Digital Commons Network^™

Cable Fixation, Axel Unger

Defensive Publications Series

No abstract provided.

Data Driven Application Store Listing Optimization, Steve Suppe, Mike Davis, Jonathan Cheung

Defensive Publications Series

A computing system (e.g., a cloud server that hosts an application store) may predict the effect of modifying marketing assets (e.g., text, an image, a screenshot, a description, a video, etc.) of an application (hereinafter referred to as an “app”) on acquisitions (e.g., installations) for the app. The computing system may generate these predictions based on historical performance data (e.g., data relating to modifications to one or more marketing assets of the app and to acquisitions for the app) of marketing assets for a variety of types (e.g., lifestyle apps, social media apps, utility apps, productivity apps, entertainment apps including …

Automatic Functional Validation Of 3d Printing System Without Use Of Consumables, Hp Inc

Defensive Publications Series

Printing systems functional validation is a need for design, manufacturing, installation and service

engineers at different phases of the product lifetime. This refers to the systems integration acceptance

after the product manufacturing or a modification is done after a repair or a redesign. The objective of

this validation is to ensure that the product is delivered to the customer with an optimal quality and

performance, meaning that the parts printed with it will accomplish all the quality standards.

This idea is planned to be part of the overall customer acceptance procedure at new installations and

during the product use. This …

Data Driven Application Store Listing Optimization, Steve Suppe, Mike Davis, Jonathan Cheung

Defensive Publications Series

A computing system (e.g., a cloud server that hosts an application store) may predict the effect of modifying marketing assets (e.g., text, an image, a screenshot, a description, a video, etc.) of an application (hereinafter referred to as an “app”) on acquisitions (e.g., installations) for the app. The computing system may generate these predictions based on historical performance data (e.g., data relating to modifications to one or more marketing assets of the app and to acquisitions for the app) of marketing assets for a variety of types (e.g., lifestyle apps, social media apps, utility apps, productivity apps, entertainment apps including …

Automation Techniques To Customize Via/Pin Printed Circuit Board Padstacks For Enhanced Signal Integrity Performance, Linda Scott, Michael Umina, Quinn Gaumer, Stephen Scearce, Stanley John, Doug Yanagawa

Defensive Publications Series

The printed circuit board (PCB) design community has a desperate need to establish a robust, automated process that enables the use of customized via and pin definitions for enhanced signal performance in current and next generation 56/112/224 gigabit per second (Gbps) designs. Presented herein is a robust, automated process (and tool) to enable the design, implementation, management, and verification of customized via and pin definitions for a variety of PCB designs. In particular, a spreadsheet format is defined to provide the exact details of each transition designed for a PCB. A software tool provides for the ability to modify a …

System For Expert-Assisted Causal Inference For Ranking Events Of Interest In Networks, Dmitry Goloubew, Nassim Benoussaid, Carlos M. Pignataro, Vladimir Yashin

Defensive Publications Series

Networks have increased in size and complexity such that the number of events occurring each day has grown drastically. Techniques of this proposal provide for the ability to infer candidates for causal relationships—in some cases, with confidence. In particular, a novel machine learning (ML) based system is described that provides for the ability to narrow-down candidate temporal patterns that may potentially explain an event of interest (e.g., a network outage). The system is trainable with a human in the loop and is highly effective even with minimal amount of prior training.

Detecting And Outlining Potential Wildfire Areas Via Satellite Imagery, Mor Schlesinger, Pete Giencke, Christophe Restif, Avi Hoffman

Defensive Publications Series

Services that track wildfire spread and trajectory can provide such information and issue related alerts. The information obtained from current services is often stale and ineffective in helping people determine whether they are likely to be harmed by the fire. Moreover, such information is not in a form that lay persons can easily grasp, thus making it difficult for them to derive actionable insight from the received information. This disclosure describes techniques to leverage satellite imagery to automatically detect and outline the boundaries of large fires. The outlined boundaries are updated regularly and surfaced as appropriate within various pertinent applications …

Predicting Wireless Link Degradation From Observation Against A Known Topology, Mike Lake, Karthik Babu Harichandra Babu, Mankamana Mishra

Defensive Publications Series

Presented herein are techniques to utilize wireless microwave network links as sensors to feed into a customized weather model for use in targeted weather forecasts that can include information regarding observed effects of weather on the wireless transmission links. Such techniques may allow quick adaptation when storms are developing faster than micro-forecasts can be updated by centralized prediction services and may improve the relevance and/or direct interpretability of forecasts to network performance. In various implementations, the weather can be computed centrally within a single router.

