Headwater Research: Ten Patents From One 2009 Provisional

Four lineages from one disclosure

The ten patents were issued between 2013 and 2024, the latest, the ’804, on 30 April 2024. They sit in four lineages: usage billing and service mediation; differential network access for background and foreground applications and for capacity protection; a shared secure message channel for many applications; and virtualised, cloud-based policy, charging and service-plan design. Every one descends from the same provisional naming Dr. Raleigh, with the earliest non-provisional filings dating from March 2009.

The enforcement model is plaintiff-selected and patent-by-patent. Each complaint pairs a defendant with the patents that map to its business, so a mobile platform, a messaging service and a cloud provider face different subsets of the same family. The accused instrumentalities differ from case to case and fall into four groups: mobile operating-system network-access controls, wireless data and messaging services, push-notification channels, and cloud service-management, metering and billing infrastructure.

Headwater has cited earlier trial and settlement outcomes in this family, reported at roughly $279 million and $175 million, as evidence that accused parties knew of the portfolio. Because the patents are in force, the damages period is still open and the claims can still be extended by continuation. That is why a challenge to each patent’s effective date matters as much as the prior art itself.

Texas complaints, one IPR, and a thin public record

Docket-level data was available for the matters below. The list covers matters that could be tied to the supplied patents from supplied and public sources, and further matters are to be added once confirmed.

Headwater is represented by Russ August & Kabat. Marc Fenster signed the most recent complaint, and Reza Mirzaie, Brian Ledahl, Dale Chang, Kristopher Davis, James Pickens, James Tsuei and Jason Wietholter appear in the matters reviewed. Venue in these complaints is typically built on the defendant’s Texas offices and personnel, and the 2026 filings were made in the Midland/Odessa division.

Procedural pushback, and pleading detail to test

The pushback so far is procedural and PTAB-driven rather than dispositive. No dispositive ruling is reported in the matters reviewed.

The last two are points an accused party can test in discovery on willfulness and on the damages start date.

The same kind of function at different layers

Across the campaign, the accused functionality is software that decides, per application, whether and when network data may flow; software that counts and reports that use; and services that deliver messages or enforce quotas on behalf of many applications. In the handset cases the focus is the operating system’s background-data and network-access controls. In the messaging cases it is a push channel shared by many applications. In the cloud cases it is service management, quota, metering and billing infrastructure. No single product serves as a benchmark, so each patent is measured against the specific feature its complaint charts.

The engineering problem was that, before this family, the operator controlled and counted data inside the network. Gateways, billing mediation and call-detail records saw aggregate traffic per subscriber but not which application on the phone produced it. A device could not be told to hold back a background update on a congested cell, or to bill one application differently, because the network could not see the application. Remedies were blunt: a single throttle for the whole subscriber, a flat cap, or traffic inspection in the core.

The shift is from aggregate, network-side control to per-application, device-side control, with the network and later the cloud acting as policy source and verifier.

Ten patents, one system

Priority is the 28 January 2009 provisional (61/206,354), confirmed for the family from publication data and for the ’117 and ’544 chains specifically. The chain for each other patent should be confirmed on Patent Center before it is relied on in a filing.

On-device classification and differential access (’976, ’359, ’445, ’544). This is the handset-facing core. An end-user device classifies its own network activity, by application, by foreground or background status, or by network condition, and applies a different access rule to each class. The ’544 adds the capacity rationale: the device monitors network service usage activities, classifies them for differential access control and associates each with a control policy that can block, throttle, delay, queue by priority, restrict to a time window, suspend or quarantine. Its published claims refer to a list of capacity-controlled services, to congestion state and to modem power-state changes that create bursts of demand. The engagement brief associates the ’359 with changes of network-access status tied to device state, such as display lock and wake; that reading should be checked against the claims.

Verifiable usage measurement and billing (’364, ’781). The ’364 covers device-assisted billing in which the device measures service use and the network reconciles it with its own accounting, across more than one account. The ’781 recites a network system in which a service policy is tied to a particular software component on the device, and a plurality of usage measurements, for example over a cellular home network, a roaming network or Wi-Fi, is used to decide whether the device is using the service in accordance with the policy. The common thread is verification: the claim does not stop at device-side counting but requires a way for the network side to check it.

