Parity Networks LLC v. Beijer Electronics Group Ab

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Parity Networks LLC (Texas)
  • Defendant: Beijer Electronics Group AB (Sweden), Beijer Electronics, Inc. (Utah), Westermo Data Communications, Inc. (Delaware), and Korenix Technology Co., Ltd (Taiwan)
  • Plaintiff’s Counsel: Farnan LLP; DiNovo Price LLP
• Case Identification: 1:22-cv-01522, D. Del., 03/27/2023
• Venue Allegations: Venue is alleged to be proper in the District of Delaware because Defendant Westermo is a Delaware corporation, the Defendants are alleged to operate as a single collective entity, Defendant Korenix is a foreign entity, and acts of infringement allegedly occurred within the district.
• Core Dispute: Plaintiff alleges that Defendants’ industrial networking switches and routers infringe five patents related to packet classification, multicast traffic management, quality of service, and network congestion control.
• Technical Context: The technology at issue involves methods for managing data packet traffic within network routers and switches, which is a foundational element for ensuring the security, performance, and reliability of modern data networks.
• Key Procedural History: The complaint notes that certain claims of the ’046 Patent were previously subject to conflicting court rulings on indefiniteness; a court in the Central District of California found them indefinite in December 2020, while a court in the Western District of Texas found them not indefinite in January 2021. The complaint also alleges that Defendant Korenix received pre-suit notice of the patents-in-suit as early as October 2016, which may be used to support allegations of willful infringement.

Case Timeline

Date	Event
2001-03-06	Earliest Priority Date for '844 and '963 Patents
2001-08-22	Earliest Priority Date for '394, '046, and '352 Patents
2004-07-13	'394 Patent Issued
2005-03-22	'844 Patent Issued
2006-09-05	'046 Patent Issued
2006-09-12	'352 Patent Issued
2010-05-18	'963 Patent Issued
2016-10-05	Pre-suit notice letter allegedly sent to Defendant Korenix
2016-11-28	Second pre-suit notice letter allegedly sent to Defendant Korenix
2020-12-22	C.D. Cal. court rules certain '046 Patent claims indefinite
2021-01-13	W.D. Tex. court rules same '046 Patent claims not indefinite
2023-03-27	First Amended Complaint Filed

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

U.S. Patent No. 6,763,394 - Virtual Egress Packet Classification at Ingress

Issued July 13, 2004

The Invention Explained

• Problem Addressed: The patent’s background section describes that network routers traditionally perform packet filtering (pass/drop decisions) at both ingress (entry) and egress (exit) ports, which requires duplicative hardware and adds processing delay (’394 Patent, col. 2:36-48). Forgoing egress filtering to save costs means losing the ability to apply port-specific rules, which is a significant functional limitation (’394 Patent, col. 2:41-48).
• The Patented Solution: The invention proposes a system to perform the egress filtering check "virtually" at the ingress port. After an incoming packet's egress port is determined by a standard forwarding lookup, a specialized Access Control List (ACL) at the ingress port makes the pass/drop decision using rules that incorporate both the packet's header information and its assigned egress port identity (’394 Patent, Abstract; col. 5:10-21). This consolidates filtering logic at a single point, aiming to reduce hardware complexity and latency.
• Technical Importance: This approach sought to streamline router architecture by eliminating the need for separate, dedicated filtering mechanisms at every egress port, potentially lowering hardware costs and improving packet throughput (’394 Patent, col. 2:41-48).

Key Claims at a Glance

• The complaint asserts at least independent claim 19 (Compl. ¶43).
• The essential elements of independent claim 19 are:
  • A method at an ingress port for determining both ingress and egress rules for packets.
  • Noting header combinations, values, and the egress port destination for incoming packets.
  • Comparing, in a "first lookup table," the packet's headers and egress port destination with a rule set that includes egress port destinations to return a rule.
  • Comparing, in a "second lookup table," a separate rule set for ingress-only determinations "without egress port numbers."
  • Returning a rule for the packet as a result of the ingress-only comparison.

