US 8098695

I'll search for current information on this patent, including any CAFC 2026 activity.

Let me run more targeted searches on litigation and docket history for this specific patent number.

Let me do one final check for any Federal Circuit 2026 appeal specifically involving this patent.

Search Results Summary — US 8,098,695

Important caveat up front: I do not have direct access to USPTO Patent Center/PAIR or PACER, so I relied on the full patent text you supplied plus live web results. I found no Federal Circuit appeal docketed in 2026 that specifically identifies US 8,098,695 as an asserted patent. The 2026 Federal Circuit Maxell dockets that surfaced (e.g., 26-1111, 26-1320, 26-1330, 2026-1625) involve Samsung/PTAB matters and reference other patents; I could not confirm '695 is among them. Treat any "CAFC 2026 activity" for this patent as unverified.

Bibliographic Data (from the authoritative full text)

Abstract (verbatim substance)

A multiplexed audio data decoder apparatus is provided in which integration of an audio decoder is easy and which has high flexibility when the number of formats to be processed increases or when specification changes. An external ROM 60 accumulates a plurality of decoding program codes corresponding to respective plural compression/encoding methods. A controller 50 transfers the decoding program code corresponding to the changed compression/encoding method from the external ROM 60 to an internal RAM 25. A DSP 22 starts decoding using the code transmitted into internal RAM 25.

Independent Claims — Plain-Language Overview

The patent has 6 claims total, with two independent claims (1 and 4) and four dependent claims.

Claim 1 — audio-only decoder (independent).
A multiplexed audio data decoder that receives a multiplexed packet stream (packets carrying compressed/encoded audio sequences) and selectively decodes a user-designated audio sequence. It requires:

• (a) a demultiplexer that pulls out the user-designated audio sequence using each packet's property/attribute information, and extracts the compression/encoding method from the audio sequence's header;
• (b) a memory holding decoding program codes (including decoding algorithms) corresponding to the compression/encoding method;
• (c) a digital signal processor that sequentially decodes the compressed audio using those program codes; and
• (d) a controller that receives the compression/encoding method from the demultiplexer, detects whether that method has changed to another, and — if it has — downloads the decoding program code for the new method into the memory from outside that memory.
  The DSP then begins decoding using the downloaded code.

Claim 4 — audio + video decoder (independent).
Same architecture and same four core elements as claim 1, but applied to a stream that multiplexes a video data sequence plus its accompanying audio data sequences, where a user designates a set of video + audio to be decoded. The additional element is a video decoder that decodes the compressed/encoded video data sequence. The demultiplexer, memory, DSP, and controller operate as in claim 1.

Dependent Claims

• Claim 2 (dep. on 1): After downloading the code for the changed method, the DSP performs a process to determine the presence of download errors.
• Claim 3 (dep. on 2): In that error determination, a specific audio data sequence is decoded using the decoding program code and compared against an expected value.
• Claim 5 (dep. on 4): Same as claim 2, applied to the audio+video embodiment.
• Claim 6 (dep. on 5): Same as claim 3, applied to the audio+video embodiment.

Notable drafting point: The granted claim language says "downloading … to said memory from outside of said memory," whereas the specification's summary/embodiments describe "transferring" decoding process codes from an external read-only memory into an internal RAM. The claim scope is therefore framed more broadly around "downloading from outside the memory" than the ROM-to-RAM narrative in the description. Anyone relying on the spec's "read-only memory" framing should confirm the prosecution history, which I have not independently reviewed.

Litigation / Docket Context (grounded in search results)

• Maxell v. ZTE Corp., No. 5:16-cv-00179 (E.D. Tex.) — Filed Nov. 2016. The '695 patent was one of eight asserted. A jury on June 29, 2018 found willful infringement of seven patents and awarded $43.3M; '695 was among the patents asserted. Reported per WIPR (https://www.ciplawyer.com/html_e/Patent1/20180704/[139509](/patent/139509).shtml) and Law360/Mayer Brown (https://www.mayerbrown.com/-/media/files/news/2019/02/maxell-zte-reach-settlement-after-43m-patent-verdi.pdf). The case settled confidentially in Feb. 2019.
• IPR2018-00240 (ZTE Corporation / ZTE (USA) Inc., petitioner) — listed on Google Patents as the PTAB proceeding associated with this patent, recorded as not instituted on the merits. (The docket declaration by counsel Jamie B. Beaber confirms pro hac vice applications across IPR2018-00235/-236/-238/-239/-240.)
• Maxell Ltd. v. Huawei Device U.S. Inc., 297 F. Supp. 3d 668 (E.D. Tex. 2018) — Claim construction addressing the '695 patent: preambles held limiting; "a demultiplexer…," "a memory," and "begins decoding processing" given plain and ordinary meaning (CourtListener: https://www.courtlistener.com/opinion/[7329083](/patent/7329083)/maxell-ltd-v-huawei-device-us-inc/).
• Other docket references listed on the Google Patents page: N.D. Ill. 1:22-cv-00256; W.D. Tex. 6:21-cv-01169 and 6:21-cv-00158; E.D. Tex. 5:16-cv-00179.

⚠️ Uncertainty flag: One Chinese-language trial summary table appeared to attribute an infringement finding of "claims 1 and 8" to an audio-decoding patent in the ZTE case. US 8,098,695 as issued has only 6 claims, so a "claim 8" finding cannot apply to '695 and likely refers to the sibling patent US 6,816,491 (a different member of the same family). I flag this rather than auto-correcting it.

Bottom Line

US 8,098,695 is the third-generation continuation of a 1998-priority Hitachi application claiming a DSP-based audio decoder that dynamically downloads the appropriate decoding program code into working memory when the incoming stream's compression/encoding format changes, optionally with a decoded-test-data error check, and optionally alongside a video decoder (independent claim 4). It is expired (maintenance-fee lapse). It saw notable enforcement in the 2016–2019 Maxell v. ZTE campaign and a 2018 claim-construction ruling in Maxell v. Huawei. No 2026 CAFC docket specifically tied to this patent was found, so I cannot confirm any 2026 appellate activity for '695.

I'll research litigation involving US patent 8098695 using multiple searches.

Let me gather more specifics on the known cases and the IPR.

Known Litigation Involving U.S. Patent No. 8,098,695 ("'695 Patent")

Important framing note: I searched specifically for U.S. Patent No. 8,098,695 and excluded results for closely related family members that are commonly confused with it — notably U.S. Patent Nos. 6,816,491 ("'491", the grandparent) and 7,447,241 ("'241", the parent), which share the same title ("Multiplexed audio data decoding apparatus and receiver apparatus"), the same inventors, and the same specification. Where an action asserted both the '491 and '695 patents together, I say so explicitly.

I also note the '695 Patent is expired — Google Patents records it as "Expired – Fee Related," lapsed for failure to pay maintenance fees effective 2020-01-17 (patent discontinuation notice posted 2020-02-24). That affects the current status of any surviving case but not the historical record below.