Personalized Routing Dynamically Adjusted To Avoid Adverse Situations, Joseph Johnson Jr, Shiblee Hasan, David L. Lee, Christopher P. Hluchan

Defensive Publications Series

Routes that are optimal in general for a particular travel need may at times be unsuitable in a specific instance because of various factors such as construction, weather, accidents, rush hour traffic, road surface conditions, presence of crowds, etc. The impact of such factors on a person’s travel plans can depend on the specifics of the person’s travel circumstances, such as travel mode, vehicle characteristics, etc. This disclosure describes techniques to proactively deliver travel routing guidance personalized to a user, based on user-permitted contextual factors. If the user permits, the personalized routing guidance is adjusted to avoid one or more …

Android Security, Nancy Mehra

Defensive Publications Series

Android is the most popular mobile operating system among the most used touch devices. Today, 80 percentage of touch devices are using Android Operating System . The universality of Android devices has a direct impact on the app store Google Play, which was first introduced under the name of Android Market. Android is an open source operating system. Thus convincingly available and recognized by various operating system and code familiar to java , its applications can be easily developed and implemented on the Smartphones. Thousands of mobile apps are released through the Google Play Store every day ,so to …

Securing Connectivity Fault Management (Cfm) Control Packets Using Stamped Passport Attestation, Karthik Babu Harichandra Babu, Eric Voit, Sujal Sheth

Defensive Publications Series

Attestation is a trusted computing technology that can be applied to Ethernet layer operation, administration, and management (OAM) protocols, such as ethernet-cfm. A Canary Stamp is a collection of security evidence that demonstrates a sender's integrity and trustworthiness. Techniques presented herein provide for extending all types of Connectivity Fault Management (CFM) frames to carry a Canary Stamp Type-Length-Value object. Once a CFM frame is obtained by a receiver MEP, the receiver is to verify a sender's integrity and trustworthiness using the Canary Stamp TLV. CFM can add this trustworthiness status on each of its services in order to maintain connectivity …

Fraud Mitigation On Digital Payment Platforms, Santhosh Kumar Edukulla

Defensive Publications Series

Digital payment platforms enable users to pay merchants and to carry out a variety of online and offline financial transactions. However, there is a risk of fraudulent transactions, whereby payment credentials are misused to post fraudulent charges. This disclosure describes techniques that obtain a fraud rating for a merchant and provide it to users of a digital payment platform prior to the user conducting a transaction with the merchant via a payment application. The merchant fraud rating can be derived based on merchant and fraud rating data obtained by the payment platform from payment networks. The payment platform can also …

Humidity Condenser For Potable And Non Potable Water Generation, And Water-Metal-Ground Heatsink, Luca Aiello

Defensive Publications Series

This working hypothesis involves the use of a humidity condenser with heat sink that works from a depth of 5 meters and beyond (up to 10 meters depending on the sizing requirements), using water, thermal conductive metals and good conductive gravels or soil. The dissipator cools the water so that, with a circulation pump, it cools the exchanger placed in the humidity condenser. The exchanger/condenser, in copper or aluminum, with tube and tube bundle, on the one hand produces condensed water (distilled) and cold air, and on the other hand introduces into the cooling circuit the water slightly heated during …

Self-Adaptive Anomaly Detection With Deep Reinforcement Learning And Topology, Qihong Shao, Carlos M. Pignataro

Defensive Publications Series

In the networking field, network topology is one of the most important perspectives as it can bring additional insights to the modeling process. Existing anomaly detection approaches do not take topology information into consideration. To address such limitations techniques are presented herein that support deep Convolutional Neural Network (CNN) modeling with Reinforcement Learning (RL) employing, for example, an Advantage Actor Critic (A2C) algorithm. Additionally, aspects of the techniques presented herein support an innovative new way to model a "customer profile" that leverages topology information.

Locating And Securing Use-After-Free Pointers, Kentaro Hara

Defensive Publications Series

Dangling pointers and/or use-after-free pointers are gateways to exploits for software such as browsers, operating systems, etc. To forestall attacks that rely on dangling or user-after-free pointers, ordinary pointers in source code can be replaced by secure pointers. However, rolling out code with such replacement requires the developer to incur testing and engineering costs. Further, due to the likely memory and runtime overheads of secure pointers, it is useful to identify ordinary pointers in source code that merit replacement. This disclosure describes techniques to determine which ordinary pointers in source code are suitable for replacement by secure pointers.