U.S. Patent No. 6,870,844 - Apparatus and Methods for Efficient Multicasting of Data Packets

Issued March 22, 2005

The Invention Explained

• Problem Addressed: The patent’s background explains that multicasting—sending a single data stream to multiple recipients—can create network bottlenecks when handled by general-purpose router processors or external servers (’844 Patent, col. 2:46-54). Additionally, conventional flow-control mechanisms used to prevent congestion can propagate "stop" signals upstream, unnecessarily halting unrelated traffic and potentially stalling large portions of the network (’844 Patent, col. 2:1-8).
• The Patented Solution: The invention describes a specialized hardware component, a "multicast-capable port" or "M-Port," integrated directly into the router's switching fabric cards (’844 Patent, Abstract; col. 6:50-54). This M-Port is designed to receive a packet designated for multicast, use a table to replicate the packet into multiple copies, assign the correct destination address to each copy, and then re-inject the new packets into the router's ingress path for normal forwarding (’844 Patent, Fig. 5; col. 7:48-67).
• Technical Importance: This architecture proposed a scalable, distributed, hardware-based solution for high-volume multicasting, moving the function from a centralized processor to the switching fabric itself to improve performance and avoid network-wide stalls (’844 Patent, col. 3:10-24).

Key Claims at a Glance

• The complaint asserts at least independent claim 24 (Compl. ¶51).
• The essential elements of independent claim 24 are:
  • A method for multicasting.
  • Providing a plurality of "multicast engines" within a router, where each engine is integrated as part of a line card's port.
  • Receiving multicast packets at a port and sending them to one of the multicast engines.
  • Replicating and/or modifying the packets according to tabled instructions associated with the engine.
  • Outputting the replicated/modified packets to the "ingress path" to individual ones of the router's ports.

U.S. Patent No. 7,103,046 - Method and Apparatus for Intelligent Sorting and Process Determination of Data Packets Destined to a Central Processing Unit of a Router or Server on a Data Packet Network

Issued September 5, 2006

• Technology Synopsis: The patent addresses the problem of a router's central processing unit (CPU) being overwhelmed by malicious or low-priority packets, which can lead to a denial-of-service attack (’046 Patent, col. 2:3-15). The solution is a system that intercepts CPU-bound packets and sorts them into different priority queues (e.g., high, medium, low) before they reach the CPU, allowing the CPU to process trusted traffic first while throttling or ignoring suspect traffic to maintain stability (’046 Patent, Abstract).
• Asserted Claims: At least independent claim 6 (Compl. ¶60).
• Accused Features: The accused products are alleged to infringe by using packet processors to categorize packets based on source, place them in queues, and have a CPU process them according to the priority of those categories (Compl. ¶¶33, 61).

U.S. Patent No. 7,107,352 - Virtual Egress Packet Classification at Ingress

Issued September 12, 2006

• Technology Synopsis: As a continuation of the application leading to the ’394 Patent, this patent targets the same problem of router hardware complexity and latency caused by performing packet filtering at both ingress and egress ports (’352 Patent, col. 2:36-48). The invention describes a method where the filtering decision is consolidated at the ingress port by using an Access Control List (ACL) that factors in both the packet's header data and its pre-determined egress port destination, thereby obviating a separate filtering step at the egress port (’352 Patent, Abstract).
• Asserted Claims: At least independent claim 25 (Compl. ¶65).
• Accused Features: The accused products are alleged to infringe by using ACLs at an ingress port to filter and drop packets based on rules that include egress determinations (Compl. ¶¶29, 66).

U.S. Patent No. 7,719,963 - System for Fabric Packet Control

Issued May 18, 2010

• Technology Synopsis: The patent addresses network congestion caused by traditional flow control methods that can unnecessarily halt traffic across an entire network fabric (’963 Patent, col. 2:5-15). The solution is a queue management system that avoids propagating stop signals; instead, when a queue begins to fill past a certain threshold, it starts proactively dropping packets (e.g., using Weighted Random Early Detection, or WRED) to signal congestion without halting traffic on other, non-congested paths (’963 Patent, Abstract).
• Asserted Claims: At least independent claim 1 (Compl. ¶73).
• Accused Features: The accused products are alleged to infringe by implementing WRED algorithms that manage congestion by dropping packets from queues based on configurable size thresholds (Compl. ¶¶35-36, 74).

III. The Accused Instrumentality

Product Identification

A broad array of industrial-grade networking switches and routers sold under Defendants' Korenix JetNet and Westermo brands, including the JetNet, MRD, BRD, Lynx, RedFox, and Viper product series (Compl. ¶¶44, 52, 61, 66, 74).