Confirmed litigation

1. Maxell, Ltd. v. ZTE Corporation and ZTE USA Inc.

• Court / Jurisdiction: U.S. District Court for the Eastern District of Texas, Texarkana Division
• Case No.: 5:16-cv-00179-RWS (consolidated for pretrial with Huawei action 5:16-cv-00178)
• Presiding Judge: Robert W. Schroeder III
• Filed: November 17–18, 2016 (Unified Patents lists filing date 2016-11-17; contemporaneous reporting cites a November 18, 2016 complaint)
• Plaintiff: Maxell, Ltd. (successor to Hitachi Maxell / Hitachi)
• Defendants: ZTE Corporation (China); ZTE USA Inc. (Richardson, TX)
• Role of the '695 Patent: The '695 Patent was one of eight patents-in-suit (5,396,443; 6,748,317; 8,339,493; 8,736,729; 6,408,193; 6,329,794; 6,816,491; and 8,098,695). It was asserted against ZTE USA only (the Huawei action asserted a different set). Note that the '695 and its sibling '491 were both asserted here, so alerts and reports frequently conflate them.
• Outcome: A jury trial began June 18, 2018. On June 29, 2018, the jury found ZTE willfully infringed seven asserted patents (including the '695) and awarded $43.3 million; the '443 patent fell out. Post-trial motions (ZTE's Rule 50/59 motions and Maxell's renewed JMOL) were pending when the parties filed a joint motion to stay and notice of settlement; the case was terminated March 11, 2019 under a confidential settlement. This is the "won a $43.3 million jury verdict for willful infringement" matter later cited by Maxell in its Samsung PTAB papers.
• Source link (Google Patents family litigation listing): https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/5%3A16-cv-00179

2. Maxell Ltd. v. ZTE Corporation (bifurcated ZTE Corp. case)

• Court / Jurisdiction: U.S. District Court for the Eastern District of Texas, Texarkana Division
• Case No.: 5:18-cv-00080
• Filed: June 15, 2018
• Plaintiff / Defendant: Maxell Ltd. v. ZTE Corporation
• Context: This is the bifurcated portion of the original action directed at the foreign parent, ZTE Corp., which had not accepted service. The '695 Patent was part of this same patent family grouping.
• Outcome: An identical notice of settlement was filed in this bifurcated case at the same time as in Case No. 5:16-cv-00179 (joint motion dated February 4, 2019). Resolved by the same confidential settlement.
• Source link: https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/5%3A16-cv-00179 (related docket); https://dockets.justia.com/docket/texas/txedce/5:2018cv00080/[182962](/patent/182962)

3. Maxell, Ltd. v. Lenovo Group Ltd., Lenovo (United States) Inc., and Motorola Mobility LLC

• Court / Jurisdiction: U.S. District Court for the Western District of Texas, Waco Division
• Case No.: 6:21-cv-01169
• Presiding Judge: Alan D. Albright
• Filed: November 12, 2021
• Plaintiff: Maxell, Ltd.
• Defendants: Lenovo Group Ltd.; Lenovo (United States) Inc.; Motorola Mobility LLC
• Role of the '695 Patent: One of eight patents asserted (the '695; 7,577,417; 7,551,209; 6,928,292; 9,420,212; 7,952,645; 8,059,177; and 7,072,673). Maxell specifically alleged infringement of claims 1 and 4 of the '695 Patent (literally and/or under the doctrine of equivalents) by Lenovo/Motorola smartphones.
• Outcome / Current Status: Terminated/closed in mid‑2023 — Docket Alarm shows termination June 29, 2023 (case flags: CLOSED, PATENT, STAYED); the Ex Parte litigation analytics record shows status Closed (06/15/23). This is consistent with a settlement/license resolving the matter. A separate Section 337 ITC complaint by Maxell against a smart-TV respondent invoked a Lenovo patent license agreement as background, indicating Maxell and Lenovo have since reached licensing arrangements.
• Source links: https://www.docketalarm.com/cases/Texas_Western_District_Court/6--21-cv-01169/Maxell_Ltd._v._LENOVO_GROUP_LTD._et_al/ ; https://unicourt.com/case/pc-db5-maxell-ltd-v-lenovo-group-ltd-et-al-[1076976](/patent/1076976)

Related PTAB proceeding (not a district-court case, but part of the '695 enforcement record)

4. ZTE Corporation and ZTE (USA) Inc. v. Maxell, Ltd. — IPR2018-00240

• Forum: USPTO Patent Trial and Appeal Board
• Petitioner: ZTE Corporation and ZTE (USA) Inc.
• Patent challenged: U.S. Patent No. 8,098,695
• Filed / effective date: November 22, 2017 (Google Patents records the IPR filing event January 2, 2018)
• Outcome: Not Instituted – Merits. ZTE filed parallel IPR petitions against each of the seven patents on which it had been sued; each asserted patent, including the '695, survived the challenges (as Maxell itself later asserted in correspondence and PTAB briefing).
• Source link: https://portal.unifiedpatents.com/ptab/case/IPR2018-00240

Items I could NOT confirm (flagged rather than guessed)

• The Google Patents "family has litigation" panel for the '695 family also lists two additional U.S. district-court matters:

  • N.D. Illinois, case 1:22-cv-00256 — https://portal.unifiedpatents.com/litigation/Illinois%20Northern%20District%20Court/case/1%3A22-cv-00256
  • W.D. Texas, case 6:21-cv-00158 — https://portal.unifiedpatents.com/litigation/Texas%20Western%20District%20Court/case/6%3A21-cv-00158

  These are reported against the same patent family on Google Patents, but I could not verify from available search results (a) the party names, or (b) whether the '695 patent specifically (as opposed to a different family member such as the '491 or '241) was asserted in them. I am therefore not attributing them to the '695 without confirming the asserted patent number. If your question requires precision on these two, they should be checked directly on the Unified Patents / PACER dockets before being listed as '695 litigation.

• I found no litigation asserting the '695 Patent against Apple, Samsung, VIZIO, TCL, or Coretronic/Optoma. Those Maxell campaigns involve different patents (e.g., the Apple and VIZIO suits assert camera/display/streaming patents, not the '695).

Summary table

Bottom line: The only district-court litigation I can confirm asserted U.S. Patent No. 8,098,695 is (i) Maxell v. ZTE (E.D. Tex. 5:16-cv-00179, plus bifurcated 5:18-cv-00080) and (ii) Maxell v. Lenovo/Motorola (W.D. Tex. 6:21-cv-01169). Both ended in settlement/license after the '695 survived a ZTE IPR challenge (IPR2018-00240, not instituted). The two additional family dockets listed on Google Patents (1:22-cv-00256 and 6:21-cv-00158) should be verified via PACER/Unified Patents before being counted as '695 cases, since I could not confirm the asserted patent number for them.

I'll investigate the PTAB record for US 8098695. The structured ODP block says no AIA trials, but the patent's own prosecution history page lists an IPR — let me reconcile that with web sources.

Let me dig into IPR2018-00240 specifically and check for any other proceedings.

I've confirmed ZTE filed IPRs against this patent family. Let me pull the specific IPR2018-00240 record and decision.

PTAB Proceedings on US 8,098,695 ("Multiplexed audio data decoding apparatus and receiver apparatus")

Reconciliation note up front. The supplied "PTAB proceedings on file" block says the USPTO ODP API returns no AIA trial proceedings for this patent. That is contradicted by the patent's own Google Patents prosecution/litigation record, which states: "PTAB case IPR2018-00240 filed (Not Instituted - Merits)" with a legal-event entry dated 2018-01-02 reading "TRIAL NO: IPR2018-00240 … Opponent name: ZTE CORPORATION AND ZTE (USA) INC … Effective date: 20171122." Independent press coverage confirms ZTE filed seven IPRs against Maxell's asserted patents and the Board denied all seven. The ODP ingest therefore appears stale or incomplete for this proceeding. I rely on the patent-page record plus public reporting and flag what could not be verified.

Second caveat: on the patent page the string "Unified Patents" attached to IPR2018-00240 is the data-source attribution for the litigation feed, not the petitioner. The petitioner of record is ZTE Corporation and ZTE (USA) Inc. Do not treat this as a Unified Patents (defensive aggregator) IPR.

Proceedings overview

There is one AIA trial proceeding on US 8,098,695 that I can identify — IPR2018-00240 (ZTE Corporation & ZTE (USA) Inc. v. Maxell, Ltd.), filed 2017-11-22 and denied institution on the merits ("Not Instituted - Merits"). Breakdown: 0 active, 0 claims invalidated, 0 claims sustained (no FWD ever issued), 1 institution-denied, 0 settled at the PTAB. The bottom line for a defendant today is that the patent is un-hardened but also un-canceled: no claim of the '695 patent has ever been adjudicated by the PTAB, so all six claims remain the validly-issued claims Maxell can point to — but the patent expired more than five years ago (term of the continuation family ending ~2021-01-09, with a maintenance-fee lapse recorded effective 2020-01-17), so there is no forward-looking infringement exposure and an IPR estoppel was never triggered (no final written decision).

