US 5048057

I'll search for this specific patent number across the sources you mentioned.

Let me check specifically for litigation/appeal activity and the USPTO record.

US Patent 5,048,057 — "Wireless Local Area Network"

Important sourcing note: I do not have direct query access to USPTO PatentCenter/PAIR or to the CAFC 2026 argument/opinion dockets; I relied on web searches plus the full authoritative patent text you supplied. Everything below is grounded in those sources. Where I could not verify something, I say so.

Bibliographic data

Note: the title is literally "Wireless local area network" (singular), even though the disclosure is often described elsewhere as a wireless LAN system. Likewise the EP family member carries a different title ("Maximum likelihood decoding with interference detection for frequency-hopped coded modulation systems"), and the JP member is titled "Data communication system and method."

Abstract (as granted)

"In a wireless local area network, predetermined interference-detection sequences of signal points, known at both the transmitter and receiver, are transmitted over a plurality of hopping frequencies along with a sequence of data signal points taken from a modulated code exhibiting built-in diversity. In the receiver, the extent to which there are errors in the received interference-detection signal points is used as a mechanism for identifying the likely occurrence of strong interference. If there are no, or few, errors, decoding proceeds normally. However, if the number of errors is above a predetermined threshold, it is assumed that the transmitted signal was subjected to significant interference and an erasure is declared. This is used as 'side' information by the decoder to improve its ability to accurately recover data in the presence of interference."

Plain-language overview

Problem: Indoor radio LANs suffer both multipath fading and interference from other systems sharing the band (e.g., the FCC ISM band). Frequency hopping plus coding mitigates fading, but conventional coding loses effectiveness when a strong interferer swamps a hop.

Solution: Combine three things the inventors say had not been combined before — (1) coded modulation with built-in diversity (per the copending Wei application Ser. No. 386,185, "Coded Modulation for Mobile Radio"), (2) frequency hopping that spreads a codeword's diversity-bearing signal points across independently fading frequencies, and (3) soft-decision decoding fed with an erasure signal.

Mechanism: A known, predetermined "interference-detection" sequence of signal points is sent alongside the data on each hop. The receiver hard-decides each of these known points and counts errors. Few errors → normal soft-decision decoding using actual received values. Many errors → raise an erasure flag; the affected received points are treated as being at the origin (effectively ignored / set to zero) so a corrupted point (e.g., point E in FIG. 6, which is far from the transmitted point and would bias the decoder) does not poison the decision.

Key design choices in the spec: The detection points may be sent as a preamble per hop (FIG. 3), but the preferred embodiment distributes them non-time-contiguously among the data points on each frequency (FIG. 4), so an interference burst occurring after a preamble can still trigger an erasure. Illustrative codes are 4-PSK-based block codes: Code I (8D, 2-fold diversity, MDX=4), Code II (16D codeword-repetition, X=4, MDX=8), Code III (8D, X=3, MDX=8, lower rate), Code IV (16D repetition of Code III, X=6, MDX=16). Decoding proceeds by "2D point metrics" (squared Euclidean distances), then 4D subset metrics, then 8D subset metrics, and is explicitly described as parallelizable. Variations include space diversity via multiple receive antennas, and M-DPSK instead of M-PSK.

Independent claims in plain language

Claim 1 (apparatus — means-plus-function form). An apparatus having four cooperating elements:

1. Generator: means responsive to a stream of input data that produces a sequence of coded modulation signal points, where the code has built-in diversity;
2. Transmitter: means for sending those signal points to a receiver, with at least the diversity-providing points carried on respective different frequencies;
3. Interference detector: means for detecting which received coded modulation signal points are likely to have been subject to strong interference; and
4. Soft-decision decoder: means for soft-decision decoding based on the received values of the signal points other than the ones detected as likely interfered-with, to decide what was transmitted.
   (In effect: soft decoding that deliberately excludes/neutralizes the erased points.)

Claim 8 (method — mirrors claim 1). A method with the same four steps: (a) generating coded modulation signal points from a code with built-in diversity; (b) transmitting them, with at least the diversity-providing points on respective different frequencies; (c) detecting received coded modulation signal points likely subject to strong interference; and (d) soft-decision decoding responsive to the received values of the points other than the detected ones, to decide the transmitted values.

Dependent claims 2–7 add: that the transmission is arranged so strong interference produces a detectable particular characteristic (cl. 2); that predetermined interference-detection signal points are transmitted along with the data (cl. 3); that the interference-indicative signal is a function of the error rate of those detection points (cl. 4); that detection points for each block are transmitted non-time-contiguously (cl. 5); that a block and its detection points share a single frequency (cl. 6); and that the indicator is a function of the error rate for each block (cl. 7). Dependent claims 9–14 are the method-side counterparts of claims 2–7.

Litigation / CAFC 2026 docket check

• I found no CAFC 2026 docket, appeal, or opinion involving US 5,048,057, and no district-court or ITC infringement action naming it.
• This is consistent with its status: the patent's anticipated expiration was 2010-01-02 and it is recorded as Expired – Lifetime, so a 2026 appeal concerning this patent would be highly unusually late (a § 112/§ 102-type appeal could in principle outlive infringement life, but I found none).
• Caveat: searches for the bare string "5048057" returned many unrelated hits that must not be confused with this patent — e.g., JP 5048057 B (a Japanese patent/publication number dated 1993/1994 and 2012), a CourtListener financial-disclosure investment record ID, a UK gazette notice, a Lithuanian insolvency notice, and a Chinese utility listing. These are different identifier spaces, not this US patent.
• The patent is cited as prior art by many later patents (Google Patents lists 136 "Cited By" entries, including US 5,301,209, US 5,403,374, US 5,471,503, US 5,809,083, and much of the Netwave/Xircom wireless-LAN family) and cites three references: US 3,423,729, US 3,605,019, and US 4,881,241. Its own family cites US 5,056,112 (Wei, "Interleaving in coded modulation for mobile radio") among others.

