US 10554763

Let me search more specifically for any CAFC appeal or litigation involving this patent.

Based on the authoritative patent text provided (Google Patents record for US10554763, fetched 2026-09-03) and corroborating web-search results, here is the summary.

Important note on scope of the searches

• USPTO/patent databases confirm the bibliographic details below.
• I found no CAFC 2026 docket entries, appeals, IPRs, or litigation involving patent number 10554763 in any search (multiple query variants returned zero results). I cannot rule out the existence of such proceedings, but I have no authoritative evidence of any CAFC 2026 case naming this patent. Treat that as a negative-result finding with normal search limitations, not a definitive legal conclusion.

Patent bibliographic summary

Abstract (as published)

A method and apparatus for virtual base station migration in a BBU pool having multiple baseband units. The migration method comprises steps performed by a first virtual base station: measuring an available bandwidth between the first and a second virtual base station (belonging to different baseband units in the BBU pool); estimating a service interruption time of the first virtual base station; and selecting a migration mechanism according to the estimated service interruption time. The migration mechanism comprises configuring a common measurement gap for user devices, or random access time slots, or a discontinuous receiving (DRX) cycle. The solution keeps connected-state user devices unaffected during the service interruption and achieves lossless migration between virtual base stations.

Plain-language overview of the independent claims

The patent has 13 claims; four are independent (claims 1, 6, 9, 12).

Claim 1 (method, first virtual base station side) — A method executed by a "first" (source) virtual base station in a multi-BBU pool that:

1. Measures the available bandwidth between itself and a "second" (target) virtual base station residing on a different BBU;
2. Estimates its service interruption time;
3. Selects a migration mechanism based on that estimated time;
4. Sends the second base station a migration request carrying the control-plane and user-plane data, causing the target to restore user-device context and buffered data; and
5. Sends a migration completion message to the migration manager.

Dependent claims 2–5 add: (2) the tiered mechanism — common measurement gap / random access slots / DRX when interruption ≤ first threshold; random access slots or DRX between first and second thresholds; DRX only above the second threshold; (3) receiving a migration preparation message before measuring; (4) sending a preparation-ack and receiving a migration command before estimating; (5) estimating interruption time from number of user devices, per-device data rate, and available bandwidth.

Claim 6 (method, migration manager side) — A method executed by a migration manager that:

1. Collects resource usage rates of BBUs in the pool;
2. Makes a migration decision from the collected usage rates and a migration threshold, determining a first and a second virtual base station on different BBUs;
3. Sends a migration preparation message to both to trigger the bandwidth measurement;
4. Receives migration preparation acknowledgement messages from both;
5. Sends the first base station a migration command message triggering the migration; and
6. Receives migration completion messages from both.

Dependent claims 7–8 add: presetting/configuring the threshold from collected usage; and sending the preparation message when the first station's BBU resource usage is below the threshold.

Claim 9 (apparatus, source-side) — Corresponding apparatus (one or more processors) to the claim-1 method: measure the inter-BBU available bandwidth; estimate service interruption time from user count, data rate, and bandwidth; select the migration mechanism; plus transmitting units to send the migration request to the target and the completion message to the migration manager. Dependent claims 10–11 mirror the tiered mechanism (claim 2) and the prepare/ack/command messaging units (claim 3/4 equivalents).

Claim 12 (apparatus, migration-manager side) — Corresponding apparatus to the claim-6 method: processors to collect BBU resource usage and make the migration decision (first/second virtual base station on different BBUs), plus transmitting/receiving units for the preparation message (sent when source-BBU usage is below threshold), preparation acknowledgements, the migration command, and migration completion messages. Dependent claim 13 adds threshold presetting/configuring from collected usage.

Technology context (plain language)

The invention addresses power savings in Cloud-RAN by consolidating lightly loaded virtual base stations (vBSs) onto fewer BBUs during idle periods, then shutting down unused BBUs. A "live migration" between BBUs inherently creates a gap in which neither the source nor target vBS serves user devices; if a UE sees no PDCCH signaling during that gap it may trigger RRC re-establishment and drop services. Rather than predicting a fixed interruption window, the patent dynamically estimates the interruption time (from number of connected UEs, their data rates, and the measured inter-BBU bandwidth) and then masks the outage from UEs using one of three LTE mechanisms sized to cover the interruption: (i) a common measurement gap (shortest interruption, ≤ ~6 ms), (ii) random access time slots (mid-range, ~6–20 ms, with preamble-retry/neighbor-context/UE-notification fallbacks against random-access failure), or (iii) a DRX cycle (longest interruption, > ~20 ms), coordinating the migration with UE sleep/measurement/access windows so the target vBS takes over seamlessly.

Uncertainty disclosure

• Bibliographic fields (title, assignee, inventors, filing 2015-04-10, issue 2020-02-04) are consistent between the provided Google Patents text and independent search hits; no conflicts found.
• The CAFC 2026 docket portion is unverified: searches for "10554763" combined with CAFC/Federal Circuit/litigation/IPR terms returned no results. I have no authoritative information indicating any pending CAFC matter for this patent; absence of results is not proof of non-existence.

Based on the searches I performed (web searches across general patent-litigation indexes, Unified Patents-style portals, and court records aggregators), I found no known litigation involving US Patent 10,554,763 ("Method and apparatus for virtual base station migration in BBU pool," Alcatel Lucent / Nokia).

Search results summary:

• Searches for "10554763" patent litigation, "US10554763" lawsuit, "US10554763B2" litigation OR complaint OR "civil action", and related queries returned no case naming this patent.
• The only "Alcatel" litigation results that surfaced (e.g., Telcordia Technologies, Inc. v. Lucent Technologies/Alcatel, D. Del. 1:04-cv-00874, and related ITC Investigation No. 337-TA-572) involve unrelated, decades-old patents (U.S. Patent Nos. 4,893,306; Re. 36,633; 4,835,763) and are not connected to patent 10,554,763.
• One hit (PatentLeaderboard) merely lists the patent in an inventory of Liyu Cai's Alcatel-Lucent patents, with no litigation data.

Conclusion: No district-court litigation, ITC investigation, or other adversarial proceeding specifically involving US Patent 10,554,763 was identified as of the search date.

