US 6636178

I'll research US patent 6636178 across the requested sources.

Let me search for litigation/docket information and the full claim set.

Let me check for specific CAFC dockets and litigation history involving this patent.

US Patent 6,636,178 B2 — Research Summary

A note on method and limits: I was able to search public web sources (Google Patents, Espacenet, Justia, Unified Patents, etc.). I could not directly query the USPTO PatentCenter/Patent Assignment systems or the CAFC's PACER/docket system in this session, and one follow-up query was cut off by a step limit. Findings below are grounded in the sources I retrieved; where I lack authoritative confirmation I say so explicitly.

Important identifier caution: The literal string "6636178" also appears in unrelated documents — OSTI ID 6636178 (a 1984 gallium-arsenide oxidation patent, US 4,433,006), Japanese Patent No. 6636178 (Siemens, turbine fault diagnosis), and JP6636178B2. Those are not US 6,636,178. All statements below refer only to US 6,636,178 B2.

Bibliographic Data

Sources: Google Patents, Espacenet INPADOC family

Minor discrepancies noted (not auto-corrected): Google Patents lists the prior-art/priority date as 1999-03-30; the Unified Patents portal listed 1999-03-29. Both are preserved as found.

Abstract (verbatim)

"A signal detector is provided in which correlation analyses of multiple segments of a received signal from separate and distinct periods of time are combined to improve the detection capability of the signal detector. In one embodiment, the signal detector is part of a GPS receiver."

Plain-Language Overview of the Independent Claims

The record I retrieved (Espacenet claim listing) shows independent claims 1, 10, 12, and 13. Claim language below is paraphrased in plain terms.

Claim 1 — System (signal detector). A receiver captures at least a first and a second segment of a signal, where the segments come from separate periods of time and the signal is "a signal of interest perturbed by noise or pseudo-noise." A correlator (which also generates the hypothesis being tested) produces first correlation data for the first segment and second correlation data for the second segment. A combiner then:

1. determines a parameter difference between the two sets of correlation data,
2. adjusts one of the two data sets based on that difference (in practice, a code-phase shift), and
3. combines the adjusted data with the other data set to obtain cumulative correlation data useful for detecting a parameter of the signal of interest.

The key inventive idea: you can add up correlation results from disjoint, differently-sized capture windows even though their absolute code phases are unknown, by computing the relative difference and compensating for it.

Claim 10 — Signal detector (PN-code-specific). A receiver captures first and second segments of a signal that represents multiple signals of interest, each being a repeating PN code modulated onto a carrier. A correlator derives first and second correlation data against a combined PN code and code phase hypothesis. A combiner determines a code phase difference between corresponding portions of the two correlation data sets, adjusts one portion based on that difference, combines the adjusted portion with the other to obtain cumulative correlation data, and generates the code phase hypothesis.

Claim 12 — Method (counterpart to claim 1). Receiving first and second time-separated segments of a signal of interest perturbed by noise/pseudo-noise; generating a hypothesis; deriving first and second correlation data against the hypothesis; determining a parameter difference between them; adjusting one data set based on the difference; and combining the adjusted data with the other to obtain cumulative correlation data useful for detecting the signal or a parameter of it.

Claim 13 — Method (counterpart to claim 10). As shown in the partial record: generating a code phase hypothesis; deriving first and second correlation data between the two segments and a combined PN code and code phase hypothesis; determining a code phase difference between corresponding portions of the two data sets; adjusting one portion; and combining to obtain cumulative correlation data for detection.

Dependent claims (from the Espacenet record):

• Cl. 2: receiver is an RF receiver
• Cl. 3: signal of interest is a carrier modulated with a repeating PN code
• Cl. 4: correlator is a matched filter
• Cl. 5: combiner is a processor
• Cl. 6: parameter difference is a code phase difference
• Cl. 7: successive combining across segments until the signal/parameter is accurately and reliably detectable
• Cl. 8: parameter detected = code phase
• Cl. 9: parameter detected = presence of the signal
• Cl. 11: correlator comprises a matched filter

Uncertainty: The Espacenet rendering numbered entries "145.–157." (an artifact of its extraction) and displayed claims 1–13. I could not confirm with full authority whether the patent contains additional claims beyond 13 (e.g., 14–16). Treat the total claim count as unverified; the independent claims are 1, 10, 12, and 13.

The Underlying Technical Concept

The patent describes combining correlation results from non-uniform, disjoint GPS sample segments. Because each capture window starts at an arbitrary offset from a frame mark, each has a different (unknown) code phase. The patent teaches computing the relative code-phase shift from the Doppler hypothesis and the time offset between captures:

• ΔCP = [(F_PN + D) × ΔT] mod 1 ms
• ΔT = (T_n − T_2 + OS_1 − OS_2) × S

The second segment's correlation row is circularly shifted by ΔCP and added element-by-element to a cumulative array. This lets a GPS receiver in a mobile phone accumulate signal energy during idle (non-transmit) windows of varying length until SNR is sufficient for detection — a technique it also credits for reducing processor RAM/throughput demands. It is related to US 6,044,105 ("Doppler corrected spread spectrum matched filter") and US 6,496,145 B2 ("Signal detector employing coherent integration"), and a sibling filing led to US 6,850,557.

Litigation / CAFC 2026 Docket

• Google Patents flags this patent family as having litigation ("Family has litigation," Darts-IP family 23078602): Darts-IP link.
• No authoritative CAFC 2026 case tied to US 6,636,178 was found. My searches for CAFC 2026 dockets returned only unrelated matters (e.g., Ironburg Inventions Ltd. v. Valve Corp., No. 2024-2088, decided 2026-06-18, an IPR-estoppel case with no connection to this patent). Queries specifically pairing "6,636,178" with Federal Circuit returned no matching results, and a further targeted query was cut off before completion.
• The patent's anticipated expiration is 2019-03-30 (status: Expired – Lifetime). While an appeal involving an expired patent is legally possible, a live 2026 CAFC docket on this number would be unusual and I have no evidence of one.