Uncertainty flags

• Current assignee is disputed between sources (Nokia Bell Labs USA / AT&T Inc. per Google Patents vs. AT&T Corp per Unified Patents). I cannot resolve this authoritatively without USPTO assignment records.
• I could not directly query USPTO PatentCenter or the Federal Circuit's 2026 docket system; the absence of CAFC activity is based on web search and the patent's expired status, not a direct docket pull. Treat "no CAFC 2026 case found" as strongly likely but not conclusively verified.
• The Google Patents text renders some subscripts awkwardly (e.g., the interference-detection sequence notation "P 1 i , P 2 i , . . . , P K i , j=1, 2, . . . , N," which should read as P₁ʲ … P_Kʲ). I have not "corrected" it in the quoted material.

Sources: https://patents.google.com/patent/US5048057/en ; https://portal.unifiedpatents.com/patents/patent/US-5048057-A ; https://uspto.report/patent/grant/5,048,057 ; https://patents.justia.com/patent/5048057

I'll search for litigation involving US patent 5048057.

Initial results are mostly noise. Let me search more targeted litigation sources.

Let me check a few more specialized litigation databases directly.

Litigation Search Report — U.S. Patent No. 5,048,057

Bottom line

No litigation involving U.S. Patent No. 5,048,057 was found. I could not identify any district court action, ITC investigation, or Federal Circuit appeal asserting this patent. If any litigation exists, it is not discoverable through the public sources I was able to reach.

What the patent is (for context on why)

Sources: https://patents.google.com/patent/[US5048057](/patent/US5048057)/en ; https://portal.unifiedpatents.com/patents/patent/US-5048057-A

Searches performed and results

Relevant non-litigation observations

• No reexamination, IPR, or PGR for 5,048,057 appeared in any of the results.
• The patent's family includes EP 0439925 B1, JP 2862682 B2, and DE 69027018 T2 (same priority date 1990-01-02). No litigation was surfaced for the family either.
• The patent is cited by later art including US 5,301,209 (AT&T Bell Labs, "Multidimensional trellis-coded modulation for fading channels") and US 5,404,374 (Apple), but citation is not litigation.
• The patent expired 2010-01-02, roughly 10 years before the present-day NPE assertion wave, which is consistent with (but not proof of) the absence of assertion activity.

Caveats (stated explicitly, per instruction)

1. I was unable to complete direct queries against PACER, the CAFC docket, or RPX Insight's litigation module (RX) before reaching the tool-call limit. My conclusion rests on the sources above.
2. Absence of records in these sources is not affirmative proof that no suit was ever filed — a suit could have been filed and settled, dismissed, or never indexed. I will not fabricate a plaintiff, defendant, jurisdiction, case number, or filing date to fill this gap.
3. One search hit referenced a different patent family (US 6,534,805 / SRAM cell) with ITC Inv. No. 337-TA-792 and a Westlaw "Litigation Search Report." That is not this patent and I have deliberately excluded it.

If you need a definitive answer, the authoritative check would be a PACER/Unified Patents Litigation tab query and a Westlaw/Lexis "KeyCite" or "CourtLink" search on U.S. Patent 5,048,057, plus a Docket Navigator "patent" search — none of which I was able to runhere.

I'll verify the structured block's "no proceedings" default against public sources before reporting.

Let me confirm with targeted docket/portal checks.

Verification performed

The structured "PTAB proceedings on file" block is the canonical list, and it is empty. Before reporting that as final, I ran independent checks for a proceeding the ODP ingest might have missed:

• Google Patents US5048057A page — no "PTAB" / AIA-trial section, no reexamination certificate among the bibliographic entries, and no petitioner list (only 136 Cited By references, none of them a Board decision).
• Unified Patents patent portal (US-5048057-A) — the page surfaces bibliographic data (Grant 1991-09-10, Expiration 2010-01-02, RUBY D, BRIX B) but no IPR/PGR/CBM entries.
• Full-text searches on the patent number plus "inter partes review," "PTAB," "Patent Trial and Appeal Board," and docket-aggregator queries — no hits in this identifier space.

Those searches surfaced only foreign/coincidental uses of the string "5048057" (a CourtListener financial-disclosure investment record 5048057, a Florida LLC FEI number, a JP publication number) and a different patent — Sandt Technology's U.S. '057 patent (5,514,654-family utility, Sandt v. Resco, Fed. Cir. 2001, a district-court invalidity case, not this patent). None of these are proceedings on US 5,048,057. I do not treat them as such.

Proceedings overview

There are zero AIA trial proceedings on US 5,048,057 — 0 IPRs, 0 PGRs, 0 CBMs — meaning no claim has ever been canceled, no claim has ever been "sustained" through a Board FWD, and the defendant's bottom line is the unusual one: the patent is neither hardened by repeated survival nor softened by an invalidation record, it is simply expired and untouched (anticipated expiration 2010-01-02, recorded "Expired – Lifetime"), so the only defensible posture is that the patent's enforcement life has run out and there is nothing for an IPR to accomplish.

No proceeding sections follow

There are no {PROCEEDING_NUMBER} — {Petitioner} v. {Patent Owner} entries to render, because fabricating one is not permitted and the record contains none. Two structural reasons explain the vacuum, and both matter to a defendant:

1. Statutory unavailability at the relevant time. The AIA trial mechanisms — IPR (35 U.S.C. §§ 311–319), PGR, and CBM — only became available for petitions on/after 2012-09-16. This patent issued 1991-09-10 and its 20-year term ran out 2010-01-02. Every possible AIA filing window opened after the patent had already expired.
2. No live assertable conduct. Infringement liability under § 271(a) accrues only for acts occurring before expiration. Any surviving damages theory would be limited to pre-2010 conduct within the § 286 six-year lookback (roughly 2004–2010). A 2026-filed IPR would be (a) untethered from any complaint served within the § 315(b) one-year window, and (b) of no practical value to a defendant, since disposing of claims on an expired patent buys nothing.

The nearest pre-AIA analogues — inter partes reexamination (available 1999–2012) and ex parte reexamination — also leave no trace: Google Patents shows no reexamination certificate in the patent's bibliographic record, and no ex parte or inter partes reexam number is associated with 07/459,880. I flag this as an inference from the absence of a certificate, not from a direct reexam-docket pull.