Computing A Distance Between A Source Code Directory And Another Source Code Directory, Armijn Hemel

Defensive Publications Series

Sometimes it is important to see how much a vendor's source code tree is deviating from an upstream open source project such as the Linux kernel, as big deviations potentially mean increased maintenance costs, as changes/fixes might need to be backported from the official version. In this document a method to compute a score to quantify possible maintenance costs is proposed.

Tags: Git, software engineering, defect discovery, tlsh, software maintenance, software maintenance cost

Determining Original Sources Of Online Content, Irina Blok, Robert Benea

Defensive Publications Series

Information found on the Internet, on websites, social media platforms, video sharing services, etc. is often fragmented and appears out of context. There is no good way to know where particular content came from or to provide credit to original creators. This disclosure describes techniques that enable a user to easily dig deeper into sources of online content to determine its provenance. Per the techniques, a given image (bitmap or vector), audio, video, or other media is digitally fingerprinted. The fingerprinting technique is selected such that the generated fingerprint is robust to formatting changes, skewing, noise, etc. The fingerprints are …

Destination Search With User-Specified Constraints, Lingyi Zhang, Cliff Chin Ngai Sze

Defensive Publications Series

When issuing destination-related queries users sometimes include on or more constraints within the query. Even though a search engine can be used to search for data regarding each constraint individually, users need to integrate the individual results manually based on separate searches for different types of data. This disclosure describes techniques to retrieve and present search results for destination-related queries based on user-specified constraints present within a user query. The results are obtained by performing separate searches based on various constraints specified in the user query and combining and filtering the results to include only those results that match all …

Predictive Road Safety Alert System, N/A

Defensive Publications Series

This disclosure describes techniques to automatically provide alerts to users driving a vehicle when unsafe driving conditions may soon occur. An alert is determined based on vehicle speed and road conditions such as tight curves, pavement conditions, and weather, and with user permission, based on an ascertained skill of the user driving the vehicle as indicated by past driving occurrences. Safe speeds to handle upcoming road conditions can also be output. Described features provide information and guidance regarding appropriate driving speeds based on road conditions, vehicle characteristics, and driver ability.

Finding Out How Close Source Code Files Are To Files In The Git Version Control System., Armijn Hemel

Defensive Publications Series

A lot of popular software is developed using a version control system. Historically systems such as RCS, CVS and Subversion were used, but a lot of developers have moved to Git. These systems have a lot of information available about the history of a file. When software is distributed, it is often distributed without this history information. In some situations it is important to find out how close a certain piece of software is to any given version in a version control system, for example for assessing copyright, security research or other provenance issues.

keywords: Git, software engineering, security, defect …

Use Of Machine Learning To Generate Estimates Of Code Review Time And Effort, Marko Ivanković, Goran Petrovic

Defensive Publications Series

Providing accurate estimates of time required to perform code reviews can enable practitioners to prioritize reviews and can help improve productivity. Currently, factors such as the number of lines of code to be reviewed, or the number of code files, are provided to the reviewer. Such factors are relatively poor predictors of the actual effort required. This disclosure describes the use of machine learning techniques to generate effort estimates of the actual human time required to perform a code review. The machine learning model is trained on a dataset that is generated based on historical code modifications and corresponding reviews. …

Identification Of Low-Value Code To Improve Efficiency Of Automated Code Analysis, Marko Ivanković, Goran Petrovic

Defensive Publications Series

Certain software testing techniques such as mutation testing are often not utilized due to the relatively high proportion of testing activity that is not deemed useful. Such techniques can benefit from identification of high-value code. This disclosure describes the classification of different parts of a codebase into high and low value code. The techniques utilize an abstract syntax tree (AST) generated from the codebase on which predefined heuristics are applied to identify low-value code. A propagation algorithm is applied to mark low value branches. Additionally, machine learning techniques are applied to classify portions of code as high or low value. …

Sponsored Next Steps In Search Engine, Lakshmi Kumar Dabbiru, Hinali Marfatia, Ramakrishnan Kandhan, Chalmers Wang, Gang Wu, Senthil Hariramasamy, Abhineet Bhardwaj

Defensive Publications Series

Search engines can be accessed by client devices to improve search experiences, return on investment for content providers, and monetization of the search engine. A data processing system can construct a journey for a client device by suggesting subsequent steps for the client device to follow. The data processing system can generate search queries associated with a search query input by the client device, where each search query can include a statistical conversion rate. A subset of the search queries can be selected based on their statistical conversion rate. Consequently, the data processing system can transmit a search result including …

A Novel Inter-Personal Area Network Communication Mechanism For Large-Scale Low-Power And Lossy Networks, Huimin She, Harbor Dong, Chuanwei Li, Feiliang Wang

Defensive Publications Series

For large-scale Low-Power and Lossy Networks (LLNs), communications between nodes in different Personal Area Networks (PANs) (i.e., inter-PAN communications) typically have to traverse long routing paths, even for cases in which two nodes may be located physically close to each other. Techniques herein provide a novel mechanism to improve inter-PAN communications by utilizing the nodes that are located at the boundary of PANs.