Caveats (stated honestly):

1. My search coverage was limited to general web/aggregator indexes. I did not directly query PACER's full-text docket database, RPX, DocketNavigator, or LexisNexis/Westlaw docket files, which would be the definitive check.
2. Absence of litigation does not rule out other proceedings such as ex parte reexamination, IPR/PGR petitions at the PTAB, or license/settlement activity, none of which would necessarily appear as "litigation."
3. The patent remains in force (maintenance fee paid in 2023, 4th year, large entity), so litigation could still be filed or pending in a source I did not reach.

If you need a definitive answer, I recommend a direct docket search on PACER for the patent number and on the PTAB's API/Unified Patents portal, which I was unable to fully query within this session's constraints.

Proceedings overview

Total AIA trial proceedings on file: zero. The USPTO Open Data Portal (ODP) returns no IPR/PGR/CBM proceedings for US 10554763 as of the most recent ingest, and web searches (multiple query variants covering "10554763," the title, the assignee Alcatel Lucent, PTAB/IPR terms, and litigation) surfaced no petitions, no institution decisions, no Final Written Decisions, and no settlements naming this patent. The bottom line for a defendant: this patent is untested at the PTAB — no claim has been canceled, but no claim has been battle-hardened either. The absence of PTAB activity (and of any identified district-court litigation) is itself notable for a patent that issued in 2020 and is owned by a large entity (Alcatel Lucent / Nokia lineage) that has paid maintenance fees through at least 2023 — an asserted, revenue-relevant patent of this vintage would ordinarily have attracted an IPR by now.

There are no individual proceedings to report, so the per-proceeding template below does not apply. Rather than fabricate proceeding numbers, panels, or decisions, I state the negative result plainly and give the defensive analysis that follows from it.

Strategic summary

Claims CANCELED vs. SUSTAINED vs. UNTESTED. All 13 claims — independent claims 1, 6, 9, and 12, plus dependents 2–5, 7–8, 10–11, and 13 — remain in force and untested. No AIA trial has been filed, so there is no PTAB determination on any claim's validity. The only validity-related scrutiny on the record is the ordinary USPTO examination history (the application received a final rejection in 2019 and was then allowed, per the legal-event history on the Google Patents record). If a defendant is facing assertion today, every claim limitation in the four independent claims must still be addressed on its merits; there is no PTAB shortcut available from a prior proceeding.

Estoppel landscape — § 315(e)(2). Because no IPR/PGR/CBM has been instituted, no petitioner exists and no § 315(e)(2) estoppel has attached to anyone. All prior-art grounds remain fully available to any defendant who has not itself been through a trial on this patent. That said, a defendant should still check the ordinary civil-litigation estoppels and, more importantly, the § 315(b) one-year bar from service of a complaint: if a defendant has been sued and more than one year has passed since service, its own IPR window has closed (absent a different petitioner). There is also no evidence of any prior § 315(e)(1) estoppel from a co-defendant's proceeding, since none exists.

Pattern signals. No repeat-petitioner pattern exists (no petitioner at all). No defensive-aggregator activity (Unified Patents, RPX, OpenSky, etc.) surfaced in searches. There is likewise no evidence of patent-owner PTAB litigation aggressiveness — Alcatel Lucent (now within the Nokia corporate family) has not needed to defend this patent at the Board. One practical note: the Google Patents record shows the family was actively prosecuted and the patent has been maintained (4th-year fee paid 2023-07-19), which signals the owner (or its licensee) still considers it commercially valuable — the kind of asset that can emerge in assertion campaigns even without a prior IPR history.

Recommended next steps

• If you are a defendant facing assertion: The absence of PTAB activity does not weaken your § 102/§ 103 defenses — it means the art is fresh and unencumbered. Before filing your own IPR, verify the § 315(b) timing: an IPR petition must be filed within one year of service of a complaint alleging infringement of this patent. If you are within that window, a petition is viable; if you have been sued and the year has passed, your path is district-court validity litigation or (if the patent owner asserts in a venue where they can be counterclaimed) declaratory relief.
• Prior-art development: The closest art the examiner and family prosecution considered is visible on the record — e.g., Wang et al., "A Study on Virtual BS Live Migration – A Seamless and Lossless Mechanism for Virtual BS Migration," 2013 IEEE PIMRC (cited as XP055225133 during prosecution), plus the cited references US 2015/0036664 A1 (LG, handover in C-RAN systems), US 2013/0183991 A1 (Alcatel Lucent, base station access), EP 2717158 A1 (Huawei, virtualized cluster integration), and US 2008/0020801 A1 (distributed virtual base station manager). A fresh IPR would need to go beyond this cited art — the "reasonable could have raised" ground is measured against what a skilled searcher would find, not just the cited references, but the cited references are a floor for what the examiner already saw and overcame.
• Monitor for new filings: Re-check the USPTO Patent Trial and Appeal Board proceeding search (PTAB E2E) and the ODP API periodically. Given the patent's maintenance-fee payment and Nokia-family ownership, an assertion or an IPR could still appear; the absence found here is dated to the most recent ODP ingest and search date (2026-09-05) and is not a guarantee of future inactivity.
• Do not over-read the negative result: This report should not be characterized in court papers or to a client as "the patent was exonerated by the PTAB." It was never there. The correct framing is that the patent is validity-untested at the PTAB, which cuts both ways: no canceled claims to exploit, but no estoppel and no affirmed claims to overcome either.

Search-limitation disclosure: PTAB proceeding numbers, panels, decisions, and settlements cannot be invented, and none were found. The ODP-based "no proceedings on file" default, corroborated by web searches returning no IPR/PGR/CBM or CAFC activity for US 10554763, is a negative result subject to normal database-ingest lag and search coverage limits — not a definitive legal representation that no proceeding has ever been filed.

Search results were limited (multiple queries returned no hits on the USPTO Assignment Center or on post-issuance transfers for this patent), so the assignment reconstruction below rests on the one recorded assignment evidenced in the Google Patents legal-events record for US10554763, cross-checked against family/publication data. Where the correspondent of record is not evidenced, I say so rather than guessing.