Strategic summary

Claim status: all 14 claims stand exactly as granted, and all 14 are expired. Claims 1 and 8 (the independent apparatus and method claims) were never tested at the Board; dependent claims 2–7 and 9–14 likewise. There is no certificate of cancellation, no FWD, no certificate of patentability from any reexamination. If a demand letter today cites claims 1–14 of 5,048,057, the correct response is not "those claims are invalid" but "those claims expired on 2010-01-02 and there is no residual enforceable right" — a § 286 limitations and laches/estoppel problem, not a § 102/§ 103 one. Contrast this with the scenario the prompt anticipates: there is no "claims 1–5 have been canceled" fact here, and it would be malpractice to assert one.

Estoppel landscape: no IPR estoppel exists, and the "still-available ground" analysis is a dead letter. Section 315(e)(2) estoppel is triggered only by a petitioner's participation in an IPR that reaches a final written decision. With zero IPRs, no petitioner, real party in interest, or privy is estopped as to this patent, and there is no estoppel map to draw. The practical corollary is that the ordinary reason to run an IPR-vs.-district-court estoppel analysis (i.e., "which art did the petitioner leave on the table?") has no application to a patent in this procedural state. Prior-art grounds are nominally "available" (the three references of record — US 3,423,729; US 3,605,019; US 4,881,241 — remain the only cited art) but pursuing them would be a solution in search of a problem given expiration.

Pattern signals: none. No repeat petitioner (there is no first petitioner). No patent owner appeals to the Federal Circuit arising from a Board decision — and note the related CAFC check in the earlier patent-summary section already found no 2026 appeal involving this patent, which is consistent. No defensive aggregator in the chain: Unified Patents' portal carries the patent as a record page only, with no challenge filed — that is a passive catalog entry, not a Unified Patents IPR. The patent's post-expiration significance is as prior-art wallpaper: it is cited by 136 later references (including AT&T 5,301,209; Apple 5,403,374 and 5,471,503; AT&T Wireless 5,809,083; and the Netwave/Xircom wireless-LAN family), and its cited-by footprint — not its claims — is where a defendant's real exposure lies, especially if the defendant is being sued on a descendant patent that claims priority to or builds on the Saleh/Wei disclosure.

Recommended next steps

1. If you are a defendant on 5,048,057: do not budget for an IPR — it is not a viable vehicle. Build the defense on expiration (2010-01-02), § 286's six-year damages window, and laches/equitable estoppel for any pre-2010 conduct. Link the record of expiration to the patent's own status entry rather than to any Board decision: Google Patents US5048057A and the Unified Patents portal page both show "Expired – Lifetime" with anticipated expiration 2010-01-02.
2. If your real dispute is over a younger patent in this family: the more valuable exercise is the forward citation tree, not this patent's PTAB history. Screen the 136 citing references for continuations, divisionals, and CIPs that are still in force. The earlier patent-summary section already flagged the current-ownership disagreement (Nokia Bell Labs USA / AT&T Inc. per Google Patents vs. AT&T Corp per Unified Patents) — resolve that against USPTO assignment records before serving any demand-response or negotiating a license, because you are dealing with an AT&T/Lucent/Nokia chain.
3. If you need a negative-certification-style statement for a file: the correct phrasing is "No IPR, PGR, or CBM proceeding has been filed against US 5,048,057; the patent expired 2010-01-02, before AIA trial practice took effect on 2012-09-16." Cite the USPTO PTAB E2E system (ptacts.uspto.gov/ptabweb) as the primary source for the null result, with the ODP structured block as corroboration.
4. Watch for false positives. Anyone searching this number will hit at least three unrelated "5048057" objects (a CourtListener investment record, a Florida FEI number, and a JP publication number) plus the Sandt/Resco '057 case — which is a different patent. Confirm any claimed hit resolves to application 07/459,880 / patent 5,048,057 / Saleh & Wei / "Wireless local area network" before relying on it.

Uncertainty flags carried forward and new ones:

• The absence of PTAB activity rests on the canonical structured block plus web/docket-aggregator corroboration. I did not directly query the PTAB E2E back-end for a case-list pull; treat the null result as high-confidence but not a raw docket export.
• The absence of any reexamination (ex parte or inter partes) is inferred from the lack of a reexamination certificate in the Google Patents bibliographic record — not verified against the IFW/reexam docket. If a defendant needs that point airtight, pull the file wrapper.
• No FWD, panel, settlement, or appeal details are reported because no proceeding exists to report them for. Every claim-level disposition field the task template asks for would necessarily be fabricated here, so all are left empty rather than filled with inference.

I'll research the assignment chain for this patent across the sources. Let me run parallel searches.

Inventors

Departure-pattern note: I found no evidence of inventors departing the original assignee within 12 months of filing. The opposite is indicated: both inventors executed an assignment of assignors' interest to their employer on the filing date itself (1990-01-02), and Wei's continued AT&T Bell Laboratories-assigned filings after the filing date (US 5,056,112, filed 1989-12-27 and granted 1991-10-08) show continuity. I did not obtain an independently verified career record for either inventor, so "no departures" is a negative finding from the record I could reach, not a positive confirmation of tenure.

⚠️ Contradiction flag: the prior-art section of this analysis describes Wei's US 5,056,112 as a "continuation-in-part of Ser. No. 386,185," while Google Patents lists US 5,056,112 as its own application filed 1989-12-27. Both can be true only if the CIP claim was added later. I could not resolve this, and it does not affect the ownership analysis — both are AT&T Bell Laboratories-assigned.

Original assignee

Entity named on the patent: the granted front page lists AT&T Bell Laboratories Inc. as original assignee (Google Patents "Original Assignee" field). The recorded assignment instrument (1990-01-02) names the assignee more fully as American Telephone and Telegraph Company, Bell Telephone Laboratories, Incorporated.

Primary line of business: AT&T Bell Laboratories was the research-and-development arm of AT&T Corp. — a large, operating telecommunications R&D organization, not a patent-holding vehicle.