Conclusion on docket question: I cannot confirm any 2026 CAFC docket naming US 6,636,178. The family has known litigation history (recorded in the Darts-IP database), but I could not retrieve case names, numbers, or outcomes — that requires direct PACER/CAFC docket or Darts-IP access, which I do not have here.

Confidence Summary

• High confidence: title, inventor, application number, filing/priority/issue dates, assignee history, abstract, CPC classes, expired status, and the substance of independent claims 1, 10, 12, 13.
• Medium confidence: exact total claim count (verified only through claim 13).
• Low/no confidence: existence of any specific 2026 CAFC appeal involving this patent; specific litigation case numbers involving this patent.

I'll research litigation involving US patent 6636178 across patent litigation databases.

Litigation Involving U.S. Patent No. 6,636,178

Summary: I found one instance of litigation that specifically names U.S. Patent No. 6,636,178 as an asserted patent — the GPS chip patent dispute between SiRF Technology and Global Locate filed in late 2006. Google Patents' bibliographic record for US 6,636,178 also carries a "Family has litigation" flag (per Darts‑ip family 23078602), which is consistent with this being the family‑level dispute. I could not confirm any other district‑court or ITC case that asserts the '178 patent itself.

Case 1 — SiRF Technology, Inc. v. Global Locate, Inc. (and distributor)

Grounding:

• A Chinese-language report of the suit, stating SiRF filed against Global Locate and its distributors in the U.S. District Court for the Central District of California, asserting U.S. Patents 7,091,904; 7,043,363; 6,850,557; and 6,636,178, and seeking damages and an injunction: http://unitalen.com/html/report/16113197-1.htm (and a mirror at https://www.lawtime.cn/info/zhuanli/zlnews/2011071267127.html).
• A secondary analysis of the SiRF/Global Locate dispute (from beidou.gov.cn) that gives the C.D. Cal. case docket number 4:2006cv06964 and states SiRF's motivation was TomTom's decision to source GPS chips from Global Locate for its "TomTom ONE" product: http://www.beidou.gov.cn/zt/zscq/[201710](/patent/201710)/t20171011_4248.html.
• Google Patents record for US 6,636,178, which flags "Family has litigation" (Darts‑ip family 23078602): https://patents.google.com/patent/US6636178/en.

Related proceedings (context only — the '178 patent is not listed as asserted)

These are part of the same SiRF/Global Locate campaign but do not name the '178 patent, so I list them only to avoid confusion:

1. ITC Inv. No. 337‑TA‑602, In re Certain GPS Devices and Products Containing Same — complaint filed by Global Locate (later joined by Broadcom) against SiRF, Pharos, MiTAC, Mio, and E‑TEN. Asserted patents: U.S. 6,417,801; 6,606,346; 6,651,000; 6,704,651; 6,937,187; 7,158,080. The '178 patent is not among them. Final determination of violation Jan. 15, 2009; affirmed in SiRF Tech., Inc. v. ITC, No. 2009‑1262 (Fed. Cir. Apr. 12, 2010). Sources: https://www.govinfo.gov/content/pkg/FR-2009-01-23/pdf/E9-1428.pdf and http://cafc.uscourts.gov/opinions-orders/09-1262.pdf.

2. ITC Inv. No. 337‑TA‑596, In re Certain GPS Chips, Associated Software and Systems, and Products Containing Same — SiRF as complainant against Global Locate. An expert declaration in a later PTAB filing lists the SiRF‑side patents‑in‑suit as including U.S. 6,304,216 (the parent of the '178 patent — the '178 patent is a continuation of Ser. No. 09/281,741, now U.S. 6,304,216), 7,043,363, 7,091,904, 7,132,980, and 7,158,080 — but does not expressly list 6,636,178. Source (Ambato Media petition exhibit): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1554022](/patent/1554022)/download-documents?artifactId=m50HwLQF2mHSppBwnpL2hBuiUKGSFlP2it0FKm-lYKhJjEGL65ESZoc.

3. Broadcom Corp. v. SiRF Technology, Inc., C.D. Cal., filed May 14, 2008 — Broadcom asserted four of its own patents (not the '178 patent). Source: SiRF Form 10‑K, https://www.sec.gov/Archives/edgar/data/[1163943](/patent/1163943)/000119312509041380/d10k.htm.

Caveats and confidence

• Confidence in the core finding (SiRF v. Global Locate, C.D. Cal. 4:2006‑cv‑06964, asserting the '178 patent): moderate‑to‑high. The asserted patent list comes from a contemporaneous press summary (SiRF's announcement was widely reported at the time). The docket number comes from a single secondary Chinese source and I was not able to independently verify it against PACER, the Unified Patents litigation portal (https://portal.unifiedpatents.com/litigation/caselist), CourtListener/RECAP, or Darts‑ip during this session. Per the strict rule, I am reporting the number literally as stated in that source (4:2006cv06964) without correction.
• Outcome of the '178 case is unverified. I did not locate a disposition (settlement, dismissal, judgment, or consolidation). The broader SiRF/Global Locate confrontation ended in corporate consolidation rather than a clean adjudication — Broadcom acquired Global Locate in July 2007, and CSR plc acquired SiRF in 2009 — which typically moots such disputes, but I am not asserting that as the case outcome.
• No later NPE litigation found. After CSR Technology (2011) and Samsung Electronics (2012) acquired the '178 patent, I found no evidence that the '178 patent was asserted in any subsequent district‑court action, ITC investigation, or PTAB proceeding. However, I could not directly query the Unified Patents litigation case list or full PACER/Darts‑ip records, so this negative is based on the searches I was able to run.
• I could not verify any litigation in jurisdictions outside the U.S. involving US 6,636,178 specifically.

If you need the docket number and outcome confirmed to a litigation‑grade standard, the next step would be a PACER pull for C.D. Cal. Case No. 4:2006‑cv‑06964 (and a check of the Darts‑ip family 23078602 via the link on the Google Patents page).

I'll verify the structured data against public sources before writing anything up.

