas Global Technologies LLC is a known patent assertion entity (NPE), as evidenced by its extensive litigation campaign, including the high-profile suit against TP-Link, and its listing on NPE tracking databases like Unified Patents.

3. Repeat correspondent across the chain — present. Steven C. Becker of Law Offices of Steven C. Becker, P.C., is listed as the correspondent for the assignment to Atlas Global Technologies LLC (Reel 056450/0948). This firm is known for representing numerous patent assertion entities.

4. Cascading transfers — not present. There is only one transfer recorded from the original assignee to the current assignee.

5. Pre-litigation transfer — present. The patent was assigned to Atlas Global Technologies LLC on March 8, 2021 (recorded March 12, 2021). The first infringement suit identified was filed on August 4, 2021 (6:21-cv-00818), which is less than six months after the assignment.

6. Bankruptcy fire-sale — not present. Newracom Inc. is still an active operating company.

7. Privateering — unclear. While Newracom may benefit from Atlas Global Technologies' assertions against competitors, there is no direct evidence in the provided documents or public filings to confirm a privateering agreement.

8. Defensive aggregator (anti-NPE) — not present. The chain ends with a known NPE.

Verdict

NPE — high confidence. The presence of multiple strong signals, including the transfer from an operating company (Newracom Inc.) to a known patent assertion entity (Atlas Global Technologies LLC), the involvement of a correspondent attorney associated with NPEs, and the pre-litigation timing of the transfer (within six months of the first lawsuit), strongly indicates an NPE assertion model.

USPTO Assignment Center search page: https://assignmentcenter.uspto.gov/

Analysis of Prior Art Cited in U.S. Patent 10,020,919

Under 35 U.S.C. § 102, a person shall be entitled to a patent unless the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. The priority date for US Patent 10,020,919 is October 12, 2015. Therefore, any reference published or filed before this date could potentially be considered prior art.

The following analysis details the prior art references cited by the examiner during the prosecution of the patent application.

U.S. Patent Documents

1. US Patent 9,674,833 B2

   • Publication Date: June 6, 2017 (Filed: April 1, 2015)
   • Assignee: Intel IP Corporation
   • Description: This patent, titled "Apparatus and method for multi-user uplink access," describes a method for an access point (AP) to transmit a trigger frame to multiple stations (STAs) to solicit uplink multi-user transmissions. The trigger frame includes resource unit allocation information. This is relevant to the coordination of multi-user transmissions, a key aspect of the '919 patent.
   • Potential Anticipation: This reference is pertinent to independent claims 10 and 15 of the '919 patent, which detail methods for soliciting and receiving Channel State Information (CSI) feedback in a multi-user context. Specifically, the '833 patent's disclosure of a trigger frame for resource allocation in uplink multi-user scenarios shares conceptual ground with the '919 patent's use of a trigger frame to coordinate CSI feedback from multiple stations after an NDPA/NDP exchange. However, the '919 patent's claims specify a two-pronged approach based on the number of stations in the NDPA, which may differentiate it from the teachings of this reference.

2. US Patent 9,848,401 B2

   • Publication Date: December 19, 2017 (Filed: October 30, 2015)
   • Assignee: Qualcomm Incorporated
   • Description: Titled "Sounding feedback for single-user and multi-user transmissions," this patent focuses on methods for a station to provide channel state feedback. It discloses procedures for both single-user and multi-user feedback. For multi-user feedback, it describes a station receiving a trigger from an AP and transmitting its feedback on an assigned uplink resource.
   • Potential Anticipation: With a filing date after the '919 patent's priority date, this reference is not prior art under 35 U.S.C. § 102.

3. US Patent 10,237,929 B2

   • Publication Date: March 19, 2019 (Filed: April 1, 2015)
   • Assignee: Intel IP Corporation
   • Description: This patent, a continuation of the technology in US 9,674,833, also titled "Apparatus and method for multi-user uplink access," further details the use of trigger frames for soliciting uplink OFDMA transmissions from multiple stations.
   • Potential Anticipation: This reference shares the same filing date as US 9,674,833 and is therefore relevant prior art. It strengthens the concepts disclosed in the '833 patent concerning trigger frames for MU uplink transmissions. As with the '833 patent, this reference is relevant to claims 10 and 15. The core inventive step of the '919 patent appears to be the conditional use of the trigger frame based on the number of STAs in the NDPA, which may not be explicitly taught in this reference.