Did they ship a product embodying the claims? Unverified — and I want to be precise here. The claims (frequency-hopping + built-in-diversity coded modulation + interference-detection sequences + erasure-aware soft-decision decoding) read as a system/architecture patent rather than a single product claim, and the disclosure itself frames the invention as a wireless LAN for ISM-band operation. AT&T-in-this-era did have a commercial wireless LAN line — NCR WaveLAN, developed in the AT&T Bell Laboratories / NCR orbit in the early 1990s — and the ISM-band framing is consistent with that program. However, I could not verify that WaveLAN practiced claims 1–14, and the patent's own illustrative codes (Code I–IV, 4-PSK block codes) look like research-constellation designs rather than a consumer product implementation. I decline to assert embodiment on this record.

Current status of the original assignee: Bell Telephone Laboratories, Inc. / AT&T Bell Laboratories as a legal entity no longer exists in that form. The historical chain is: AT&T Corp. → Lucent Technologies Inc. (equipment/R&D spin-off, effective 1996-03-29 per Lucent-originated assignment instruments) → Alcatel-Lucent USA Inc. (merger/change of name, effective 2008-11-01 in the Lucent-to-Alcatel-Lucent change-of-name records I surfaced) → Nokia (acquisition of Alcatel-Lucent, 2016). AT&T Inc. is the separate, surviving telecom services company.

Assignment timeline

⚠️ Critical sourcing caveat, stated up front. I was unable to execute a direct query against USPTO Assignment Center (https://assignmentcenter.uspto.gov/) or the legacy interface (https://assignment.uspto.gov/patent/index.html) for this patent within this session — my search budget was exhausted before I reached those records. Everything below is what the accessible secondary sources expose. I have not filled in reel/frame numbers that I could not actually observe for this patent, and I have explicitly labelled every reel/frame I quote from sibling patents as such.

What I can confirm from the record:

• 1990-01-02 (executed) / recorded 1990-01-02 — Reel NNNNNN/NNNN not exposed in my sources
  • Conveyance: ASSIGNMENT OF ASSIGNORS' INTEREST
  • Assignor: WEI, LEE-FANG; SALEH, ADEL A. M.
  • Assignee: AMERICAN TELEPHONE AND TELEGRAPH COMPANY, BELL TELEPHONE LABORATORIES, INCORPORATED
  • Correspondent: not exposed in the sources I could reach (Google Patents legal-event rendering does not surface the correspondent field; the printed assignment's attorney/agent of record must be pulled from Assignment Center or the file wrapper)
  • Context: Pre-issuance inventor-to-employer assignment — the standard AT&T Bell Laboratories practice of taking title from employee-inventors on the filing date. This is the only assignment event associated with US 5,048,057 that I could verify.

Post-issuance assignments on THIS patent (07/459,880): none found.

Google Patents' legal-events block for US5048057A (as captured in the authoritative full text supplied for this analysis) lists exactly one ownership event — the 1990-01-02 assignment above — followed by the EP/DE/JP priority entries, the 1991-09-10 grant, and the 2010-01-02 anticipated expiration. No Lucent, Alcatel-Lucent, Nokia, or third-party assignment event appears in that legal-events list. I cannot rule out that a post-issuance assignment exists but is simply not rendered in the Google index, which is why the caveat above matters.

Corporate-chain events documented on sibling AT&T/Lucent patents — illustrative of the chain, NOT confirmed against this patent:

• Effective 1996-03-29 — AT&T Corp. → Lucent Technologies, Inc. — recorded under reels 012059/0893, 012754/0365, and 012946/0578 (observed on other AT&T-origin patents, not on US 5,048,057). Note the effective date is 1996-03-29 while the recording reels are 12xxxx, i.e. recorded years later (roughly 2001–2002) — consistent with a batch re-recordation of the 1996 spin-off transfers.
• Effective 2008-11-01 — Lucent Technologies Inc. → Alcatel-Lucent USA Inc., CHANGE OF NAME, recorded under reel 049887/0613 (observed on a different Lucent patent via a PTAB exhibit reproducing a full USPTO Assignment Abstract of Title). A change-of-name (not a transfer of title) is the correct conveyance type for the Lucent→Alcatel-Lucent step.
• 2017-09-13 — a Nokia-origin assignment naming PROVENANCE ASSET GROUP LLC as assignee and NOKIA TECHNOLOGIES OY / NOKIA SOLUTIONS AND NETWORKS BV / ALCATEL LUCENT SAS as assignors, reels 043877/0001 and 043967/0001, with a SECURITY INTEREST to PROVENANCE ASSET GROUP HOLDINGS, LLC at reel 043879/0001 (observed in that same PTAB exhibit on a different patent). ⚠️ This is a Nokia-family transfer pattern, not a transfer of US 5,048,057, and I found no record placing this patent in that transfer. I flag it only because it is the single most relevant watch item for a Nokia-owned 1990s portfolio — see signal 2 and signal 7 below.

Timeline diagram

timeline
    title Ownership of US 5048057
    1990 : Application filed 1990-01-02
         : Inventors assign to AT&T Bell Labs
    1991 : Patent issued 1991-09-10
    1996 : AT&T spin-off forms Lucent Technologies
    2006 : Lucent merges into Alcatel-Lucent
    2010 : Patent expired 2010-01-02
    2016 : Nokia acquires Alcatel-Lucent

NPE / troll-pattern signals

One additional non-troll observation worth putting on the record: both accessible assignee indices list an AT&T-affiliated entity (Google Patents: "Nokia Bell Labs USA; AT&T Inc."; Unified Patents: parent company "Nokia Bell Labs Usa / AT&T", current assignee "AT&T Corp"). The persistence of "AT&T Corp" as a current assignee designation is at least consistent with the hypothesis that no post-1996 assignment to Lucent was ever recorded against this particular patent, with Google inferring the Nokia connection from corporate-tree data rather than from a recorded instrument. This is a hypothesis, not a finding — the prior Patent summary section of this analysis already flagged the same Nokia-vs-AT&T disagreement as unresolved, and the present assignment review does not resolve it.

Verdict

Insufficient data.

