Book a demo

RAN Intelligent Controller AI Patents: Who Leads, Gaps 2026

RAN Intelligent Controller AI Patents: Who Leads, Gaps 2026
https://www.patsnap.com/resources/blog/rd-blog/ran-intelligent-controller-ai-and-machine-learning-patent-landscape/ · Patsnap · data cut-off 2026-07-31 · downloaded from the live page
Patent Landscape · Telecom & Wireless
RAN Intelligent Controller Patents: Mapping AI and Machine Learning Claims in Open RAN
  • Filings peaked in 2022 at 82 records and have since flattened, suggesting the core RIC architecture claims are largely staked out rather than still being built.
  • H04W wireless-network claims dominate at 202 of 208 records while only 36 records sit in G06N, showing most filings protect the RAN integration layer, not the underlying ML models.
  • The two most-cited filings share one title "Radio access network intelligent application manager" (WO2023091664A1 and its US continuation) account for 237 and 118 citations, anchoring the field's prior art.
Get a prior-art report on your approach
208
Published Records
75%
Top-5 Share of All Records
+25%
Filing Growth 2021→2024
US
Leading Jurisdiction

Filing growth compares 2021 (24 records) with 2024 (30) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 208 records in scope (CR5), not by the ranked leaders only.

Published byPatsnap Research··7 min readSourced from Patsnap Eureka
Overview

What this landscape covers

This dataset tracks 208 patent families published between 2015 and mid-2026 that combine RAN Intelligent Controller (RIC) architecture with machine learning functions such as traffic prediction, self-optimizing network control, and anomaly detection. The search string ties title/abstract mentions of RIC, Open RAN controllers or RIC platforms to classification codes covering wireless network management (H04W24/02), AI model computation (G06N20) and wireless communication arrangements (H04W28), so the set is narrow to RIC-specific ML claims rather than general Open RAN filings.

Publication activity is concentrated in a single filing office triad: the United States, the European Patent Office and WIPO's PCT route together account for the large majority of receiving-office activity, with India a distant but notable fourth. Because publication lags filing by roughly eighteen months, the apparent slowdown after 2022 should be read with that lag in mind rather than as a definitive peak.

Filing activity, 2017-2026
  1. 1RAKUTEN SYMPHONY INC44
  2. 2RAKUTEN MOBILE INC43
  3. 3SAMSUNG ELECTRONICS CO LTD43
  4. 4INTEL CORP14
  5. 5AT&T INTELLECTUAL PROPERTY I L P12
  6. 6RAKUTEN MOBILE USA LLC10
  7. 7VMWARE INC8
  8. 8DELL PROD LP7
  9. 9AT&T MOBILITY II LLC6
  10. 10BOOST SUBSCRIBERCO LLC4
Source: Patsnap Eureka. Assignee ranking and totals. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP

Let an AI agent run this analysis on your own technology

Pick a task. Every answer cites the patents behind it.

10,000 free credits to start
Filing Data

Trend and technology composition

The filing curve and the IPC breakdown together show where claim activity actually sits: in RAN-side control logic rather than in the machine learning models themselves.

Filings rose sharply through 2022, then flattened

Annual filings moved from zero in 2017 to a peak of 82 in 2022, the dataset's midpoint year. Volumes since then have not exceeded that peak, consistent with a technology whose core architectural claims (non-RT and near-RT RIC interfaces, SMO orchestration) were substantially filed during the 3GPP and O-RAN Alliance standardization push of 2020-2022.

Filings rose sharply through 2022, then flattened025507510002017201820192020202182202220232024202522026Most recent year is partial — publication lag means later filings are not yet visible.

H04W wireless networking dominates the classification mix

202 of 208 records carry an H04W code, versus 84 in H04L (digital transmission) and 36 in G06N (AI computing). That gap indicates most applicants are claiming how ML output is applied inside the RAN control loop rather than novel model architectures, which sit more often in dedicated G06N filings elsewhere.