IPR2018-00240 — ZTE Corporation & ZTE (USA) Inc. v. Maxell, Ltd.

• Type: Inter Partes Review (35 U.S.C. §§ 311–319)
• Filed: 2017-11-22 (accorded filing date / "effective date" per the patent's legal-events record; the IPR was docketed in the 2018 IPR series alongside ZTE's other petitions, IPR2018-00235 through -00240)
• Status: "Not Instituted - Merits" (verbatim label). Plain-English gloss: the Board did not institute trial and disposed of the petition on the substantive merits (i.e., not a discretionary § 314(a)/§ 325(d) denial). This is the strongest form of loss for a petitioner and the best form of win for the patent owner.
• Judge panel: Not confirmed. I could not verify the panel that decided IPR2018-00240. (For contrast, the sibling ZTE petition on a different Maxell patent, IPR2018-00237, was decided by APJs Lynne Pettigrew, Minn Chung, and Terrence McMillin — do not attribute that panel to the '695 petition.)
• Petition grounds: Not confirmed. I could not retrieve the petition and could not verify which claims were challenged, which references were relied on, or whether grounds were § 102 and/or § 103. The '695 patent has six claims: independent apparatus claims 1 and 4, and dependent claims 2–3 and 5–6. For context, the patent owner's meet-and-confer correspondence in the co-pending E.D. Tex. litigation shows ZTE had also been asserting JP Hei 6-295195 ("Hirose") and US 5,765,136 ("Fukuchi") against the '491/'695 family in the district court case — but those references were expressly withdrawn from the district court case, and I cannot say whether they appear in the '240 petition. Do not assume they do.
• Institution decision: Denied (merits). I could not pin down the exact decision date or the panel's specific reasoning for -00240. Public reporting indicates the final one of ZTE's seven denials issued on Monday, 2018-07-02 ("the Patent Trial and Appeal Board turned away the last of ZTE's seven administrative challenges to the validity of the patents" — Law.com, 2018-07-02). I cannot confirm that the 2018-07-02 denial was the '695 petition specifically, as opposed to one of the sibling petitions.
• Final Written Decision: None — none exists. Because institution was denied, the Board never reached a final written decision, so no claim of the '695 patent was canceled, and no claim was held patentable in an FWD. Any statement that "claims 1–5 were canceled" or that "the patent survived an IPR on the merits" at a claim level is unsupportable: there is no FWD for this patent.
• Settlement / termination: Not applicable at the PTAB (the IPR ended by denial of institution, not settlement). The underlying district court litigation settled — Maxell and ZTE filed a joint notice of settlement in February 2019 in E.D. Tex. Nos. 5:16-cv-00179 and 5:18-cv-00080, following a $43.3M willful-infringement jury verdict (2018-06-29) in which the '695 patent was one of the seven asserted patents. The settlement terms are confidential and were filed under seal.
• Appeal: None. No FWD means nothing to appeal to the Federal Circuit. Separately, 35 U.S.C. § 314(d) generally bars appeal of an institution denial, subject to only narrow exceptions (compare the appeal attempted in the unrelated Apple v. Maxell IPR2020-00597 on the '493 patent). I found no CAFC docket for IPR2018-00240.
• Defensive value: Mixed, and time-sensitive. On the one hand, this is a petitioner's loss — institution was denied on the merits, so the '695 patent's claims carry no PTAB blemish and Maxell can argue it "survived" ZTE's challenge just as it told Apple it had. On the other hand, the patent has expired, so that validity strength buys a defendant nothing prospectively; the only live question for anyone still receiving demand letters is historical damages.

Strategic summary

Claim status: all six claims UNTESTED, none canceled, none sustained. The '695 patent issued with claims 1–6 (independent claims 1 and 4; dependents 2–3 and 5–6). No PTAB FWD ever issued, so the PTAB has never canceled or affirmed a single claim of this patent. The only claim-scope adjudication of record is judicial, not administrative: the E.D. Tex. Markman decision in Maxell Ltd. v. Huawei Device U.S. Inc., 297 F. Supp. 3d 668 (E.D. Tex. 2018), which construed terms of '695 claims 1 and 4 (including the dispute over whether "controller for receiving a method of compression and encoding …" is a § 112 ¶ 6 means-plus-function term) — that is district-court claim construction, not a PTAB outcome, and it does not invalidate anything. Claims 1–6 must be treated as intact but never PTAB-tested.

Estoppel landscape: no IPR estoppel attached — § 315(e)(2) is a non-issue here. IPR estoppel is triggered only by a final written decision. ZTE's petition died at institution, so neither ZTE nor its privies are estopped from raising in district court any invalidity ground they raised or reasonably could have raised in IPR2018-00240. Practically, this means a defendant today can freely re-run the same art ZTE assembled in 2017 (and any art ZTE chose not to include), with no § 315(e)(2) trap. The one real ceiling is the expired term: any invalidity work would only serve to defeat past damages in a pending case, not to clear future products.

Pattern signals. The same petitioner (ZTE) mounted a coordinated seven-petition campaign against Maxell's E.D. Tex. patents — IPR2018-00235 through -00240 and beyond — and lost all of them, which is a meaningful signal that this portfolio resisted institution-stage attack. No defensive aggregator appears in this patent's chain: the "Unified Patents" string is a dataset label, and Unified's portal merely indexes the district-court cases (N.D. Ill. 1:22-cv-00256; W.D. Tex. 6:21-cv-01169 and 6:21-cv-00158; E.D. Tex. 5:16-cv-00179). Apple did later file IPRs against other Maxell patents (e.g., IPR2020-00597 on the '493 patent; IPR2020-00407 on the '317 patent), but I found no Apple petition on the '695 patent. Maxell does not appear to have litigated any PTAB appeal on this patent because there was nothing to appeal.

Termination of rights. The '695 patent shows a maintenance-fee lapse ("Lapsed due to failure to pay maintenance fee," effective 2020-01-17; "PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES," 2020-02-24), with an adjusted expiration of 2021-01-09. Either way, the patent's enforceable term has run. Maxell subsequently reorganized the asset (Maxell Holdings → Maxell, Ltd.), but the patent itself is expired.

Recommended next steps

• If you are a defendant today: the most useful document to cite is not a decision (there is none) but the proceeding record showing denial of institution — IPR2018-00240, status "Not Instituted - Merits," petitioner ZTE Corporation & ZTE (USA) Inc., filed 2017-11-22, as reflected on the patent page (https://patents.google.com/patent/US8098695/en) and the PTAB portal entry (https://portal.unifiedpatents.com/ptab/case/IPR2018-00240). You can verify the file directly at PTAB E2E / PTAB Center (https://ptacts.uspto.gov/ptacts/) by searching proceeding number IPR2018-00240.
• Lead with expiration, not validity. Because the patent lapsed effective 2020-01-17 (adjusted expiration 2021-01-09), there is no injunctive or forward-looking exposure. Any demand letter asserting the '695 patent against current products should be met with the maintenance-fee lapse and expiration before you spend anything on invalidity.
• On validity, the coast is open but the value is limited. With no FWD, there is no § 315(e)(2) estoppel against any ground; a defendant facing a past-damages claim can raise § 102/§ 103 in court using ZTE's 2017 art, the '491/'695 family references (Hirose, Fukuchi), or anything else. If you do file an IPR or PGR, note that with the patent expired there is limited practical value in a PTAB challenge — prioritize district-court invalidity and damages defenses.
• If a settlement/licensing demand leans on the ZTE verdict: the $43.3M E.D. Tex. judgment was settled confidentially in February 2019 and never became a Federal Circuit affirmance; the willfulness finding died with the settlement. It is not a validity holding and does not bind you.
• Things I could not verify (do not treat as confirmed): the specific grounds/claims/art in IPR2018-00240; the identity of the deciding panel; and the exact date and reasoning of the -00240 institution denial. If precision on those points matters, pull the IPR2018-00240 papers directly from PTAB E2E, since public secondary sources aggregate the seven denials without consistently mapping decision dates to patent numbers.