Justification: the only ownership event I could verify for US 5,048,057 is the 1990-01-02 assignment of assignors' interest from inventors Wei and Saleh to American Telephone and Telegraph Company / Bell Telephone Laboratories, Incorporated, with no reel/frame or correspondent exposed in the sources I could reach; no post-issuance assignment, LLC transferee, or repeat correspondent is documented for this patent. That is the textbook posture the rubric assigns to Insufficient data ("no records, or only the original assignment") — and it is reinforced by the fact that the strictly negative NPE findings (no shell entity, no known asserter, no cascading LLC transfers, no pre-suit transfer, no bankruptcy sale, no defensive aggregator) are all supported and none of the eight signals is present on this record. I deliberately do not upgrade to "operating-company assertion" (the assignees ship products generally, but I could not verify embodiment of claims 1–14, and no competitor suit exists) and I do not upgrade to "NPE – moderate confidence" on the strength of the unrelated Nokia→Provenance pattern, which I could not tie to this patent number.

Verification link: USPTO Assignment Center search — https://assignmentcenter.uspto.gov/ (legacy interface: https://assignment.uspto.gov/patent/index.html). Search by patent number 5048057 or application number 07/459,880 to pull the reel/frame, conveyance type, and correspondent of record that I could not retrieve here.

Carried-forward uncertainty flags: (1) I did not execute a direct Assignment Center / Assignment Search API query — my post-issuance "no records" conclusion rests on Google Patents' legal-events rendering, which does not render every recorded instrument; (2) no reel/frame or correspondent is reported for the 1990-01-02 assignment because none was exposed in my sources, and I have not invented one; (3) the "AT&T Corp as current assignee" hypothesis is an inference from a data inconsistency already flagged in the earlier Patent summary section, not a documented assignment event; (4) all Nokia→Provenance reel/frame citations above are from other patents' records and must not be represented as US 5,048,057's chain of title.

I'll verify the USPTO record and research each cited reference. Let me run several searches in parallel.

Now let me research the foreign-family citations and the IEEE references that were considered during EP prosecution of the same family.

I ran the USPTO-side searches and cannot complete further queries (search budget reached), so I'll flag the two Japanese family references where I could not retrieve full text.

Prior-art analysis for US 5,048,057 ("Wireless local area network")

Note on the record: The patent identified literally as US 5,048,057 is the one at issue here — application 07/459,880, granted 1991-09-10, inventors Saleh and Wei, original assignee AT&T Bell Laboratories. Search hits for the bare string "5048057" that resolve to other identifier spaces (e.g., JP 5048057 B, a CourtListener financial-disclosure ID) are not this patent and are excluded, per the operating rule to interpret identifiers literally. Confirmed record: https://patents.google.com/patent/[US5048057A](/patent/US5048057A)/en ; https://portal.unifiedpatents.com/patents/patent/US-5048057-A

0. Important framing: which "citations" the task is asking about

There are three distinct sets of references associated with this patent, and they must not be conflated:

1. References cited on the face of US 5,048,057 (Google Patents: "Patent Citations (3)") — these are the references the examiner actually used against this patent. This is the set the task asks about.
2. References cited in the family ("Family Cites Families") — lifted from the sibling EP/JP/DE prosecution, some of which postdate the priority date.
3. References cited during EP prosecution of the counterpart (EP 0 439 925 B1) — including two 1985 IEEE papers. These are not on the US face but are the most technically on-point art, and I treat them below because they bear directly on § 102/§ 103 of claim 1.

Applicable statute: this is a pre-AIA patent (priority 1990-01-02), so pre-AIA § 102 governs. The one-year § 102(b) cutoff is 1989-01-02.

1. References cited on the face of US 5,048,057 (the three of record)

1.1 US 3,423,729 A — "Anti-fading error correction system"

What it discloses: Data is error-correction encoded, then passed through a "time-spread permutation device" that separates adjacent bits of a codeword by bits belonging to other codewords, so a fade (or burst of errors) of B×T_F bits corrupts no more than one bit of any codeword; an inverse permutation device restores codeword order for decoding. The described benefit is explicitly framed as a form of time diversity ("compares favorably with double space diversity"). Source: https://patents.google.com/patent/US3423729/en

Anticipation analysis: This is the classic time-interleaving/time-diversity reference. It supports the concept of built-in time diversity exploited by interleaving, but it does not disclose:

• a coded modulation code with built-in diversity in the sense defined in the '057 specification (codewords differing in at least X signal-point positions);
• transmission of the diversity-bearing signal points on respective different frequencies (this reference is single-channel);
• interference-detection signal points or any erasure/strong-interference detection; or
• soft-decision decoding excluding/zeroing the detected signal points.

Anticipates under § 102? No claim of the '057 patent is anticipated. It lacks every limitation of independent claims 1 and 8 and the interference-detection limitations of claims 3–7/10–14. Its only pertinence is as § 103 background to the interleaving/diversity concept.

1.2 US 3,605,019 A — "Selective fading transformer"

What it discloses: Errors caused by selective frequency fading are converted into time-localized (burst) errors by transmitting the Fourier transform of the baseband signal. Preferred embodiment: FDM of several channels, error-correction encoding per channel, multilevel phase modulation (expressly "four-level phase modulation," i.e., 4-PSK), real-time digital Fourier transform transmission, and inverse transform + error correction at the receiver. Source: https://patents.google.com/patent/[US3605019A](/patent/US3605019A) ; https://uspto.report/patent/grant/[3605019](/patent/3605019)

Anticipation analysis: Relevant to the 4-PSK / multilevel PSK modulation and to transmitting coded data over a frequency-selective fading channel, but it discloses no frequency hopping, no coded modulation with built-in diversity, no interference-detection sequence, and no erasure/soft-decision decoding that ignores interfered points. Its "burst errors" are re-localized fade errors, not interference.

Anticipates under § 102? No. Not a single claim's complete set of limitations is present. Best characterized as § 103 art on the modulation/fading-mitigation backdrop.