PTAB Activity Report — U.S. Patent No. 6,636,178

Patent: Signal detector employing correlation analysis of non-uniform and disjoint sample segments
Application: 09/971,293 · filed 2001-10-04 · granted 2003-10-21
Priority: 1999-03-30 (continuation of Ser. No. 09/281,741, now U.S. Pat. No. 6,304,216)
Original assignee: SiRF Technology, Inc. → CSR Technology Inc. → Samsung Electronics Co., Ltd. (2012-10-08 assignment)
Legal status: Expired – Lifetime; anticipated expiration 2019-03-30
Source of record: USPTO Open Data Portal structured "PTAB proceedings on file" block (returned no AIA trial proceedings as of latest ingest); corroborated by Google Patents, https://patents.google.com/patent/[US6636178](/patent/US6636178)/en

Proceedings overview

Total AIA trial proceedings on file: 0. No IPRs, no PGRs, no CBM reviews — nothing active, nothing instituted-then-settled, nothing where claims were canceled or sustained. Because the count is zero, there is no breakdown by outcome to report, and the bottom-line defensive posture is unusual and worth stating precisely: this is not a "hardened" patent that has survived attacks — it is an unattacked patent, and it is also an expired one. No PTAB panel has ever construed a single claim of 6,636,178, so there is no estoppel, no claim-construction record, and no FWD to point to. That cuts both ways for a defendant: you get no free invalidity roadmap, but the patent owner equally gets no affirmation of validity to wave at you. The dominant defensive fact is the 2019-03-30 expiration — the patent is in the public domain, and any live dispute can only be about pre-expiration past damages, not prospective infringement or injunctive relief.

No proceedings to summarize

I searched for AIA trial activity and found none. Specifically:

• The structured ODP block supplied with this task reports no AIA trial proceedings.
• Web searches for an IPR/PGR/CBM involving 6,636,178, and for a "Samsung v. CSR/SiRF," "Unified Patents," or similar challenge to this patent, returned no PTAB proceeding on this patent. Search hits referencing "6636178" resolved to an unrelated Japanese patent (JP 6636178 B2, a Siemens turbine-diagnostics patent) and to Unified Patents' patent profile pages — which are library entries, not proceedings.
• The only proceeding-adjacent signal in the record is Google Patents' "Family has litigation" flag, via Darts-ip family ID 23078602. That is a litigation flag for the family (district court and/or ITC), not a PTAB trial. I could not tie it to a specific case number from the sources retrieved, so I am not naming a case I did not verify.

Confidence and caveat: I did not have direct query access to PTAB E2E or the PTAB Decisions database in this session, so this conclusion rests on the ODP structured data plus web search. A zero count is a negative finding, and negative findings are the easiest kind to get wrong. Before relying on it in a brief or an opinion letter, confirm independently at:

• PTAB E2E: https://e2e.uspto.gov/ (search by patent number 6636178)
• USPTO PTAB Decisions: https://developer.uspto.gov/ptab-web/
• Patent Public Search "PTAB" tab: https://ppubs.uspto.gov/

What I will not do is manufacture a proceeding number to fill the template. There is no ### IPR20XX-XXXXX section below because there is no such proceeding.

Strategic summary

Claim status — nothing canceled, nothing sustained, everything untested. All claims stand unadjudicated: independent claims 1, 10, 12 and 13, and their dependents through claim 13 (see the claim text at Espacenet, https://worldwide.espacenet.com/publicationDetails/claims?CC=US&NR=[6636178B2](/patent/6636178B2)&KC=B2&FT=D). There is no IPR-narrowed claim set to work from and no surviving-claims list, because no tribunal has narrowed anything. If you are being asserted against on this patent today, the validity fight is entirely un-fought — you would be the first challenger, with a clean slate and no adverse claim-construction precedent binding you.

Estoppel landscape — there is none, which is the good news for a defendant. IPR estoppel under 35 U.S.C. § 315(e)(2) runs only against "the petitioner, or the real party in interest, or privy of the petitioner" in an IPR that "results in a final written decision." With zero proceedings, no party is estopped from anything. Every printed publication and patent you can find is available as an invalidity ground, in the PTAB or in district court, without the SAS/Enthen estoppel analysis that normally consumes a defense budget. Practically, that means you are not confined to "prior art a petitioner reasonably could have raised" — there is no petitioner.

Pattern signals — none of the usual ones fire. No serial petitioner, no defensive aggregator (Unified Patents, RPX, et al.) in the chain for this patent, no patent-owner appeal history to the Federal Circuit arising from a PTAB case on this patent. The current assignee, Samsung Electronics, acquired the portfolio through the 2012 CSR assignment but has not had to defend it at the PTAB. Note the related-family context: the parent, U.S. Pat. No. 6,304,216, and the co-filed U.S. Pat. No. 6,496,145 ("Signal detector employing coherent integration") share this specification, and the 1998 U.S. Pat. No. 6,044,105 is incorporated by reference. I did not verify PTAB activity on those siblings in this session — if you are clearing a product against the SiRF/CSR GPS acquisition portfolio, check the whole family separately, because a challenge to a sibling may generate estoppel or art findings that matter here.

Recommended next steps

If you are a defendant:

1. Run the expiration math first. The patent expired 2019-03-30 and is marked "Expired – Lifetime" in the ODP data. There is no prospective injunction and no ongoing royalty. Any exposure is limited to past damages within the 6-year lookback of 35 U.S.C. § 286, measured from the complaint. Get the asserted damages period on the record early; a demand letter citing this patent for current products is asserting a right that no longer exists.
2. Confirm the zero-PTAB finding yourself at PTAB E2E and the PTAB Decisions database (links above) before you tell a court or an adversary that no proceeding exists.
3. Because nothing is estopped, build the invalidity case from scratch. The relevant prior art is the GPS receiver art of the mid-1990s: Krasner (U.S. Pat. Nos. 5,874,914 and 5,841,396), Lomp (U.S. Pat. No. 5,912,919), Davis (U.S. Pat. No. 5,877,724), Lau (U.S. Pat. No. 5,883,594), and Kondo (U.S. Pat. No. 5,832,021) all appear on the face of this patent as cited references — a useful starting map, but the Examiner's citations are not an invalidity opinion.
4. Watch the § 315(b) one-year clock if (and only if) a proceeding ever appears. There is none today, so the clock has not started. But if the patent owner files suit over pre-expiration conduct, and a co-defendant files an IPR, you would have one year from service of your complaint to file — and intervening rights under § 252 become relevant on any amended claim.
5. Do not cite an FWD. There is no Final Written Decision for this patent. If opposing counsel hands you one, verify the patent number — the "6636178" string collides with a Japanese Siemens patent, and it would be an easy error to import the wrong document.

