PulseLink Systems LLC v. Hewlett Packard Enterprises Co

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: PulseLink Systems, LLC (Texas)
  • Defendant: Hewlett Packard Enterprise Company (Delaware)
  • Plaintiff's Counsel: Skiermont Derby LLP
• Case Identification: 2:26-cv-00475, E.D. Tex., 06/15/2026
• Venue Allegations: Plaintiff alleges venue is proper because Defendant Hewlett Packard Enterprise Company (HPE) maintains a regular and established place of business in the district at its Frisco, Texas office, and has committed acts of infringement in the district.
• Core Dispute: Plaintiff alleges that Defendant's enterprise networking products, including its Aruba switches, access points, and EdgeConnect SD-WAN appliances, infringe four patents related to network tunneling, power management over Ethernet, and dynamic network path optimization.
• Technical Context: The lawsuit involves foundational technologies in modern enterprise networking, including data center fabric design (EVPN-VXLAN), Power over Ethernet (PoE) management for edge devices, and Software-Defined Wide Area Networking (SD-WAN) for optimizing application performance.
• Key Procedural History: The complaint alleges that Plaintiff is a non-practicing entity. It further alleges that HPE and its predecessors/acquired companies (HP Development Company, L.P. and Juniper Networks, Inc.) had pre-suit knowledge of certain asserted patents or their applications through their own patent prosecution activities. The complaint also states that Plaintiff provided HPE with notice of the asserted patents via correspondence on May 6, 2026, prior to filing the lawsuit.

Case Timeline

Date	Event
2006-10-03	'619 Patent Application Filing Date
2007-01-05	'677 Patent Application Filing Date
2007-05-15	'845 Patent Earliest Priority Date
2007-09-27	'845 Patent Application Filing Date
2010-09-10	'230 Patent Priority Date
2010-11-01	'230 Patent Application Filing Date
2011-05-10	'677 Patent Issue Date
2011-06-30	HPE predecessor (HPDC) allegedly knew of '677 application
2011-10-25	'619 Patent Issue Date
2012-10-16	'845 Patent Issue Date
2013-10-22	'230 Patent Issue Date
2015-01-08	Juniper allegedly knew of '230 patent application
2025-07-02	HPE acquisition of Juniper Networks
2026-05-06	Plaintiff sent notice letter to HPE
2026-06-15	Complaint Filing Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 8,565,230 - "Shared Virtual Tunnels Supporting MAC Learning in Communication Networks" (issued October 22, 2013)

The Invention Explained

• Problem Addressed: The patent describes a problem in Ethernet Virtual Private Network (EVPN) dual-homing, where a single source device is connected to a network through two separate physical switches. A remote switch receiving traffic from this source may see packets arriving over two different tunnels, causing it to repeatedly learn and re-learn the same source MAC address against different tunnels. This "MAC-learning churn" increases processing load and can lead to packet delays, packet loss, and device failure (Compl. ¶28, citing '230 Patent, col. 2:66-3:25).
• The Patented Solution: The invention proposes creating a "shared virtual tunnel" that uses a "single virtualized source device address" to represent both physical switches in the dual-homed pair (Compl. ¶29, citing '230 Patent, claim 1). This causes the remote switch to learn only one virtual tunnel for the source, regardless of which physical path the packet travels. This is intended to eliminate MAC-learning churn, minimize CPU loads, and improve overall network stability (Compl. ¶30, citing '230 Patent, col. 9:47-54; '230 Patent, col. 10:54-11:4).
• Technical Importance: This technique aims to improve the scalability and resilience of dual-homed network architectures, which are fundamental for building highly available data center and campus networks.