I'll verify what's retrievable before writing. Let me cross-check the assignment records, litigation, and correspondent data.

Good — I recovered the reel/frame numbers embedded in the Google Patents legal events and one correspondent. Let me check Maxell's corporate status and the remaining cases.

Inventors

The US application (Ser. No. 09/433,049, filed 1999‑11‑03) named Hitachi, Ltd. and Hitachi Video and Information System, Inc. as the original assignees, which establishes that all four inventors were Hitachi‑group employees at filing. The record does not allocate individual inventors between the two Hitachi entities, and I could not verify per‑inventor employers from the assignment abstracts, so I am not asserting them.

Priority: JP 10‑313254, filed 1998‑11‑04.

Unusual patterns: none on the inventor side. There is no recorded inventor assignment, no inventor‑to‑third‑party transfer, and no evidence any inventor departed within 12 months of filing. The unusual feature of this family is prosecution span, not inventor churn: the JP priority is 1998, the first US case issued as US 6,816,491 (2004‑11‑09), the second as US 7,447,241 (2008‑11‑04), and this third continuation issued as US 8,098,695 on 2012‑01‑17 — roughly 13 years from priority to the last member. The patent owner still asserted the '695 patent in 2021–2022 litigation. I found no public record of inventor interviews, depositions, or inventor-founded entities.

Original assignee

Hitachi, Ltd. (Tokyo) and Hitachi Video and Information System, Inc. (JVIS), per the "Original Assignee" field on Google Patents (https://patents.google.com/patent/US8098695/en).

• Primary line of business: Hitachi, Ltd. is a diversified Japanese industrial conglomerate (TSE: 6501) — information/telecommunication systems, power and industrial systems, plus the consumer media/electronics businesses that this patent grew out of. JVIS was the group company handling video and information equipment.
• Did they ship a product embodying the claims? Plausibly, but not documented in the record. The specification itself is drafted around receiver/recorder products — FIGS. 10–13 describe a broadcast receiver, a video-capable receiver, a recording/reproducing-apparatus-built-in receiver, and a communication apparatus in which the decoder is installed. Hitachi was a major supplier of digital broadcast receivers and set‑top boxes in the relevant window. I could not find a specific Hitachi product identified as practicing the claims, so treat "shipped an embodying product" as likely but unverified.
• Current status: Hitachi, Ltd. is operating (and is no longer a party to the chain — its '695 interest was assigned out in 2013, see below). Hitachi Video and Information System, Inc. is not a distinct entity today: it merged into Hitachi Advanced Digital, Inc. effective 2009‑04‑03 (recorded as reel 037859/0049), which in turn merged into Hitachi Industry & Control Solutions, Ltd. effective 2014‑04‑01 (reel 037776/0708). Hitachi was never in bankruptcy.

Assignment timeline

Sourcing note (please read before relying on any single entry). The reel/frame identifiers, conveyance types, execution dates and recording dates below are taken from the Google Patents legal‑events record for US 8,098,695, which is populated from the same USPTO assignment abstracts served by the Assignment Center (https://assignmentcenter.uspto.gov/). I was unable to load the Assignment Center UI itself in this session, so I could not independently verify correspondents for every link. I verified one correspondent — reel 045142/0208 — via a reprinted Patent Assignment Abstract of Title for a sibling patent recorded on the same reel (US 10,128,855 / app. 12/028,885), which shows the shared reel/frame entry and its correspondent. Because that reel covers a batch of patents recorded together, attributing the correspondent to the '695 entry on that reel is an inference from a shared reel number, not a direct read of the '695 abstract. For the remaining reels I have no correspondent data and say so rather than guessing.

Recorded in order of execution date:

• 2009-04-03 (executed) / recorded 2016-02-19 — Reel 037859/0049

  • Conveyance: Merger
  • Assignor: Hitachi Video & Information System, Inc.
  • Assignee: Hitachi Advanced Digital, Inc.
  • Correspondent: not retrieved
  • Context: internal reorg (JVIS folded into Hitachi Advanced Digital). Note the seven‑year recording lag — a 2009 merger not papered at the USPTO until 2016, i.e. chain‑of‑title cleanup timed to the run‑up to the ZTE suit.

• 2013-06-07 (executed) / recorded 2013-06-12 — Reel 030668/0719

  • Conveyance: Assignment
  • Assignor: Hitachi, Ltd.
  • Assignee: Hitachi Consumer Electronics Co., Ltd.
  • Correspondent: not retrieved
  • Context: internal reorg — Hitachi moved its consumer‑product‑facing IP (this patent included) into the consumer electronics company; Maxell's own complaint describes this as part of a 2009–2013 reorganization (W.D. Tex. 6:21‑cv‑00158 Compl. ¶4).

• 2014-04-01 (executed) / recorded 2016-02-19 — Reel 037776/0708

  • Conveyance: Merger
  • Assignor: Hitachi Advanced Digital, Inc.
  • Assignee: Hitachi Industry & Control Solutions, Ltd.
  • Correspondent: not retrieved
  • Context: internal reorg (second leg of the JVIS lineage).

• 2014-08-26 (executed) / recorded 2014-09-08 — Reel 033694/0745

  • Conveyance: Assignment (recorded as "ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Hitachi Consumer Electronics Co., Ltd. (the abstract lists the assignor twice, once with a stray comma — a clerical duplication in the record, not a second assignor)
  • Assignee: Hitachi Maxell, Ltd.
  • Correspondent: not retrieved
  • Context: internal reorg / transfer-to-licensing-aligned entity — this is the link where the asset left the consumer‑electronics operating company and landed in the Maxell side of the group, which is the entity that later ran the licensing and litigation program.

• 2016-01-29 (executed) / recorded 2016-02-19 — Reel 037776/0765

  • Conveyance: Assignment
  • Assignor: Hitachi Industry & Control Solutions, Ltd.
  • Assignee: Hitachi Maxell, Ltd.
  • Correspondent: not retrieved
  • Context: internal reorg — closes the JVIS lineage into Hitachi Maxell, completing sole ownership ~9 months before the first suit naming this patent.

• 2017-10-01 (executed) / recorded 2018-01-25 — Reel 045142/0208

  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNOR'S INTEREST")
  • Assignor: Hitachi Maxell, Ltd.
  • Assignee: Maxell, Ltd.
  • Correspondent: Nadeen Nsouli, 71 South Wacker Drive, Chicago, IL 60606 — that address is Mayer Brown LLP's Chicago office, and Mayer Brown (Jamie B. Beaber) was Maxell's lead trial counsel in the ZTE, Apple and Lenovo campaigns. Flag: this is the same firm litigating the portfolio, recorded here as correspondent; within this chain the name appears only once, so it is not by itself a "repeat correspondent" finding — but it means assignment recording and enforcement were handled out of the same shop.
  • Context: transfer-to-asserter / corporate separation — executed the same day Hitachi Maxell renamed itself Maxell Holdings, Ltd. and the operating company took the Maxell name (per Hitachi's 2017‑03‑21 release and Maxell's corporate history page). Maxell's complaint characterizes the move as aligning the IP "with the licensing, business development, and research and development efforts of Maxell."

• 2021-10-01 (executed) / recorded 2021-11-29 — Reel 058255/0579

  • Conveyance: Merger
  • Assignor: Maxell, Ltd.
  • Assignee: Maxell Holdings, Ltd.
  • Correspondent: not retrieved
  • Context: change of corporate structure (collapse of the holding‑company structure begun in 2017).

• 2021-10-01 (executed) / recorded 2021-12-03 — Reel 058666/0407

  • Conveyance: Change of Name
  • Assignor: Maxell Holdings, Ltd.
  • Assignee: Maxell, Ltd.
  • Correspondent: not retrieved
  • Context: change of name only — completes the 2017–2021 holding‑company round trip; the surviving entity is once again Maxell, Ltd., the named current assignee.

