Samsung · Filed Jun 13, 2025 · Published Jul 23, 2026 · verified — real USPTO data

Samsung Patent Covers AI/ML Data Reporting Methods for 5G Dual Connectivity Networks

Your phone already juggles two separate cell connections at once in parts of 5G. Samsung wants those two connections to feed real-time data to AI models running inside the network, so the network can make better decisions about how to manage your call or download.

Samsung Patent: AI/ML Data Collection in 5G Dual Connectivity — figure from US 2026/0214486 A1
Figure from the official USPTO publication.
Publication number US 2026/0214486 A1
Applicant Samsung Electronics Co., Ltd.
Filing date Jun 13, 2025
Publication date Jul 23, 2026
Inventors Donggun KIM
CPC classification 370/252
Grant likelihood Medium
Examiner CENTRAL, DOCKET (Art Unit OPAP)
Status Docketed New Case - Ready for Examination (May 13, 2026)
Parent application is a National Stage Entry of PCTKR2023095067 (filed 2023-11-02)
Document 15 claims

What Samsung's dual-network AI data collection actually does

Imagine your phone is connected to two cell towers at the same time, which is something modern 5G networks actually do to boost speed and reliability. One tower is the main one; the other is a helper. Samsung's patent describes a way for your phone to act as a data collector for AI tools running inside the network, gathering measurements from both connections and reporting them back.

The network sends your phone a configuration message that tells it exactly what data to gather and where to send it. For the main tower's AI, your phone reports to the main tower. For the helper tower's AI, it reports there. The setup even works when your phone isn't in an active call, covering the low-power "idle" and "inactive" states most phones spend a lot of time in.

The end goal is giving network operators a steady stream of real-world signal data so their AI systems can optimize coverage, predict congestion, and allocate bandwidth more efficiently, all without you noticing a thing.

How the UE collects and routes AI/ML data across two cell groups

The patent covers a protocol-level procedure for how a phone (called a user equipment, or UE in cellular standards language) participates in AI/ML data collection inside a dual connectivity network. Dual connectivity means the phone maintains simultaneous radio links to two separate cell groups: a Master Cell Group (MCG) and a Secondary Cell Group (SCG).

The phone first establishes a control connection (called an RRC connection, which is the standard signaling channel 5G phones use to talk to a base station) with the master node. Over that channel, the master node sends at least one configuration message. That message can include:

  • Master measurement configuration: instructions for the phone to collect AI/ML-relevant data (signal quality, load metrics, interference readings) and report it to the master node.
  • Secondary measurement configuration: instructions to do the same thing but report to the secondary node instead.

Once configured, the phone gathers the requested measurements and routes them to whichever node the configuration specified. The patent also handles the case where the phone is in IDLE or INACTIVE state (not in an active session), so AI data collection doesn't have to pause just because no call or download is happening.

The technical contribution here is the signaling architecture: defining exactly how configuration messages are structured and how reporting responsibilities are split between the two nodes.

What this means for AI-driven 5G and future 6G networks

AI-driven network management is a major theme in both 5G rollouts and the early planning for 6G. Operators want AI models that can predict where congestion will hit, pre-position resources, and cut interference, but those models need fresh, real-world data from actual devices. This patent carves out the signaling scaffolding that makes that data collection possible in the specific case of dual-connectivity setups, which cover a growing share of premium 5G deployments.

For everyday users, the impact would be indirect: better-tuned networks, fewer dropped connections in crowded areas, and more consistent speeds. For Samsung as a network infrastructure vendor (it sells base stations and core network gear, not just phones), locking down this signaling design in a patent is a way to shape how the industry standardizes AI data collection at the radio layer.

Editorial take

This is deep cellular-standards plumbing, the kind of patent that matters to network equipment vendors and 3GPP standards committees far more than to consumers. Samsung sells both handsets and base stations, so owning pieces of how AI data collection is defined at the protocol level is a legitimate business move. It's not exciting reading, but it's exactly the kind of filing that shapes what 6G looks like.

Which company should we read for you?

We track 17 companies here. Pro is the same weekly breakdown for any company you choose, delivered privately. Type a name and we'll scope it and send you a quote.

Get one Big Tech patent every Sunday

Plain English, intelligent commentary, no hype. Free.

Source. Full patent text and figures from the official USPTO publication PDF.

Editorial commentary on a publicly published patent application. Not legal advice.