IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MARYLAND
: AUTHPOINT LLC :
v. : Civil Action No. DKC 25-683
: GL COMMUNICATIONS INC. :
MEMORANDUM OPINION Presently pending and ready for resolution in this patent- infringement case is the motion to dismiss for failure to state a claim filed by GL Communications, Inc. (“GL” or “Defendant”). (ECF No. 25). The issues have been briefed, and the court now rules, no hearing being deemed necessary. Local Rule 105.6. For the following reasons, the motion to dismiss will be granted. I. Background A. Factual Background Plaintiff AuthPoint LLC (“AuthPoint” or “Plaintiff”) is the assignee of United States Patent No. 8,699,395 (the “’395 patent”), which is titled “Method and device for inverse multiplexing of multicast transmission.” (ECF No. 23 ¶¶ 8–9). GL makes and sells at least four products that allegedly infringe the method described in Claim 1 of the ’395 patent.1
1 AuthPoint does not describe GL’s general business model. GL describes itself as a telecommunication company, (ECF No. 25- 1, at 9), and its website indicates that it offers “a comprehensive suite of Voice testing and Network testing solutions for all 1. Preliminary Terminology Before diving into the prior art and patented method, it is helpful to explain some of the terms used in the claim, including inverse multiplexing, inverse demultiplexing, multicast
transmission, and subscriber. What these terms mean, as a general matter, is not in dispute.2 GL describes inverse multiplexing as the process of “breaking data into smaller units[] . . . [and] sending those smaller units to a destination along different paths.” (ECF No. 25-1, at 8). Inverse demultiplexing, GL explains, is the process of “reassembling the data for the end user.” (Id.). GL further contends that these two processes together help “send large amounts of data over a network.” (Id.). By spreading out pieces of the large message across different paths, inverse multiplexing offers “bandwidth benefits.” (See ECF No. 23 ¶ 11).
telecom networks,” Company Overview, GL Comm’cns Inc., https://www.gl.com/about.html [https://perma.cc/C6ZV-T6DD]. The information on GL’s website is a proper subject of judicial notice. Rodgers v. Eagle All., 586 F.Supp.3d 398, 409 n.1 (D.Md. 2022) (taking judicial notice of information on a party’s website regarding the party’s business, which had not been discussed in the papers).
2 Plaintiff contends that Defendant engages in claim construction at various points. The definitions provided in this subsection, however, come from Plaintiff’s amended complaint, the ’395 patent, or uncontested explanations in Defendant’s motion to dismiss. 2 Messages transmitted across a network, such as the internet, can be unicast, broadcast, or multicast. U.S. Patent No. 8,699,395 col. 1 l. 15–17 (filed Sep. 9, 2005).3 “A unicast message is
addressed at a single terminal of the network, a broadcast message is addressed at all terminals of the network[,] and a multicast message is addressed at a group of a plurality of the terminals of the network.” Id. col. 1 l. 18–22. A subscriber is a “terminal [that] has subscribed to the multicast stream.” Id. col. 2 l. 2– 3. Putting these concepts together, an inversely multiplexed multicast transmission to multiple subscribers is a message to various terminals that is broken down into smaller units along different pathways for easier transmission. 2. Prior Art Prior to the invention of the ’395 patent, the method of inversely multiplexed multicast transmission proceeded as follows:
(1) An “upstream” multicast router would send a multicast message, (2) the multicast message would be inversely multiplexed, (3) the multicast message would be inversely demultiplexed, and (4) a second, “downstream” multicast router would send copies of the
3 The patent is also filed at ECF No. 23-1. For precision in citation to particular text within the patent, the court will cite to lines of the patent itself rather than to pages of the ECF filing. 3 inversely demultiplexed message to the various subscribers. (See ECF No. 23 ¶¶ 11–12); ’395 Patent col. 2 l. 3–11. That conventional method was “unable to efficiently deliver the same multicast stream