H04W wireless networking dominates the classification mixH04W · Wireless communication networks20297.1%H04L · Digital information transmissi…8440.4%G06N · Computing based on AI models3617.3%H04B · Transmission (general)199.1%G06F · Electric digital data processi…125.8%B64C · Aeroplanes & helicopters21.0%G05B · Control & regulating systems21.0%G06K · Data recognition & presentation10.5%Other10.5%

Shares are the percentage of the 208 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.

Source: Patsnap Eureka. Filing trend and technology composition. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.

Go deeper on RAN Intelligent Controller AI and Machine Learning with Eureka

This page is one run against one query. Ask Eureka your own question about ran intelligent controller ai and machine learning and every answer comes back with the patent numbers behind it.

Try Eureka
Key Patents

The filings anchoring this space

Representative filing
US20210337420A12021-10-28

Functional architecture and interface for non-real-time RAN intelligent controller

SAMSUNG ELECTRONICS CO., LTD.

A service management and orchestration (SMO) entity enabling a functional split between a non-real-time RAN intelligent controller and an external AI/ML server will, during a data collection phase, collect and process RAN and non-RAN data and, during a data transfer phase, transfer that data to the external server via an interface. During a training model input phase, the SMO entity receives a trained AI/ML model, metadata and training results back via an interface.Filed by Samsung Electronics; published 2021-10-28.

US20210337420A1 — patent drawing 1US20210337420A1 — patent drawing 2
View full filing
Most-cited records in the dataset
#Publication no.Patent titleCitations
1WO2023091664A1Radio access network intelligent application manager237
2US20240259879A1Radio access network intelligent application manager118
3US20220167236A1Intelligence and Learning in O-RAN for 5G and 6G Cellular Networks99
4US20210337420A1Functional architecture and interface for non-real-time ran intelligent controller47
5US20210029580A1Flexible buffer management for optimizing congestion control using radio access network intelligent controlle…46
6US20210266907A1Facilitation of radio access network intelligent controller resource preservation framework for 5g or other n…29
7US20200314622A1Beam as a service for 5g or other next generation network21
8US20220286914A1RIC sdk18
9US20220286915A1Distributed ric17
10US20240098568A1Automated detection and mitigation of intra- and interdomain conflicts in open radio access networks16

Citation counts are drawn from within this searched corpus and favour earlier-filed records; treat them as a measure of influence on later applicants, not of current commercial relevance.

Each row carries its publication number; clicking a row searches Eureka by that number.

Source: Patsnap Eureka. Citation counts and representative records. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Run it yourself

Put your own technology through the same analysis

 
Where to run it
Fastest

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 →
For builders

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 →
Insights

What the numbers mean for strategy

Three patterns stand out once filing volume, classification mix and citation concentration are read together.

Filing curve
82 in 2022
peak filing year

The architecture land-grab has slowed

Filings rose from zero in 2017 to 82 at the 2022 midpoint and have not surpassed that level since. New entrants are more likely to face occupied claim space on core non-RT/near-RT RIC interface architecture than an open field.

Recency caveat: the most recent year is understated due to publication lag.
Classification mix
202 vs 36
H04W records vs G06N records

Integration claims outnumber model claims six to one

Only 36 of 208 records fall in G06N (AI computing), against 202 in H04W (wireless networks). Applicants are overwhelmingly claiming how predictions and anomaly scores get applied to RAN parameters, not the learning algorithms that produce them.

A gap this wide suggests model-layer claims remain comparatively open.
Citation concentration
237 citations
top-cited record

Two related filings anchor the prior art

WO2023091664A1 and its US continuation, both titled "Radio access network intelligent application manager," carry 237 and 118 citations respectively — far ahead of the next-ranked records. Any freedom-to-operate review in this space should start there.

Older filings accumulate citations by tenure; treat the count as influence, not necessarily relevance today.
Collaboration
10 co-assignee pairs
joint-filing relationships

Filing partnerships are narrow and carrier-centric

Only 10 co-assignee pairings appear across 208 families, and the strongest links sit within single corporate groups rather than across competitors — an operator and its own subsidiaries, or a carrier and its own IP holding entity.