Also on the record and relevant to the chain's use: the patent lapsed for non‑payment of the 7.5‑year maintenance fee on 2020‑01‑17 (fee due 2019‑07‑17 plus the 6‑month grace period; USPTO legal‑event codes LAPS/STCH recorded 2020‑02‑24, FRAMED FP entry effective 2020‑01‑17). Google Patents lists the adjusted expiration as 2021‑01‑09 and status "Expired – Fee Related." Maxell nonetheless continued to assert the '695 patent in complaints filed 2021‑11‑12 (W.D. Tex. 6:21‑cv‑01169, Lenovo/Motorola) and 2022‑03‑30 (W.D. Tex. 6:22‑cv‑00334, Motorola), consistent with seeking pre‑lapse past damages.

Timeline diagram

timeline
    title Ownership of US 8098695
    1998 : JP priority application filed
    1999 : US application 09 433 049 filed
    2009 : JVIS merges into Hitachi Advanced Digital
    2012 : US 8098695 issues
    2013 : Hitachi Ltd assigns to Hitachi Consumer Elec
    2014 : Hitachi Consumer Elec assigns to Hitachi Maxell
    2016 : Chain title recorded before first suit
         : Hitachi unit assigns to Hitachi Maxell
    2017 : Hitachi Maxell assigns to Maxell Ltd
    2018 : ZTE jury verdict for Maxell
    2020 : Patent lapses for unpaid maintenance fee
    2021 : Holding company merger and name change
         : Maxell sues Lenovo and Apple
    2022 : Maxell sues Motorola again

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
Every assignee in the chain is a named operating corporation in the Hitachi/Maxell group: Hitachi Advanced Digital (037859/0049), Hitachi Consumer Electronics (030668/0719), Hitachi Industry & Control Solutions (037776/0708), Hitachi Maxell (033694/0745, 037776/0765), Maxell Holdings (058255/0579), Maxell (045142/0208, 058666/0407). None is an "IP/Licensing/Holdings/Ventures" LLC, none uses a registered‑agent service address, and none is a Delaware or Texas single‑purpose vehicle. Maxell Holdings is a Japanese kabushiki kaisha holding company, not a shell. The closest thing to a shell signal is the 2017‑10‑01 move into Maxell, Ltd. (reel 045142/0208) — but that entity is a listed Japanese manufacturer with products, so the naming tells you nothing by itself, and I am not treating it as a shell.

2. Known asserter in the chain — NOT PRESENT (with a qualification).
No entity in the chain matches the named NPE lists: no Acacia, Marathon, Intellectual Ventures, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or Spangenberg entity appears in any reel/frame above. Qualification: Maxell is nonetheless a repeat, high‑volume plaintiff, and it appears as plaintiff in the Unified Patents litigation portal for the case that first asserted this patent (https://portal.unifiedpatents.com/litigation/Texas%20Eastern%20District%20Court/case/5:16-cv-00179) and in PTAB activity (IPR2018‑00240). Its campaigns include ZTE (5:16‑cv‑00179, jury verdict 2018‑06‑29, $43.3M, settled 2019‑02), Huawei (5:16‑cv‑00178), Apple (5:19‑cv‑00036; 6:21‑cv‑00158), Lenovo/Motorola (6:21‑cv‑01169, 6:22‑cv‑00334), TCL (5:23‑cv‑00108), Fandango, and Samsung (2023–2025, plus ITC 337‑TA‑1432 and UPC actions). That is an enforcement program, but Maxell is not a listed NPE, and I will not stretch the list to make the signal present. Signal 2 as defined: not present.

3. Repeat correspondent across the chain — UNCLEAR (data gap, not a finding).
Only one correspondent is recoverable: Nadeen Nsouli, 71 South Wacker Drive, Chicago, IL 60606 on reel 045142/0208 (recorded 2018‑01‑25). That address is Mayer Brown LLP, the firm that tried the Maxell campaigns. Because it appears on only one link, the recurrence test is not met on the evidence I have. I could not retrieve correspondents for reels 037859/0049, 030668/0719, 037776/0708, 033694/0745, 037776/0765, 058255/0579, or 058666/0407, so I cannot rule recurrence in or out — this needs a manual pull from the Assignment Center abstracts. I am marking it unclear rather than not present for that reason, and I am not asserting Mayer Brown handled the other recordings.

4. Cascading transfers — NOT PRESENT.
There is one same‑day batch: three recordings all dated 2016‑02‑19 (reels 037859/0049, 037776/0708, 037776/0765 — two of them sharing the 037776 reel prefix, i.e. one recording batch). But the underlying events are two intra‑group mergers with backdated execution dates (2009‑04‑03 and 2014‑04‑01) plus one real assignment (2016‑01‑29) — this is a chain‑of‑title consolidation, not a daisy chain of LLCs flipping the asset. The remaining links are separated by 1–5 years. No shared‑correspondent-address pattern is visible (correspondents unknown). No <24‑month cascade of distinct assignees.

5. Pre-litigation transfer — NOT PRESENT (but the timing is worth noting).
First suit naming this patent: Maxell v. ZTE, 5:16‑cv‑00179, filed 2016‑11‑18. The most recent recorded acquisition before that is the 2016‑01‑29 execution (recorded 2016‑02‑19, reel 037776/0765) — roughly 9 months before the suit, outside the 6‑month window. The 2017‑10‑01 transfer to Maxell, Ltd. (reel 045142/0208) post‑dates the ZTE complaint and precedes the Apple 2019 suit by ~20 months and the Lenovo 2021 suit by ~46 months. So no link sits inside 6 months of a first assertion. Still, the pattern — 2009 and 2014 mergers papered in a single 2016 batch, then the first suit nine months later — has the shape of a standing record assembled for enforcement, and should be noted even though it fails the strict test.

6. Bankruptcy fire‑sale — NOT PRESENT.
Hitachi never filed Chapter 7/11 relating to these assets. The Maxell separation was a consensual corporate demerger: Hitachi sold 7,800,000 Hitachi Maxell shares (14.76%) on 2017‑03‑22, and ran a secondary offering of 6,194,100 Maxell Holdings shares on 2017‑11‑17, cutting its stake to 3.01% (Hitachi releases at hitachi.com, 2017‑03‑21 and 2017‑11‑17). No distressed sale, no §363 auction, no creditors' committee.

7. Privateering — NOT PRESENT.
For privateering you need an operating company funding an NPE to assert against its competitors. Here Hitachi severed the relationship (stake down to ~3.01% by late 2017, no equity‑method consolidation, no personnel overlap recorded), and the defendants in the Maxell campaigns (Apple, Lenovo, Samsung, ZTE, TCL) are not Hitachi's competitors in its core Social Innovation businesses. Maxell is asserting its own portfolio, not acting as Hitachi's proxy, and I found no SEC disclosure, Patent Progress or EFF item describing a Hitachi‑funded assertion program. There is a structural echo of privateering — an operating parent seeding a licensing entity with legacy IP — but no evidence of direction or benefit.

8. Defensive aggregator — NOT PRESENT.
The chain terminates at Maxell, Ltd., an asserting plaintiff, not at RPX, AST, LOT, Unified Patents, or OIN. The one defensive‑flavored event, the ZTE IPR of this patent (IPR2018‑00240, filed by ZTE Corporation and ZTE (USA) Inc.), was not instituted on the merits, and ZTE is an accused infringer, not an aggregator. The patent has not been neutralized by aggregation; it was abandoned by fee lapse instead.

Verdict

NPE — moderate confidence.