If you are the patent owner's counsel: the absence of PTAB activity is not an endorsement of the claims. Validity has never been tested by an Article III or PTAB tribunal on this patent, and the specification's heavy reliance on the "Magna"/"Scorpio" implementation example and the incorporated U.S. Pat. No. 6,044,105 means written-description and § 112 positions on the combining algorithm (equations (1), (2), and (6)) are untested as well.

Nothing above is legal advice, and no proceeding number, panel composition, institution date, or claim-level disposition has been invented or inferred. Where the record is silent, this report says so.

I'll reconstruct this from the authoritative patent text first, then try to corroborate the assignment records and reel/frame numbers with searches.

I have the patent's internal legal-events record. Now let me try to pin down real reel/frame numbers and correspondents from the USPTO assignment records and corroborating sources.

Let me try to pin down the actual reel/frame records and check for any litigation or later transfers.

Ownership & Assignment Analysis — US 6,636,178 B2

Patent: US 6,636,178 B2 — "Signal detector employing correlation analysis of non-uniform and disjoint sample segments"
App. No.: 09/971,293 · filed 2001-10-04 · granted 2003-10-21
Priority: 1999-03-30 (continuation of App. 09/281,741, now US 6,304,216)
Family: US 6,304,216 (parent), US 7,002,516, US 6,496,145 (coherent integration sibling), EP 1,173,778, WO 00/58746, JP 3964621, TW 500927, ATE 418079
Status: Expired – Lifetime (anticipated expiration 2019-03-30, per Google Patents legal events)

Data-quality note up front (please read before relying on the timeline): I was able to retrieve the existence and sequence of the recorded assignment events, but not the reel/frame numbers or the correspondent-of-record fields. Every field below marked "not retrieved" is a genuine gap in the sources I could reach in this session, not an absence of a record. Do not treat the missing reel/frames as meaning "no assignment" — the ODP/assignment legal-events mirror on Google Patents clearly shows post-issuance transfers. Verify at the USPTO Patent Assignment Center before filing anything that depends on reel/frame precision.

Inventors

Unusual-pattern check: The inventor-side history is not a classic "all inventors bail within 12 months" case — there is only one inventor and the rights did not move until ~4 years after the 1999 priority filing. What is notable is the unit-of-work pattern: the invention was made inside Conexant's GPS operation, and the patent rights followed the business unit, not the individual, when Conexant's GPS business was sold to SiRF in 2003. I could not verify whether Gronemeyer personally transferred to SiRF with the unit (he appears on many downstream SiRF-era GPS filings, which is consistent with a transfer, but I have no primary-source confirmation in the sources reviewed — flag as unverified).

Original assignee

SiRF Technology, Inc. — named on the face of the issued patent (freePatentsOnline lists "Assignee: SiRF Technology, Inc."; Espacenet lists applicant "SIRF TECH INC [US]").

• Primary line of business: Fabless GPS receiver/chipset vendor — the SiRFstar family of GPS chipsets used in handsets, PNDs, and automotive. This patent is core acquisition/signal-detection technology (matched-filter correlation across disjoint sampling windows), i.e., the exact capability a handset-integrated GPS receiver needs to accumulate SNR in the gaps between cellular transmit bursts.
• Shipped a product embodying the claims? Yes — SiRF's operating business was commercial GPS silicon, and the specification itself describes the "Magna" (matched filter), "Scorpio" (processor), and "Gemini/Pices Monopack" (RF) chips as product implementations. (Those specific chips were developed under Conexant; they passed to SiRF with the 2003 GPS-business acquisition.)
• Current status: The original assignee no longer exists as such. SiRF Technology Holdings, Inc. merged into CSR plc in 2009 (deal announced 2009-02-10; completed 2009-07-24) after being described in trade press as financially troubled. The US operating entity was renamed CSR Technology Inc. (change-of-name recorded 2011-12-22). CSR plc was subsequently absorbed by Qualcomm (2015). CSR's handset connectivity/location business — including the GPS patent group relevant here — was sold to Samsung Electronics Co., Ltd. in 2012.

Assignment timeline

Chronological, from the ODP/Google Patents legal-events mirror of the USPTO assignment records. Reel/frame and correspondent fields could not be retrieved in this session and are flagged accordingly.

1. 2002-10-22 (recorded/effected) — Reel/frame: not retrieved

   • Conveyance: Assignment (assignment of assignors' interest)
   • Assignor: SiRF Technologies Holdings, Inc.
   • Assignee: SiRF Technology, Inc.
   • Correspondent: not retrieved
   • Context: Internal reorganisation / intra-group transfer between affiliated SiRF entities (holding entity → operating entity).

2. 2003-01-07 (recorded/effected) — Reel/frame: not retrieved

   • Conveyance: Assignment
   • Assignor: SiRF Technology Holdings, Inc.
   • Assignee: SiRF Technology, Inc.
   • Correspondent: not retrieved
   • Context: Internal reorganisation / intra-group transfer (the second leg of the same holding-to-operating tidy-up; note the assignor name here is spelled "TECHNOLOGY" while the 2002-10-22 record spells it "TECHNOLOGIES" — see anomaly note below).

3. 2003-01-07 (recorded/effected) — Reel/frame: not retrieved

   • Conveyance: Assignment
   • Assignor: Conexant Systems, Inc.
   • Assignee: SiRF Technology Holdings, Inc.
   • Correspondent: not retrieved
   • Context: Asset acquisition — SiRF's purchase of Conexant's GPS business in 2003; this is the link that brought the patent from the original inventive entity (Conexant) into the SiRF family. Corroborated by SiRF's own corporate statements ("early GPS patents that SiRF acquired through its acquisition of Conexant's GPS business in 2003").