1.3 US 4,881,241 A — "Method and installation for digital communication, particularly between and toward moving vehicles"

What it discloses: Convolutional encoding, then time interleaving and frequency distribution ("frequency interlacing") of the coded symbols across subchannels at different frequencies, with de-interlacing and maximum-likelihood (Viterbi) decoding at the receiver for mobile/vehicle reception. It is a COFDM-family broadcasting scheme. Source: https://patents.google.com/patent/US4881241

Anticipation analysis: Closest of the three face references to the system architecture (frequency-spread coded symbols + ML decoding) and arguably maps loosely onto the "interleaver 16" and multi-frequency transmission concepts of claims 5–6/12–13. But it is convolutional coding, not coded modulation with block-coded built-in diversity; it has no interference-detection signal points, no error-rate-based erasure flag, and no soft-decision decoding that excludes detected interfered points.

Anticipates under § 102? No claim is anticipated. It is, however, the strongest § 103 combination candidate among the three face references for the "interleave coded symbols across different frequencies" sub-features.

Summary table — face citations vs. '057 claims (pre-AIA § 102)

2. References cited in the family (not on the US face)

These appear under Google's "Family Cites Families" and originate from the sibling EP/JP prosecution. Two are relevant; two are not prior art to '057.

3. The most on-point art: EP-citation IEEE papers (search-report art, EP 0 439 925 B1)

During EP prosecution of the counterpart, the examiner relied on two 1985 IEEE papers. They are not named on the US face, but they are the references most directly aimed at the interference-detection/erasure heart of claim 1, and any § 102 challenge to '057 would run through them:

• Milstein et al., "Coding and Modulation Techniques for Frequency-Hopped Spread-Spectrum Communications Over a Pulse-Burst Jammed Rayleigh Fading Channel," IEEE J. Sel. Areas Commun., vol. SAC-3, no. 5, Sept. 1985, pp. 644–651. Discloses concatenated codes operated in an error-and-erasure mode, where the receiver erases a received symbol it determines was jammed by monitoring the channel, and treats that side information as perfectly reliable. This maps closely onto the "detecting ... likely to have been subject to strong interference" + "decoding responsive to the received values ... other than the detected [erased] ones" language of claim 1.
• Wayne, "Coding for Frequency-Hopped Spread-Spectrum Communication with Partial-Band Interference — Part II: Coded Performance," IEEE Trans. Commun., vol. COM-33, no. 10, Oct. 1985, pp. 1045–1057. Discloses hard or soft decoding given side information on the presence of a partial-band jammer, and — critically — a soft decoder that "treats any received symbols during jamming with less weight" than un-jammed symbols.

§ 102 assessment against claim 1 / claim 8: Even these closest references do not appear to disclose the complete combination — specifically coded modulation whose code has built-in diversity (X-fold, per the '057 definition), combined with the predetermined interference-detection signal points whose error rate produces the erasure decision. Milstein/Wayne teach the erasure mechanism (channel monitoring, weighting/erasing jammed symbols) but not the known-signal-point-sequence error-rate detector of claims 3–4/10–11 nor the built-in-diversity coded-modulation layer. So at most they support a § 103 obviousness attack on the broad claim 1/8 concept; standing alone they do not anticipate claims 1, 3, 4, 8, 10, or 11 on the record I can verify.

(Source for the EP citations as quoted: the EP 0 439 925 B1 specification, https://patentimages.storage.googleapis.com/11/8c/ad/eaf02d9224f899/EP0439925B1.pdf)

4. Bottom line

• The three references cited on the face of US 5,048,057 are US 3,423,729 (Heller/Westinghouse, 1969), US 3,605,019 (IBM, 1971), and US 4,881,241 (CNET, 1989). All predate the 1990-01-02 priority date and qualify as prior art (§ 102(b) for the first two; § 102(e) for the CNET reference).
• None of the three anticipates any claim of the '057 patent under § 102. Each lacks at least the interference-detection-signal-point and erasure-driven soft-decision-decoding limitations that define independent claims 1 and 8, and none discloses a codeword code with defined built-in diversity transmitted over independently fading hop frequencies.
• The genuinely dangerous art is Milstein (1985) and Wayne (1985), cited during EP prosecution — they disclose erasure/side-information decoding for frequency-hopped, jammed spread-spectrum, and are the natural basis for a § 103 attack, but on my review they still do not, alone, anticipate the built-in-diversity + interference-detection-sequence combination of claim 1/8.
• US 5,056,112 (Wei) is family/copending same-inventor art rooted in the very Ser. No. 386,185 application incorporated by '057 — a § 102(e)-type reference as to the code element, and a commonly-owned-art / double-patenting consideration, not a third-party anticipatory reference for the whole claim.
• FR 2 658 016 B1 postdates the '057 priority date and is not prior art against US 5,048,057.

Verification caveats I want on the record:

1. I could not retrieve the full text of JPS 62-152239 A or JPS 63-50218 A within this session; their disclosures are unverified and should be pulled directly before any validity conclusion.
2. The § 102(e)/§ 102(a) characterizations of US 4,881,241 and US 5,056,112 depend on the actual invention date of the '057 subject matter (not just its filing date); treat those subsections as likely rather than certain.
3. I could not complete a direct USPTO PatentCenter/PAIR pull for the face-citation list within this session; the three-reference face list is taken from the Google Patents record (which mirrors the printed patent) and should be cross-checked against the granted front page before being relied on in a filing.

Sources: https://patents.google.com/patent/US5048057A/en · https://portal.unifiedpatents.com/patents/patent/US-5048057-A · https://patents.google.com/patent/US3423729/en · https://patents.google.com/patent/US3605019A · https://uspto.report/patent/grant/3605019 · https://patents.google.com/patent/US4881241 · https://patents.google.com/patent/US5056112A/en · https://patentimages.storage.googleapis.com/11/8c/ad/eaf02d9224f899/EP0439925B1.pdf

I'll ground the obviousness analysis in the actual prior-art references listed on the patent page. Let me pull specifics on each.

Good grounding. Let me check a few more anchors: the parent Wei application (Ser. No. 386,185), the Saleh/Cimini background papers, and whether any cited reference teaches interference detection/erasure.