Inventors

• Cheng Wang — employer at filing not directly evidenced; co-author (with Liyu Cai et al.) of the PIMRC 2013 paper "A Study on Virtual BS Live Migration," and the CN priority applications trace to Alcatel-Lucent's China operation (Alcatel-Lucent Shanghai Bell). Moderate confidence, not certain.
• Yun Deng — employer at filing not directly evidenced; same Alcatel-Lucent China context inferred from the assignment and CN family.
• Liyu Cai — listed in assignee profiles as an Alcatel Lucent inventor based in Shanghai, China; also co-author of the PIMRC 2013 paper on virtual BS live migration (the technical precursor to this patent).

Pattern note: no evidence any inventor departed the original assignee within 12 months of filing; all three assigned their interests to Alcatel Lucent in the ordinary course (reel 040159/0343, effective 2016-09-28).

Original assignee

• Alcatel Lucent SAS (named on the issued patent as original/current assignee; the recorded 2016 assignment names the assignee as "ALCATEL LUCENT, FRANCE").
• Line of business: telecommunications equipment and software (wireline/wireless networking, including C-RAN/Cloud-RAN research and products — the invention's subject matter).
• Whether a product shipped embodying the specific claims: unclear — C-RAN/virtualized-BBU technology was in Alcatel-Lucent's portfolio and research pipeline, but I found no evidence tying a specific shipping product to these claims.
• Current status: operating as a Nokia subsidiary. Nokia completed its acquisition of Alcatel-Lucent in January 2016 (before this patent issued in 2020); Alcatel Lucent SAS continues to exist as a Nokia-controlled French entity and is the recorded owner on the patent.

Assignment timeline

The USPTO Assignment Center could not be queried directly in this session (search attempts returned no retrievable records), so this timeline reflects the single assignment evidenced in the Google Patents legal-events record. That record is authoritative for this one entry but should be verified against Assignment Center before relying on it for litigation standing.

• 2016-09-28 (executed) / recorded 2016-10-28 — Reel 040159/0343
  • Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS INTEREST)
  • Assignor: Cheng Wang, Yun Deng, Liyu Cai (the inventors)
  • Assignee: ALCATEL LUCENT, France (recorded owner name; patent names Alcatel Lucent SAS)
  • Correspondent: not evidenced in the records I could retrieve — no attorney name or firm available for this entry
  • Context: standard inventor-to-employer assignment, executed ~17 months after filing and ~3 years before grant, during prosecution.

No post-issuance assignments were found. The absence of recorded post-issuance transfers in the retrievable record suggests the original assignee (Alcatel Lucent SAS, now Nokia-controlled) still owns the patent — consistent with the Google Patents status showing the patent Active with the 4th-year maintenance fee paid in 2023. This is itself a finding: no evidence of any transfer into an NPE, licensing LLC, or aggregator chain.

Timeline diagram

timeline
    title Ownership of US 10554763
    2014 : Priority filed in China by Alcatel Lucent
    2015 : PCT filed by Alcatel Lucent
    2016 : Inventors assign to Alcatel Lucent
         : Nokia completes Alcatel Lucent acquisition
    2020 : Patent issued to Alcatel Lucent SAS
    2023 : Maintenance fee paid 4th year

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No recorded transfer to any "IP / Patents / Licensing / Holdings / Ventures" LLC; the only recorded assignment (reel 040159/0343, 2016-10-28) runs from the inventors to the operating company Alcatel Lucent.
2. Known asserter in the chain — not present. No assignee matches Acacia, Marathon, Intellectual Ventures, Conversant, or any Unified Patents / RPX high-frequency plaintiff list; the assignee is Alcatel Lucent SAS / Alcatel Lucent France.
3. Repeat correspondent across the chain — unclear / not determinable. The correspondent of record for reel 040159/0343 was not retrievable in my searches. With only one recorded assignment, no recurrence analysis is possible.
4. Cascading transfers — not present. No chained LLC assignments in any time window; only a single inventor→company conveyance.
5. Pre-litigation transfer — not present. No infringement litigation naming this patent was found, and the single assignment (recorded 2016-10-28) predates grant (2020-02-04) by three years — it is a prosecution-stage formalities filing, not a standing-arrangement transfer.
6. Bankruptcy fire-sale — not present. Alcatel-Lucent did not go through bankruptcy; it was acquired by Nokia in a negotiated transaction completed January 2016. No court-supervised patent sale is evidenced.
7. Privateering — not present. No transfer of this patent from Nokia/Alcatel Lucent to a third-party NPE for competitor assertion is evidenced, and no SEC-filing or press coverage to that effect surfaced.
8. Defensive aggregator (anti-NPE) — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN; it appears to remain with the operating-company assignee.

Verdict

Insufficient data — the only recorded assignment found (reel 040159/0343, recorded 2016-10-28, inventors → Alcatel Lucent) is the original assignment, and no post-issuance transfer, litigation, or assertion activity surfaced in any search. The totality of available evidence points to the patent remaining with the original operating-company assignee, Alcatel Lucent SAS (now Nokia-controlled) — i.e., there are no NPE signals — but because the USPTO Assignment Center could not be directly queried in this session, I cannot certify the assignment record is complete. Verification link: https://assignmentcenter.uspto.gov/ (search by patent number 10554763).

I'll verify the USPTO record for 10554763 and gather details on each cited prior-art reference to assess § 102 relevance.

I have solid data on the major references. Let me verify the remaining cited references before compiling the prior-art analysis.

Prior-Art Analysis for US 10554763 B2 (Patent No. 10554763)

Scope note and legal framework applied

• Patent under analysis: US 10554763 B2, "Method and apparatus for virtual base station migration in BBU pool," Alcatel Lucent SAS; inventors Cheng Wang, Yun Deng, Liyu Cai; filed 2015-04-10 (US 15/307,527, from PCT/IB2015/000634); effective filing date / earliest priority 2014-04-29 (CN201410177997.8; a CIP priority CN201410605357.2 of 2014-10-29 is later and does not help); issued 2020-02-04.
• Granted claims: 13 claims; independent claims 1 (source-vBS method), 6 (migration-manager method), 9 (source-side apparatus), and 12 (manager-side apparatus).
• Because the application is governed by the AIA, references are § 102 prior art if they were patented, described in a printed publication, or publicly available before 2014-04-29 (§ 102(a)(1)), or are US patents/USPGPs with an effective filing date before 2014-04-29 naming different inventors (§ 102(a)(2)).
• Caveat on "anticipation": § 102 anticipation requires a single reference to disclose every claim limitation, expressly or inherently. For the references below I identify which claims are potentially anticipated and which limitations are clearly missing, so the analysis is usable for both § 102 and § 103 positions. An "X" characterization does not mean a claim is dead — most of these references are strongest in § 103 combinations. I did not perform a full-text element-by-element review of every Chinese-language reference; where confidence is lower I say so explicitly.