Functionality and Market Context

The accused products are described as hardware and software for switching and routing data in industrial network environments (Compl. ¶¶11, 27). The complaint alleges, through citation to product manuals, that these devices provide functionalities central to the asserted patents. This includes configurable Access Control Lists (ACLs) for packet filtering, IGMP Snooping for managing multicast traffic, Quality of Service (QoS) for traffic prioritization, and Weighted Random Early Detection (WRED) for managing queue congestion (Compl. ¶¶29-36). The screenshot of an IP Filter configuration screen shows settings for Source IP, Destination IP, Protocol, and an "Egress Port" (Compl. ¶8). Another screenshot describes how IGMP Snooping allows ports to "detect IGMP queries, report packets, and manage multicast traffic through the switch" (Compl. ¶32).

IV. Analysis of Infringement Allegations

'394 Patent Infringement Allegations

Claim Element (from Independent Claim 19)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
(a) noting header combinations and values, and egress port destination for incoming packets;	The accused products' ACL feature is allegedly configured using packet header information (e.g., source/destination IP, protocol) and allows for the specification of an egress port as a filtering criterion.	¶¶8, 29-30	col. 8:5-7
(b) comparing, in a first lookup table, the header combinations and values and egress port destinations with a rule set associated with the header combinations and values including egress port destinations;	The accused devices allegedly check Access Control Entries (ACEs) which contain rules that include egress port information, and then permit or deny packets based on a match with that rule set.	¶¶8, 30	col. 8:8-14
(c) returning a rule for the packet;	The "Action" field in the accused devices' ACL configuration (e.g., Permit, Deny) is alleged to be the "rule" that is returned and applied to the packet.	¶¶8-9	col. 8:15-16
(d) comparing, in a second lookup table, a separate rule set association for accomplishing ingress rule application determinations without egress port numbers;	The complaint alleges that the accused devices' ACLs can also be configured to filter packets based on header information "with or without an egress port identity," which is alleged to constitute a separate comparison.	¶30	col. 8:17-21
(e) returning a rule for the packet as a result of step (d).	The ACL's "Permit" or "Deny" action is allegedly returned based on rules that do not specify an egress port, thereby satisfying this limitation.	¶¶9, 30	col. 8:22-23

Identified Points of Contention

• Scope Questions: Does the accused ACL feature, which appears to be a single configurable system, meet the claim limitation of a "first lookup table" and a "second lookup table"? A court may need to determine whether a single, flexible software module can embody two structurally distinct "tables" as recited in the claim.
• Technical Questions: The patent’s "virtual egress" concept implies that the egress port is first determined by a forwarding lookup, and that destination is then used in the ACL comparison. The complaint provides a screenshot where "Egress Port" is a filter criterion in the ACL rule itself (Compl. ¶8). This raises the question of whether the accused products perform the steps in the sequence required by the patent’s inventive concept.

'844 Patent Infringement Allegations

Claim Element (from Independent Claim 24)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
(a) providing a plurality of multicast engines within a router, each having one or more first ports for communicating with second ports of the router... wherein the multicast engine is integrated as a part of a port of a line card in the router.	The complaint alleges the accused switches contain "multicast-capable component[s]" that use IGMP Snooping technology, which are alleged to be the claimed "multicast engines" integrated within the devices.	¶¶31-32	col. 9:57-60
(b) receiving multicast packets at one of the second ports and sending the multicast packets to one of the multicast engines via the first ports;	The IGMP Snooping feature allegedly allows ports to "detect IGMP queries, report packets, and manage multicast traffic through the switch," which is alleged to be the receiving and sending of packets to the engine.	¶32	col. 9:61-65
(c) replicating and/or modifying the data packets for multicasting according to tabled instructions associated with the multicast engine;	The accused devices' IGMP Snooping functionality allegedly "track[s] subscribed multicast groups," which implies a table of instructions used to manage the forwarding of multicast packets.	¶32	col. 10:1-5
(d) outputting the replicated or modified packets to the ingress path to individual ones of the second ports;	By managing multicast traffic, the IGMP Snooping feature is alleged to output packets to the appropriate ports. The complaint does not provide specific evidence of the "outputting... to the ingress path" architecture.	¶32	col. 10:6-10

Identified Points of Contention

• Scope Questions: Does the industry-standard feature of "IGMP Snooping," which typically filters or prunes an existing multicast data stream, constitute a "multicast engine" that performs the active replication described in the patent? The defense may argue that snooping is a passive filtering mechanism, whereas the patent claims an active packet-creation engine.
• Technical Questions: The claim requires a specific internal data flow: outputting replicated packets "to the ingress path" of a port for re-processing. This is a core architectural element of the patent's "M-Port" design. What evidence does the complaint, which relies on high-level user manuals, provide to show that the accused products share this specific and unconventional hardware architecture?

