Network Slicing Patents: Who Leads, Where the Gaps Are 2026
- Filing has flattened, not grown. Filings peaked at 35 in 2022 and 2026 sits back at 3, the same level as 2017 — this is a field that has already had its filing wave, not one still accelerating.
- No single assignee dominates recent activity. The latest-year momentum table shows most tracked assignees at 0 or 1 filing, including a -100% YoY drop for one major holder — leadership here is unsettled, not locked in.
- Claims cluster in two IPC subclasses. H04L (109 records) and H04W (98 records) carry almost all the activity, while AI-assisted orchestration (G06N, 2 records) is barely claimed at all.
What the patent record shows about network slicing and orchestration
Network slicing and orchestration patents cover the mechanisms that let a single physical network run multiple logical, isolated services — resource isolation, lifecycle management of slices, multi-domain coordination across operator boundaries, and quality-of-service guarantees enforced per slice. The dataset behind this page covers 145 patent families filed between 2015 and 2026, drawn from filings that combine network-slicing or service-based-architecture terminology with claim language on resource isolation, lifecycle management, multi-domain operation or intent-based networking, classified under H04L41, H04W28 or H04L47.
Filing activity rose through the late 2010s, peaked in 2022, and has since pulled back to levels last seen at the start of the window. Because publication typically lags filing by around eighteen months, the final one to two years in any such trend will always look thinner than they eventually turn out to be — but a decline from a mid-window peak back to early-window levels is a distinct pattern from simple under-counting at the edge.
Filing trend and technology composition
145 patent families, filed mainly through the USPTO with a secondary cluster at the EPO and via PCT, concentrated in two IPC subclasses.
A peak-and-pullback filing curve
Annual filings ran 3 in 2017, climbed through the late 2010s and early 2020s, peaked at 35 in 2022, and have since fallen back to 3 in 2026 (a partial year). This is not a technology still in its growth phase; the bulk of foundational claim-staking already happened.
Concentration in transmission and wireless subclasses
H04L (digital information transmission, 109 records) and H04W (wireless communication networks, 98 records) dominate the classification mix. G06F (electric digital data processing, 17), H04B (transmission general, 6) and H04J (multiplex communication, 5) form a smaller supporting tier, while G06N (AI models, 2) and G06Q (business data processing, 2) are barely represented — a sign that AI-driven orchestration and commercial/billing logic for slices remain lightly claimed relative to the core transmission mechanics.
Shares are the percentage of the 145 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Network Slicing and Orchestration with Eureka
This page is one run against one query. Ask Eureka your own question about network slicing and orchestration and every answer comes back with the patent numbers behind it.
Try EurekaThe most-cited filings and a representative claim
Resilient radio resource provisioning for network slicing (US20220124560A1, Intel Corporation, 2022-04-21)
The filing describes a resilient radio access network slicing framework combining a resource planning engine with distributed, slice-aware scheduling at access nodes, edge compute nodes or cloud services. It covers resource planning, slice-aware scheduling, and the signalling exchanges needed to provision resilient RAN slicing, with the stated aim of achieving resource isolation more efficiently and agilely than prior network slicing approaches.Filed by Intel Corporation; published 2022-04-21.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20170164349A1 | Method and system for performing network slicing in a radio access network | 311 |
| 2 | US20180343567A1 | Private multefire network with SDR-based massive MIMO, multefire and network slicing | 308 |
| 3 | US20170142591A1 | System and methods for network management and orchestration for network slicing | 219 |
| 4 | US11711759B1 | On-demand application-driven network slicing | 121 |
| 5 | US20230179974A1 | Locality-based network slicing in radio-based networks | 119 |
| 6 | US20220124560A1 | Resilient radio resource provisioning for network slicing | 114 |
| 7 | US20230006889A1 | Flow-specific network slicing | 103 |
| 8 | CN108260169A | 一种基于QoS保障的服务功能链动态部署方法 | 77 |
| 9 | US20240305533A1 | Radio resource planning and slice-aware scheduling for intelligent radio access network slicing | 66 |
| 10 | US9369387B2 | Segment routing based wide area network orchestration in a network environment | 66 |
Citation counts favour older filings simply because they have had more time to accumulate citations within this searched corpus — read them as a signal of influence on the field, not as a ranking of current importance.
Patent titles are shown in the language they were filed in, not translated, so that each record stays verifiable against the original filing — a translated title will not match in Eureka or in any national register. Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the filing pattern signals for R&D and IP strategy
Three things stand out once the raw counts are put in context: where activity is concentrated, how recent momentum has shifted, and where claim density is thin enough to be worth entering.
The wave has already crested
Filings rose from 3 in 2017 to a peak of 35 in 2022, then declined back toward single digits by 2026. A field that peaks mid-window and pulls back is different from one still climbing — the core architectural claims (slice isolation, lifecycle state machines, QoS enforcement) were largely staked between roughly 2019 and 2022.
No assignee is pulling away
Latest-year figures show most tracked assignees at 0 or 1 filing, and one major prior filer down 100% year-on-year. Combined with a flat co-assignee count (2 pairs total across the dataset), this reads as a fragmented field with little sustained collaboration or single-firm dominance in current filing.
AI-assisted orchestration is barely claimed
Against 109 H04L and 98 H04W records, only 2 records fall under G06N and 2 under G06Q. Intent-based networking and AI-driven slice orchestration are named in the search criteria but show up in only a handful of the underlying classifications, suggesting the claim space around automated, model-driven orchestration decisions is still open.