Summary table

Per-reference analysis

NPL-1 — Most relevant: Cheng Wang, Yang Wang, Chaohua Gong, Yan Wan, Liyu Cai, Qinglin Luo, "A Study on Virtual BS Live Migration – A Seamless and Lossless Mechanism for Virtual BS Migration," 2013 IEEE 24th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), pp. 2803–2807, XP055225133, DOI 10.1109/PIMRC.2013.6666624.

• Dates: Published 2013 (PIMRC conference held September 2013), well before the 2014-04-29 effective filing date. It is a printed publication under § 102(a)(1). Note it shares three co-authors with the patent (Wang, Cai; also Gong/Wan/Luo), so it is inventor-authored but still § 102 art because it was publicly available >1 year before the earliest non-provisional US filing (2015-04-10) — indeed it predates even the CN priority date.
• Description: The paper is the direct precursor study to the claimed invention. It analyzes live migration of virtual base stations within a C-RAN BBU pool for power savings, and proposes a "seamless and lossless" mechanism to hide the inherent service interruption from connected UEs — the same problem the patent's Background section describes.
• Claims potentially anticipated: It is the strongest candidate against claims 1, 2, and 5 (and apparatus equivalents 9 and 10) because it concerns the same vBS-migration-in-BBU-pool subject matter and the same goal of making migration transparent to UEs. Whether it discloses each specific claim limitation — e.g., measuring available bandwidth between source/target vBSs, estimating interruption time from UE count/data rate/bandwidth, the three-tiered choice among a common measurement gap / random-access slots / DRX cycle with thresholds, the migration-request-and-completion messaging — depends on the paper's full text, which I did not retrieve in full.
• Honest assessment: I can state with high confidence it is the closest prior art and should be treated as an X/Y reference against the independent source-side method/apparatus claims; I cannot confirm complete anticipation of claim 1 (with its post-selection messaging steps) without the full paper text. The ISR for PCT/IB2015/000634 (dated 2015-11-19, cited in the record) would show how the examiner treated it.

1. EP 2717158 A1 — Huawei Technologies, "Method and device for integrating virtualized cluster, and virtualized cluster system" (inventor: Zhao Yanbin)

• Dates: Priority 2012-08-21 (CN201210298873.6); published 2014-04-09 — ten days before the 2014-04-29 effective filing date, so it is § 102(a)(1)/(a)(2) prior art. (Verified via EPO/Google Patents and the EPO GPI record; abstract corroborated.)
• Description: Selects, based on cluster load, a physical machine whose virtual machines are to be consolidated; determines a migration time and an interrupt time for migrating each VM; computes a migration cost from the migration/interrupt times; selects the VM(s) to migrate and a target physical machine by load; then migrates. It is a data-center virtualization/power-management technique (G06F9/455, G06F1/32) for consolidating VMs onto fewer hosts — conceptually the same "consolidate-and-shut-down-idle-hardware" motivation as the BBU-pool energy savings.
• Claims potentially anticipated: It most directly overlaps the interruption-time estimation concept of claim 5 (estimating a service-interruption parameter used to decide migration) and the resource-usage-based migration decision of claim 6/12. However, it does not disclose the BBU-pool/cellular context, virtual base stations serving user devices, measuring available bandwidth between first and second virtual base stations, the migration-manager message protocol (preparation/ack/command/completion), UE context/buffer restoration, or any of the three LTE masking mechanisms (measurement gap / random access / DRX) of claims 2 and 10. It therefore does not by itself anticipate claims 1, 2, 6, 9, or 12, and is best used in § 103 combinations (e.g., with the PIMRC paper or LG's C-RAN references).

2. US 2015/0036664 A1 — LG Electronics (Yuk, Kim, Chung), "Method for performing handover in C-RAN systems, and apparatus therefor" (granted as US 9,526,049 B2)

• Dates: Priority 2012-02-23; published 2015-02-05. Because its publication postdates the 2014-04-29 effective filing date, it is not § 102(a)(1) art, but it qualifies under § 102(a)(2) as a USPGP with a 2012 effective filing date naming different inventors. (Verified via Google Patents and USPTO.report.)
• Description: A UE in a C-RAN (cloud RAN) receives candidate-RRH information, measures reception signal strengths of the serving cell and candidate RRHs, and reports when the measurements satisfy prescribed threshold relationships (serving-RRH strength ≥ first threshold and serving-cell-minus-RRH difference ≤ second threshold); a changed serving RRH is then indicated. It is a UE-measurement/handover procedure between RRHs in a C-RAN, not a virtual-base-station migration between BBUs.
• Claims potentially anticipated: It supplies C-RAN architecture context and threshold-based decisions but none of the claimed BBU-pool vBS migration elements: no available-bandwidth measurement between vBSs on different BBUs, no service-interruption-time estimation, no migration mechanism selection, no context/buffer transfer, and no manager messaging. No claim is anticipated. Useful as secondary C-RAN background art in § 103 grounds.

3. US 2008/0020801 A1 — Nestor Alexis Fesas, "Managing Wireless Base Stations Using a Distributed Virtual Base Station Manager"

• Dates: Priority 2006-07-20; published 2008-01-24 — § 102(a)(1)/(a)(2) prior art. (From the '54763 record; no independent full-text fetch was completed within the search budget.)
• Description: Based on the title and publication record, it concerns a distributed virtual base station manager that manages wireless base stations, i.e., an early virtualization/management layer over base stations. Its exact teaching (whether it covers load monitoring, re-homing of base-station functionality between compute resources, or manager-initiated control) could not be fully verified in this pass.
• Claims potentially anticipated: Superficially analogous to the migration-manager side (claims 6, 8, 12) in that a "manager" entity governs virtual base station resources. Nothing in the record indicates it discloses bandwidth measurement between vBSs on different BBUs, service-interruption-time estimation, the tiered measurement-gap/random-access/DRX selection, or the specific preparation/ack/command/completion messaging. No claim appears anticipated on the available record, but a full-text check is warranted before relying on that conclusion.