to multiple subscribers without first reassembling the inversely multiplexed stream at a central location and then re-distributing it.” (ECF No. 23 ¶ 12). In other words, the second multicast router was a “bottleneck” that “negated the bandwidth benefits of inverse multiplexing.” (Id. ¶¶ 10–11); ’395 Patent col. 2 l. 14– 15. Consequently, the conventional method “required additional network infrastructure, introduced latency, and created single points of failure in the message delivery path—problems that consumed network resources and degraded the quality of service for bandwidth-intensive applications such as video and audio streaming.” (ECF No. 23 ¶ 12). 3. The ’395 Patent The ’395 patent, filed in 2005 and issued in 2014, devised a
new method that eliminated the second multicast router. This patented method retains the first two steps of the conventional method but then decentralizes the inverse demultiplexing and distribution of copies. The patent presents two configurations of the method. A representation of the first configuration, as depicted in Figure 1 of the patent, is shown below: 4 % PY
14 14 ee ee 14 18 16. 18 16 18 16
□□
‘395 Patent fig. 1. The patent offers a description of this configuration with bolded numbers that correspond to the above depiction: Fig. 1 shows a part of a network, comprising a multicast router 10, an inverse multiplexing device 12, a plurality of functionally parallel connections 14, a plurality of inverse demultiplexing/ forwarding devices 16, a local network 17 and a plurality of [subscribers] 18. Multicast router 10 has a network connection 11 to a further part of the network (not shown) and a connection to inverse multiplexing device 12. Inverse multiplexing device 12 is connected to the inverse demultiplexing/forwarding devices 16, each via a respective one of the functionally parallel connections 14. Inverse demultiplexing/forwarding devices 16 are
coupled to each other via local network 17. Each inverse demultiplexing/forwarding device 16 is coupled to a respective one of the [subscribers] 18.
Id. col. 3 l. 36–48. The patent goes on to describe how this configuration of the method works: A stream of messages for [subscribers] 18 is received at network connection 11 and sent [via the multicast router 10] to inverse multiplexing device 12, which distributes the stream over connections 14 by means of inverse multiplexing, e.g. according to a round robin distribution scheme. Each inverse demultiplexing/forwarding device 16 receives those messages of the stream that have been sent over the connection 14 to which the inverse demultiplexing/forwarding device 16 is connected. The inverse demultiplexing/ forwarding device 16 forwards these messages to other inverse demultiplexing/forwarding devices 16 if the messages are needed by [subscribers] 18 other than the [subscriber] that is connected to the inverse demultiplexing/forwarding device 16. The connected inverse demultiplexing/forwarding device 16 of that [subscriber] 18 reassembles a stream from the messages that it receives from its connection 14 and from the other inverse demultiplexing/forwarding devices 16, by means of inverse demultiplexing.
Id. col. 3 l. 59–67, col. 4 l. 1–8. In other words, after the multicast message is inversely multiplexed, each subscriber’s inverse demultiplexer/forwarding device receives a portion of the message, retains that portion, and forwards a copy of that portion to the other subscribers in the network. At the conclusion of this process, each subscriber’s inverse demultiplexer will contain 6 all portions of the message. Each subscriber can then inversely demultiplex all portions of the message and thereby reassemble the original message. The second configuration, as depicted in Figure 2 of the patent, contemplates the inverse demultiplexer and the forwarding device as separate devices rather than a combined device, and is shown below: ~~
14 14 14 22, 22 22
fr tC‘ HS ( 20 20
18 18 18
Id. fig. 2. As with Figure 1, the patent offers a (condensed) description of this configuration and how it works with bolded numbers that correspond to the above depiction:
In this [configuration] the functions of forwarding and inverse demultiplexing have been separated. Forwarding units 22 and inverse demultiplexing devices 20 are provided[.] Forwarding units 22 are coupled between respective connections 14 and local network 17. Inverse [de]multiplexing devices 20 are coupled between [subscribers] 18 and local network 17. In this [configuration] forwarding units 22 use the destination addresses of messages to determine to which of the demultiplexing devices 20 messages will be forwarded via local network 17.