Mechanism To Speed Up Secure Communication Handshakes In Constrained Conditions, Panos Kampanakis, Rupak Chandra

Defensive Publications Series

It is desirable to speed up secure tunnel negotiations in constrained mediums where numerous clients are "competing" to form secure connections to destination servers or endpoints. In support of that objective, techniques are presented herein that minimize the authentication data that is transferred within encrypted tunnel handshakes under constrained conditions where the data can introduce unacceptable slowness or failures. The techniques may apply to Internet of Things (IoT) constrained environments in which bandwidth is sparse and multiple devices are competing for bandwidth. The techniques may also be used on the Internet with post-quantum algorithms, which can introduce unnecessary slowness due …

Enhanced Network Slicing For Industrial And Energy Protocols, Pascal Thubert, Patrick Wetterwald, Jonas Zaddach, Eric Levy-Abegnoli

Defensive Publications Series

With the development of industry 4.0 and the recent evolution of the substation automation, as prescribed at least by the International Electrotechnical Commission (IEC) 61850 Standard, the network is becoming one of the key element of these trends. Network design and network architecture are becoming more and more complex and leading to challenging problems and issues, such as network security, multiplication of unmanaged broadcast domains, and bandwidth limitations. Recent tools have been introduced to help network engineers visualize different industrial Internet of Things (IIoT) protocol flows and characterizations for devices connected to the network. However, visualization is not enough and …

Individualized Lifelong Learning And Career Portal, Krishna Chytanya Ayyagari

Defensive Publications Series

This disclosure describes a career portal that provides career and job related information to a user based on insights into the learning style, aptitude, and career goals of the user. A recommendation engine includes modules for learning style determination and aptitude determination that utilize user-permitted data, e.g., from the user’s online activity and other data sources, to provide individualized course and career recommendations. A dominant learning style for the user is determined and is utilized together with their career goals to suggest coursework for the user to upgrade their skills. The recommendation engine also suggests suitable job opportunities. The described …

Docking Station With On-Demand Physical Memory (Ram) Enhance Processing Power And User Experience, Hp Inc

Defensive Publications Series

Notebook computers conventionally operate on volatile memory that is RAM for accessing multiple

programs at once with Speed and efficiency, as RAM is significantly faster than hard disk. When

operating a device with multiple heavy application that requires heavy usage of volatile memory to

process information immediately, RAM helps to succeeded with the task when there is enough of

memory available with the device’s RAM.

In designing notebook computers, there is trade‐off between the size of the RAM to be fitted for

performing better operations. Often devices have limited RAM and the applications that are being

used/ operated needs more …

Novel Secured Inter-Personal Area Network Group Management In Low-Power And Lossy Networks, Nan Yi, Wenjia Wu, Huimin She, Lele Zhang

Defensive Publications Series

Low-power and Lossy Network (LLN) environments may comprise, possibly among other things, different personal area networks (PANs) resulting in, for example, node communication challenges across or between PANs. To address these types of challenges, techniques are presented herein that support a novel secure group management method to self-solve the inter-PAN problem that is both low-cost and customer-friendly. Aspects of the techniques presented herein encompass establishing a secure node-to-node (N2N) communication link between involved inter-PAN nodes, automatically looking for the relay neighbors between different PANs (as the inter-PAN node can help with forwarding the local the Routing Protocol for LLN (RPL) …

Embedded Backup Path Information For Stateless Srv6 Micro Segment Identifier Node Protection, Nagendra Kumar Nainar, Kamran Raza, Carlos M. Pignataro, Rajiv Asati

Defensive Publications Series

Techniques presented herein provide for the ability to improve Topology Independent Loop-Free Alternate Fast Re-route (TI-LFA) with node protection by embedding backup path information (on a per segment basis) in a stack itself. In particular, techniques herein introduce a Node protected Prefix/Adjacency Segment Identifier (NP-SID) with a new (yet simple) forwarding semantic that can be used to embed backup path information directly in a segment stack. The NP-SID is always followed by the backup node information in the segment stack and the forwarding semantic to the NP-SID will involve a lookup on the backup information only if a protected node …