4. US 2009/0191874 A1 — Texas Instruments, "System and method for managing radio link failures" (granted as US 8,280,375 B2)

• Dates: Priority 2008-01-25 (provisional 61/023,704); published 2009-07-30 — § 102(a)(1)/(a)(2) prior art. (Verified.)
• Description: LTE handover and radio-link-failure management. A WTRU begins a handover upon receiving an HO command specifying a first time; if that time expires before handover completion, the WTRU reconfigures to a radio link interruption procedure and first attempts to re-establish the original link. The reference is precisely about the failure mode the '54763 patent seeks to avoid (a UE that loses PDCCH contact and triggers re-establishment during a service interruption).
• Claims potentially anticipated: It addresses the same problem as claim 1's background (avoiding UE-side RRC re-establishment during a network-side switchover) but solves it UE-side with timers/counters, not network-side by estimating interruption time and configuring measurement gaps / random-access slots / DRX to mask the migration. No claim anticipated; useful in § 103 as evidence that managing interruption-induced RLF was known.

5. US 2013/0183991 A1 — Alcatel Lucent (Bosch, Mullender), "Method and base station system for providing access to a mobile communication network" (granted as US 9,173,201 B2; EP 2434822 B1)

• Dates: Priority 2010-09-27; published 2013-07-18 — § 102(a)(1) prior art. (Verified; assignee Alcatel Lucent, inventors Peter Bosch and Sape Mullender — a different inventive entity from the '54763 inventors.)
• Description: Virtualizes a base station at the logical-channel layer so multiple virtual operators can offer service through virtual machines instantiated on a shared base-station computing platform; on a UE channel request, the station creates/instantiates a virtual machine for the UE's home operator, establishes a channel through it, and allocates resources. It teaches the "virtual base station as a VM/logical entity" concept but no migration between hosts/BBUs.
• Claims potentially anticipated: No migration decision, no interruption-time estimation, no bandwidth measurement, no masking mechanisms, no migration-manager protocol. It is background art for the general notion of virtualized base stations (relevant to claim construction terms like "first/second virtual base station"), but anticipates no claim.

6. CN 101321393 A — Tsinghua University, "A method and system for seamless handover"

• Dates: Filed 2008-07-03; published 2008-12-10 — § 102(a)(1) prior art. (From the '54763 record; full Chinese text not retrieved.)
• Description: A method/system for seamless handover, i.e., maintaining service continuity across a source-to-target radio switch. It is in the same problem space as claims 1–2 (avoiding user-perceived interruption during a base-station-side switch) but in the ordinary handover context.
• Claims potentially anticipated: Without the full text I cannot map elements precisely; nothing in the record indicates it discloses vBS migration between BBUs of a BBU pool, available-bandwidth measurement, interruption-time estimation from UE count/rate/bandwidth, or the measurement-gap/random-access/DRX selection. Anticipation unlikely on the available record; treat as background/§ 103 art.

7. CN 101267249 A — Huawei, "Antenna selection method, terminal and network side equipment in distributed wireless communication system"

• Dates: Filed 2007-03-13; published 2008-09-17 — § 102(a)(1) prior art. (From the '54763 record.)
• Description: Antenna selection in a distributed wireless communication system (the distributed-RRH-style architecture family). It does not concern base-station virtualization or migration.
• Claims potentially anticipated: None. It appears cited only as general distributed-architecture background; no overlap with the claimed migration mechanism.

8. US 2016/0192181 A1 — LG Electronics, "Method of obtaining RRU information by BBU, and BBU"

• Dates: Priority 2013-09-10; published 2016-06-30. Publication postdates the effective filing date, so it is not § 102(a)(1) art, but with a 2013-09-10 effective filing date (before 2014-04-29) and different inventors it qualifies under § 102(a)(2).
• Description: A BBU obtains RRU information in a C-RAN BBU–RRU architecture — relevant to the general C-RAN architecture (BBU pools, RRHs) but not to migrating a virtual base station between BBUs.
• Claims potentially anticipated: None. It does not disclose the migration decision, bandwidth measurement, interruption-time estimation, masking mechanisms, or manager messaging of claims 1–13. Background/§ 103 art only.

9. JP 2013-541289 A — Alcatel-Lucent, Japanese publication corresponding to US 2013/0183991 A1

• Dates: Priority 2010-09-27; published 2013-11-07 — § 102(a)(1) prior art.
• Description/Analysis: Same disclosure as reference 5 (JP family member). Anticipates no claim, for the same reasons.

10. US 2008/0069088 A1 — Matsushita Electric Industrial, "Hierarchy Encoding Apparatus and Hierarchy Encoding Method"

• Dates: Priority 2004-04-29 (per the Google Patents record — note this numeric date coincidence with the '54763 CN priority year, 2014-04-29; I am not treating them as related); published 2008-03-20 — § 102(a)(1) prior art.
• Description: Hierarchical (layered) encoding apparatus/method — an audio/video or signal-encoding reference. On its face it is outside the wireless-base-station-migration field.
• Claims potentially anticipated: None identified. Its presence in the citation list is unexplained from the available record (it may relate to an encoding or data-rate concept used in a dependent limitation, or be a miscataloged/third-party citation). Flagging uncertainty: I could not verify why the examiner or a third party cited it; do not rely on it as substantive art without pulling the file wrapper.

Ranking of most relevant prior art

1. Cheng Wang et al., PIMRC 2013 (NPL-1) — closest art; direct precursor study on lossless vBS live migration in a BBU pool by the inventors; primary candidate against claims 1, 2, 5, 9, 10. Highest priority for full-text retrieval.
2. EP 2717158 A1 (Huawei) — closest third-party art for the load-based migration decision and migration/interruption-time-cost computation (claims 5, 6/12); best-in-class § 103 anchor. Note its publication date (2014-04-09) narrowly precedes the priority date, so confirm the priority chain if relying on it.
3. US 2015/0036664 A1 / US 9,526,049 (LG) — best C-RAN-context reference (§ 102(a)(2)-eligible); secondary in § 103 grounds.
4. US 2009/0191874 A1 (TI) — best art on the UE re-establishment/RLF problem the invention solves; supports obviousness of the "mask the outage" motivation.
5. US 2008/0020801 A1 (Fesas) and US 2013/0183991 A1 (Alcatel) — background on virtual-base-station management/virtualization; no anticipation.