Key Claims at a Glance

• The complaint asserts independent claim 1 Compl. ¶26
• Claim 1 recites a method comprising the following essential elements:
  • Identifying a transport network with switching devices that use tunnels to transport data.
  • Identifying a first and second switching device that provide a dual-homed connection for a first access network to the transport network.
  • Creating a shared virtual tunnel connecting the first and second switching devices to a third switching device.
  • The shared virtual tunnel defining a single virtualized source device address that represents both the first and second switching devices.
  • In response to receiving a data packet from the access network, encapsulating the packet with the single virtualized source device address and transmitting it via the shared virtual tunnel.
  • The shared virtual tunnel extends from a first and second edge device (comprising the first and second switching devices) to a third edge device.
• The complaint expressly reserves the right to assert additional claims Compl. ¶32, fn. 1

U.S. Patent No. 8,046,619 - "Apparatus and Methods for Data Distribution Devices Having Selectable Power Supplies" (issued October 25, 2011)

The Invention Explained

• Problem Addressed: The patent notes that Power over Ethernet (PoE) sources can become overburdened when too many connected devices draw power, potentially causing the source to malfunction and reducing network reliability (Compl. ¶50, citing '619 Patent, col. 1:28-33). This creates a need for devices that can select power from multiple sources to improve reliability (Compl. ¶50, citing '619 Patent, col. 1:46-50).
• The Patented Solution: The patent discloses a network apparatus that can be powered from two different sources: a first input that provides both data and power (e.g., a PoE port) and a second input that provides only power (e.g., a connection for an AC/DC adapter) (Compl. ¶51, citing '619 Patent, col. 1:59-2:3). The apparatus includes sensors to detect the presence of power on each input and a controller that selects which source to use based on a set of logic (Compl. ¶¶51-52, Compl. ¶citing '619 Patent, Compl. ¶¶col. 8:3-33; '619 Patent, Compl. ¶Fig. 3).
• Technical Importance: This invention provides a mechanism for power source redundancy and intelligent power management in edge network devices, increasing their reliability and deployment flexibility.

Key Claims at a Glance

• The complaint asserts independent claim 1 Compl. ¶49
• Claim 1 recites a network apparatus comprising the following essential elements:
  • A data distribution device with a first input (power/data), second input (power-only), first output (power/data), and second output (data-only).
  • A first sensor to detect a power signal at the first input.
  • A second sensor to detect a power signal at the second input.
  • A controller connected to both sensors.
  • The controller electrically connects the first input to the first output when only the first power signal is present, and electrically connects the second input to the first output when only the second power signal is present.
• The complaint reserves the right to assert additional claims Compl. ¶54

Multi-Patent Capsule: U.S. Patent No. 7,941,677

• Patent Identification: U.S. Patent No. 7,941,677, "Apparatus and Methods for Managing Power Distribution over Ethernet," issued May 10, 2011 Compl. ¶16
• Technology Synopsis: The patent addresses power management for networked devices, like wireless access points, that have multiple power sources (e.g., PoE and an independent adapter) and also provide power to auxiliary devices (e.g., via a USB port) (Compl. ¶¶72-73, Compl. ¶citing '677 Patent, Compl. ¶¶col. 1:8-17; '677 Patent, Compl. ¶¶col. 1:35-48). The invention describes a device with a power detection unit to sense the active power source and a controller that controls the operation of an auxiliary device based on which source is active, thereby preventing power overloads (Compl. ¶73, citing '677 Patent, claim 16). The patent specification describes this control, for example, by enabling or disabling a USB port based on real-time power measurements (Compl. ¶74, citing '677 Patent, col. 16:13-29).
• Asserted Claims: Independent claim 16 Compl. ¶70
• Accused Features: HPE wireless access points, such as the AP-505H, that have both a PoE-in port and an external power adapter input (Compl. ¶¶77; 80). The complaint alleges these devices can determine the active power source and, based on that determination, control auxiliary functions like disabling a USB port or restricting power to a Power Sourcing Equipment (PSE) output port Compl. ¶83

