IP Library Granted Patent US 12,284,097
Granted Patent B1
US 12,284,097 · App. 18/962,607 · Granted Apr 22, 2025

Network linker for tracking network traffic across a multiple protocol network

Inventor: Lev Zelenin (New York, NY)
Assignee: Morgan Stanley Services Group Inc.
H04L43/08
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,284,097
App. No.
18/962,607
Granted
Apr 22, 2025
Kind
B1
Abstract

A device, system and method is provided for tracking network communication in a multi-protocol network. During the offline phase, a linkage graph may be generated representing network communication pathways of recorded messages comprising nodes representing network communication devices and edges representing direct communication paths pairwise connecting parent-child nodes over which the messages are transmitted. The linkage graph may be optimized by eliminating or reducing branching nodes. During a real-time online phase, for each new recorded message associated with a device node, tracing a communication pathway along edges of the optimized linkage graph by linking a sequence of nodes that connect sequential pairs of child-parent nodes extending from the device node to a root node. Simultaneous to linking each node to the device node, a network traffic metric for the linked node(s) may be accumulated to quickly and efficiently track the network traffic metric for the new recorded message.

Claims (42)

1. A method for tracking network communication in a multi-protocol network, the method comprising:

In an offline phase:

tapping communication circuits in the multi-protocol network to record communication messages;

generating a linkage graph representing network communication pathways comprising generating an initial candidate arrangement of a plurality of nodes each representing a distinct network communication device uniquely identified in the recorded messages and generating a plurality of edges each connecting a pair of parent-child nodes representing a potential direct communication path from a parent device to a child device identified by recorded messages therefrom having matching content that are sequenced at the parent device directly prior to the child device;

optimizing the linkage graph by eliminating or reducing branching nodes that connect one node to multiple other nodes;

In a real-time online phase, for each new recorded message:

identifying a node representing a network communication device uniquely identified in the new message;

tracing a communication pathway along edges of the optimized linkage graph by linking a sequence of nodes that connect sequential pairs of child-parent nodes extending from the identified node to a root node; and

simultaneous to linking each node to the identified node, accumulating a network traffic metric for the linked node to the network traffic metric for the new recorded message.

2. The method of claim 1 comprising inputting the linked network traffic metrics into a machine learning model to output an altered linkage graph with an altered arrangement of nodes and edges that is predicted to improve network traffic metrics.

3. The method of claim 2 comprising inputting the altered linkage graph into one or more network routers or gateways to automatically alter the communication paths of the plurality of network communication devices to follow communication paths represented by the altered arrangement of nodes and edges.

4. The method of claim 2 , wherein the altered linkage graph is optimized to alter the communication paths to reduce hot spots or bottlenecks of network traffic at network communication devices.

5. The method of claim 2 , wherein the altered linkage graph is optimized to reduce underperforming nodes representing network communication devices with substandard latency.

6. The method of claim 2 , wherein accumulated network traffic metrics are used to assign weights to the altered linkage graph configured for optimal flow to rebalance network traffic.

7. The method of claim 1 comprising iteratively repeating the offline phase to update the graph by a new set of tapped recorded messages.

8. The method of claim 1 comprising:

if the communication pathway does not branch at a node, linking the node by following a unique pointer to a single parent node;

if the communication pathway branches at a node to multiple other parent nodes indicated by a multi-node pointer thereto, linking the node by retrieving from a lookup table a unique pointer to a single parent node; and

if a node can not be associated with a single parent node, searching all potential parent nodes in the linkage graph for recorded messages with matching content that are sequenced at the parent device directly prior to the device associated with the node.

9. The method of claim 1 , wherein the network traffic metrics are selected from the group consisting of: a number of nodes and edges the transmission traversed, message size, message latency associated with the identified node's single parent node, total latency as a sum of latencies associated with the sequence of linked nodes extending along the traced communication pathway extending from the identified node to the root node, and classification or priority of the recorded message associated with the identified node's linked node or root node.

10. The method of claim 1 comprising, in the linkage graph, replacing edges with nodes, such that each node represents the potential direct communication path from a parent device to a child device, and connecting the nodes with edges sequencing the potential direct communication paths.