Uncertainty disclosure

• Full texts of the Chinese references (CN 101267249 A, CN 101321393 A), US 2008/0020801 A1 (Fesas), and the Matsushita reference (US 2008/0069088 A1) were not retrieved in full within the search budget; their element-level mappings rest on titles, the '54763 record, and (where available) abstracts. Element-by-element anticipation conclusions for those should be treated as provisional.
• The PIMRC 2013 paper's exact disclosure (in particular whether it teaches the measurement-gap/random-access/DRX threshold selection of claim 2 and the bandwidth-measurement step) needs full-text verification; the PCT ISR of 2015-11-19 would show the examiner's X/Y designation.
• The star/asterisk markers in the Google Patents citation lists distinguish "cited by examiner" from "cited by third party"; I did not rely on those markers for any substantive conclusion.
• This analysis addresses § 102 potential anticipation only, not § 103 obviousness (where several of these references combine strongly) and not claim construction.

§103 Obviousness Analysis — US 10,554,763 B2 ("Virtual base station migration in BBU pool")

0. Scope note and ground rules

This analysis uses only the references appearing in the "Prior Art" (Patent Citations / Non-Patent Citations) sections of the Google Patents record for US 10,554,763, plus the family/citation data printed on that record. I treat the record's own cited art as the universe for this exercise. I verified content via live search for the key references (Cheng Wang et al. 2013 PIMRC paper; Huawei EP2717158A1; Alcatel US20130183991A1; LG US20150036664A1; Fesas US20080020801A1). For references where only the title was available from the record, I say so rather than inventing disclosure content.

AIA regime assumed. The patent's earliest priority application (CN201410177997.8, 2014-04-29) postdates the March 16, 2013 AIA transition, so §102/§103 analysis is under AIA. Relative to the 2014-04-29 effective filing date, the following cited items are citable prior art (with one important caveat flagged in §7):

The later "Families Citing" documents (e.g., CN105472681A, 2016; US11533376B2, 2022) and the patent's own family members (EP3138357B1, CN105025541B) are not prior art and are excluded.

1. Person of ordinary skill in the art (PHOSITA)

A PHOSITA at the 2014 priority date would be a team: (i) a wireless systems engineer with 3–5 years' LTE/3GPP RAN protocol experience (RRC, RLC/PDCP, HARQ, DRX, measurement gaps, PRACH, X2/S1, C-RAN BBU-pool architectures) and (ii) a systems/virtualization engineer familiar with VM live migration, load-based consolidation, and resource/threshold management. This team definition matters because the claimed invention sits at the LTE-protocol × datacenter-virtualization intersection, and §103 combines knowledge from both fields (KSR, v. Teleflex, 550 U.S. 398, 417 (2007)).

2. Claim inventory (for cross-reference)

• Claim 1 (independent, method at source vBS): measure inter-BBU available bandwidth → estimate service interruption time → select a migration mechanism in accordance with the estimated time → send migration request with C-/U-plane data → send completion message to migration manager.
• Claim 2 (dependent): tiered masking: (≤1st threshold) common measurement gap or random-access slots or DRX; (1st–2nd threshold) random-access slots or DRX; (>2nd threshold) DRX.
• Claims 3–5: preparation message; prep-ack then migration command; interruption-time estimate from #UEs, per-UE data rate, available bandwidth.
• Claim 6 (independent, migration-manager method): collect BBU resource usage → decide (threshold) source vBS + target vBS on different BBUs → prepare both → receive prep-acks → command source → receive completions.
• Claims 7–8: threshold preset/dynamic; send prep when source BBU usage < threshold.
• Claims 9 & 12 (independent, apparatus): processor-implemented mirrors of claims 1 and 6; claims 10, 11, 13 mirror claims 2, 3–4, 7.

3. Closest prior art and what each actually teaches

3.1 Huawei EP2717158A1 — the strongest §103 anchor (consolidation + interrupt-time-based selection)

Verified from the record and live search (abstract and description): the method selects, according to a cluster load, a physical machine to be integrated (i.e., workload consolidation); determines a migration time and an interrupt time used to migrate each VM; computes migration cost from those times; selects which VM to migrate; selects a target physical machine; migrates; and powers off the emptied physical machines. The interrupt time is modeled from the memory volume to be transferred before migration, the memory dirtying rate, and the network transmission rate during migration (Tn = M·Dn/Rn+1 style expressions). The stated purpose is load-drop consolidation for power saving and service-level preservation.

This maps with near one-to-one correspondence onto the problem the target patent addresses (consolidate lightly loaded base stations during idle hours, power off unused BBUs — the patent's Background section and the PIMRC paper both say exactly this) and onto two of the patent's mechanism elements: estimate an interruption time from a data volume and a transfer rate, and select a migration course of action based on that estimated interruption time. EP2717158 is in the VM/datacenter art, but the patent itself defines a virtual base station as "a logic entity realized based on … virtual machines," which is the express bridging admission a challenger needs for analogous-art arguments.

3.2 Cheng Wang et al., PIMRC 2013 (NPL) — the system-level blueprint

Verified: same Bell Labs (China)/Alcatel-Lucent team as the patent (inventors C. Wang and L. Cai are co-authors). The paper discloses, for a C-RAN BBU pool: a Migration Manager issuing an execution command; activation of the target eNB; IQ sample-path switching; data forwarding of four kinds of C-/U-plane data (including PDCP/RLC/HARQ buffers and status — the exact data the patent's claims 1/9 "migration request" carries); S1 downlink path switching; consolidation "on idle hours like midnight … so that multiple vBSs can be consolidated onto a single BBU board" (the patent's identical premise); a user-perspective definition of service interruption time ("interval between the time when the UE receives the last packet from the source eNB and the time when the UE receives the first packet from the target eNB" — verbatim in the patent); and an analysis that interruption time T is dominated by data-forwarding time X where X depends on the amount of data to forward and hence on the number of active users and traffic rate.