Maxell, Ltd. is the current owner (final link: reel 058255/0579 merger and reel 058666/0407 name change, both executed 2021‑10‑01), it does not practice the claimed audio‑decoder technology, and it monetizes this asset exclusively through licensing demands and litigation across at least eight unrelated defendants over 2016–2025 — the classic licensing‑entity function. The strongest evidence is the 2017‑10‑01 transfer on reel 045142/0208, which Maxell's own complaint describes as aligning the IP with its "licensing, business development" efforts, plus the abandonment of the patent for non‑payment of the 2019/2020 maintenance fee (lapse effective 2020‑01‑17) while still asserting it in 2021–2022 — conduct a practicing manufacturer would not exhibit. Confidence is only moderate, not high, because the chain contains no shell entity and no listed NPE: every link is a named Hitachi/Maxell corporate entity, the 2009–2021 movements are genuine intra‑group mergers and a demerger recorded at reels 037859/0049, 030668/0719, 037776/0708, 033694/0745, 037776/0765 and 058255/0579, and the assignee continues to make and sell products (batteries, storage media, projectors) — so a stricter reader could call this a hybrid operating/licensing company rather than an NPE, and I have flagged that disagreement rather than hiding it.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 8098695; the Google Patents mirror used above is https://patents.google.com/patent/US8098695/en). The two open data gaps that would move this verdict are (a) the correspondent of record on reels 030668/0719, 033694/0745 and 037776/0765, and (b) whether a single attorney/firm recurs across them.

I'll research this systematically. Let me start by verifying the patent and searching for the prior-art references.

Let me dig into the IPR proceeding and remaining key references.

Prior-Art Analysis — U.S. Patent No. 8,098,695

Analyst note on scope: I was asked to search USPTO for patent number 8098695 literally and not to auto-correct. I am treating US 8,098,695 B2 as the subject. ⚠️ One naming collision must be flagged: a Japanese publication JP 8098695 A exists (a 1996 laid-open publication, cited in an unrelated PCT search report). That is not the U.S. patent at issue and I have excluded it. Likewise US 6,816,491 and US 7,447,241 are the same family (parents of '695), not prior art.

A. Patent Verification

The patent was asserted and construed in Maxell Ltd. v. Huawei Device U.S. Inc., 297 F. Supp. 3d 668 (E.D. Tex. 2018), where the court held the preambles limiting and gave plain meaning to "a memory" and "begins decoding processing."

B. The Claims Being Tested

Critical date cutoffs (priority 1998-11-04):

• § 102(a)/(b): a reference must predate the invention / be >1 year before filing (i.e., before ~1997-11-04 for § 102(b)).
• § 102(e): a U.S. patent/application counts only if its filing date precedes the '695 priority date. US 6,378,101 B1 (filed 1999-01-27) post-dates 1998-11-04 and therefore cannot be § 102 prior art at all — it was cited but is non-anticipatory.

C. Citation-by-Citation Review (all 26 references listed on the '695 face)

Legend for "§102 risk": H = high/plausible single-reference anticipation of at least one claim; M = moderate (anticipates a dependent claim or sub-combination only); L = low (background / § 103 art only); N/A = date-barred or non-enabling.

Non-patent citation (1): "Design and Implementation of AC-3 Coders," IEEE Trans. Consumer Electronics, Vol. 41, No. 3, August 1995 — the DSP-based AC-3 decoder described in the '695 Background. §103 background only; it does not disclose downloadable process codes or a packet demultiplexer.

D. Most Relevant Prior Art — Ranked

1. JPH 06-295195 A (Hitachi Ltd, pub. 1994-10-21) — primary reference.
Discloses the exact architectural move the '695 patent claims as inventive: rather than burning the decode algorithm into the DSP's on-chip ROM, store a plurality of compression/decompression algorithm programs in an external ROM, select one under host-device control, and boot it into the DSP's internal program memory to run. This is the § 103 primary reference that ZTE relied on in IPR2018-00240 (not instituted on the merits), and it is the reference whose translated specification appears in the PTAB petition record. § 102 gap: no MPEG-TS packet demultiplexing by PID, and no automatic "detect whether the compression/encoding method changes" trigger derived from stream header information — the '695 patent's own point of novelty over this teaching.

2. US 6,167,521 A (IBM) — best § 102(a) candidate for the "downloading … from outside of said memory" limitation of claim 1 (element d), albeit in a secure-code-distribution context.

3. US 6,256,487 B1 (Ericsson) — best art for claim 1's "extracting a method of compression and encoding … from header information" plus receiver-side mode selection; multi-mode codec with the coding mode conveyed in-band.

4. US 5,845,249 A / US 6,430,533 B1 (LSI Logic) — the two LSI references together supply the "audio core + external RAM + ROM + MPEG-2/AC-3 dual-format" architecture and expressly motivate moving decode firmware off-chip (6,430,533: fully-firmware decoders "require large sized Code ROMs … and are expensive"). Strong § 103 combination art against claims 1 and 4.

5. JPH 05-313908 A (Fujitsu) — most relevant to claims 2 and 5 (download-error handling/verification of a downloaded program).

6. US 5,765,136 A (Nippon Steel) and US 5,420,866 A (Scientific-Atlanta) — multiplexed audio/video expansion and packet-based decoder control, respectively; § 102 background for the demultiplexer/AV-decode elements of claims 1 and 4.

E. Bottom Line on § 102

• No single listed reference appears to disclose every element of independent claim 1 or claim 4. Each is missing at least one of: (i) PID/attribute-based demultiplexing of one user-designated audio sequence from a multiplexed packet group, (ii) extraction of the compression method from the audio header, and (iii) controller-side detection of a change in that method followed by download of the matching decoding program code into the on-board memory. The '695 claims are therefore best challenged under § 103, not § 102.
• The strongest § 102(a) single-reference positions are narrow: JPH 06-295195 A against the download/execute core of claim 1, US 6,167,521 A against the "download from outside said memory" limitation, and JPH 05-313908 A against claims 2/5.
• US 6,378,101 B1 is date-barred (filed 1999-01-27, after the 1998-11-04 priority) and should be removed from any anticipation theory.

F. Caveats / Confidence

• The citation dates above are reproduced as listed on the '695 face (Google Patents labels the first column "priority date"); for several references this is the original filing date.
• Content descriptions for US 5,845,249, US 6,430,533, JPH 06-295195 A, US 5,765,136 A, US 6,256,487 B1, US 6,378,101 B1 and US 6,167,521 A were corroborated by live retrieval of their Google Patents/PDF records. For the remaining references (e.g., JPH 05-313908 A, JPH 08-289255 A, US 5,420,866 A, US 7,375,761 B2) I reached the step limit before retrieving full text, so those descriptions rest on the titles and assignee data on the '695 face and should be confirmed against the actual specifications before being relied on in a validity contention.
• I could not retrieve the full IPR2018-00240 petition PDF (institution was denied on the merits), so I cannot state its final grounds with certainty; the translated JPH 06-295195 material I retrieved from the PTAB docket is the basis for ranking that reference first.

I'll ground the key prior-art references before building the combination analysis.

Let me ground the remaining key references cited on the page.

Obviousness Analysis — U.S. Patent 8,098,695 B2

"Multiplexed audio data decoding apparatus and receiver apparatus"
Assignee of record: Maxell, Ltd. (originally Hitachi Ltd./Hitachi Video & Information System, Inc.) · Priority: 1998‑11‑04 (JP 10‑313254) · Continuation of US 7,447,241 (Ser. No. 10/945,905), itself a continuation of US 6,816,491 (Ser. No. 09/433,049, filed 1999‑11‑03) · 6 claims, all independent-claim 1 or 4 with dependents.

Sources used. The patent text, claim set, and the complete "Patent Citations (26)" / NPL list are taken from the Google Patents record you supplied (https://patents.google.com/patent/[US8098695](/patent/US8098695)/en). Where I verified a reference's actual disclosure I cite its own record (US5845249, US6430533, US6256487, US5765136). Where I only had the citation metadata on the '695 page (number, title, assignee, date), I say so explicitly and treat my characterization as an inference from the title — I have flagged every such instance rather than inventing disclosure content.

1. Framing: what the claims actually require

The issued claims are narrower in number of memories but broader in substance than the "Summary of the Invention" in the specification. Compare:

Two consequences matter for §103:

• The ROM (60) element was deleted. The issued claim is not tied to "plural codes permanently resident in a ROM exchanged to add formats." It covers any architecture in which decoder code is downloaded from outside the decoder memory in response to a detected change in the compressed-audio coding method. That is the classic "firmware-loadable / boot-loaded DSP codec" concept.
• "Said memory" is ambiguous (the spec's FIG. 1 uses two memories — external ROM 60 accumulated the codes, internal RAM 25 received the transfer). Under a broadest-reasonable-interpretation reading, the claim reads on (i) a RAM loaded from an external store (FIG. 1), or (ii) a single large memory holding all codec programs and refreshed from outside (FIG. 6). Either reading must be addressed below.

