Digital Commons Network^™

Custom Print Settings Using Annotations, Hp Inc

Defensive Publications Series

Print settings for documents often are set using the print dialog and print preferences UI of the

printer selected. These settings will be applied at document or page level only. Here, we

propose a novel solution where user can customize print settings using annotations in the

document at content level.

Control Plane Shared Risk (Cpsr) Factor In A Converged Packet-Optical Network, Praveen Kumar, Navdeep Sood

Defensive Publications Series

Presented herein is a "control plane type" along with numbers in the calculation of diverse paths for Label Switch Path (LSP) diversity. The control plane type is published along with the Shared Risk Link Group (SRLG) value at the termination endpoint. This helps in determining the source control plane of the SRLG value. The control plane type publication is local to the node/endpoint and does not need any additional signaling.

Generation Of Synthetic Crash Data And Application Of Crash Algorithms With Deep Learning Methods, Verena Blunder

Defensive Publications Series

No abstract provided.

App-Centric Distributed Messaging Bus For Performance Monitoring, Ted Hulick

Defensive Publications Series

A novel messaging system simplifies integration of dissimilar devices and instrumentation to enable a "distributed" application performance intelligence.

System And Method For Validating A Multi-Digit Number, Kim Wagner, Chackan Lai, Simon Hurry

Defensive Publications Series

The present disclosure provides a modified type of MOD 10 algorithm for validating a multi-digit number. Such MOD 10 algorithm may also be referred to as modified Luhn check or modifies Wagner check. The proposed modified Luhn check extends range of the Luhn check from 1 to 10 while preserving integrity of the multi-digit number when the multi-digit number is manually captured or typed. Also, by the proposed method, 9 times more PANs may be introduced without increasing the length of the PAN.

Methods And Systems To Use Relational Data To Assess Group Credit Worthiness, Joseph York, Nicholas Kurlas, Samantha Chatkin, Anastasiya Zarko

Defensive Publications Series

A computer-implemented method for aggregating and managing credit worthiness of a community including: selecting, with at least one processor, a group of individuals from the community; generating, with at least one processor, an individual risk score for each individual of the group based on individual data; generating, with at least one processor, a relational risk score for each individual of the group based at least partially on relational data; determining, with at least one processor, a weighted individual risk score and a weighted relational risk score based on a weighing parameter; generating, with at least one processor, a multi-party risk …

An Efficient Approach To Control Job Cancellation, Hp Inc

Defensive Publications Series

This disclosure focuses on providing an enhanced job cancellation approach in printing devices.

A concept is disclosed here, wherein when a job is being processed/queued, the user is enabled to

cancel printing of specific pages or a range of pages. This option thus provides a more usercontrolled

cancellation of the ongoing job in printing devices.

Method And System For Anomaly Detection In Large-Scale Networks, Fang Chen, John Garrett, Dave Zacks, Vladimir Yashin

Defensive Publications Series

Effectively spotting anomalies in network or application operations can be challenging in very large, complex networks, making it difficult to be alerted to their presence in order to take action to remediate such anomalies. Proper anomaly detection is impeded by too much data from too many disparate sources, which may manifest in too many different ways on various network devices ("the curse of dimensionality" familiar to many machine learning (ML) practitioners).

Proposed herein is a novel implementation of a Long Short-Term Memory based Variational Autoencoder (LSTM-VAE) to detect such anomalies, and an associated visualization technique to display them to the …

Power Backoff Mechanism For Intermodulation Distortion And Harmonic Distortion, Thomas Klingenbrunn, Fernando Selma, Hung Tsang, Aamir Akram

Defensive Publications Series

This publication describes systems and techniques for power backoff to address intermodulation distortion and harmonic distortion. A wireless communication device, which engineers often refer to as user equipment (UE), includes power backoff logic and a transmit backoff table. In a dual-connectivity environment with operating conditions known to result in potential distortion, a processor of the UE looks up a power backoff level in the transmit backoff table and directs a transceiver to back off the transmit power to mitigate intermodulation distortion or harmonic distortion. The UE can use similar systems and techniques to address distortion issues in carrier aggregation or …

A Mechanism To Dynamic Adjusts Usb Type C Ports' Bandwidth, Hp Inc

Defensive Publications Series

USB 3.2 spec provides 2 sides (flip-flop side) data transfer path called 3.2x2 (10Gx2) and

USB 4.0 (20Gx2) following. However, it is challenged to implement the solution into a real

system because of 2 side signal request more circuits and PCB/A space. Therefore, it will

increase the system design complexity and cost either.