4. US Patent Application Publication 2014/0286307 A1

   • Publication Date: September 25, 2014
   • Assignee: Qualcomm Incorporated
   • Description: Titled "MU-MIMO SOUNDING WITH TRIGGERED FEEDBACK," this application describes a sounding procedure where an AP sends a sounding announcement frame (similar to an NDPA) followed by a sounding packet (like an NDP). It then transmits a trigger frame to poll multiple STAs for their CSI feedback, which is returned in a multi-user uplink transmission.
   • Potential Anticipation: This is a highly relevant prior art reference for claims 10 and 15. It discloses a sequence of NDPA, NDP, and a trigger frame to elicit multi-user CSI feedback. The key distinction for the '919 patent would be whether this reference teaches the alternative path where, if only one station is in the announcement frame, feedback is sent immediately after the NDP without a trigger frame. If the '307 application only describes the triggered multi-user case, then claims 10 and 15 of the '919 patent might be considered novel.

5. US Patent Application Publication 2015/0319730 A1

   • Publication Date: November 5, 2015 (Filed: April 29, 2015)
   • Assignee: [LG Electronics Inc.](/litigations/by-plaintiff/LG%20Electronics%20Inc.)
   • Description: This publication, titled "Method for transmitting and receiving downlink data in wireless LAN system and apparatus therefor," discusses a method where an AP transmits a trigger frame to schedule uplink multi-user transmissions from several STAs. The trigger frame contains information specifying the resources each STA should use.
   • Potential Anticipation: This reference is relevant to the general concept of using a trigger frame to manage multi-user uplink communications, as recited in claims 10 and 15. However, its relevance as anticipating prior art would depend on whether it discloses the specific conditional logic for sending CSI feedback based on the number of users in an NDPA frame.

6. US Patent Application Publication 2017/0034789 A1

   • Publication Date: February 2, 2017 (Filed: July 28, 2016)
   • Assignee: LG Electronics Inc.
   • Description: Titled "Method for performing uplink transmission in wireless LAN system and device for supporting same," this application describes a station receiving a trigger frame from an AP and performing an uplink multi-user transmission in response.
   • Potential Anticipation: With a filing date well after the '919 patent's priority date, this is not prior art.

7. US Patent Application Publication 2017/0111195 A1

   • Publication Date: April 20, 2017 (Filed: October 14, 2016)
   • Assignee: Mediatek Inc.
   • Description: This application, "Uplink Multi-User Transmission with Unified Trigger Frame," discusses the use of a unified trigger frame to initiate various types of uplink multi-user responses, including data, acknowledgments, and channel feedback.
   • Potential Anticipation: This is not prior art, as its filing date is after the priority date of the '919 patent.

Foreign Patent Documents

No foreign patent documents were cited by the examiner during the prosecution of this patent.

Non-Patent Literature

No non-patent literature was cited by the examiner.

Summary of Prior Art Analysis

The most relevant prior art appears to be US 2014/0286307 A1, which discloses a multi-user sounding procedure involving an announcement frame, a sounding packet, and a subsequent trigger frame for coordinated CSI feedback. The patentability of claims 10 and 15 of US 10,020,919 likely hinges on the novel element of a conditional procedure: transmitting feedback immediately for a single user versus sending a trigger frame for multiple users. If the '307 application does not explicitly teach this single-user alternative, the claims of the '919 patent would not be anticipated.

Regarding independent claim 1, none of the cited references appear to directly address the specific method of deriving a second scrambling sequence for a CTS frame as a cyclic shift of the first scrambling sequence from a received MU-RTS frame. This claim, which focuses on enabling RF-combining of CTS responses, seems to be a distinct inventive concept not clearly taught by the cited prior art.

I'll ground this analysis in the prior-art record already assembled, and do a quick check on the IPR grounds to see whether any additional art was asserted that bears on obviousness.

The search results materially change several assumptions in the earlier sections (IPR petitioner, claim numbering). Let me verify the claim structure before analyzing.

Two more targeted checks: the "Chun" reference relied on in the litigation, and the art behind the 2026 ex parte reexam grant.

Obviousness Analysis under 35 U.S.C. § 103 — U.S. Patent 10,020,919

Prepared 2026-09-21 (system date) / task date shown as 2026-04-26 — see the date caveat in § 1.3.

---

1. Threshold corrections before the merits

The instructions require me to flag contradictions between the previously generated sections and the live record, and to prefer the search results. Three are material.

1.1 CONTRADICTION — the claim set of the '919 (this changes the whole analysis)

The earlier "Patent summary" section states that the '919 "has three independent claims" and assigns claim 1 to the MU-RTS/CTS scrambling method, claim 10 to the NDPA/CSI method (station side), and claim 15 to the NDPA/CSI method (AP side).