Id. col. 5 l. 8–17. Put another way, after the multicast message is inversely multiplexed, a forwarding device on each parallel connection forwards copies of that connection’s portion of the message to the inverse demultiplexer coupled with each subscriber. At the conclusion of this process, each subscriber’s inverse demultiplexer will contain all portions of the message. Each subscriber can then inversely demultiplex all portions of the message and thereby reassemble the original message. This decentralized architecture captured in Figures 1 and 2 is described in Claim 1: A method of forwarding a stream of multicast messages through a network by transmitting a stream of multicast messages from a multicast router to a multicast subscriber device and a 8 further multicast subscriber device, the method comprising:
inverse multiplexing the stream of multicast messages, thereby converting the stream of multicast messages into multiple parts, each part being transmitted via one of a plurality of communication channels[;]
inverse demultiplexing the multiple parts of the multicast messages of the inversely multiplexed stream with an inverse demultiplexer for the multicast subscriber device; and
forwarding, by a plurality of forwarding devices coupled to respective ones of the plurality of communication channels, respective ones of the multiple parts of the inversely multiplexed stream from the plurality of communication channels to a further inverse demultiplexer of the further multicast subscriber device.
Id. col. 9 l. 53–67, col. 10 l. 1–4. AuthPoint describes this method as a “concrete improvement over prior art.” (ECF No. 23 ¶ 18). Its “distributed forwarding of inversely multiplexed parts for decentralized reassembly” avoids the bottleneck issue while retaining the bandwidth benefits of inverse multiplexing. (Id.). 4. GL’s Products AuthPoint alleges that GL has infringed the ’395 patent by making, selling, and internally testing and using the following four products: (1) Inverse Multiplexing for ATM (IMA) Emulator (“IMA Emulator”); (2) MAPS ED-137 Radio Emulator (“Radio 9 Emulator”); (3) STM-1 Multiplexer; and (4) ATM Analyzer.4 (See generally ECF No. 23-2). AuthPoint uses color-coded boxes to identify which portion of the product allegedly performs the
patented method in Claim 1, but the color coding tends to confuse more than clarify. Accordingly, the court will primarily refer to the textual allegations in the claim chart. The IMA Emulator is the clear focus of the claim chart and the only product AuthPoint attempts to tie to every element of the claim. The claim chart provides no textual allegations regarding the Radio Emulator; its relation to the other products, and frankly the claim itself, is too unclear to proceed with any discussion of this product. As for the STM-1 Multiplexer and ATM Analyzer, AuthPoint discusses them sporadically in the claim chart and they will be noted where relevant accordingly. Below is GL’s depiction of the IMA Emulator:
4 These product names include a variety of acronyms. AuthPoint does not explain the ATM acronym, but GL states that ATM stands for “Asynchronous Transfer Mode.” (ECF No. 25-1, at 9). The record contains no clear explanation of ATM technology, but it presumably constitutes one way to transmit data over a public network. Documents attached to Defendant’s motion to dismiss suggest that MAPS stands for “Message Automation & Protocol Simulation” and enables air-to-ground and ground-to-ground calls in the air traffic context. (ECF No. 25-7, at 3–4). It is not evident what STM means. 10 Inverse Multiplexing for ATM (IMA) Emulator Using Client-Server WAN Access Switch me WAN Aces Suriich Cee ae ee i / Es ATM ATM CS Access ACCESS cd aaa Voice event tes Voice a, ae Data Data o-oo rs □□ Gi’s ATM IMA Simulator Pes
(ECF Nos. 23-2, at 2; 25-2, at 2). AuthPoint identifies the blue disks as multicast routers. (ECF No. 23-2, at 2). Once a multicast message leaves one of the routers, AuthPoint believes that the IMA Emulator, presumably via the ATM (IMA) device connected to that router, inversely multiplexes the message and sends it over the plurality of physical communication channels (identified in the diagram by three lines circled by a dotted line labeled IMA). (Id. at 4-5). The ATM Analyzer is relevant to this step. It “analyzes data that has been inverselly] multiplexed by companion GL products including the IMA Emulator.” (Id. at 5). GL describes the ATM Analyzer as able to “capture and reassemble frames that were transmitted with Inverse Multiplexing,” which AuthPoint interprets to mean that “the GL Communications product ecosystem is designed to work with inverse[ly] multiplexed multicast streams.” (Id.).