Combined Approach For Building A Domain-Specific Natural Language Semantic Parser, Michael Laor, Tal Maoz, Aviva Vaknin

Defensive Publications Series

Proposed herein is a combined approach for building a domain-specific natural language semantic parser. The disclosed technique uniquely combines several known approaches with a new, original rule-based approach to obtain accurate mappings of utterances in a given domain into a single, pre-defined, domain-specific data structure that can be used for further processing. An implementation of the semantic parser has been built for the networking domain, including the QoS, filter (ACL) and set-path sub-domains, and has been integrated with a dashboard-managed L3 firewall for end-to-end implementation of filtering commands.

Driver Safety Improvements Using Data From Smartphone Sensors, N/A

Defensive Publications Series

A safe and smooth driving experience is one devoid of stressful events. To improve overall trip smoothness and driver safety, with user permission, the techniques of this disclosure utilize environmental factors (e.g., road geometry, weather, narrow lanes, etc.) and identify risky driver actions (e.g., speeding, sharp deceleration, etc.) to predict the likelihood of an unfavorable driving outcome. With user permission, dangerous driving maneuvers are identified using data from smartphone sensors such as GPS and inertial sensors such as the accelerometer, gyroscope, and magnetometer, e.g., on a smartphone or other device in the vehicle. A machine learning model, provided as part …

Driver Recognition And Data Retention Based On Driving Behavior, Verena Blunder

Defensive Publications Series

No abstract provided.

Artificial Intelligence Based Hyper-Realistic Deep Fakes For Participant Video In Collaboration Meetings, Rajarathinam Chidambaram, Karthikeyan Muthusamy

Defensive Publications Series

Techniques are described for Artificial Intelligence (AI) based deep fakes for a participant view in the meeting. A deep fake video may be built from poor quality and noisy/low-light video. A participant can eliminate others in the video that the participant did not intend to see but are in the camera field of view. Quality of video from poor quality cameras can also be eliminated. Moreover, a deep fake algorithm is described to build a high quality hyper realistic video from a low quality/no video scenario from the audio. This can be achieved by capturing only the face metadata on …

Contextual Information Triggered By Deep Click, Donny Chen Reynolds, Kejia Shao

Defensive Publications Series

Users often open new browser tabs or applications to look up something, e.g., additional information on a word found in the present tab. Such lookup of contextual information requires the user to frequently jump out of and return to the present tab or application, which is a tedious and distracting process. This disclosure describes techniques that provide user-interface cards that include contextual information based on the cursor position at the instant of a deep click of a haptic trackpad or touchscreen.

Security And Other Vulnerability Prediction Using Novel Deep Representation Of Source Code With Active Feedback Loop, Krishna Sundaresan, Anshul Tanwar, Prasanna Ganesan, Sriram Ravi, Sathish Kumar Chandrasekaran, Parmesh A.

Defensive Publications Series

Since the cost of fixing vulnerabilities can be thirty times greater after an application has been deployed, it is recognized that properly-written code can yield potentially large savings. Accordingly, approaches presented herein apply machine learning and Artificial Intelligence (AI) techniques to improve developer experience by enabling developers to avoid introducing potential bugs and/or vulnerabilities while coding. Billions of lines of source code, which have already been written, are utilized as examples of how to write functional and secure code that is easy to read and to debug. By leveraging this wealth of available data, which is complemented with state-of-art machine …

Cache Coherence Protocol, Long Nguyen, Mohsen Vakilian

Defensive Publications Series

Disclosed herein is a cache coherence protocol for a distributed cache and a distributed strongly-consistent database in which an improved mechanism is provided for determining the validity of cached profile values and determining whether to update cached profile values. The mechanism can store profile values in a cluster. The mechanism can read a profile value from the cluster and store the profile value in a cache in connection with a read timestamp and a staleness value. The mechanism can detect an event for which the profile value is to be used. The mechanism can then determine, based on the read …

Intent-Based Service Automation For 5g Core Network Function Slicing, Viktor Leijon, John Mullooly, Scott Wainner, Fredrik Jansson

Defensive Publications Series

The techniques presented herein maintain network function (NF) resource adjacencies regardless of the shared or dedicated nature of the NF and independent of the segmentation of the NF's cluster resources across geographic management domains. To achieve this, the techniques use intent-based networking models in order to create NF resources that facilitate the network slice creation and resource allocation. More specifically, the techniques utilize a Yet Another Next Generation (YANG) model representation of a network slice to facilitate creation and state management of all NFs associated with a network slice instance. The YANG model enables inter-site and intercluster adjacencies where cloud-native …

Method And Apparatus For Automated Link Failure Detection In Networks Using Machine Learning, Karthik Babu Harichandra Babu, Xinyuan Huang, Ajay Anand, Mankamana Mishra

Defensive Publications Series

Techniques are presented for automatic link failure prediction and detection with an automatic troubleshooting capability to quickly identify switches/ports involved in the failure before the failure leads to a complete outage. A framework to predict the failure and easily handoff all routing protocols to an alternate path is provided. Such prediction helps a network to react before a failure occurs. Almost zero traffic loss during a failure may be achieved by predicting the failure in advance.

