Dynamic Content Fulfillment associated with Realtime Bidding

Patent #:US11227314B2

Issued: Jan 18, 2022

An approach for creating dynamic content. The approach receives advertiser data associated with activities of one or more advertisers and receives publisher data associated with activities of one or more publishers. Approach manages the one or more DSPs activities associated with the received advertiser data and publisher data. Furthermore, approach manages the one or more SSPs activities associated with the received advertiser data, publisher data and the one or more DSPs activities and selects one or more advertisement for one or more website. Finally, the approach manages the one or more consumer behaviors associated with the selected one or more advertisement

See Patent

Dynamic network management based on user, device, application, and network characteristics

Patent #:US11012332

Issued: May 18, 2021

In an approach to dynamic network management based on user, device, application, and predicted network characteristics, one or more computer processors identify one or more network interfaces. The one or more computer processors determine one or more network performance parameters and capabilities of the identified network interfaces. The one or more computer processors determine a highest ranked network connection based on the determined network performance parameters and capabilities of the identified network interfaces. The one or more computer processors generate one or more network templates based on the determined highest ranked network connection parameters

See Patent

Generative Experiments for Application Deployment in 5G Networks

Patent Reference #: P202010210US01

Filed: May 01, 2021

The core idea is to determine how to deploy the best container to the right node and ensure that the new deployment will ensure the entire solution which consists of multiple containers spread across multiple nodes functions optimally. The system will also ensure once deployed that any potential issues that may occur in the future are quickly identified and corrections made by deploying or stopping relevant containers.

Problem is graph matching: matching a (logical) microservice graph to a (physical) network topology. In general, this matching problem is NP-hard. Several prior approaches have examined optimization solutions (quadratic integer program) and heuristics to solve this problem. Such approaches incorporate the graph structure (in both logical and physical graphs), placement constraints (e.g., this logical node can only be placed on a physical node that has at least 16 GB RAM), (anti-)affinity constraints (e.g., logical nodes A and B should (not) be placed on the same physical node). However, such approaches often fail to consider the richer semantics of nodes/links in the logical/physical graphs

Delayed Instantiation of Network Slices

Patent Reference #: P202007699

Filed: Feb 01, 2021

Solve the problem of potential degraded performance in virtual network. This can occur if creating + finding right “network slicing” can take greater time compared to actual packet forwarding. Proposed solution focuses on need to have the “Lazy Network Slice” method approach to enable consistent lower latency & introduce slice differentiation on as ‘as needed’ basis

Many network architectures provide support for creating virtual networks on top of physical networks, e.g. the concept of network slices in 5G networks. Virtual networks allow customization of network functions, but creating virtual networks can add significant performance overhead in the implementation of the network functions. In many cases, implementation of virtual networks can degrade performance worse than just supporting best effort traffic due to the inherent inefficiencies in the implementation. We want to have a smart implementation of virtual networks which avoids the inefficiencies due to network slices support

Omnichannel Virtual Assistant Using Artificial Intelligence

Patent Reference #: P201806738US01

Record ID #: 95128569

Filed: Apr 2019

Chatbots are becoming increasingly pervasive. However, there are several limitations including trustworthiness, limited information derived from them, lack of personalization and inability to rapidly move between different chatbot channels. It would be good if user could control or at least understand the data source and external entities the chatbot is using to get the information from. In addition, as chatbots get more prevalent, consumer may need to interact with multiple chatbot channels. Even better would be if user does not even go to the company chatbot to get the information about the product but instead interact with their trusted chatbot which is personalized for user on user’s language (including grasp of the ‘localized’ language), trusted, has the relationships with the other chatbots out there to enable user to do what they want. This invention presents an assistant that users can trust and the underlying features on how to make it a reality 

Method for Managing Edge Nodes Across Multiple Edges Including Mobile Edge

Patent Reference #: P201909481

Record ID #: 95880523

Filed: Oct 2018

Edge computing is beginning to have a big impact on how systems are designed. There are currently multiple definitions of Edge each with different characteristics.
• The Local Edge (aka Gateway Edge) is the edge location where larger capacity and capability infrastructure, such as Kubernetes clusters can be deployed.
• The Far Edge (aka Device Edge) is the edge location at the furthest point from the centralized data center and is often an IoT device or camera with minimal compute capabilities. In most cases it is stationary and therefore able to communicate with the local edge without issue.
• The Mobile Edge is the Edge location populated by devices which are mobile. These may be small IOT devices or a more complex device such as a car or train. These devices must be able to interact with different local edges for compute and data.
The significant growth in the number and type of all edge devices is greatly increasing the complexity of the interactions. A far or mobile edge device must now interact with different local edges for compute, data and connectivity. In addition, far edges may need to become local edges communicating with each other in a mesh environment if a local edge is not available.
This disclosure addresses the issue of how such mobile and far edge devices can continue to work with local edges in an effective manner