AuthPoint alleges upon information and belief that GL performs this step when testing the IMA Emulator and ATM Analyzer. (Id.). AuthPoint next alleges upon information and belief that the
ATM (IMA) device “at the far-end,” or “the receiving side,” of the physical links “functions as the inverse demultiplexer for the multicast subscriber devices connected downstream, including the PCs, voice equipment, and data devices shown in the diagram.”5 (Id. at 7–8). The STM-1 Multiplexer is relevant to this step because it also performs inverse demultiplexing. (Id. at 8). AuthPoint alleges upon information and belief that GL “sells its IMA Emulator and STM-1 Multiplexer products as matched pairs for use at both ends of a communication link, with the receiving-end device performing inverse demultiplexing.” (Id.). Again, AuthPoint believes that GL performs this step when testing the IMA Emulator and ATM Analyzer. (Id.).
As for the final step, AuthPoint alleges upon information and belief that the ATM Access Switches in the IMA Emulator diagram “function as a plurality of forwarding devices that are coupled to respective ones of the plurality of [the physical] communication channels, forwarding the inverse[ly] multiplexed ATM cells toward
5 As just one of many examples of the confusion AuthPoint’s color coding creates, it codes the data device as a “multicast router” in one row of the claim chart and then in later rows describes the same device as a “multicast subscriber device.” (Compare ECF No. 23-2, at 2, with id. at 8, 10). 12 the further inverse demultiplexer at the far end of the network.” (Id. at 9–10). AuthPoint identifies the further inverse demultiplexer as “the receiving-end ATM (IMA) device.” (Id. at
10). It also alleges upon information and belief that the diagram “depict[s] symmetrical configurations with inverse demultiplexers (ATM (IMA) devices) at both ends of the network, with the device at the second site functioning as the ‘further inverse demultiplexer’ for the ‘further multicast subscriber device’ comprising the PCs and other endpoint equipment at that site.” (Id.). Finally, AuthPoint alleges upon information and belief that the IMA Emulator can be configured with multiple IMA Emulator devices (presumably the ATM (IMA) device), each functioning as an inverse demultiplexer and thus at least one functioning as a further inverse demultiplexer. (Id.). AuthPoint believes that GL performs this step and employs such configurations when testing the IMA Emulator and ATM Analyzer. (Id.). B. Procedural History
On March 2, 2025, AuthPoint filed a complaint against GL asserting a single claim of direct infringement of the ’395 patent. (ECF No. 1). After various extensions, GL moved to dismiss the complaint for failure to state a claim on December 5, 2025. (ECF No. 14). Rather than respond to the motion to dismiss, AuthPoint filed an amended complaint on January 9, 2026. (ECF No. 23). The 13 amended complaint, which contains additional factual allegations and a more developed claim chart, continues to assert a single claim of direct infringement of the ’395 patent. On January 23,