Cross-company joint filings remain rare in this dataset.
Eureka AI Agent
Looking for what nobody has claimed yet?

Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to ran intelligent controller ai and machine learning, with the prior art for and against each one.

Find the white space →
Source: Patsnap Eureka. Co-assignee relationships and derived observations. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Players

Who is filing, and where momentum is shifting

Recent-year momentum figures show a field where even the more active filers are pulling back, not accelerating.

Momentum
-50% YoY
Rakuten Mobile

Rakuten group entities are still filing, but slower

Rakuten Mobile and Rakuten Symphony each show a single filing in the latest year, down 50% and 75% year-on-year respectively. Their strongest co-assignee link in the dataset (11 shared filings) confirms tight internal coordination on RIC claims across the group.

Group-internal collaboration is the dominant pattern here, not cross-company alliances.
Momentum
0 in latest year
Samsung, Intel

Large incumbents show no new filings in the latest year

Samsung Electronics and Intel both register zero filings in the most recent year captured. Given the 18-month publication lag, this likely reflects reporting delay rather than an actual halt, but it means recent competitive signal from these filers is currently invisible in the public record.

Re-check this cohort once later filing years backfill.
Momentum
-100% YoY
AT&T IP entities

AT&T's filing entities have gone quiet

AT&T Intellectual Property I and AT&T Mobility II, which co-file together (6 shared records), show a full year-on-year drop to zero in the latest year. Combined with the broader post-2022 plateau, this points to a filer that front-loaded its RIC claims during the standardization window.

Watch for renewed filing once next-gen RIC specs (e.g. 6G-era O-RAN work) mature.
🔍
Under-claimed sub-areas worth a closer FTO check
Branches where filing density is thin relative to the core RIC architecture claims.
xApp/rApp conflict resolution logicnon-RT RIC to near-RT RIC model handoffRAN anomaly root-cause attributionenergy-aware traffic prediction modelsmulti-vendor RIC interoperability testing
Rank all filers by momentum →
Recent-year filing momentum by assignee
AssigneeRecent yearYoY
Rakuten Mobile, Inc.1-50%
Rakuten Symphony, Inc.1-75%
Samsung Electronics Co., Ltd.0
Intel Corporation0
RAKUTEN MOBILE USA LLC0
AT&T Intellectual Property I, L.P.0-100%
Rakuten Symphony Singapore Pte. Ltd.0
VMware, Inc.0
Source: Patsnap Eureka. Assignee-level momentum. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
What's Next

Where to take this analysis

The dataset points to specific follow-up work depending on whether the goal is freedom-to-operate, licensing, or identifying open filing space.

Run a claims-level FTO check on the top-cited pair

WO2023091664A1 and US20240259879A1 carry the highest citation counts in the set and share a title, indicating a continuation family worth reading claim-by-claim before filing adjacent architecture claims.

Open claim comparison in Eureka

Track the post-2022 plateau against later filing years

Because publication lags filing, the apparent flattening after 2022 needs re-checking as 2025-2026 records continue to backfill into the public record.

Set up a filing alert in Eureka

Map the G06N-only subset separately

Only 36 of 208 records sit in G06N; isolating that subset shows which applicants are claiming the ML model itself rather than its RAN integration.

Build a G06N subset view in Eureka
Source: Patsnap Eureka. Forward-looking reading of the same dataset. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
FAQ

Common questions on RIC AI patents

Answers are grounded in the same dataset. Derived from a Patsnap search on RAN Intelligent Controller AI and Machine Learning covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP

Research RAN Intelligent Controller AI and Machine Learning in depth with Eureka

Go past this page: query the whole ran intelligent controller ai and machine learning corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.

Try Eureka

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.

Help us improve this page

Found incorrect or outdated information? Let us know and we'll get it fixed.