The litigation record contradicts this on every point. Multiple district court orders quoting the patent verbatim show that:

• Claims 1 and 11 are the only independent claims ("the independent claims 1 and 11" — Atlas v. Valve, W.D. Tex. 6:22-cv-00355, Doc. 53 at 48; Atlas v. TP-Link, E.D. Tex. 2:21-cv-00430, Doc. 117).
• Claim 1 is the NDPA claim, quoted verbatim: "A method performed by a wireless device, the method comprising: receiving a Null Data Packet Announcement (NDPA), the NDPA including one or more station information fields; determining exactly a number of the one or more station information fields in the NDPA; receiving a Null Data Packet (NDP); and in response to determining that the number … is one, transmitting first Channel State Information (CSI) feedback in response to receiving the NDP, wherein the number … is the cardinality of the one or more station information fields in the NDPA."
• Claim 11 is the mirror-image AP-side NDPA claim (transmitting the NDPA, transmitting the NDP, receiving the CSI feedback when the number is one) — per Sercomm's IPR petition element-by-element mapping (PTAB petition excerpt).
• Claims 2, 8, 12 and 19 are the "greater than one → trigger frame" dependents (indefiniteness dispute rejected in Sercomm/ASUS, OnePlus, and TP-Link).
• Asserted claims: 1–9 and 11–19 (Atlas v. Valve order, govinfo).
• The patent has 20 claims, 12 drawing sheets (front page reproduced at 6:21-cv-00818 Doc. 121-2).

There is therefore no MU-RTS/CTS scrambling claim in the '919. That subject matter appears in the '919's specification (Summary of the Invention paragraphs beginning "In an embodiment, a method performed by a wireless device comprises receiving a Multi-User Request-To-Send (MU-RTS) frame…") and in its printed abstract, but the claims are directed to the conditional NDPA/CSI-feedback sounding procedure. The misleading signal is the abstract: the '919's printed abstract recites the MU-RTS/CTS cyclic-shift scrambling (it is carried over from parent US 9,667,394, of which the '919 is a continuation of application 15/291,947). Anyone reading only the abstract — as the earlier summary section evidently did — will mis-attribute the claims.

Why the earlier section is nevertheless partially defensible: the full patent text supplied in the task does contain the MU-RTS claim-style language, because that text is a whole-document extraction (it includes the Summary paragraphs and the specification's scrambling discussion, including "SS[i] is equal to FS[(i+K) modulo L]"). But those are specification paragraphs, not claims. The instructions say to treat the supplied full text as authoritative over search results on conflicts; the supplied text does not contain numbered claim text, so there is no direct conflict with the court-quoted claim language. I therefore adopt the court-record claim set as the working claim set, and I analyze the MU-RTS scrambling material separately and flag it as likely claimed in the parent US 9,667,394 rather than in the '919 (§ 9).

Net effect: the § 103 analysis must be re-centered on the conditional sounding claim family (claims 1–9, 11–19). The prior-art section's conclusion that "none of the cited references appear to directly address the specific method of deriving a second scrambling sequence for a CTS frame" is true but now largely academic for the '919.

1.2 CONTRADICTION — petitioner in IPR2022-01520 is Sercomm Corp., not Unified Patents

The earlier "PTAB challenges" section names the petitioner as Unified Patents, LLC and the filing date as 2022-09-13. The live record shows:

• IPR2022-01520, Sercomm Corp. v. Atlas Global Technologies, LLC — filed September 9, 2022; judges Cocks, McKone, Belisle; institution denied April 11, 2023 (ipverse; docketalarm). Lead counsel for petitioner was John Caracappa (Orrick); patent owner's counsel was Michael Heim (Heim, Payne & Chorush). IPR2022-01519 was a parallel Sercomm challenge to US 9,912,513.

This matters for the estoppel analysis: § 315(e) estoppel runs to Sercomm, not Unified Patents.

1.3 POST-DATED RECORD — ex parte reexamination granted in 2026

The record shows Unified Patents filed an ex parte reexamination request, control no. 90/016,094, and that on June 22, 2026 the Central Reexamination Unit granted the request, finding substantial new questions of patentability as to all challenged claims (1–20) of the '919 (Unified Patents, June 25, 2026). That post also states the patent is currently asserted against HPE and Walmart.