The paper does not disclose the UE-masking ladder (measurement gap / PRACH / DRX) — that is the patent's delta. Caveat on citable status: see §7.

3.3 US20080020801A1 (Fesas) — distributed virtual base-station manager

Verified: wireless base stations self-organize into clusters; a master/base-station "management point" collects information from member WBSs; the cluster "collaborates" for fast handoff, load balancing; the master makes the handoff/load-balancing decision and coordinates by sending instructions to the affected base stations. This supplies the "manager entity that gathers resource/state information, decides, and issues commands" architecture for claim 6/12 — albeit in a Wi-Fi-like, non-C-RAN context.

3.4 US20130183991A1 (Alcatel-Lucent) — virtualized base station platform

Verified: a base transceiver station with a computation platform hosting a plurality of virtual machines each dedicated to a virtual operator, with VM instantiation, resource allocation to VM instances, and air-interface exchange through the VM instance. Same assignee as the patent; establishes the "virtual base station = VM on base-station compute platform" architecture and shows that vBS resource/VM management was a live problem in Alcatel's own prior work.

3.5 LG US20150036664A1 — C-RAN virtual base station sharing and mobility

Verified: C-RAN handover with BBU/RRH split, "RRH and a virtual base station sharing scenario in a C-RAN," software-based virtual base station clusters, and UE measurement/reporting/feedback procedures controlling RRH-level mobility. Useful to show C-RAN virtual-BS environments and UE-measurement-driven serving-point changes were known.

3.6 US20090191874A1 (TI) — UE-side radio-link-failure management

By title and field: managing RLF. Useful only as evidence that the consequence the patent seeks to avoid — a UE that observes a signaling outage and declares RLF/RRC re-establishment, breaking service — was a known, named problem, and that UE-side timers/behaviors around gaps are managed by configuration.

3.7 Remaining references

CN101321393A (Tsinghua, "seamless handover") plausibly teaches buffering/forwarding to hide handover interruption, but I verified no content; I do not rely on it beyond its title. CN101267249A (antenna selection) and US20080069088A1 (hierarchy encoding) appear tangential; a challenger would not rely on them for anything claim-relevant.

4. Element-by-element obviousness: independent claims

4.1 Claim 1 (source-vBS method) — Combination A

Primary: PIMRC 2013 paper (or, if it is held non-citable, Huawei EP2717158A1 + Fesas). Secondary: EP2717158A1; US20080020801A1; Alcatel US20130183991A1; LG US20150036664A1.

Why a PHOSITA would combine: The PIMRC paper and EP2717158 solve the identical problem — consolidate idle virtualized workloads onto fewer physical hosts to power off machines — in the identical way — model the interruption (data volume ÷ transfer rate) and plan the migration accordingly. EP2717158 provides the missing "estimate first, then choose what/how to migrate based on the estimate" algorithm; the paper provides the C-RAN-specific data set (UE contexts, RLC/PDCP/HARQ buffers) and the Migration-Manager orchestration. Both are power-consolidation art, so combining them is combination of prior-art elements "according to known methods to yield predictable results" (KSR, 550 U.S. at 416). The architecture references (Alcatel '991: vBS-as-VM; Fesas: distributed manager with decision-and-command messaging) fill any gap on virtualization platform and manager messaging. No element of claim 1 requires an inventive leap once this combination is assembled.

4.2 Claim 6 (migration-manager method) — Combination B

Primary: EP2717158A1 (cluster-level load-triggered consolidation decision). Secondary: Fesas US20080020801A1; PIMRC paper; Alcatel US20130183991A1.

Claim 6 is a dressed-up version of a load-threshold-triggered consolidation controller — precisely EP2717158's cluster integrator — with the generic message names (prepare/ack/command/complete) that any implementer would use to sequence a two-phase migration. That claim form (processor + named message units) does not rescue it: the message functions correspond one-for-one to steps already performed by EP2717158's integrator and Fesas's master/manager.

4.3 Claims 9 and 12 (apparatus mirrors)

Claim 9's "one or more processors configured to [measure / estimate / select]" plus transmitting units tracks claim 1 element-for-element, and claim 12's units track claim 6. Under §103, where a method is obvious, a conventional apparatus implementing it with standard processors/transceivers is equally obvious (functional claiming of conventional hardware; no non-obvious structural limitation is recited). Same combinations apply.

5. Dependent claims

5.1 Claim 5 (and its apparatus counterpart's limitation) — estimate from #UEs, data rate, available bandwidth

Combination A1 = PIMRC paper + EP2717158A1. The paper states the forwarding volume X depends on the number of active users and their traffic rate (the #UEs and per-UE data-rate inputs); EP2717158 teaches the arithmetic — interruption/migration time = f(volume to transfer, transfer rate) — where EP's "memory before migration + dirtying rate" is the VM-domain analogue of "UE contexts/buffered data + incoming data rate," and EP's "network transmission rate" is the patent's "available bandwidth." Substituting the C-RAN data objects into EP's known rate-and-volume model is an obvious application of an established formula to the paper's already-quantified inputs. No new math is invented: the patent itself computes nothing beyond volume ÷ bandwidth (spec: "calculates the required data forwarding time … based on the number of user devices … data rate and available bandwidth").

5.2 Claim 2 — the tiered ladder (the real battleground)

This is the only claim element with no single-reference disclosure. The honest analysis:

For the challenger. The three "mechanisms" are pre-existing, standardized LTE mechanisms with standardized time envelopes, and the patent admits this: measurement gaps are network-configured silent intervals for inter-frequency measurement (6 ms gaps per LTE gap patterns); the PRACH configurations in the patent's own Table 1 (reproduced from 3GPP TS 36.211) yield random-access sub-frames spaced up to 20 ms; DRX cycles are standardized up to 2560 ms with the patent reciting the standard value set (10…2560 ms). The TI '874 reference shows that the problem the ladder solves — a UE that sees no PDCCH for a period triggering RLF/RRC re-establishment — was a known UE-behavior problem managed by network configuration. Once the PHOSITA knows (i) the migration interruption lasts an estimated number of milliseconds (PIMRC/EP2717158), and (ii) the UE must not notice an absence of serving-cell signaling for that long (TI '874; standard DRX/gap behavior), the design task reduces to: pick the smallest standardized UE-silence mechanism whose standardized envelope covers the estimated outage — measurement gap for single-digit-ms outages, PRACH-backed gaps for tens of ms, DRX sleep for longer. That is a predictable, "obvious to try" selection among a finite set of known options with known parameter ranges, driven by the well-known engineering motivation of minimizing UE disruption and data loss. The specific thresholds (6 ms / 20 ms) are not critical; they simply track the maximum standardized envelope of the adjacent mechanism (a 6 ms gap pattern; a 20 ms PRACH periodicity), which the specification itself discloses as examples ("for instance," "typically greater than the estimated service interruption time"). Under KSR, a "finite number of identified, predictable solutions" with a "design need or market pressure" supports obviousness.