GL moved to dismiss the amended complaint. (ECF No. 25). AuthPoint responded on February 20, (ECF No. 31), and GL replied on March 6, (ECF No. 34). II. Standard of Review A motion to dismiss under Fed.R.Civ.P. 12(b)(6) tests the sufficiency of the complaint. Presley v. City of Charlottesville, 464 F.3d 480, 483 (4th Cir. 2006). The court “must accept the complaint’s factual allegations as true and construe the facts in the light most favorable to the plaintiff.” Barnett v. Inova Health Care Servs., 125 F.4th 465, 469 (4th Cir. 2025) (citing Barbour v. Garland, 105 F.4th 579, 589 (4th Cir. 2024)). Ordinarily, a plaintiff’s complaint must only satisfy the standard of Rule 8(a)(2), which requires a “short and plain statement of
the claim showing that the pleader is entitled to relief.” “[W]here the well-pleaded facts do not permit the court to infer more than the mere possibility of misconduct, the complaint has alleged—but it has not ‘show[n]’—that the pleader is entitled to relief.’” Ashcroft v. Iqbal, 556 U.S. 662, 679 (2009) (quoting Fed.R.Civ.P. 8(a)(2)). A Rule 8(a)(2) “showing” requires “stat[ing] a claim to relief that is plausible on its face.” Bell 14 Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007). “A claim has facial plausibility when the plaintiff pleads factual content that allows the court to draw the reasonable inference that defendant is liable for the misconduct alleged.” Mays v. Sprinkle, 992 F.3d 295, 299–300 (4th Cir. 2021) (quoting Iqbal, 556 U.S. at 678).
Nearly all of Plaintiff’s allegations regarding Defendant’s products are made upon information and belief. “Under the pleading standard the Supreme Court [of the United States] articulated in Twombly and Iqbal, a complaint’s conclusory allegations based solely ‘upon information and belief’ are ‘insufficient to defeat a motion to dismiss.’” Van Buren v. Walmart, Inc., 611 F.Supp.3d 30, 36 (D.Md. 2020) (quoting Harman v. Unisys Corp., 356 F.App’x 638, 640–41 (4th Cir. 2009)), aff’d, 855 F.App’x 156 (4th Cir. 2021). That said, “pleading on the basis of information and belief is generally appropriate where information is particularly within defendants’ knowledge and control.” Id. at 37 (quoting Kajoshaj v. N.Y.C. Dep’t of Educ., 543 F.App’x 11, 16 (2d Cir. 2013)). Even
then, however, the plaintiff must have “sufficient data to justify interposing an allegation on the subject.” 5 Wright & Miller’s Federal Practice & Procedure § 1224 (4th ed. 2026). The United States Court of Appeals for the Federal Circuit has provided specific guidance regarding the assessment of patent- infringement claims on a motion to dismiss: 15 A plaintiff is not required to plead infringement on an element-by-element basis. [Nalco Co. v. Chem-Mod, LLC, 883 F.3d 1337, 1350 (Fed. Cir. 2018)] (“[T]he Federal Rules of Civil Procedure do not require a plaintiff to plead facts establishing that each element of an asserted claim is met.”); see Disc Disease Sols. Inc. v. VGH Sols., Inc., 888 F.3d 1256, 1260 (Fed. Cir. 2018) (finding that a plaintiff need not plead every element, but must only give the alleged infringer fair notice of infringement). Instead, it is enough “that a complaint place the alleged infringer ‘on notice of what activity . . . is being accused of infringement.’” Lifetime Indus., Inc. v. Trim-Lok, Inc., 869 F.3d 1372, 1379 (Fed. Cir. 2017) (quoting K-Tech Telecomms., Inc. v. Time Warner Cable, Inc., 714 F.3d 1277, 1284 (Fed. Cir. 2013)).