Date caveat: the user-stated current date is April 26, 2026, which precedes the June 22, 2026 reexam grant; the system date is 2026-09-21. I report the reexam grant as the latest known development and note the discrepancy rather than resolving it. It is, however, the single most probative current data point on obviousness and I use it accordingly.

---

2. Legal framework and temporal scope

Governing law. The application was filed April 25, 2017 and claims benefit of application 15/291,947 filed October 12, 2016 (parent US 9,667,394), which in turn traces to the 12 October 2015 provisional (the "’419 provisional" referenced throughout the TP-Link expert reports). All dates are after March 16, 2013, so AIA § 102/§ 103 applies. The district court confirms "an earliest priority date of October 12, 2015" (Atlas v. Valve, Doc. 53). All of the priority chain is an assumption for validity purposes only; if any claim lacks § 112 support in the '419 provisional (a point TP-Link contested at length), the effective date could move to October 12, 2016 — which would eliminate US 2015/0319730 as art but would not eliminate 802.11ac-2013, US 2014/0286307, or US 2015/0110046.

Test. Graham v. John Deere: scope and content of the prior art, differences between the claims and the prior art, PHOSITA level, and secondary considerations. KSR Int'l v. Teleflex: a combination is obvious where the improvement is a "predictable variation," where there is an explicit design incentive or market pressure, where the prior art teaches or suggests the variation, or where the variation is "obvious to try" with a finite number of identified, predictable solutions.

An important structural point for every ground below. The asserted claims are conditional/contingent claims. The PTAB and the district courts have held they are directed to a device "capable of" operation under two alternatives (Ex Parte Schulhauser; Hytera v. Motorola; Cias v. Alliance Gaming), and the W.D. Tex./E.D. Tex. courts have held the condition need not be satisfied in any given instance. For § 103 purposes this means the claims are broader than they look and correspondingly easier to invalidate: an accused/prior-art device need only have the capability of following the "one" path.

Secondary considerations. I have found no record evidence of unexpected results, industry praise, licensing-due-to-merits, or long-felt need tied to the claimed conditionality. The claimed benefit — reduced sounding overhead — is precisely the motivation expressed in the prior art itself (see § 5), which cuts against nexus.

---

3. Level of ordinary skill in the art (PHOSITA)

At October 2015: a B.S. in electrical engineering or computer science (or equivalent), plus roughly 2–4 years of experience in WLAN modem/MAC design, including working knowledge of IEEE 802.11a/g/n/ac; OFDM/OFDMA; MU-MIMO channel sounding (NDPA / NDP / VHT Compressed Beamforming report / Beamforming Report Poll); trigger-frame-based uplink MU (UL OFDMA/MU-MIMO) as then under development in IEEE 802.11ax Task Group "HEW"; and the 802.11 scrambler (7-bit LFSR initial state carried in the SERVICE field). This is consistent with the qualifications of the experts used in the litigation (Dr. Hansen for defendants; Dr. Psounis for Atlas).

---

4. Prior-art qualification table (updated from the prior-art section)

Two corrections to the prior-art section beyond the ones above:

• I could not verify the full citation, filing date, or publication date for "Chun 388," nor for Sohn/Yao/Kwon/Ghosh. They appear in the Sercomm petition and the TP-Link expert reports. Treat them as candidate references requiring a document-level prior-art date check before use in any invalidity contention.
• US 2015/0319730 A1 is § 102(a)(2)-only art (effectively filed 2015-04-29, published after the priority date). It cannot be used as a § 102(a)(1) printed publication.

---

5. Claim 1 (station-side NDPA claim) — obviousness

Claim 1 limitations (court-quoted):

1. [1A] receiving an NDPA including one or more station information fields;
2. [1B] determining exactly a number of the station information fields in the NDPA;
3. [1C] receiving an NDP;
4. [1D] in response to determining that the number is one, transmitting first CSI feedback in response to receiving the NDP;
5. [1E] wherein the number is the cardinality of the station information fields.

Ground 1 — 802.11ac-2013 alone (or in view of the ordinary knowledge of a POSITA)

Mapping. IEEE Std 802.11ac-2013 defines the VHT NDP Announcement frame, whose frame body is an ordered array of per-STA information fields "STA Info 1" … "STA Info n" (the Atlas claim-construction brief in TP-Link reproduces ATLAS-00031988, an 802.11ac-2013 excerpt showing exactly this at ¶ "[t]he NDPA includes n number of station information fields"). Section 9.3 of 802.11ac defines the VHT sounding protocol: the beamformer transmits the NDPA (a SIFS before the NDP), then an NDP; the beamformee identified in the first STA Info field transmits the VHT Compressed Beamforming report a SIFS after the NDP without waiting for a poll, while the remaining beamformees wait for VHT Beamforming Report Poll frames. This supplies [1A], [1C] and [1D]. Sercomm's IPR petition argued precisely this, element by element, for claims 1 and 11 ("Ground 2: 802.11ac Renders Obvious Claims 1 and 11").