PHOSITA (priority date, 1998). A person with a BS/MS in EE/CS and 2–5 years in digital audio/video decoder or MPEG‑2 systems IC design, familiar with (a) MPEG‑1/‑2 Audio and ATSC AC‑3 decoding algorithms, (b) MPEG‑2 Transport Stream/PES header and PMT/PID parsing, and (c) DSP/embedded firmware, including field-upgradeable ("flash") code as used in set-top boxes.

2. Element chart for claim 1

No single reference appears to disclose the whole combination, so the question is combinations under §103.

3. The cited pool, sorted by statutory weight (dates as listed on the '695 page — treated as assumptions)

Solid 102(b) art (published >1 yr before 1998‑11‑04):

• US 4,817,146 (GE, 1989‑03‑28) — cryptographic digital signal transceiver (download/security context only).
• JP H05‑313908 A (Fujitsu, 1993‑11‑26) — program download + "guarantee" of downloaded program.
• JP H06‑295195 A (Hitachi, 1994‑10‑21) — "Speech codec."
• US 5,420,866 (Scientific‑Atlanta, 1995‑05‑30) — conditional‑access info to decoders in a packet‑based multiplexed system.
• US 5,502,496 (Thomson, 1996‑03‑26) — audible instructions/status in a digital TV system.
• JP H08‑289255 A (Sony, 1996‑11‑01) — encoding/decoding device.

102(a)/102(e) art (published or US‑filed before the critical date but within one year, or post‑grant):

• US 5,845,249 (LSI Logic; filed 1996‑05‑03, granted 1998‑12‑01) — MPEG‑2/AC‑3 audio core. 102(e) only.
• US 6,430,533 (LSI Logic; priority 1996‑05‑03, granted 2002‑08‑06) — algorithm partitioning hardware/firmware. 102(e) only.
• US 5,765,136 (Nippon Steel; 1994‑10‑28 priority, granted 1998‑06‑09) — image/audio multiplexed decoding apparatus.
• US 6,256,487 (Ericsson; 1998‑09‑01 priority) — multi‑mode codec, mode conveyed to receiver.
• US 5,910,or US 6,026,088 (LSI Logic); US 5,684,527 (Fujitsu); US 5,847,763 (Oki); US 6,249,609 (Canon); US 6,449,596 (Matsushita); US 7,002,999 (Toshiba); US 2001/0028780 A1 (Samsung); US 6,287,748 (Hitachi, display compatible with digital broadcasts); US 7,375,761 (Hitachi, "Receiver having motion picture data decoder"); US 7,356,828 (BSkyB, "Receivers for television signals"); US 6,108,697 (Powerquest); US 7,389,287 (Burke).

Post‑priority on its face: US 6,378,101 B1 (priority 1999‑01‑27, "Multiple program decoding for digital audio broadcasting and other applications," Agere). On the '695 page's own dates this is after 1998‑11‑04, so it is not available as prior art unless the JP priority is broken — which is precisely the issue the issued claim language raises (see §7).

NPL: "Design and Implementation of AC‑3 Coders," IEEE Trans. CE, Vol. 41, No. 3, Aug. 1995 — DSP implementation of an AC‑3 decoder.

4. Grounds of rejection

Ground A — Claims 1–3: US 6,430,533 in view of JP H05‑313908 A, further in view of US 6,256,487

US 6,430,533 discloses a digital audio decoder for both MPEG‑1/MPEG‑2 and AC‑3 in which the decoding algorithm is deliberately partitioned between hardware (the "audio core," doing sub‑band synthesis/matrixing/windowing) and firmware running the preceding algorithm steps, with samples stored in an input RAM "coupled to the audio core." Critically, it diagnoses exactly the problem the '695 patent addresses: "fully software/firmware-based audio decoders require large sized Code ROMs to achieve the decoding functionality, and are expensive," while a complete hardware implementation "comes at the expense of large number of transistor device gates that consume a large chip area." The '695 patent's stated problem — built‑in ROM scale preventing integration of a multi‑format audio decoder, and inflexibility when formats are added (col. 1–2 of the spec) — is therefore the same problem, and US 6,430,533 supplies the motivation to move code storage off the core and keep the core format‑agnostic.

JP H05‑313908 A (title verified: "Program guarantee method for program download type encoder," Fujitsu) supplies the second half: downloading a program into a programmable signal‑processing device, plus a method for guaranteeing/validating the downloaded program. Caveat: I could not open this reference's full text within this session; I am relying on its title as listed on the '695 page, and my characterization of its content is an inference from that title. If its full text confirms download‑and‑verify as its title states, the combination yields claim 1 (download of the format‑specific program in response to format detection) and claim 2/3 (verifying the download).

US 6,256,487 supplies the receiver‑side logic: a transmitter conveys a mode indicator with the payload, and the receiver "employ[s] the appropriate channel decoding/speech decoding techniques," with the ability to "rapidly change between these different coding modes." This is analytically the "detecting a change in the method of compression and encoding and then loading the corresponding decoder" step. Its field (GSM speech/channel codecs) differs from broadcast audio, but under KSR a reference may be used where "the problem itself suggests" the combination — and the problem of a receiver that must match its decoder to the bitstream's signaled coding format is universal.

Ground B — Claims 1–3: US 5,845,249 in view of JP H05‑313908 A and US 5,765,136

US 5,845,249 discloses a reusable MPEG‑2/AC‑3 "audio core" including a data path (multiplier + accumulator), control logic with separate MPEG and AC‑3 state machines, a ROM with ROM‑addressing logic, and input/output RAMs located outside the core; the reference teaches the core is "particularly useful for expeditiously designing 'system' chips containing multiple cores on a single chip … for example, the audio core of this invention, a video core, and a CPU core." US 5,765,136 adds the multiplexed image‑audio decoding architecture with a header‑information decoding circuit separating the audio coding/banding info from payload. Adding the download‑and‑verify teaching of JP H05‑313908 A to this pair yields claims 1–3; adding the "video core"/video decoder teaching of US 5,845,249 (or US 7,375,761, also Hitachi) yields claim 4, and US 5,765,136 supplies the shared‑memory multiplexed video/audio decode arrangement of FIG. 6 of the '695 patent.

Ground C — Claims 2, 3, 5, 6 (download error checking)

These dependents add only verification of the downloaded code, with claim 3/6 specifying "a specific audio data sequence is decoded by using the decoding program code, and [is] compared with an expected value." Two independent supports:

1. The '695 specification itself concedes these are known methods: "As a methodology for the error check, there can be applied a method, in which a CRC (Cyclic Redundancy Check) code or a test data is decoded actually, so as to be compared with an expected value thereof, etc." That is a §103 admission that CRC checking and test‑vector (golden‑vector) execution are conventional in this art.
2. JP H05‑313908 A (program‑download guarantee) and US 6,167,521 (Securely downloading and executing code) independently establish that validating downloaded executable code before use was a familiar technique. Under KSR, "the improvement is the work of a skilled artisan … where the improvement is nothing more than the predictable use of prior art elements according to their established functions."

Ground D — Claim 4 (video decoder + audio decoder)

Claim 4's only additions over claim 1 are "a video decoder for decoding said video data sequence" and a DSP/video‑decoder memory. Both are squarely old: US 5,845,249 expressly contemplates an integrated "system" chip with audio core + video core + CPU core and describes the commercial HDM8211M single‑chip MPEG‑2 audio+video+transport decoder with a RISC controller; US 5,765,136 claims an "image audio multiplexed data decoding apparatus" with a demultiplexer, an image decoding circuit, an audio decoding circuit, and shared buffer memory; US 7,375,761 (Hitachi, same corporate family as the '695 patent) is a "receiver having motion picture data decoder." Combining a video decoder with the claim‑1 audio decoder requires no change in the operation of either — the quintessential KSR "combination of familiar elements according to known methods" with "predictable results."