Multi-Patent Capsule: U.S. Patent No. 8,289,845

• Patent Identification: U.S. Patent No. 8,289,845, "Assured Path Optimization," issued October 16, 2012 Compl. ¶17
• Technology Synopsis: The patent addresses the problem of network "brownouts," where performance degradation (e.g., high latency or packet loss) makes an application unusable even though network connectivity technically still exists-a problem traditional routing protocols like BGP do not handle well (Compl. ¶96, citing '845 Patent, col. 2:21-55). The invention describes a method to actively monitor multiple communication paths using test packets, calculate performance parameters, compare them to thresholds, and, upon detecting degradation, move the communication session in real-time to a better-performing path (Compl. ¶95, citing '845 Patent, claim 1). The patent explains this allows for control over the return path of measurements and ensures they mirror the actual data traffic path (Compl. ¶99, citing '845 Patent, col. 4:18-44).
• Asserted Claims: Independent claim 1 Compl. ¶94
• Accused Features: HPE Aruba Networking EdgeConnect SD-WAN products, which are alleged to automate traffic steering across multiple WAN transports like MPLS and broadband (Compl. ¶¶101; 104). The complaint alleges these products use processors to select and monitor path parameters (e.g., loss, latency), compare them against performance thresholds defined in "Business Intent Overlays," and, upon detecting degraded performance, move application traffic in real-time to a different path (Compl. ¶¶105; 106; 113).

III. The Accused Instrumentality

Product Identification

The primary accused instrumentalities are HPE's Aruba-branded enterprise networking products. For the '230 Patent, the accused products are the Aruba Networking CX Switch Series running the AOS-CX operating system Compl. ¶32 For the '619 and '677 Patents, the accused products are various Aruba wireless access points and network switches with dual power capabilities, such as the AP-505H (Compl. ¶¶54; 77). For the '845 Patent, the accused products are HPE Aruba Networking EdgeConnect SD-WAN appliances and associated software Compl. ¶101

Functionality and Market Context

• The accused Aruba CX switches are high-performance switches used to build modern data center and enterprise campus networks. The complaint focuses on their "EVPN-VXLAN with VSX logical VTEP" functionality, which allows two physical switches to be virtualized and appear as a single logical switch to the rest of the network, a key feature for building resilient, high-availability network fabrics (Compl. ¶¶34; 37). The complaint includes a network topology diagram from HPE's documentation showing a pair of switches (Leaf1A/Leaf1B) forming a "Logical VTEP#1" Compl. p. 12, Fig. 1
• The accused Aruba access points and edge switches are devices deployed at the edge of a network to provide wireless or wired connectivity to end-user devices. The complaint highlights their ability to be powered either via PoE from a network switch or a local DC power adapter, with documentation stating that "When both PoE and DC power sources are available, the DC power source takes precedence" (Compl. ¶57; Compl. ¶59, citing HPE documentation). The complaint provides a screenshot of an HPE product page for the accused AP-505H access point, showing it is available for purchase Compl. p. 30
• The accused EdgeConnect SD-WAN products are appliances and software used to manage and optimize traffic across a wide area network (WAN). Their core function is to create "Business Intent Overlays" that define application-specific quality-of-service requirements and automatically steer traffic across different WAN links (e.g., MPLS, internet) to meet those requirements based on real-time path performance Compl. ¶¶103-104

IV. Analysis of Infringement Allegations

'230 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
A method for packet switching and routing in a computer network that provides dual homed access, the method comprising: identifying a transport network...the transport network using tunnels to encapsulate and transmit data packets between respective switching devices;	The Aruba EVPN-VXLAN overlay fabric and supporting physical links are identified as the transport network. The overlay network is created using VXLAN tunnels between VTEPs (Virtual Tunnel Endpoints).	¶35	col. 8:36-46
identifying a first switching device that connects a first access network to the transport network;	A first VSX peer (e.g., "Leaf1A" in HPE's diagrams) is identified as the first switching device, connecting an access network to the EVPN-VXLAN transport network.	¶36	col. 8:47-50
identifying a second switching device that also connects the first access network to the transport network, the first and second switching devices providing the first access network with a dual homed connection to the transport network;	A second VSX peer (e.g., "Leaf1B") is identified as the second switching device. The two VSX peers together are alleged to provide the dual-homed connection.	¶37	col. 8:51-56
creating a shared virtual tunnel that connects the first switching device to a third switching device... the shared virtual tunnel defining a single virtualized source device address representing both the first switching device and the second switching device;	The network system uses "logical VTEP" functionality where both VSX peers use a single "anycast VXLAN source identity." This common identity is alleged to be the "single virtualized source device address" and creates the "shared virtual tunnel." A diagram shows a "Logical VTEP" spanning two physical switches (Compl. p. 18).	¶38	col. 8:57-67
in response to receiving a data packet from the first access network via the dual homed connection, encapsulating the data packet with the single virtualized source device address and transmitting the encapsulated data packet via the shared virtual tunnel to the third switching device...	An ingress VSX peer (either the first or second) receives a packet and encapsulates it using the common "Logical VTEP/anycast VXLAN source identity" before transmitting it through the shared virtual tunnel to a remote switch/VTEP.	¶39	col. 9:1-8