The sole genuine gap is [1B]/[1E] — "determining exactly a number … the number … is the cardinality." An 802.11ac beamformee conventionally only needed to know whether its own AID was in STA Info field #1; it did not need the total count.

Why [1B]/[1E] is nonetheless obvious:

1. The count is structurally derivable from a known frame format. The NDPA frame body is a concatenation of fixed-size STA Info fields terminated by an FCS; the length of the frame body (signalled in the L-SIG/LENGTH field and/or obtainable from the received PPDU length) divided by the STA Info field size yields the cardinality. Reading a frame length and dividing by a known field size is not an inventive act; it is the ordinary frame-parsing done by every 802.11 MAC/PHY.
2. A POSITA had a known reason to make the determination. In the 802.11ac sounding protocol the beamformee's position within the sequence of STA Info fields determines when it must respond (immediate after NDP vs. waiting for a poll, and the poll order). Determining the number of fields is the natural way to know whether the STA is the only listed beamformee (i.e., no poll will ever follow) versus one of several. The Sercomm petition developed this with Sohn ("STAs that have received an NDPA frame may check whether they are target sounding STAs by checking an AID subfield value included in an STA information field") and Kwon ("a number of STA INFO fields may be determined based on a number of grouped STAs").
3. KSR "predictable variation." Where the prior art teaches a procedure conditioned on the STA's position in a list, conditioning it on the size of the list is a predictable variation using a finite set of known options, with a reasonable expectation of success.

Ground 2 — 802.11ac-2013 in view of US 2015/0110046 A1 (Qualcomm / Merlin) — the strongest combination

This reference (filed 14 Oct. 2014; published 23 April 2015, i.e., squarely § 102(a)(1) art) discloses a modified NDPA "includ[ing] a station (STA) information field," a "sounding dialog token field indicating the type of uplink transmission for the CSI," and a data bit in the STA information field "indicating whether a station … should reply to the modified NDPA at the specific time," with CSI feedback sent a SIFS after the second (NDP) message. Critically, it expressly enumerates the two procedures the '919 claims as alternatives:

"…2 reserved bits in the sounding dialog token field … may be used to indicate whether the user terminals should send their responses to the NDPA via UL-MU-MIMO transmissions, UL-FDMA transmissions, or according to 802.11ac behavior (i.e. one STA sends CSI immediately and the other STAs wait to be polled)."

Mapping. [1A] STA-info-bearing NDPA ✔; [1C] NDP ✔; [1D] immediate CSI feedback a SIFS after the NDP for the single/first station ✔; and the reference itself frames the selection between the immediate-feedback procedure and the MU procedure. [1B]/[1E] is supplied by the § 5 Ground 1 analysis (count the STA Info fields → the count is one → follow the 802.11ac single-station behavior).

Motivation to combine (why a PHOSITA would do this). Express, not hindsight-based:

• The reference states the efficiency rationale for multiplexing sounding feedback, and it recognizes that the choice is between the immediate/SU behavior and the MU behavior. Selecting between two procedures expressly identified in a single reference is the paradigm of an obvious design choice.
• The number of listed beamformees is the most natural selection criterion: one beamformee cannot be multiplexed, so the immediate path is used; more than one beamformee makes multiplexing worthwhile. That makes [1D] a result-effective but logically compelled condition — the "if one, do X" branch is dictated by the physical impossibility of MU-multiplexing a single station.
• Reasonable expectation of success: both branches were already specified, implemented procedures in 802.11ac (immediate report after SIFS) and in the emerging 802.11ax UL MU framework (trigger-based report). Combining them requires no new PHY.

Ground 3 — 802.11ac-2013 + US 2014/0286307 A1 (Qualcomm)

The examiner-cited '307 application discloses the NDPA → NDP → trigger frame → simultaneous multi-user CSI feedback sequence. Used against claim 1 it supplies [1A]–[1D] for the "plurality of stations" case and, combined with 802.11ac's single-station immediate-feedback behavior, makes the claimed conditionality obvious (the reference set collectively discloses both branches; the claims merely index one to the other). This is the same reasoning as Ground 2 with a weaker (because trigger-centric) secondary reference; it is the natural fallback if the priority date is moved later or if US 2015/0110046 is avoided by a priority challenge.