A shared secure message channel (’117). It replaces one connection per application with a network message server that receives requests from many application servers, each naming a device and an application. The server turns each request into a message carrying an application identifier, and a messaging agent on the device maps the identifier to the right application over a secure inter-process channel. Claim 1 is directed to this network system, and its purpose is to consolidate many small transmissions into one managed secure connection.

Virtualised policy, charging and plan design (’518, ’837, ’804). The ’518 and ’837 move every policy function that is not on the data path into a “service policy cloud”, leaving a policy enforcement function in the network that talks to policy elements through a network element. The specification describes stateless processes and the ability to harmonise networks with different hardware. The ’804 extends the line to the design of service plans and their enforcement, and the complaint charts claim 1 against cloud service-control, identity, quota, metering and billing functionality.

By tying all ten patents to one 2009 disclosure, the campaign creates a multi-layered legal net: the same specification supports claims on the handset, the messaging service, the billing system and the cloud.

What the twelve months before 2009 show

The review covers 28 January 2008 to 27 January 2009, using a supplied 50-record reference set. Of the 50 unique records, 32 belong to Headwater’s own family and are excluded. Eighteen come from other owners. Of those, six have a verified priority date inside the window, one has a priority date after it (Oracle, charging enabler), and eleven could not be dated from available sources or concern unrelated fields, and are not relied on.

Earlier work senses, routes or prices at the network, at an intermediary or inside a secure module, but not through a device agent that classifies application activity and applies differential access.

The claims reviewed combine three elements that none of the in-window references shows together: identification of the application or activity on the device; a policy that treats that activity differently by network type, foreground or background state or congestion; and a way for the network or cloud to verify or bill what the device reports. The in-window set is only a slice, however. Its closest member, the Irdeto reference, is not about data services, and a motivated accused party will search earlier and wider than twelve months.

Claim scope, effective date, and ten patents in parallel

Claim scope. Terms such as “differential network access control,” “service activity,” “foreground classification” and “capacity-controlled services” will be fought over. A broad reading reaches ordinary operating-system background-data settings, while a narrow reading ties them to the verified, policy-driven system described in the 2009 specification.

Effective date. This is the most important point for an invalidity strategy. Most asserted claims issued years after the 2009 provisional, some as late as 2022 and 2024. Each claim is entitled to the 28 January 2009 date only if the original disclosure supports it. If an accused party shows that a later claim adds subject matter not described in 2009, the relevant date moves forward and a large body of intervening art, including the operating-system and messaging features of 2010 to 2015, becomes available. The ’518, ’837 and ’804, the furthest from the root, are the natural first targets. Other likely lines are obviousness combinations built from device-metering and network-policy art, subject-matter eligibility challenges to the most abstract claims, and inter partes review.

Parallel assertion. The patents address different layers of the same system, so each can be pointed at a different feature of a different defendant, and each defendant faces separate validity and non-infringement questions on each patent.

A family-driven campaign with the clock still running

This looks like a mature, family-driven enforcement effort rather than an opportunistic one: new complaints in 2025 and again in late August 2026, a continuation issuing as recently as 2024, and prior reported outcomes that give the plaintiff leverage. Because the patents remain in force, the damages period continues to run. The six-year look-back in 35 U.S.C. § 286 and the marking rules in § 287 limit recovery before suit, and both vary by patent and by defendant.

Four developments are likely to move the campaign: PTAB practice, meaning the outcome of IPR2026-00138, institution decisions on other family patents and whether district courts grant stays; marking, in light of VDPP and the reported settlement licenses in this family; ownership and pleading detail, such as the plaintiff’s inconsistent formation dates and the chain of title for each asserted patent; and claim construction of the terms noted above.

Any device maker, platform, carrier, messaging or cloud provider should treat active or threatened assertion of this family as a live exposure, and expect further complaints while the later continuations remain pending.