• Identified Points of Contention:
  • Scope Question: A central issue may be whether HPE's software-based "logical VTEP" and "anycast" IP address, which allow two switches to share a tunnel endpoint identity, meet the claim limitation of "creating a shared virtual tunnel" that defines "a single virtualized source device address." The defense may argue that the accused implementation is a routing and addressing configuration, not the creation of a distinct "tunnel" structure as contemplated by the patent.
  • Technical Question: The complaint alleges the system "creates" a shared virtual tunnel. The evidence for this is based on high-level technical and configuration guides. A key factual question for the court will be how this "creation" process actually occurs in the accused system's code and control plane, and whether that technical reality aligns with the patent's description.

'619 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
A network apparatus comprising: a data distribution device having a first input configured to receive data distribution device operating power and data, a second input configured to receive only data distribution device operating power...	The accused apparatus is the AP-505H access point, which is alleged to be a data distribution device (Compl. ¶56). The first input is the PoE-in port E0, and the second input is the 48V DC power input Compl. ¶57 An installation guide diagram shows both inputs on the device (Compl. p. 27).	¶¶56-57	col. 1:59-2:3
...a first output configured to output external device operating power and data and a second output configured to output data without external device operating power;	The first output is alleged to be the PSE-enabled Ethernet ports E1 and E2, which can power other devices. The second output is alleged to be the non-PSE ports E3 and E4 (Compl. ¶58). A diagram shows these four output ports (Compl. p. 28).	¶58	col. 7:38-45
a first sensor for determining whether a first power signal is present at said first input; a second sensor for determining whether a second power signal is present at said second input; and	The complaint alleges the AP-505H "can sense whether a power signal is present at the first input, i.e., input port E0, and whether a power signal is present at the second input, i.e., DC power input," implying the existence of the claimed sensors Compl. ¶59	¶59	col. 8:7-12
a controller operably connected to said first sensor and said second sensor,	The complaint alleges the AP-505H includes a controller connected to the sensors that determines which power source to use (Compl. ¶60).	¶60	col. 8:23-26
wherein, when said controller determines that a first power signal is present at said first input and a second power signal is not present...electrically connecting said first input to said first output, when said controller determines that a first power signal is not present...electrically connecting said second input to said first output.	The complaint alleges that when DC power is not available, the PoE input can be used to provide PoE power to output port E1, and when PoE is not available, the DC input can be used to provide PoE power to port E1, satisfying the claimed logic (Compl. ¶60).	¶60	col. 8:26-33

• Identified Points of Contention:
  • Scope Question: The claim recites specific logic for when only one of two power sources is present. The complaint's evidence (HPE's documentation) describes the device's behavior when both sources are present ("the DC power source takes precedence") or when one is not available (Compl. ¶59; Compl. ¶60). A point of dispute may be whether the documented behavior inherently includes or implies the specific logic recited in the claim for single-source scenarios.
  • Technical Question: The complaint infers the existence of a "first sensor," "second sensor," and "controller" from the product's external behavior described in user manuals. The defense may argue that the actual hardware implementation (e.g., a single, highly integrated Power Management Integrated Circuit or PMIC) does not have the distinct, separable components as claimed, raising a question of structural and functional correspondence.