Bot M8 LLC v. Sony Corp. of Am., 4 F.4th 1342, 1352 (Fed. Cir. 2021) (second alteration in original). “The level of detail required in any given case will vary depending upon a number of factors, including the complexity of the technology, the materiality of any given element to practicing the asserted claim(s), and the nature of the allegedly infringing device.” Id. at 1353. And although claim construction is not generally appropriate at the motion-to-dismiss stage, Nalco, 883 F.3d at 1350, a court “may dismiss a complaint prior to claim construction when the complaint rests on an implausible claim construction,” ALD Soc., LLC v. Verkada, Inc., 654 F.Supp.3d 972, 979 (N.D.Cal. 2023) (citing Ottah v. Fiat Chrysler, 884 F.3d 1135, 1141–42 (Fed. Cir. 2018); Nalco, 883 F.3d at 1349). Thus, “while a patentee’s 16 pleading obligations are not insurmountable, a patentee may subject its claims to early dismissal by pleading facts that are inconsistent with the requirements of its claims.” Bot M8, 4 F.4th at 1346 (citing Nalco, 883 F.3d at 1348–50). III. Analysis
35 U.S.C. § 271(a) imposes liability for direct infringement on anyone who “without authority makes, uses, offers to sell, or sells any patented invention.” A patented invention, in turn, may be a particular method. 35 U.S.C. §§ 100(b), 101. But there are several limitations on asserting a claim for direct infringement of a method. “There is no established recognition in patent law of direct infringement by ‘making’ a ‘method,’” Brumfield v. IBG LLC, 97 F.4th 854, 879 (Fed. Cir. 2024), nor does the “sale of an apparatus that is capable of infringing use” suffice for a method claim, Ormco Corp. v. Align Tech., Inc., 463 F.3d 1299, 1311 (Fed. Cir. 2006) (citing Joy Techs., Inc. v. Flakt, Inc., 6 F.3d 770,
773 (Fed. Cir. 1993); Standard Havens Prods., Inc. v. Gencor Indus., Inc., 953 F.2d 1360, 1374 (Fed. Cir. 1991)). “[I]ndeed, [the Federal Circuit] ha[s] indicated that direct infringement is limited to using the method.” Brumfield, 97 F.4th at 879 (emphasis added). To state a claim for direct infringement of a method under § 271(a), then, a plaintiff must plausibly allege that “all steps of a claimed method are performed by or attributable to” the 17 accused infringer. Akamai Techs., Inc. v. Limelight Networks, Inc., 797 F.3d 1020, 1022 (Fed. Cir. 2015) (en banc) (per curiam) (citing BMC Res., Inc. v. Paymentech, L.P., 498 F.3d 1373, 1379–
81 (Fed. Cir. 2007)). Although Plaintiff asserts that Defendant has directly infringed Claim 1 of the ’395 patent by making, selling, and using the IMA Emulator, Radio Emulator, STM-1 Multiplexer, and ATM Analyzer, only use is potentially cognizable. As for use, Plaintiff alleges that Defendant infringes the ’395 patent when it tests its products internally. Plaintiff must therefore plausibly allege that Defendant performs all steps of the patented method in Claim 1 when it tests its products internally. The amended complaint falls short of doing so. As a threshold matter, the IMA Emulator is the only product Plaintiff attempts to read onto each step of Claim 1. It is
unclear how the other three products partake in infringement of the method in Claim 1. Most neglected are the Radio Emulator and STM-1 Multiplexer. After Defendant noted in its motion to dismiss Plaintiff’s failure to explain the relevance of these products, (ECF No. 25-1, at 18–20), Plaintiff declined to discuss them at all in its opposition. “In failing to respond to th[ese] argument[s,] Plaintiff concedes the point.” Stenlund v. Marriott Int’l, 172 F.Supp.3d 874, 887 (D.Md. 2016) (citing Ferdinand- 18 Davenport v. Children’s Guild, 742 F.Supp.2d 772, 777 (D.Md. 2010); Kissi v. Panzer, 664 F.Supp.2d 120, 123 (D.D.C. 2009)). Defendant made a similar argument regarding the ATM Analyzer, to which
Plaintiff offered only an oblique response in its opposition, explaining that the ATM Analyzer can “decode ATM frames constituting Classical IP over ATM.” (ECF No. 31, at 5, 11). Assuming this means that the ATM Analyzer can inversely demultiplex, Plaintiff does not explain how the ATM Analyzer interacts with the other products, most importantly the IMA Emulator, to infringe on Claim 1. Thus, the only method to consider is the internal testing of the IMA Emulator. Defendant attacks Plaintiff’s claim regarding the IMA Emulator on myriad grounds. It argues that Plaintiff fails to allege that the IMA Emulator has a multicast router, a further multicast subscriber device, a plurality of forwarding devices, or
a further inverse demultiplexer. (ECF No. 25-1, at 14). It also invokes the seven layers of computer networking to posit that Plaintiff’s theory of infringement is legally impossible. (Id. at 21–26). Among these seven layers, the first layer is the physical layer, the second layer is the data layer, and the third layer is the network layer. (Id. at 23). Because ATM technology occupies only the lower physical and data layers, and multicasting occurs on the higher network layer, Defendant contends that the IMA 19 Emulator cannot possibly infringe on a multicasting method patent. (Id. at 23–26). Finally, Defendant argues that Claim 1 of the ’395 patent is not eligible for protection. (Id. at 26–31).