11. A system for tracking network communication in a multi-protocol network, the system comprising:

one or more memories configured to store record communication messages recorded by tapping communication circuits in the multi-protocol network; and

one or more processors configured to:

in an offline phase:

generate a linkage graph representing network communication pathways comprising generating an initial candidate arrangement of a plurality of nodes each representing a distinct network communication device uniquely identified in the recorded messages and generating a plurality of edges each connecting a pair of parent-child nodes representing a potential direct communication path from a parent device to a child device identified by recorded messages therefrom having matching content that are sequenced at the parent device directly prior to the child device, and

optimize the linkage graph by eliminating or reducing branching nodes that connect one node to multiple other nodes, and In a real-time online phase, for each new recorded message:

identify a node representing a network communication device uniquely identified in the new message,

trace a communication pathway along edges of the optimized linkage graph by linking a sequence of nodes that connect sequential pairs of child-parent nodes extending from the identified node to a root node, and

simultaneous to linking each node to the identified node, accumulate a network traffic metric for the linked node to the network traffic metric for the new recorded message.

12. The system of claim 11 , wherein the one or more processors are configured to input the linked network traffic metrics into a machine learning model to output an altered linkage graph with an altered arrangement of nodes and edges that is predicted to improve network traffic metrics.

13. The system of claim 12 , wherein the one or more processors are configured to input the altered linkage graph into one or more network routers or gateways to automatically alter the communication paths of the plurality of network communication devices to follow communication paths represented by the altered arrangement of nodes and edges.

14. The system of claim 12 , wherein the altered linkage graph is optimized to alter the communication paths to reduce hot spots or bottlenecks of network traffic at network communication devices.

15. The system of claim 12 , wherein the altered linkage graph is optimized to reduce underperforming nodes representing network communication devices with substandard latency.

16. The system of claim 12 , wherein the one or more processors are configured to use the accumulated network traffic metrics to assign weights to the altered linkage graph configured for optimal flow to rebalance network traffic.

17. The system of claim 11 , wherein the one or more processors are configured to iteratively repeat the offline phase to update the graph by a new set of tapped recorded messages.

18. The system of claim 11 , wherein the one or more processors are configured to:

if the communication pathway does not branch at a node, link the node by following a unique pointer to a single parent node,

if the communication pathway branches at a node to multiple other parent nodes indicated by a multi-node pointer thereto, link the node by retrieving from a lookup table a unique pointer to a single parent node, and

if a node can not be associated with a single parent node, search all potential parent nodes in the linkage graph for recorded messages with matching content that are sequenced at the parent device directly prior to the device associated with the node.

19. The system of claim 11 , wherein the network traffic metrics are selected from the group consisting of: a number of nodes and edges the transmission traversed, message size, message latency associated with the identified node's single parent node, total latency as a sum of latencies associated with the sequence of linked nodes extending along the traced communication pathway extending from the identified node to the root node, and classification or priority of the recorded message associated with the identified node's linked node or root node.

20. The system of claim 12 , wherein the one or more processors are configured to replace edges in the linkage graph with nodes, such that each node represents the potential direct communication path from a parent device to a child device, and connecting the nodes with edges sequencing the potential direct communication paths.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 18, 2025
From: ZELENIN, LEV
To: MORGAN STANLEY SERVICES GROUP INC.
Reel/Frame 070542/0798 →
References Cited (16)
US 7523462B1 · Nesamoney et al. · 2009 [cited by applicant]
US 10242016B2 · Gorelik · 2019 [cited by applicant]
US 11528283B2 · Yadav · 2022 [cited by examiner]
US 11941065B1 · Li et al. · 2024 [cited by applicant]
US 20080189277A1 · Meijer et al. · 2008 [cited by applicant]
US 20080320075A1 · Livshits et al. · 2008 [cited by applicant]
US 20100070448A1 · Omoigui · 2010 [cited by applicant]
US 20130031331A1 · Cheriton et al. · 2013 [cited by applicant]
US 20140358932A1 · Brown et al. · 2014 [cited by applicant]
US 20150222557A1 · Bhattacharya · 2015 [cited by examiner]
US 20160217187A1 · Iesiev et al. · 2016 [cited by applicant]
US 20190089725A1 · Anachi · 2019 [cited by examiner]
US 20200076584A1 · Steele · 2020 [cited by applicant]
US 20200162503A1 · Shurtleff · 2020 [cited by examiner]
US 20200177485A1 · Shurtleff · 2020 [cited by examiner]
US 20230412620A1 · Crabtree · 2023 [cited by examiner]
Cited By (1)
US 12,580,929