4. 2011-12-22 (recorded) — Reel/frame: not retrieved

   • Conveyance: Change of Name (not a transfer of title)
   • Assignor: SiRF Technology, Inc.
   • Assignee: CSR Technology Inc.
   • Correspondent: not retrieved
   • Context: Change of name only — CSR's rebranding of the acquired SiRF US entity following the 2009 merger; no change in beneficial ownership.

5. 2012-10-08 (recorded/effected) — Reel/frame: not retrieved

   • Conveyance: Assignment
   • Assignor: CSR Technology Inc.
   • Assignee: [Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd.
   • Correspondent: not retrieved
   • Context: Carve-out M&A / strategic patent transfer — the 17 July 2012 Samsung–CSR transaction (US$310 M for CSR's handset connectivity & location business + 21 US connectivity patents and their international counterparts, with a royalty-free licence back to CSR). Closing was "about three months" out, which matches an October 2012 recordation. This is an operating-company-to-operating-company sale, not a transfer to an asserter.

6. 2019-03-30 — anticipated expiration (statutory term), not an assignment.

Anomaly to verify: Google Patents lists two current assignees — Samsung Electronics Co., Ltd. and CSR Technology Holdings Inc. The latter is likely an artefact of the change-of-name record and/or a retained-licence entry rather than co-ownership, but I could not confirm the legal effect without the reel/frame and the recorded cover sheet. Also note the assignor-name inconsistency ("SiRF Technologies Holdings, Inc." vs "SiRF Technology Holdings, Inc.") across the 2002-10-22 and 2003-01-07 records; these are probably the same entity with a database typo, but the Assignment Center images would settle it.

Corroborating SEC/press context (not assignment records, but consistent):

• SiRF–CSR merger announced 2009-02-10, completed 2009-07-24 (SiRF described in trade press as financially distressed).
• Samsung–CSR handset connectivity/location transaction announced 2012-07-17; 21 US patents + international counterparts transferred, licence back to CSR.
• SiRF and Qualcomm signed a mutual patent non-assertion agreement (Oct 2008) covering their respective GPS portfolios — relevant because it shows SiRF's patents were being used defensively/cross-licence-wise by an operating company, not monetised through an NPE.

Timeline diagram

timeline
    title Ownership of US 6636178
    1999 : Priority application filed
         : Invented in Conexant GPS unit
    2001 : Continuation application filed
    2002 : SiRF Holdings to SiRF Technology
    2003 : Patent issued Oct 21
         : Conexant GPS unit sold to SiRF
         : SiRF Holdings to SiRF Technology
    2009 : SiRF merged into CSR plc
    2011 : SiRF renamed CSR Technology Inc
    2012 : 21 patents sold to Samsung
    2019 : Patent expired

NPE / troll-pattern signals

Verdict

Operating-company assertion

The full recorded chain runs Conexant Systems → SiRF Technology Holdings → SiRF Technology → CSR Technology Inc. (change of name, 2011-12-22) → Samsung Electronics Co., Ltd. (2012-10-08), with the only clustering being routine intra-group SiRF transfers on 2002-10-22 and 2003-01-07. Every assignee is a shipping, product-generating semiconductor/GPS company, and the terminal assignee (Samsung) is a vertically integrated manufacturer that acquired this patent group as part of the July 2012 CSR handset connectivity/location carve-out (21 US patents plus international counterparts, with a licence back to CSR) — i.e., the patent was acquired to be practised, not to be asserted by a licensing shell. Caveat: the "assertion" limb of this label is unproven for this specific patent — I found no suit naming US 6,636,178 — so read this as operating-company ownership with no NPE link, and note that the correspondent signal (item 3) is unresolved and could only ever reinforce the not-an-NPE conclusion.

Verify at: USPTO Patent Assignment Center — https://assignmentcenter.uspto.gov/ (alternate interface: https://assignment.uspto.gov/patent/index.html). Search by patent number 6636178 to pull the reel/frame images and correspondent-of-record fields this report could not obtain.

Prior-Art Analysis — US 6,636,178 B2

Important scope/limitation statement up front. A literal patent-number search of the public USPTO/patent databases for 6636178 resolves unambiguously to US 6,636,178 B2, "Signal detector employing correlation analysis of non-uniform and disjoint sample segments" (SiRF Technology, Inc.), which matches the record already built in the prior sections. I was not able to retrieve the complete front-page "References Cited / U.S. Patent Documents" list for US 6,636,178 B2 itself within this session (the Google Patents full-text view, Espacenet, Justia pre-grant publication and RPX Insight pages returned the description, claims and family data, but not a clean, authoritative rendering of the examiner's cited-reference block for the '178 patent). Under the operating rules I must not fabricate that list. What follows therefore separates (A) what I could confirm and (B) what I could not, and only maps §102 where I have a firsthand basis.

A. Confirmed bibliographic / prosecution-record facts about '178

CPC: G01S19/29, G01S19/30. Status: Expired – Lifetime (anticipated 2019-03-30).

B. References I could confirm are associated with this family

Because I could not obtain the '178 examiner citation block, the items below are the family-internal and incorporated references that the record itself establishes. I identify the correct host document for each so I do not mis-attribute another patent's citations to '178.

B1. US 6,044,105 — "Doppler Corrected Spread Spectrum Matched Filter"

• Citation: U.S. Pat. No. 6,044,105 (Ser. No. 09/145,055, filed 1998-09-01; issued 2000-03-28).
• Status relative to '178: expressively incorporated by reference in '178's specification (Figs. 5 / matched-filter Doppler-correction description and steps 109/1004).
• Brief description: Doppler-corrected spread-spectrum matched filter; samples corrected for Doppler hypotheses before correlation against a PN code.
• §102 potential: As a document with an earlier filing date (1998-09-01) than '178's 1999-03-30 priority, it is eligible as §102(e) art, but only for subject matter it actually discloses. Its disclosure is directed to Doppler correction in a matched filter — i.e., structure recited in the dependent claims (e.g., claim 4 "the correlator is a matched filter"; claim 11 equivalents), not to the combining-of-disjoint-segments logic of independent claims 1, 10, 12, 13. In practice it functions as incorporated subject matter rather than an anticipation reference.