Ground E — Alternative framing: US 5,845,249 or US 6,430,533 + US 6,256,487, with the download step supplied by the general knowledge of flash/boot‑loadable set‑top firmware

Even without JP H05‑313908 A, the record shows that (i) multiple coding formats must coexist in one broadcast receiver (US 5,845,249/US 6,430,533 backgrounds; the ATSC AC‑3 and MPEG‑2 standards documents cited therein), (ii) the receiver can be told the coding mode via the bitstream (US 6,256,487 for the analogous air interface; the MPEG‑2 PES stream_id for broadcast), and (iii) code for a signal processor is routinely stored off‑chip and loaded on demand in consumer electronics of the era. The Federal Circuit and KSR permit reliance on such "background knowledge of a skilled artisan" where the reference itself frames the problem; the '695 specification's own "Description of Prior Art" section goes further and admits that "in an audio decoder based upon the DSP … steps for the decoding processes are memorized in the built‑in ROM as the program codes," which is the very practice the claim departs from only by relocating the store.

5. Motivation to combine — why a PHOSITA would have done this

Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), the following rationales apply with particular force here:

1. Identified, predictable, finite solutions. A designer needing an MPEG‑2-Audio + AC‑3 (+ future format) receiver had a small, enumerated menu: N fixed hardware blocks (excessive gate area — US 6,430,533), N code images permanently in an on‑chip ROM (excessive ROM area and mask cost — US 6,430,533; and the '695 spec's stated integration problem), or one core plus load‑on‑demand code. Choosing the third is "obvious to try."
2. Design incentive / market pressure. ATSC/DVB receivers had to handle channel‑by‑channel format changes (AC‑3 vs MPEG Audio) and had to be upgradable as new formats (e.g., AAC) emerged without a silicon respin. US 6,256,487: "Systems can rapidly change between these different coding modes." The '695 patent's own stated purpose ("corresponding to the plural kinds of compression code formats by means of one DSP chip … without exchanging the audio decoder") is exactly this incentive.
3. Simple substitution of a known element. DVD/STB firmware was already routinely delivered in external ROM/flash and loaded to RAM at boot. Substituting a downloaded program image for a mask‑ROM program image was a well‑known engineering choice with a predictable benefit (reduced die area; field upgradeability).
4. Known technique improving a similar device in the same way. Program download with integrity verification was known (JP H05‑313908 A "program guarantee method … for program download type encoder"; US 6,167,521), as was one‑to‑many code image distribution (US 6,108,697). Applying it to a multi‑format audio decoder improves the device "in the same way" — no new function, just relocation and control.
5. Reasonable expectation of success. Downloading and re‑running a codec program is trivially verifiable (CRC, golden vector), and both the '695 spec and JP H05‑313908 A say so. There was no technological barrier to the claimed result.
6. Common ownership/analogous art. Several of the strongest references are Hitachi's own (US 6,287,748; US 7,375,761; JP H06‑295195 A "Speech codec"), addressing the same digital‑broadcast receiver problem — further evidence of the field's recognition of these elements as combinable.

6. Anticipated non‑obviousness arguments, and rebuttals

7. Date and priority caveats that materially affect the analysis

• Most of the best art is 102(e)/(a), not 102(b). US 5,845,249 (granted 1998‑12‑01) and US 6,430,533 (granted 2002‑08‑06) post‑date the 1998‑11‑04 priority date in publication; their use depends on their earlier US filing/priority dates (1996‑05‑03) as 102(e) art (and, for pre‑AIA §103, as §102(e) art with the Schwarcz-type limits on which claims of the reference may be relied upon). That is a solid position for US 5,845,249/US 6,430,533 but should be pleaded explicitly.
• US 6,256,487's 1998‑09‑01 priority is only ~9 weeks ahead of the '695 priority date, so any successful showing that the '695 claims are not entitled to the JP 10‑313254 priority is unnecessary for this reference — but conversely, if the '695 invention date is pushed earlier, US 6,256,487 falls away. It should be used as a secondary, not primary, reference.
• US 6,378,101 B1 (priority 1999‑01‑27) — arguably the most on‑point reference of all ("Multiple program decoding for digital audio broadcasting and other applications," Agere) — is post‑priority on the '695 page's own dates. It is usable only if the '695 claims lose the 1998‑11‑04 benefit. That defense‑side argument exists here: the issued claim language ("downloading … from outside of said memory") differs from the originally disclosed "read‑only memory … transferred … to said first memory" architecture, and the deleted ROM element may be worth testing for written‑description/priority support. This is the single most valuable further inquiry for a challenger.
• The 1999‑11‑03 filing date of the original application (Ser. No. 09/433,049) limits 102(b) art to publications before 1998‑11‑03 and 102(a) art to activity before the invention date; the JP publications of 1993–1996 and US patents granted through 1996 are therefore the most robust §102(b) anchors.

8. Procedural/litigation posture (context only)

• Family litigation reported on the '695 page: N.D. Ill. 1:22‑cv‑00256; W.D. Tex. 6:21‑cv‑01169 and 6:21‑cv‑00158; E.D. Tex. 5:16‑cv‑00179.
• IPR2018‑00240 (Petitioner: ZTE Corporation/ZTE (USA) Inc., filed 2017‑11‑22) — Not Instituted (merits) (https://portal.unifiedpatents.com/ptab/case/IPR2018-00240).
• Status: Expired – Fee Related (lapsed for failure to pay maintenance fees, effective 2020‑01‑17; adjusted expiration 2021‑01‑09). Current assignee Maxell, Ltd. This moots most practical invalidity work but is relevant to any retrospective §103 opinion.

9. Bottom line

1. Claims 1 and 4 are, in my assessment, more likely than not obvious over US 6,430,533 (or US 5,845,249) in view of JP H05‑313908 A, with US 6,256,487 and/or US 5,765,136 as secondary references. The primary references are in the same field, address the identical MPEG‑2‑Audio/AC‑3 multi‑format decoding problem, and expressly motivate moving format‑specific code off a fixed on‑chip ROM; the secondary references supply the "signaled coding mode → reconfigure the receiver's decoder" and "download-and-verify program" concepts.
2. Claims 2, 3, 5, and 6 are strongly obvious, given both JP H05‑313908 A / US 6,167,521 and the '695 specification's own admission that "a CRC code or a test data … compared with an expected value" is the conventional verification method.
3. Weaknesses in the case I would flag: (i) the best art is 102(e)/(a) rather than 102(b); (ii) I could not verify the full disclosure of JP H05‑313908 A, JP H06‑295195 A, JP H08‑289255 A, US 6,167,521, US 5,420,866, US 7,356,828, US 6,287,748, US 7,375,761, or US 7,389,287 in this session — those characterizations rest on the titles/dates in the '695 citation list and must be confirmed against the reference texts before any opinion is relied upon; (iii) the "controller receives the method from the demultiplexer" limitation should be mapped to an actual demux/controller disclosure (US 5,765,136's header‑information decoding circuit is the cleanest candidate) rather than inferred; and (iv) US 6,378,101 is the most directly on‑point reference but is date‑barred unless the priority claim is successfully attacked.
4. Recommended next steps if this analysis is to be hardened: pull the full texts of JP H05‑313908 A and JP H06‑295195 A (JPO/J-PlatPat); obtain the ZTE IPR2018‑00240 petition and the Board's non‑institution decision to see what was and was not argued; and run a priority/written‑description test on the issued "downloading … from outside of said memory" language against JP 10‑313254 to determine whether the 1999‑01‑27 US 6,378,101 becomes available.

This is a technical obviousness analysis, not a legal opinion, and no attorney‑client relationship is created. All statements about validity are probabilistic and depend on the full reference texts, the file history (including the amendments that produced the issued claims from the originally disclosed ROM‑based architecture), and any secondary‑considerations evidence.