V. Key Claim Terms for Construction

'394 Patent

• The Term: "first lookup table" / "second lookup table" (Claim 19)
• Context and Importance: The validity and infringement of claim 19 depend on whether the accused ACL system, which can be configured with or without an egress port, constitutes two distinct lookup tables. Practitioners may focus on this term because if it is construed to require two physically or structurally separate components, the infringement allegation against a single, configurable software feature may be weakened.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification suggests the function could be implemented in different ways, stating that "the ingress and egress pass/drop lookups" could be "combined in some embodiments" (’394 Patent, col. 5:27-29). This may support an argument that the "tables" are logical constructs, not necessarily separate physical ones.
  • Evidence for a Narrower Interpretation: The claim explicitly recites a "first" and a "second" table, and the specification describes the preferred embodiment as performing the lookups "separately" (’394 Patent, col. 5:32-35). This language supports a construction requiring two distinct and separate processes or structures, one for egress-aware rules and one for ingress-only rules.

'844 Patent

• The Term: "multicast engine" (Claim 24)
• Context and Importance: The infringement case for this patent hinges on whether the accused products' standard "IGMP Snooping" functionality can be considered a "multicast engine." Practitioners may focus on this term because IGMP snooping is a common industry feature, and a narrow construction tied to the patent's specific architecture could place most standard implementations outside the claim's scope.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The term is not explicitly defined with a precise boundary. Plaintiff may argue that any dedicated component that "manages multicast traffic" (Compl. ¶32) by applying rules from a table falls within a plain and ordinary meaning of "engine."
  • Evidence for a Narrower Interpretation: The specification consistently describes the invention as a specific "M-Port" architecture that actively replicates packets, assigns new addresses, and re-injects them into the switch fabric (’844 Patent, col. 7:14-20, 8:1-12). The detailed description and figures show a dedicated hardware block for creating new packet copies, suggesting the term "multicast engine" is not a generic component but refers to this specific packet-replicating apparatus (’844 Patent, Fig. 5, element 339).

VI. Other Allegations

Indirect Infringement

The complaint alleges both induced and contributory infringement for all asserted patents. The inducement allegations are based on Defendants providing user manuals, datasheets, and other instructions that allegedly guide and encourage customers to configure and use the accused functionalities (e.g., ACLs, IGMP Snooping, WRED) in an infringing manner (Compl. ¶¶45, 53, 67).

Willful Infringement

While not pleaded as a separate count, the complaint lays the groundwork for a willfulness claim by alleging that Defendant Korenix had actual knowledge of the patents-in-suit from notice letters sent as early as October and November 2016, long before the suit was filed (Compl. ¶40). Willfulness is also alleged based on knowledge gained from the service of the original complaint in this action.

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

• A core issue will be one of architectural equivalence: do the accused products' functionalities, which appear to be common industry-standard features like IGMP Snooping and configurable ACLs, operate in a manner that maps to the specific, and in some cases unconventional, hardware architectures and data flows required by the patent claims? For instance, does standard IGMP snooping perform the active packet replication and ingress-path re-injection of the ’844 patent’s claimed “multicast engine”?
• A second central question will be one of definitional scope: can claim terms such as “first lookup table”/“second lookup table” (’394 Patent) and "multicast engine" (’844 Patent) be construed broadly enough to encompass widely used, off-the-shelf software features, or will the detailed descriptions in the patent specifications limit the claims to the specific, novel hardware implementations disclosed?
• Finally, for the ’046 Patent, a threshold validity question will be the impact of prior conflicting judgments: how will the court treat a patent claim that one district court has found indefinite while another has found not indefinite? This presents a significant legal hurdle that must be addressed before the merits of the infringement allegation can be fully considered.