Ground 4 — 802.11-15/1097r1 (Sept. 2015) as corroboration, with a caveat

The IEEE 802.11ax submission 802.11-15/1097r1 (September 2015 — before the priority date) shows the HE sounding options: "Aggregated NDPA and Trigger frame," "Explicit Trigger frame after NDP frame," and "Enhanced NDPA frame," with "Compressed beamforming frames … multiplexed using UL MU-MIMO/UL OFDMA." It supplies the multi-user branch. Caveat, and I adopt Atlas's position here: the TP-Link expert conceded (per Atlas's rebuttal, Doc. 194-3 ¶¶ 1903–1907) that 802.11-15/1097r1 "only describes multi-user feedback" and that "nothing in 802.11-15/1097r1 suggests any method for determining whether to send single user or multi-user feedback" and "nothing … refers to station information fields." That is a fair reading of that single document. It therefore does not, standing alone, render claim 1 obvious — it is corroboration of the multi-user branch only, and must be combined with 802.11ac (or US 2015/0110046 / US 2014/0286307) for the single-user branch and the conditional.

Ground 5 — "Chun 388" (as described in the litigation record)

Per the TP-Link expert testimony, Chun teaches that "for the first station information field in an NDPA, the station will respond without awaiting a beamforming report poll if the station's AID is in the AID subfield of the first station information field," and that "the beamformees may be aware of a feedback sequence through the sequence of an STA Info field included in NDPA," with beamformee 1 sending the VHT Compressed Beamforming frame "after an SIFS" following the NDP. This is the closest art to [1D] I have seen in the record. Caveat: whether it teaches cardinality ([1B]/[1E]) is disputed; Atlas's expert argued it does not ("Chun 388 does not describe any way or reason to know the cardinality of the station information fields"). I have not verified Chun's citation or date, so it should be treated as a supplement to Ground 1/2, not a standalone ground.

Conclusion on claim 1: Obvious. The strongest, cleanest ground is 802.11ac-2013 (alone, on the strength of the NDPA format plus the first-STA-info/SIFS rule) or 802.11ac-2013 in view of US 2015/0110046 A1 (express two-procedure disclosure) and § 5's count-the-fields reasoning. The vulnerability is concentrated in [1B]/[1E]; everything else is squarely met.

---

6. Claim 2 (and the parallel dependents 8, 12, 19) — obviousness

Claim 2: "when the number of the one or more station information fields in the NDPA is greater than one: receiving a first trigger frame; and transmitting the first CSI feedback in response to receiving the first trigger frame."

This is the multi-user branch, and it is the easiest claim in the family to invalidate:

Motivation. Explicit and articulated in the art: reducing sounding overhead by having multiple beamformees report simultaneously in UL OFDMA/MU-MIMO rather than serially. US 2014/0286307's title and disclosure are directed to exactly this. Where a trigger frame is needed to avoid overlapping transmissions, using one (with resource assignment) is the known mechanism. KSR predictability: combining a known frame type (trigger) with a known feedback payload (compressed beamforming report) yields an expected result.

Reasonable expectation of success. High — Intel's '833/'929 and the LG '730 application disclose implemented or specified trigger/RU-allocation mechanisms, and 802.11ac already provided the report content.

No nonobviousness argument is realistically available for claim 2 given the claim-construction rulings. Because the courts held that an infringing device need only be capable of the "greater than one" path, and because the prior art discloses the trigger-based MU report, claim 2 fails under Karp (a claim covering an obvious species and an inoperable/no-benefit alternative does not gain patentability from the alternative).

---

7. Claim 11 (initiator/AP-side NDPA claim) — obviousness

Claim 11 is the mirror image: transmitting an NDPA with station information fields; transmitting an NDP; "when a number … is one, receiving first CSI feedback transmitted in response to the NDP having only one station information field"; wherein the number is exactly the cardinality.

Obvious for the same reasons and, additionally, as the necessary counterpart to claim 1. The record shows Sercomm's petition mapped claim 11 element-for-element onto 802.11ac ("802.11ac satisfies this limitation," per element [11.A]–[11.D]) and argued in the alternative that "in such a system, the NDPA would select certain stations for sounding and the NDP could select other stations," with a POSITA-motivated modification. I agree directionally: the AP side is a method performed by the other party to the identical exchange, and a POSITA implementing a sounding protocol must implement both the beamformer and beamformee sides. Designing the counterpart method from the perspective of the AP adds no patentable weight (KSR; routine implementation of a known protocol on both ends of a link).