For the patentee. No cited reference teaches coordinating a base-station migration with UE measurement-gap/PRACH/DRX scheduling at all, much less a threshold-based mapping. The cited art addresses VM migration interruption (EP2717158) and C-RAN mobility (LG '664) but not "mask the migration outage from connected UEs by aligning the migration with standardized UE sleep/measurement/access windows." Whether that step is "routine optimization" or an "unpredictable" cross-domain insight (moving the UE-visible silence before the outage so the UE is never actually starved) is the crux; a validity challenge would need expert testimony that a protocol engineer would naturally think of scheduling the outage inside a UE's already-silent window. This is the patent's strongest surviving feature and the most likely reason the examiner allowed it after the 2019 final rejection.

5.3 Claims 3, 4, 7, 8, 10, 11, 13 — conventional sequencing and thresholds

• Claim 3 (preparation message before measuring): EP2717158 (parameter determination before migration), Fesas (parameter distribution to the group before acting). Trivial.
• Claim 4 (prep-ack → command): generic two-phase handshake; PIMRC's command sequence and Fesas's coordinator-instruction protocol; LG '664's C-RAN handover signaling environment. Trivial.
• Claims 7–8 (threshold preset/configurable; trigger when source-BBU usage below threshold): EP2717158's load-drop-triggered integration ("demands … drop … physical machines with an empty load are controlled to be powered off") is literally a below-threshold consolidation trigger. Trivial.
• Claims 10, 11, 13: mirror 2, 3–4, 7; same analysis.

6. Motivation-to-combine synthesis (Graham/KSR)

1. Same problem, same motivation. Every primary reference shares the patent's stated objective: idle-hour workload consolidation to power off unused hardware (PIMRC paper: consolidate vBSs "onto a single BBU board" to cut BBU-pool power; EP2717158: power off emptied physical machines to cut cluster power; the patent's own Background says exactly this). An examiner/challenger needs no "teaching, suggestion, or motivation" beyond the shared, express goal of energy-efficient consolidation.
2. Analogous art. The patent itself defines vBS as a VM ("logic entity realized based on threads/processes or virtual machines"), making EP2717158's VM-live-migration interrupt-time modeling and Fesas's virtual base-station manager directly analogous. Alcatel '991 (same assignee) already treated a base station as a VM-hosting platform.
3. Known, bounded problem; finite, known solutions. Service interruption during vBS migration was disclosed and quantified (PIMRC paper, ~tens of ms); the UE-visible failure mode (no PDCCH → RRC re-establishment) was known (TI '874); the masking tools (measurement gap, PRACH, DRX) were standardized LTE mechanisms with standardized envelopes that the patent itself recites. Selecting among them by outage duration is the kind of "predictable variation" and "obvious to try" design work KSR treats as obvious.
4. Reasonable expectation of success. The mechanisms are standards-defined; combining them with a migration coordinator requires only configuring existing RRC/PHY parameters around a known outage window — no new protocol, no new hardware.

7. Honest limitations and counterweights (stated plainly)

1. The PIMRC paper's citable status is genuinely contestable. It was authored by two of the patent's named inventors and published 2013-09-08/11. If the 2014-04-29 CN priority application validly supports the claims, the paper falls inside the §102(b)(1)(A) one-year inventor-disclosure grace window and is not prior art for those claims. If the effective filing date is the 2014-10-29 continuation-in-part (CN201410605357.2) or the 2015-04-10 PCT filing — which a challenger would argue for the masking/threshold subject matter that appears to be the CIP's contribution — the paper is >1 year old and is §102(a)(1) prior art. Either way, the paper is powerful evidence of what the same team considered routine, but its formal status must be resolved before relying on it.
2. No single reference (or combination in the cited record) discloses the claim-2 ladder — common measurement gap / PRACH slots / DRX selected by interruption-time thresholds. The strongest challenger theory is KSR "obvious to try"/design-choice; the strongest patentee theory is that aligning a network-initiated migration with UE sleep/measurement/access windows is an insight absent from all cited art. This is the claim most likely to survive.
3. Claim 1's breadth cuts both ways. It is easier to attack (any "migration mechanism" selection based on an estimated time; EP2717158 arguably reads on it), but if "migration mechanism" is construed through the specification as limited to the three masking options, claim 1 inherits claim 2's vulnerability question.
4. Prosecution history caution. The file wrapper shows a final rejection (2019-02-27) followed by allowance (2019), indicating claims were contested and likely amended/narrowed; without the rejection grounds I cannot confirm what prior art was applied, and this analysis should not be read as predicting how those grounds were overcome.
5. Tangential citations. CN101267249A (antenna selection) and US20080069088A1 (hierarchy encoding) appear unrelated and add nothing; CN101321393A and LG US20160192181A1 are characterized only by title/field because I could not verify deeper content — do not rely on unverified details.

8. Bottom line

Overall assessment: Under §103, independent claims 1, 6, 9, and 12 and the mechanical dependent claims (3, 4, 7, 8, 11, 13) present a substantial obviousness exposure on combinations of EP2717158A1 (interrupt-time estimation and load-threshold consolidation), the PIMRC 2013 paper (vBS migration orchestration and forwarding data set), and Fesas (manager-based coordination), with the Alcatel '991 and LG '664 references supplying the virtualized base-station/C-RAN environment. The genuinely contestable feature — and the one most likely to carry the patent if litigated — is claim 2's tiered selection of UE-masking mechanisms (measurement gap/PRACH/DRX) against estimated service-interruption time, which is not disclosed in any cited reference and whose obviousness rests entirely on a KSR design-choice theory.