Plaintiff contests each of these positions. Although it is true that Plaintiff need not plead its method claim “element-by-element,” the use of a further inverse demultiplexer is particularly material to Claim 1. Bot M8, 4 F.4th at 1352–53. The principal problem with the prior art was the bottleneck created by requiring centralized inverse demultiplexing and forwarding. The ’395 patent resolved that problem by decentralizing inverse demultiplexing and forwarding. Therefore, the absence of plausible decentralization of inverse demultiplexing in Defendant’s products cuts at the heart of Claim 1 and results in a lack of fair notice to Defendant as to the infringing nature of its internal testing. Because Plaintiff fails
to allege that Defendant’s internal testing of the IMA Emulator includes use of a further inverse demultiplexer, the direct infringement claim fails, and Defendant’s other arguments need not be addressed. Plaintiff makes, in essence, three arguments regarding the existence of a further inverse demultiplexer in the IMA Emulator. None is plausible. First, Plaintiff argues that the ATM (IMA) device at the receiving end of the multicast message could contain 20 both the inverse demultiplexer and the further inverse demultiplexer, and whether the two must be at different network locations is a question of claim construction. (ECF No. 31, at
5). Defendant correctly observes that this construction of Claim 1 is plainly implausible. (ECF No. 34, at 4). If all inverse demultiplexing occurs at the same network location, that architecture creates the precise bottleneck that the ’395 patent eliminates. This argument fails at the threshold. Second, Plaintiff contends that the ATM (IMA) devices on both ends of the ATM network are capable of inverse demultiplexing, and one of them thus functions as the further inverse demultiplexer. As Plaintiff sees it, “the ATM (IMA) device at the first site [i]s the ‘inverse demultiplexer for the multicast subscriber device’ (the PCs and equipment at Site 1), and the ATM (IMA) device at the second site [i]s the ‘further inverse demultiplexer of the further
multicast subscriber device’ (the PCs and equipment at Site 2).” (ECF No. 31, at 8; see also ECF No. 23-2, at 10). Defendant asserts that “[t]his allegation is not plausible because . . . the [two] ATM (IMA) devices cannot share the same stream of multicast messages as the method requires. One multicast message would be inversely multiplexed by the first device and inversely demultiplexed by the second device.” (ECF No. 25-1, at 17–18). Defendant highlights a key deficiency in Plaintiff’s theory: If 21 both ATM (IMA) devices operate as inverse demultiplexers, then no device appears to function as an inverse multiplexer. In other words, the IMA Emulator diagram contradicts this theory of the
further inverse demultiplexer. Plaintiff is left with only conclusory information and belief to support the possibility that both devices serve as inverse demultiplexers in the transmission of the same multicast message. (ECF No. 23-2, at 10). Although that information is within Defendant’s control, Plaintiff lacks “sufficient data to justify interposing an allegation on the subject,” 5 Wright & Miller’s Federal Practice & Procedure § 1224 (4th ed. 2026), such as evidence that some of Defendant’s customers use such a configuration of the IMA Emulator or a statement from an employee of Defendant that it tests such a configuration. “[T]he Federal Circuit has made clear that a plaintiff’s inability to plead facts without discovery does not excuse compliance with
Rule 8 or permit a purely speculative complaint to proceed past a motion to dismiss.” Auth Token LLC v. City Nat’l Bank, 817 F.Supp.3d 211, 221 (S.D.N.Y. 2026). Plaintiff’s complaint cannot proceed on this theory. Third, and relatedly, Plaintiff argues that when Defendant internally tests the IMA Emulator, Defendant “employs configurations with multiple receiving endpoints to verify multicast delivery functionality.” (ECF Nos. 23-2, at 10; 31, at 22 6, 9). To the extent Plaintiff refers to configurations beyond the one discussed in the previous paragraph, its allegations are only more implausible. Far from being a “factual issue,” (ECF No.