Obviousness Analysis — U.S. Patent No. 5,048,057 under 35 U.S.C. § 103

Analyst note on scope. This analysis uses only the prior-art references appearing in the "Prior Art" portions of the Google Patents record for US 5,048,057 (the examiner Citations, the Family Cites Families list, and the references discussed in the patent's own Background), plus what those references actually disclose as verified by search. Where I could not verify a reference's content, I say so rather than infer it. Patent numbers and identifiers are reproduced literally.

I. Governing law and framework

The application was filed 1990-01-02 — before the March 16, 2013 effective date of the AIA first-inventor-to-file changes — so pre-AIA § 103(a) governs, and the pre-AIA § 102 definitions of prior art supply the reference set. The controlling analysis is the Graham v. John Deere framework (scope and content of the prior art; differences between the prior art and the claims; level of ordinary skill; secondary considerations), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) and MPEP § 2143. Under KSR, a combination of familiar elements according to known methods is obvious when it does no more than yield predictable results, and the motivation may come from the references themselves, from the nature of the problem, or from ordinary creativity — but the analysis must be performed from the perspective of the PHOSITA at the time of invention, without hindsight drawn from the patent's own disclosure. That hindsight caution is important here, because the patent's specification itself supplies a tidy decomposition of the invention that a challenger might be tempted to read back into the references.

Level of ordinary skill (assumed): a B.S./M.S. in EE with 2–5 years' experience in digital communications / coded modulation and radio systems; familiarity with PSK constellations, interleaving, Viterbi/ML decoding, and frequency-hopping spread spectrum.

Claim construction note. Claim 1 is drafted in means-plus-function form; under § 112 ¶ 6 each "means" is limited to the corresponding structure disclosed in the specification and equivalents — e.g., block encoder/mapper 13, interleaver 16, interference-detection-sequence inserter 19, frequency-hopping modulator 21 (transmit side); frequency-hopping demodulator 54, checker/remover 56, de-interleaver 58, decoder 61 (receive side). This matters because a broader "any structure that performs the function" reading makes the claim easier to invalidate; the narrower reading makes it harder.

II. The independent claims decomposed

Claims 1 (apparatus) and 8 (method) recite the same four limitations. For § 103 purposes I elementize them as:

III. The prior art of record and what each reference teaches

Two important record observations: (1) the applicant's own Background expressly acknowledges the frequency-hopping-plus-coding-plus-interleaving art (the Saleh/Cimini papers), which is a specification admission usable in an obviousness analysis; and (2) the "copending application Ser. No. 386,185 … 'Coded Modulation for Mobile Radio'" that the patent identifies as the source of the built-in-diversity coded modulation is US 5,029,185, i.e., the element-A core of this patent is, by the patent's own words, not the applicants' contribution.

IV. The obviousness case

Combination 1 — Wei '185 in view of CNET '241 (the substrate combination)

Teaches: Wei '185 supplies limitation A in full — the very "coded modulation with built-in diversity" the patent recites as its starting point. CNET '241 supplies limitation B in full — coded symbols distributed over multiple different frequencies with time interlacing — and supplies the soft/ML decision portion of limitation D (the "closest permitted mosaic" search).

Motivation: Both references are aimed at the same problem (transmission over fading channels), and both are in the same field of endeavor (digital communication over fading/mobile radio channels). CNET '241's stated objective of digital broadcasting to moving vehicles and Wei '185's stated objective of mobile radio performance make the combination a routine engineering pairing, and the patent's own Background tells us the applicant considered frequency hopping + coding + interleaving to be an established approach to countering fading. Under KSR, this is "a combination of familiar elements according to known methods" yielding the predictable result of diversity against fading.

Result: Renders the substrate of claims 1 and 8 obvious — a diversity-coded modulation stream transmitted over different frequencies and recovered by soft-decision decoding. It does not reach limitation C.

Combination 2 — Combination 1 + Westinghouse '729 and/or IBM '019 (strengthening the interleaving and impairment-detection elements)

• US 3,423,729 reinforces that interleaving/permutation is the known answer to fading-induced errors in a coded system and is expressly a "time diversity" technique, matching this patent's use of an interleaver (16) to separate diversity-bearing points.
• US 3,605,019 is the more interesting reference for limitation C. It teaches the concept of transmitting a known reference signal (a swept pilot tone) and observing the received signal to locate where the channel is impaired, and it teaches deliberately arranging the transmission so that the impairment appears as a particular, easily detected error signature (burst errors) that can then be handled by the decoder. That is structurally the same idea the patent uses at claim 2 — "transmit … in such a way that said strong interference will cause the received signal to have a particular characteristic" that the receiver then detects. A PHOSITA seeking to protect a frequency-hopping LAN in a shared band would have been motivated to adapt IBM '019's known-reference/tone-probing channel-sounding technique to characterize impairments on a hop and act on them.

However: IBM '019 detects fading and feeds information back to the transmitter; it does not declare an erasure at the receiver, and it is not about an interferer stronger than the signal. The mapping to limitation C is analogical, not literal.

Combination 3 — the hard limitation: detecting "strong interference" and erasing

This is where the record is thin, and I will state that plainly rather than construct a rejection the art does not support.

• None of the three examiner Citations (US 3,423,729, US 3,605,019, US 4,881,241) discloses declaring an erasure for signal points corrupted by strong interference and then soft-decision-decoding around them. Each addresses noise or fading, not an interference source comparable to or stronger than the transmitted signal.
• The other listed sources are either post-dated (FR 2658016) or of unverified content (JPS 62152239, JPS 6350218).
• What can be argued: within a frequency-hopping LAN in a shared band, using a known transmitted sequence to detect a channel impairment (IBM '019's swept tone) combined with the well-known coding-art practice of erasure declaration (an ordinary tool in the error-control-coding art by 1990) would have been an obvious mechanical extension. This is the KSR "known technique, known problem" argument: the problem of a hop being swamped by an interferer is inherent in the ISM-band environment the patent itself identifies, and the remedy of telling the decoder "this point is unreliable — ignore it" is a conventional erasure technique. But no reference in the record actually teaches the interference-versus-fading discrimination that limitation C requires, so a prima facie case on C rests on the applicant's own admitted problem statement plus general knowledge, not on an identified reference. That is a materially weaker footing than the case on A/B/D.