B2. US 6,496,145 B2 — "Signal Detector Employing Coherent Integration"

• Citation: U.S. Pat. No. 6,496,145 B2 (Ser. No. 09/971,297, filed 2001-10-04; issued 2002-12-17).
• Status: expressively incorporated by reference; same-day filing as '178.
• §102 potential: Because it shares the same priority/filing timing, it is not available as §102(e) art against '178 on its independent claims, and its subject matter (coherent integration) is distinct from the disjoint-segment combining claimed here. Low §102 relevance.

B3. US 6,304,216 B1 — the parent (Ser. No. 09/281,741)

• Filed 1999-03-30, issued 2001-10-16. This is the parent of '178 (same disclosure). It is the priority document, not prior art to '178.

B4. Foreign / PCT family members (search-report context only)

Espacenet INPADOC lists the same-invention family: WO 00/58746 A1, EP 1 173 778 A1/A4/B1, JP 2002540434 A / JP 3964621 B2, TW 500927 B, AU 9659701 A. Their examination (e.g., during EP 1 173 778) is where prior art citations for this invention actually surface in a searchable form — but those are the EPO's citations, and I could not pull the EP 1 173 778 search report in this session, so I cannot present them as "citations for 6636178."

B5. Sibling/continuation family

The '178 disclosure also gave rise to/relates to US 7,002,516 B2 (from US 2005/0035905 A1, "…non-uniform and disjoint sample segments") and US 6,850,557 — these are forward family members, i.e., they cite '178, not vice-versa.

Forward citations (for completeness, not prior art): Google Patents lists only one citing patent for '178 itself, US 9,444,539 B2 (Locata Corp., 2016); the broader "families citing this family" list (72 entries) includes US 6,546,040 (Trimble), US 6,301,545, US 6,542,538, etc. These are later than '178 and cannot anticipate it.

C. What a §102 mapping requires — and the honest gap

To do the task exactly as specified ("look at each patent citation for 6636178 … which claim(s) it potentially anticipates under 35 U.S.C. § 102"), I need the examiner's cited-references block (front page, field 56) of US 6,636,178 B2. I did not retrieve it. Consequently I will not invent specific US patents and pin §102 assertions on them, because doing so would risk exactly the fabrication the rules prohibit.

Two categories I can still speak to with confidence:

1. The named/incorporated references (B1–B2). Independent claims 1, 10, 12, 13 of '178 are directed to receiving first and second time-separated segments, deriving correlation data for each, determining a parameter (code-phase) difference, adjusting one data set, and combining into cumulative correlation data. Neither '105 nor '145 discloses that segment-difference-and-combine step, so neither anticipates claims 1/10/12/13. At most, '105 touches dependent claim 4 / claim 11 ("correlator is a matched filter") and the Doppler-correction subject matter.

2. The parent '216 is not art (it is the priority document), so a §102 attack on '178 cannot rest on it.

D. Why the family-level prior art surface is worth a targeted follow-up

The known family litigation flag (Darts-IP family 23078602) and the fact that '178 was asserted-era SiRF/Conexant GPS IP means the real §102/§103 references most likely appear in the EP 1 173 778 B1 grant file and the US 2002/0041252 A1 prosecution. Two reference classes recur in the GPS matched-filter art of this vintage and are the natural §102 candidates if they appear on '178's front page: (i) fast-acquisition / high-sensitivity GPS receiver patents (e.g., Krasner-family "Fast Acquisition, High Sensitivity GPS Receiver," Ser. No. 09/021,854, filed 1998-02-11) and (ii) Doppler-corrected / matched-filter GPS correlator patents (e.g., related spread-spectrum matched-filter work cited in the family's EP prosecution such as WO 97/14053 and US 5,577,025). I am flagging these as leads to verify, not as confirmed §102 references for '178 — I have not verified them against '178's field-56 list.

E. Confidence summary

• High confidence: number→US 6,636,178 B2 mapping; continuation/parent relationship to US 6,304,216; the two incorporated-by-reference patents (US 6,044,105; US 6,496,145 B2) and their dates; the foreign family list; independent claims 1, 10, 12, 13 and their scope.
• Medium confidence: that '178 issued with only the family's internal references plus a small examiner set (inferred from the sparse forward-citation profile, not confirmed).
• Low / no confidence — explicitly not asserted: any specific front-page examiner reference list, and any specific §102 anticipation mapping built on references I could not verify were cited. I did not retrieve the '178 field-56 citation block within this session, so I am not supplying a fabricated per-reference §102 table.

Recommended next step (single targeted pull): open the USPTO PatentCenter/Public PAIR "References Cited" view for application 09/971,293 (or the granted '178 front page) and the EP 1 173 778 B1 search/opposition file; those two documents will contain the authoritative citation lists the task asks for, after which a per-reference §102 mapping to claims 1, 10, 12, 13 (and dependents 4, 6–9, 11) can be completed reliably.

I'll analyze obviousness of US 6,636,178. First, a necessary flag, then research.

Important gap: The task says to "use the results from the Prior Art section of this page," but the previously generated sections supplied to me are only the Patent summary and Litigation summary — there is no Prior Art section in what I received. I'll flag this contradiction rather than invent one. To ground the § 103 analysis, let me retrieve the reference set actually cited on the patent's face and look for any post-grant challenges that mapped prior art.

I have the full claim set now (18 claims per RPX/Justia — a discrepancy with the earlier summary's ~13, which I'll flag). Let me pull the patent's actual cited references and check for reexamination.

Obviousness Analysis — US 6,636,178 B2 under 35 U.S.C. § 103

I. Two threshold caveats before the analysis

1. The "Prior Art section" referenced in the task was not supplied. The previously generated sections I received are the Patent summary and the Litigation summary. Neither contains a Prior Art section. Rather than fabricate one, I reconstructed the prior-art record from the authoritative source documents themselves — the "References Cited" list on the face of US 6,636,178 (retrieved from FreePatentsOnline and Justia) and the patent's Related Applications section. I flag this as a contradiction between the instruction and the supplied materials.