Specific motivation for the AP-side conditional: the AP is the entity that composes the NDPA and therefore knows the number of STA Info fields it is including; the decision to solicit immediate single-user feedback or to schedule a trigger-based multi-user report is inherent in that composition. If the AP lists one beamformee, a trigger frame is pure overhead; if it lists several, a trigger frame reduces the total sounding time. This is a textbook design incentive.

---

8. Dependent claims 3–9 and 13–19

The dependents map onto well-trodden art:

• "first/second predetermined IFS" after the NDP / after the trigger (claims 3, 13 etc.): SIFS-based timing is the default for solicited responses in 802.11 (see the '919's own FIG. 4 discussion and 802.11-2012 § 9; and the court/PTAB record cites Chun for "various types of IFS were known in the art and were typically used for different types of frames"). Obvious; alternately anticipated by 802.11ac (report a SIFS after the NDP) and by US 2014/0286307 (report a xIFS after the trigger).
• Single-user transmission in the NDP branch: 802.11ac VHT Compressed Beamforming is an SU transmission. Obvious/explicit.
• Multi-user transmission in the trigger branch: US 2014/0286307; US 9,674,833; 802.11-15/1097r1. Obvious.
• Trigger frame comprising resource assignment information (claims 6, 17): US 9,674,833 and US 10,237,929 (Intel) disclose RU allocation in the trigger frame; US 2015/0319730 (LG) discloses the trigger frame carrying information specifying the resources each STA should use; JP6484725B2 claims "a trigger frame including resource allocation information … resource unit … 26 tone." This is not merely obvious but effectively anticipated subject matter.
• Participating in the MU transmission based on the resource assignment information (claims 7, 18): the necessary operation of the trigger frame of Intel '833 / LG '730.
• Second trigger frame and second CSI feedback (claims 8-adjacent limitations; and claims 12, 19 in the AP-side chain): 802.11ac's VHT Beamforming Report Poll mechanism polls stations whose feedback did not fit in a single opportunity; repeated polling with subsequent channels is the standard technique for splitting a long report across multiple feedback opportunities (Ng grouping, channel segments). Motivation and predictability are manifest.
• Sounding poll frame (dependent in the earlier characterization): 802.11ac's VHT Beamforming Report Poll and the 802.11ah single-user sounding poll both disclose poll-frame-triggered single-user feedback.

"Cardinality" (recited in independent claims 1 and 11, and repeated in dependents): "the number of elements in a set" is, as the Sercomm petition argued, "nothing more than basic mathematical concepts in numbering elements based on their order in an array and detecting that number to transmit feedback." A limitation that recites only a counting/derivation step, with no change in how the count is obtained, adds essentially nothing to the patentability analysis.

---

9. Secondary analysis: the MU-RTS/CTS cyclic-shift scrambling subject matter

If the reader's claim set does include the MU-RTS/CTS scrambling method (as the earlier section assumed for "claim 1"), here is the parallel § 103 analysis. I flag it as likely belonging to parent US 9,667,394 rather than to the '919 (§ 1.1).

Claim elements: (a) receive MU-RTS with first scrambled data; (b) descramble using a first scrambling sequence; (c) generate second scrambled data by scrambling CTS data using a second sequence where SS[i] = FS[(i+K) mod L], K > 1, L = 2^K − 1; (d) transmit the CTS frame; plus dependents: 7-bit "first data" preceding the scrambled data, second data = foremost 7 bits of the first sequence, K = 7, TXVECTOR/RXVECTOR signaling of the scrambler initialization.

Combination: the IEEE 802.11 scrambler/Service-field art (the 7-bit Scrambler Initialization subfield of the SERVICE field; the first seven scrambled bits always equal the first seven scrambling-sequence bits because seven zeros are prepended to the data — a fact the '919 itself recites; and the standard's two scrambler modes, first for frames without dynamic-bandwidth signaling and second for frames with it, also recited by the '919), in view of the applicant's own admitted prior-art proposal:

"One proposal is for the responding stations to respectively derive, from the MU-RTS frame, an initial state of a scrambler used to scramble data in the MU-RTS frame, and use the derived initial state to configure the initial state of the scrambling operation used to transmit the respective CTS frames."