31, at 9), this guesswork is a serious pleading issue. Even if the IMA Emulator is capable of transmitting multicast messages, it does not follow that testing of the IMA Emulator would include multiple ATM (IMA) devices on the receiving end. While a multicast message requires multiple subscribers, it does not require multiple inverse demultiplexers. As Plaintiff itself explains, the prior art for transmission of multicast messages lacked multiple inverse demultiplexers. (See ECF No. 23 ¶ 17 (alleging that the prior art required “reassembling the inversely multiplexed stream at a central router”)). Hence, the only thing separating Defendant’s internal testing of the IMA Emulator from the prior art is Plaintiff’s unsupported speculation. As already
established, that is not enough. In short, the IMA Emulator does not plausibly perform the step of Claim 1 requiring a further inverse demultiplexer. The amended complaint must be dismissed. IV. Leave to Amend In the event that the amended complaint fails to state a claim, Plaintiff requests leave to amend at the end of its opposition. (ECF No. 31, at 24). Fed.R.Civ.P. 15(a)(2) provides 23 that courts “should freely give leave [to amend] when justice so requires.” Accordingly, the United States Court of Appeals for the Fourth Circuit has instructed that “leave to amend a pleading
should be denied only when the amendment would be prejudicial to the opposing party, there has been bad faith on the part of the moving party, or the amendment would be futile.” Johnson v. Oroweat Foods Co., 785 F.2d 503, 509 (4th Cir. 1986) (citing Foman v. Davis, 371 U.S. 178, 182 (1962)). Leave to amend is not warranted in this instance. To begin, Defendant’s initial motion to dismiss asserted that Plaintiff’s original complaint failed to identify a further inverse demultiplexer, (ECF No. 14-1, at 12, 19), yet the amended complaint filed thereafter did not remedy that defect. There is little reason to think the third time will be more successful. Plaintiff’s proffer only confirms that doubt:
AuthPoint could likely add factual allegations regarding the specific network configurations in which GL Communications deploys, tests, and demonstrates its IMA Emulator products, including configurations involving multiple receiving-end IMA devices serving multiple subscriber endpoints; the manner in which multicast IP traffic is encapsulated within ATM cells and inverse[ly] multiplexed by the IMA Emulator; and/or the role of ATM Access Switches and other forwarding devices within 24 the Public ATM Network depicted in GL Communications’ system diagrams.
(ECF No. 31, at 24). These proposed amendments contain no specific factual allegations and offer no promise of fixing the identified defect. Moreover, Plaintiff’s request does not comply with Local Rule 103.6, which requires the attachment of the proposed (second) amended complaint and a redline showing the proposed changes. Leave to amend will be denied, and the amended complaint will be dismissed with prejudice. V. Conclusion For the foregoing reasons, Defendant’s motion to dismiss will be granted. A separate order will follow.
/s/ DEBORAH K. CHASANOW United States District Judge