2. Claim-count discrepancy (flagged, not auto-corrected). The earlier summary said independent claims are 1, 10, 12, 13 and that it "could not confirm" claims beyond 13. The RPX/Insight and Justia records now show 18 claims, adding independent claims 14 (means-plus-function system), 15 (system with "parameter determiner"), 17 (method), and 18 (computer-readable medium), plus dependent claim 16. This discrepancy may reflect a reexamination certificate, or simply the earlier extraction stopping early. I could not verify a reexamination certificate in this session (the confirming search was cut off), so I report the 18-claim set as found without correcting the earlier statement, and analyze both readings.

3. Practical posture. The patent expired 2019-03-30 (status "Expired – Lifetime"). A § 103 analysis is therefore hypothetical/academic; there is no live IPR or validity challenge available. I found no PTAB proceeding asserting this patent. This is a technical-analyst opinion, not a legal conclusion.

4. Effective prior-art date. Priority 1999-03-30 (or 1999-03-29 per Unified Patents — discrepancy preserved). Prior art must predate that date under § 102(a)/(b), or qualify under § 102(e) as US filings. Several face-cited patents issued after the priority date (e.g., US 6,091,361 issued 2000-07-18; US 6,075,808 issued 2000-06-13) and therefore qualify only as § 102(e) art based on their earlier US filing dates — a point I cannot fully verify here.

II. The prior-art landscape on the face of US 6,636,178

The examiner-cited references cluster into four groups, all in the same field (spread-spectrum/GPS signal acquisition, correlation, and duty-cycled receiving):

(a) GPS/DS-CDMA receivers that capture discrete, time-separated signal segments (the "separate periods of time" element):

• US 5,781,156 (Krasner) — "GPS receiver and method for processing GPS signals" (snapshot processing)
• US 5,841,396 / 5,874,914 (Krasner) — GPS receiver utilizing a communication link
• US 5,812,087 (Krasner) — SPS-based time measurement
• US 5,883,594 (Lau) — GPS receiver using a message system for reducing power consumption
• US 6,075,808 (Tsujimoto) — "Spread spectrum time diversity communications system" (combining energy received in separated time intervals)
• US 5,854,605 (Gildea) — GPS receiver using data-bit timing for fast TTFF

(b) Correlators / matched filters generating correlation data against PN-code and code-phase hypotheses (the "correlator … generating said hypothesis" element):

• US 4,701,934 (Jasper), US 4,894,842 (Broekhoven), US 5,293,398 (Hamao), US 5,253,268 (Omura)
• US 5,832,021 (Kondo) — "Correlation detector for use in a spread spectrum communications receiver"
• US 5,912,919 (Lomp) — efficient multipath centroid tracking (combining correlation taps)

(c) Doppler/time-base handling and timing:

• US 5,878,725 (Kalafus) — WAAS retrofit receiver
• US 6,091,361 (Davis) — joint space-time array signal processing

(d) Non-patent literature: D.J.R. Van Nee & A.J.R.M. Coenen, "New Fast GPS Code-Acquisition Technique Using FFT," Electronics Letters, vol. 27, no. 2 (Jan. 17, 1991).

Commonly owned related applications (not prior art to the '178 claims, which share the priority chain): US 6,044,105 (Ser. 09/145,055, "Doppler corrected spread spectrum matched filter") and US 6,496,145 B2 (Ser. 09/971,297, "Signal detector employing coherent integration"). The parent US 6,304,216 is likewise not § 102 prior art (same family/priority).

External references not on the face (candidates, with date caveats):

• WO 99/41846 (Ericsson, pub. 1999-08-19) — teaches segmenting a sequence of samples, correlating each segment with a known sequence, and summing the correlation values to produce a total correlation output. Caveat: a publication date of Aug 1999 is after the Mar 1999 priority date; it would qualify only via an earlier US § 102(e) filing.
• US 5,579,338 (Kojima) and US 5,663,734 (Krasner, snapshot GPS) — adjacent art.
• Sources: FreePatentsOnline 6,636,178, Justia 6,636,178, Google Patents, RPX Insight.

III. Claim 1 — element-by-element mapping and combinations

Claim 1 recites three elements: (A) a receiver receiving first and second segments representing separate periods of time, the signal being "a signal of interest perturbed by noise or pseudo-noise"; (B) a correlator deriving first/second correlation data against a hypothesis it generates; (C) a combiner that determines a parameter difference, adjusts a selected one of the two data sets responsive to that difference, and combines the adjusted data with the other to obtain cumulative correlation data.

Mapping

The three strongest combinations

Combination 1 — Krasner + Kondo + Tsujimoto (all on the face).
Krasner's snapshot receiver supplies the "finite segment captured at an arbitrary time" and the weak-signal/noise environment; Kondo supplies hypothesis-sweeping correlation; Tsujimoto supplies the concept of combining energy from separated time intervals to improve SNR. A POSITA seeking to detect weak GPS signals within short, intermittently available capture windows (e.g., between cellular transmit bursts) would combine these to sum correlation energy across snapshots. This maps cleanly onto (A)+(B) and the "cumulative correlation data" of (C).

Combination 2 — Tsujimoto + Kondo + a code-phase-propagation teaching (Gildea '605 or the admitted art).
Time-diversity combining is only physically meaningful once the relative code phase between separated intervals is accounted for — otherwise the PN codes add destructively. Gildea '605 (data-bit timing to achieve fast TTFF) and the general GPS/DS-CDMA knowledge that code phase advances at the chipping rate plus Doppler (F_PN + D) over the elapsed interval (ΔT) supply the "parameter difference" and "adjust" steps. This is the strongest single-combination theory, because it supplies a reason (destructive-vs-constructive addition) that itself motivates the adjustment.