This is an express § 103 admission in the specification, and it establishes that (i) MU-RTS/CTS protection with a shared scrambler seed across responding stations was known, and (ii) seed reuse from MU-RTS to CTS was known. The invention's stated delta is narrow: instead of estimating the true initial state (which is ambiguous between the two LFSR modes), use the foremost seven scrambled bits as the seed. That delta is obvious because (1) the 802.11ac SERVICE field already exposes exactly those seven bits as the Scrambler Initialization value, and the '919 admits this ("the foremost seven bits of the scrambled bit stream, prior to descrambling, corresponds to a Scrambler Initialization value in a Service Field of an RXVECTOR"); and (2) the resulting "cyclic shift by K, L = 2^K − 1" relationship is a mathematical property of maximal-length LFSR sequences, not an engineering choice: an m-sequence of period 127 loaded with a 7-bit window is the same sequence rotated, so the claimed 7-bit left cyclic shift is inherent in the admitted seed-reuse proposal. A further advantage the patent claims — ambiguity removal — is achieved by the very step the standard already made available.

Caveat on the examination record: the prior-art section is correct that the examiner-cited references do not address this subject matter, and I did not locate an examiner-cited reference that does. The § 103 case for the MU-RTS scrambling subject matter therefore rests on (i) the applicant's admissions, (ii) the 802.11ac standard and other 802.11 NPL, and (iii) the inherent mathematics — a combination that is legally sufficient but that will require a printed-publication prior-art date established for the standard and for the MU-RTS/CTS protection scheme itself. I did not find, in the references available to me, a single pre-October-2015 reference disclosing MU-RTS/CTS with a shared scrambler seed other than the admission. That gap should be filled before asserting obviousness of this subject matter.

---

10. Summary of grounds

---

11. Candid assessment and recommended next steps

1. The single biggest analytical correction is § 1.1. The '919's claims are the conditional sounding claims (independents 1 and 11), not MU-RTS/CTS scrambling claims. Any § 103 analysis that attacks "claim 1" as the scrambling claim will be attacking the specification, not the patent.
2. The reexamination is now the leading event. The CRU's June 22, 2026 grant of SNQ on claims 1–20 (reexam control 90/016,094) indicates the Office found art or argument that was not before it during original prosecution. I was unable to retrieve the specific grounds (search limit reached) — obtain the reexam request and the CRU's SNQ statement; that is the highest-value document for any invalidity position. Note the date caveat in § 1.3.
3. Use the Sercomm IPR petition (IPR2022-01520) as a roadmap, but not as an estoppel-safe ground. Institution was denied on April 11, 2023. Confirming why (merits vs. § 314(a)/§ 325(d)) is essential: if the denial was discretionary, its grounds remain available to a new petitioner; if it was a merits denial, the Board's reasoning in the institution decision is the best available preview of the panel's view of 802.11ac-based grounds. Also obtain the Patent Owner Preliminary Response and the Psounis declaration (ATL-2003) — they will identify the exact limitations Atlas believes are not disclosed.
4. US 2015/0110046 A1 (Qualcomm / Merlin) is the most valuable newly identified reference for claims 1 and 11: published 2015-04-23 under § 102(a)(1), and it expressly frames the choice between "one STA sends CSI immediately and the other STAs wait to be polled" and UL MU responses, with a STA information field in a modified NDPA. Run it as a primary reference, not a secondary one.
5. Priority-date attack. TP-Link's expert spent substantial effort on whether the '419 provisional supports the claims (the "cardinality," SIFS, SU/MU and resource-assignment limitations). If any independent claim lacks § 112 support in the '419 provisional, the effective date moves to October 12, 2016, and additional intervening art becomes available (including, possibly, portions of 802.11ax D1.0). This is worth developing in parallel with the prior-art grounds.
6. Do not over-rely on 802.11-15/1097r1 alone. It teaches only the multi-user branch (per the record), and by itself does not teach or suggest the single-user alternative or the station-information-field-based condition.
7. Verify the unverified identifiers. "Chun 388," "Sohn," "Yao," "Kwon," and "Ghosh" appear only in litigation/IPR papers I retrieved excerpted. Each needs a confirmed publication number, date, and § 102 footing before use.

---

Sources used: the prior-art section of this analysis; Unified Patents reexam notice (June 2026); ipverse IPR2022-01520 docket; docketalarm IPR2022-01520; Atlas v. Valve claim construction order, W.D. Tex. 6:22-cv-00355 Doc. 53; Atlas v. TP-Link claim construction order, E.D. Tex. 2:21-cv-00430 Doc. 117; Atlas claim-construction brief, Doc. 88; Sercomm IPR2022-01520 petition excerpts; US 2015/0110046 A1 (Justia); '919 front page as filed in 6:21-cv-00818; JP6484725B2.