V. Claim-by-claim dispositions

The dependent claims do not rescue obviousness — the question is whether the independent claim's combination as a whole would have been obvious, and each dependent claim is a narrower, readily envisioned implementation of the same concept. Conversely, the dependent claims 3, 4, 7, 10, 11, 14 are the claims a challenger would have the hardest time invalidating on this record, because they capture the specific mechanism (known detection sequence + error-rate threshold) that no listed reference supplies.

VI. Motivation to combine — articulated

A PHOSITA in January 1990, facing the patent's own stated problem (indoor multipath fading and interference from other systems in a shared band), would have had explicit motivations:

1. Same field, same problem. Wei '185/'112 (coded modulation for fading channels) and CNET '241 (coded symbols over multiple frequencies for mobile reception) are both directed at fading-channel reliability; CNET '241's "moving vehicles" and Wei '185's "mobile radio" framing make the pairing natural.
2. Express teachings in the references. IBM '019 states that converting fading into time-localized burst errors "result[s] in easy implementation of error correcting schemes" — a direct suggestion to couple channel-impairment handling to the decoder. Wei '112 expressly pairs the interleaver with the code's diversity so that time-diverse points fade independently.
3. Design incentive / market pressure. The patent itself identifies the FCC ISM band and the expectation of strong interference. Once the environment is known to contain interferers stronger than the signal, improving the decoder by marking corrupted points as unreliable (an erasure) is a predictable, well-motivated application of a known coding technique.
4. Predictable results. Spreading diversity-bearing points across independently fading frequencies, then soft-deciding with corrupted points ignored, yields nothing more than the expected aggregate improvement in error performance — the hallmark of an obvious combination under KSR.

VII. Legal and evidentiary caveats a challenger must confront

1. Common ownership / § 103(c). The two backbone references (US 5,029,185 and US 5,056,112, both Wei, both AT&T Bell Laboratories) are, on their face, commonly owned with the application that issued as US 5,048,057 (also AT&T Bell Laboratories), and share inventor Wei. If those references qualify as prior art only under § 102(e) (they are U.S. patents/applications, unpublished at the time, whose filing dates precede the filing date), pre-AIA § 103(c) may disqualify them from use in a § 103 rejection. This is a serious obstacle: at the time of the original 1990–91 prosecution the then-current version of § 103(c) reached only § 102(f) and (g) art, so § 102(e) art could be used — but the § 103(c) safe harbor was later extended to § 102(e) art. A modern challenger would need to resolve which version applies and whether the common-ownership condition is met. I flag this as unresolved rather than asserting a conclusion.
2. Applicants' own work (Saleh/Cimini). The Saleh and Cimini papers are the work of the patent's own inventors (Saleh is a named inventor here). Their use as § 103 prior art depends on whether they are statutory bars (§ 102(b)) — which turns on the precise January 1989 publication date relative to the January 2, 1989 critical date — or merely § 102(a) art that the applicants could antedate. The specification's citation of them is at minimum a usable admission about the state of the art.
3. Post-dated family citation. FR 2658016 B1 (priority 1990-02-06) post-dates this patent's priority date and therefore is not prior art; its appearance in the "Family Cites Families" list is a prosecution artifact of the EP member and should not be used.
4. Unverified references. I could not verify the disclosures of JPS 62152239 and JPS 6350218; titles alone are insufficient to map them to any claim limitation, and I decline to characterize them.
5. Hindsight risk. The clean elementization above (A/B/C/D) comes from the patent's own specification; the references must be shown, on their own terms, to suggest the combination, not merely to contain the individual pieces.

VIII. Bottom line

• Claims 1 and 8 (substrate). There is a strong § 103 case that limitations A, B, and the soft-decision component of D were obvious over US 5,029,185 (Wei, Ser. No. 386,185) in view of US 4,881,241 (CNET), optionally reinforced by the Saleh/Cimini frequency-hopping-and-coding art admitted in the patent's own Background and by the interleaving teachings of US 3,423,729 and US 3,605,019. The patent's own specification concedes that the coded-modulation-with-diversity core and the frequency-hopping/interleaving environment were known.
• The interference-detection-and-erasure limitation (C). This is the weakest link in the record. No reference listed on the patent page squarely discloses detecting signal points corrupted by strong interference (as opposed to fading) and erasing them for a soft-decision decoder. A rejection on C would have to rest on the analogy to IBM '019's known-reference impairment detection plus the general-knowledge erasure technique, and on the applicant's admitted identification of the interference problem — an argument that is plausible under KSR but materially less robust than the case on A/B/D.
• Dependent claims 3, 4, 7, 10, 11, 14 (predetermined interference-detection signal points; error-rate-threshold indicator; per-block/per-frequency variants) are the most resistant to the record art, because they capture the specific detection mechanism the references do not supply.
• Practical takeaway: if this patent were being challenged today, the efficient path is a § 103 attack on claims 1 and 8 built on Wei '185/'112 + CNET '241 + the admitted frequency-hopping art, with the KSR "known problem / known technique" argument carrying limitation C — while the challenger must first clear the pre-AIA § 103(c) common-ownership hurdle for the Wei references and the own-work issue for the Saleh/Cimini papers. Claims 3, 4, 7 and their method counterparts would likely require art outside the record assembled here.

Sources: https://patents.google.com/patent/US5048057/en · https://patents.google.com/patent/US5029185 · https://patents.google.com/patent/[US5056112A](/patent/US5056112A)/en · https://patents.google.com/patent/US4881241 · https://patents.google.com/patent/US3423729 · https://patents.google.com/patent/[US3605019A](/patent/US3605019A) · https://uspto.report/patent/grant/[5,056,112](/patent/5056112)

I did not verify the disclosures of JPS 62152239, JPS 6350218, or FR 2658016 B1, and I have not attributed any specific teaching to them. If a claim-by-claim invalidity content chart with exact quotations is needed, the remaining step is to pull the full texts of those three family citations and of the Saleh/Cimini JSAC paper.