Filing is concentrated in the US, with a European and PCT tail
The United States receives 88 of the tracked filings, more than four times the EPO total of 21 and the WIPO/PCT count of 19. India (7), China (6) and the Netherlands (3) trail well behind, indicating that enforcement and freedom-to-operate risk in this dataset is weighted heavily toward US practice.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to network slicing and orchestration, with the prior art for and against each one.
Assignee activity and where the field is still open
Recent-year filing counts are low across the board, and the strongest recent signals come from assignees with just one filing in the latest year rather than an established leader extending a lead.
Several firms tied at the top of a thin field
Ericsson, Samsung and Amazon each show 1 filing in the latest tracked year, with Ericsson flat year-on-year. None of the assignees in the momentum table show growth beyond that single-digit level, which is consistent with a field past its peak filing year rather than one still building.
A prior active filer has gone quiet
Intel, which appears as the assignee on the representative filing discussed above, shows 0 filings in the latest year against a -100% year-on-year change, and Huawei similarly shows 0 in the latest year. Both remain part of the historical filing base but are not currently adding to it at pace.
Co-filing is rare
Only two co-assignee pairs appear across the full 145-family dataset, both involving Huawei alongside individual named inventors. This is a field of largely single-assignee filings rather than joint ventures or cross-licensed development programmes.
| Assignee | Recent year | YoY |
|---|---|---|
| Telefonaktiebolaget LM Ericsson | 1 | 0% |
| Samsung Electronics Co., Ltd. | 1 | — |
| Amazon Technologies, Inc. | 1 | — |
| Intel Corporation | 0 | -100% |
| Huawei Technologies Co., Ltd. | 0 | — |
| Verizon Patent and Licensing Inc. | 0 | — |
| Nicigen Partners IP LLC | 0 | — |
| AT&T Intellectual Property I, L.P. | 0 | — |
Where to take this from here
The filing curve and classification mix point to specific next checks rather than a single conclusion.
Check freedom-to-operate against the most-cited filings
The five most-cited records in this corpus, several from 2017, set foundational claim scope on slicing architecture and orchestration that later filings had to design around. Any new filing in core resource-isolation or slice-management claims should be checked against these first.
Run a claim comparison in EurekaTrack the momentum shift among named assignees
With Intel down and Huawei flat at zero in the latest year, and no assignee showing sustained growth, the competitive picture is unsettled. Monitoring quarterly filing activity rather than annual snapshots will catch a new leader emerging earlier.
Set up assignee monitoring in EurekaExplore the thin AI-orchestration classification
Only 2 of 145 records fall under G06N, despite intent-based networking appearing in the search criteria. This gap is worth a deeper prior-art search before committing to claim language in that space.
Search white space in EurekaCommon questions about network slicing patents
In this dataset, a network slicing patent is a filing whose title or claims reference network slicing, network orchestration or service-based architecture, combined with claim language on quality-of-service guarantees, resource isolation, lifecycle management, multi-domain operation or intent-based networking, and classified under H04L41, H04W28 or H04L47. That combination captures the mechanisms for carving a shared physical network into isolated logical services rather than every telecom patent that happens to mention '5G' or 'slicing' in passing. Analysts should always check the underlying search string before comparing counts across different landscape reports, since a broader or narrower definition changes the totals substantially.
The dataset does not show one dominant, currently-active leader; several assignees including Ericsson, Samsung and Amazon each hold just one filing in the most recent tracked year, and historically active filers like Intel and Huawei show zero filings in that same year. Intel's presence in the most-cited and representative filings reflects earlier activity rather than current output. Anyone assessing 'leadership' here should separate historical filing volume, which is concentrated among a handful of large telecom and cloud vendors, from recent-year momentum, which is thin and spread across many single-filing entrants.
No, not on the filing-count evidence: filings rose from 3 in 2017 to a peak of 35 in 2022, then fell back to 3 by 2026. Some of that recent-year drop is an artefact of publication lag, since filings typically take about eighteen months to publish and so the last one to two years always look artificially thin. But a decline all the way back to early-window levels, rather than a modest dip at the very edge, is a stronger signal that the bulk of core architectural claiming has already occurred.
The classification data points to AI-assisted and business-logic aspects of orchestration: only 2 of 145 records fall under G06N (AI models) and 2 under G06Q (business/commerce data processing), against 109 in H04L and 98 in H04W. That suggests intent-based, model-driven slice orchestration decisions and billing-aware or SLA-reconciliation logic across multi-domain slices are named in the field's terminology but not heavily claimed yet. A prior-art search focused specifically on those narrower branches, rather than the core resource-isolation mechanics, is the more productive place to start a new filing.
US20220124560A1, filed by Intel Corporation and published 2022-04-21, describes a resilient radio access network slicing framework: a resource planning engine paired with distributed, slice-aware scheduling modules at access nodes, edge compute nodes or cloud services, plus the signalling needed to provision that resilience. It targets more efficient and agile resource isolation than prior slicing approaches. It is a useful reference point for anyone building resilient RAN-level slice scheduling, but it is one filing among 145 tracked families, and its scope should be checked against the independent claims directly rather than the abstract before assuming it blocks a specific implementation.
Research Network Slicing and Orchestration in depth with Eureka
Go past this page: query the whole network slicing and orchestration corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.