V. Key Claim Terms for Construction

For U.S. Patent No. 8,565,230

• The Term: "shared virtual tunnel"
• Context and Importance: This term is the central inventive concept of the '230 patent. The entire infringement case for this patent hinges on whether HPE's "logical VTEP" functionality, which uses a shared "anycast" IP address for two physical switches, constitutes a "shared virtual tunnel" as the claim language is construed by the court.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification describes the invention in functional terms, stating its purpose is to "virtualize[] the address of transport devices" to solve the MAC churn problem (Compl. ¶31, citing '230 Patent, col. 9:62-10:52). This could support a construction that covers any technical implementation achieving this virtualized, shared identity, including HPE's software-defined approach.
  • Evidence for a Narrower Interpretation: Claim 1 requires that the tunnel "connects" the first, second, and third switching devices. The specification also discusses specific protocols like PBB (IEEE 802.1ah) as a way to implement the transport network '230 Patent, col. 6:15-18 A defendant may argue that a "shared virtual tunnel" must be a specific, discrete logical construct created in a particular way, rather than an emergent property of an anycast routing configuration.

For U.S. Patent No. 8,046,619

• The Term: "electrically connecting"
• Context and Importance: Claim 1 requires the controller to be "electrically connecting" an input to an output. The validity of the infringement allegation depends on whether the power path in the accused AP-505H meets this definition. Practitioners may focus on this term because modern power management systems often re-regulate power rather than simply switching a direct connection.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: A plaintiff may argue the term simply means to establish a path for power to flow from the selected input to the designated output, regardless of any intervening power conditioning or regulation circuits.
  • Evidence for a Narrower Interpretation: A defendant may argue that "electrically connecting" implies a direct, low-impedance path, akin to closing a physical switch or relay. The patent's block diagram (Fig. 3) shows a "PSE Controller and Power Controller" (370) between the power selection logic and the powered output port, which may suggest that the connection is not a direct electrical one, but a controlled, re-sourced power delivery, potentially supporting a narrower construction.

VI. Other Allegations

• Indirect Infringement: The complaint alleges induced infringement for all four patents. The allegations are based on HPE providing product documentation, installation guides, user manuals, and technical support that allegedly instruct and encourage customers and end-users to configure and operate the accused products in an infringing manner Compl. ¶¶43-44 Compl. ¶¶63-64 Compl. ¶¶87-88 Compl. ¶¶117-118
• Willful Infringement: The complaint alleges willful infringement based on both pre-suit and post-suit knowledge. Pre-suit knowledge is alleged based on Plaintiff's notice letter of May 6, 2026 Compl. ¶20 It is also alleged that HPE's predecessors or acquired companies (HPDC and Juniper) were aware of the applications leading to the '677 and '230 patents because they were cited during the prosecution of their own patents, years before the lawsuit Compl. ¶42 Compl. ¶86 The complaint also pleads willful blindness, alleging HPE deliberately avoided learning of its infringement despite a subjective belief in a high probability of the same Compl. ¶22

VII. Analyst's Conclusion: Key Questions for the Case

This case presents a multi-front dispute over core technologies in enterprise networking. The outcome will likely depend on the court's resolution of several key questions:

1. A question of software abstraction versus claimed structure: Can the '230 patent's term "shared virtual tunnel," which is described with specific structural connections, be construed to read on HPE's "logical VTEP" - a software-defined abstraction that relies on anycast IP addressing to achieve a similar functional outcome?

2. A question of hardware implementation: For the '619 and '677 power management patents, does the inferred operation of the accused devices, based on high-level documentation, provide sufficient evidence of the specific "sensors," "controllers," and "electrical connections" recited in the claims, or will an examination of the actual integrated circuitry reveal a different architecture that falls outside the claim scope?

3. A question of real-time performance: Will the '845 patent's requirements for "real-time" session movement based on monitoring with "test packets" be met by the accused SD-WAN products, which use automated "path-characterization" traffic and "traffic steering" based on pre-configured policy thresholds? This will likely turn on a detailed factual analysis of how quickly and in what manner the accused products react to network degradation.