Combination 3 — WO 99/41846 + Kondo + Krasner.
WO 99/41846 expressly teaches "segmenting the sequence of samples … correlating each of the plurality of segmented sequences … summing the plurality of correlation values to produce a total correlation output" — i.e., non-coherent accumulation of disjoint segment correlations, which is the literal "combining … to obtain cumulative correlation data." Add Krasner for the GPS/noise context and a code-phase/time-offset teaching for the "determine a parameter difference … adjust" limitation. Date caveat: WO 99/41846 was published 1999-08-19, after the 1999-03-30 priority; it can only be used if its underlying US filing predates that date as § 102(e) art — I could not verify this.

Motivation to combine (KSR / Graham)

• Same field, common problem: every reference addresses acquisition of weak spread-spectrum/GPS signals with limited sampling windows. Weak GPS signal detection is a recognized, long-standing problem (the patent itself notes the ~11,000-mile path loss and cross-satellite interference).
• Predictable result: non-coherent integration of correlation magnitudes to raise SNR is a textbook, expectation-of-success technique; KSR ("a known technique applied to a known device") and the "toolbox" line of cases apply directly.
• The adjustment step is a predictable variation: accounting for the fixed relationship ΔCP = [(F_PN + D)·ΔT] mod 1 ms is the routine kinematics of a periodic PN code; a POSITA would implement it without undue experimentation.

Claim 1 is therefore vulnerable to § 103 on Combination 2 (strongest), with Combinations 1 and 3 as supporting/alternative theories. The weakest link in the mapping is that the cited references do not expressly teach computing a relative code-phase difference and circularly shifting one correlation row before element-wise summation; that specific step is where a patentee would focus a non-obviousness argument (see § VII).

IV. Claims 10, 12, 13 (and 14, 15, 17, 18)

These are the PN-code-specific and method/system counterparts, adding "combined PN code and code phase hypothesis," "code phase difference between portions corresponding to the PN code hypothesis," and "generating said code phase hypothesis."

• The PN-code + code-phase hypothesis limitations are squarely met by the face-cited correlator references Kondo '021, Hamao '398, Broekhoven '842, and Jasper '934, each of which sweeps combined PN-code/code-phase hypotheses.
• The "portions corresponding to the PN code hypothesis" limitation (i.e., the per-(PN, Doppler) row structure) is a data-organization choice; Lomp '919 (combining correlation taps across hypotheses) and Kalafus '725 teach hypothesis-keyed correlation result structures.
• The method/CRM/system variants (12, 13, 17, 18) and means-plus-function claims (14, 15) recite the same functional steps; under § 112(f) they are met by the same structure and are obvious for the reasons above. Claim 18's computer-readable-media format adds only a conventional storage medium.

So the same Combination 2 disposes of claims 10–18, subject to the same "relative code-phase adjustment" caveat.

V. Dependent claims 2–9, 11, 16 — trivially obvious

Each adds only a design choice or a known GPS parameter; no independent § 103 argument is needed beyond Combination 2.

VI. Motivation-to-combine summary (the § 103 "glue")

1. Shared problem: weak-signal GPS acquisition within short, externally constrained capture windows — explicitly the patent's own stated motivation (mobile-phone transmit blanking).
2. Known solution architecture: snapshot/duty-cycled capture (Krasner, Lau) + matched-filter correlation (Jasper, Broekhoven, Kondo) + cross-segment energy combining (Tsujimoto, WO 99/41846).
3. Predictable adjustment: the code-phase advance over ΔT as a function of the nominal PN rate and Doppler is standard GPS/DS-CDMA knowledge; no new physical principle was required.
4. Reasonable expectation of success: non-coherent integration to raise SNR is routine; the combination yields nothing more than the sum of the references' known benefits (avoiding transmitter interference and increasing SNR).

VII. Where the obviousness case is weakest (candidate non-obviousness arguments)

To be balanced, the strongest rebuttal a patentee could raise:

• The relative-phase-combination without knowing absolute code phases. The '178's distinctive move is computing the relative code-phase difference between two captures from the elapsed time and Doppler hypothesis — without ever knowing either absolute code phase — and applying it as a circular shift of one correlation row, then element-wise adding. If the cited references combine segments only coherently (where absolute phase is known/estimated) or only non-coherently (magnitudes, phase discarded), the specific "shift-then-add" combination may not be squarely disclosed.
• The per-(PN, Doppler) row granularity, i.e., deriving a different code-phase shift for each Doppler hypothesis before combining, is a non-trivial implementation specific.
• Secondary considerations (Graham factors): SiRF's commercial success in GPS chips, the long-felt need for weak-signal GPS in handsets, and the reference in the earlier Litigation summary to the '178 being asserted in SiRF v. Global Locate (C.D. Cal. 4:2006-cv-06964) suggest the industry valued this family. However, no nexus is established here to the claimed subject matter, and commercial success is weak evidence on an expired patent.

I could not obtain the prosecution history (office actions/applicant arguments) in this session, so I cannot say which of these the examiner actually relied on to allow the claims.

VIII. Bottom line

• Most likely § 103 combination: Tsujimoto (US 6,075,808) + Kondo (US 5,832,021) + Gildea (US 5,854,605), optionally with Krasner (US 5,781,156) and/or WO 99/41846, renders claim 1 obvious, and by extension claims 2–18.
• Alternative combinations: Krasner + Kondo + Tsujimoto (all on the face), and WO 99/41846 + Kondo + Krasner (subject to the WO date caveat).
• Confidence: Moderate that claims 1 and 10–18 are obvious over this combination; high for the dependent claims and for the "system architecture" elements; lower that any single reference expressly discloses the relative-code-phase "shift-then-add" step, which is the real locus of patentability.
• Caveats honored: I did not auto-correct any patent number or date (both 1999-03-29 and 1999-03-30 preserved; the OSTI ID 6636178 / JP 6636178 look-alikes excluded). I flagged the missing Prior Art section and the 13-vs-18 claim-count discrepancy rather than silently resolving them. The 2026 CAFC docket question remains unanswered from the prior sections and is unaffected by this analysis.

If you can supply the actual "Prior Art section" intended by the task, or the prosecution-history office actions and any reexamination certificate, I can tighten this to the specific references and rejections actually on record.