Spitting Algorithm To Improve Drop Detectability Of Optical Drop Detector Sensor Used In Print Systems, Hp Inc

Defensive Publications Series

A solution is presented to improve drop detectability of optical drop detector sensors used in print systems. The

disclosed approach uses the concept of intermediate, cleaning drops between microcirculation and drop detected

sequences in order to prepare the nozzle to be able to produce a reliable and well-formed drop that will be

detected by the sensor.

Surfacing Biased Portions Of Multimedia Content Using Machine Learning, Daniel Keysers, Victor Carbune

Defensive Publications Series

Multimedia enables content creators to communicate information by adding nuance which is difficult to convey through written language via voice tone, camera angle, content highlighting, etc. However, it can be difficult for content consumers to discern biased opinions included within multimedia content. This disclosure describes techniques to automatically detect and surface such biased opinions within multimedia content. The process involves examining publicly available multimedia and/or text content related to a given piece of multimedia content to identify and flag biased portions. The identified biased portions are surfaced to the user via a suitable user interface mechanism.

User Interface To Present Query Responses Separated By Data Sources, Peter F. Ciccotto

Defensive Publications Series

Pieces of information that are responsive to user queries are typically personalized to the user. Information delivered in response to the user’s query can be derived from a variety of data sources, some of which are on the user’s device and some that are online. However, the response is typically delivered independent of the underlying data source from which the information is obtained. As a result, the user interface that shows the results does not enable a user to determine whether the response is based on public or user-specific information and whether it was retrieved from online or local data …

Client Authorization Grants With Account Creation Capability Using Oauth 2.0, Rohey Livne, Vitalii Tomkiv, Jeff Craig, Jason Stoops

Defensive Publications Series

A user of a web or mobile application, e.g., a video-sharing application, may want to access another online service, e.g., a third-party blogging or social media service, without leaving the application. The OAuth 2.0 standard is a common mechanism for authorizing a web application to access an account owned by the user on another online service. However, if the user does not have an account on the online service, accessing the service from within the application is at best tedious, and at worst, not presently possible. This disclosure describes OAuth-based techniques that enable a user of a web or mobile …

China Utility Model Patent Trash Or Treasure - A Data-Based Analysis, Patent Liu

Defensive Publications Series

The Chinese Utility Model Patent (CUMP) has existed since the implementation of China’s Patent Law in 1985. Previous studies on the CUMP focus on the patentability standards (subject matter, novelty, and non-obviousness), and the nature of the system itself. Data-based empirical studies are not often seen. This paper fills this gap by analyzing the CUMP from the data perspectives of citation, litigation, and finance. Derived conclusions show the growing importance of the CUMP, and its high value, as a treasure to CUMP holders, contrary to the traditional misconception that the CUMP is trash.

Mechanism For Changing The Mode Of A Shower, Anonymous

Defensive Publications Series

Disclosed is a mechanism for changing the spray mode of a shower. The mechanism includes a rocker and linkage system to translate a force applied directly to the shower face into a rotational force to rotate a selector thereby allowing the user to select different spray modes from the shower.

Multipart, Multipath Entropy Query Based On Egress Link Combination, Nagendra Kumar Nainar, Carlos M. Pignataro, Sagar Soni

Defensive Publications Series

Dynamic entropy ranges are runtime operational state pre-use cached results of entropy calculations based on an actual state of links. These results may include proactive entropy calculations based on the up- and down-states of links. A non-binary state for links that describes their load can also be included. This information enables network optimization and traffic engineering by introducing the capability of querying additional details in the multipart-multipath entropy query. Querying can be accomplished using Label Switched Path (LSP) Ping or any other Operations, Administration, and Maintenance (OAM) extensions, such as Yet Another Next Generation (YANG). Upon receiving such a query, …

Using Fresh Web Content To Generate Subscribable Search Notifications, Yew Jin Lim, James Kunz, Jon Dreyfus, Mauricio Monico, Tao Feng, Maya Palem

Defensive Publications Series

Content subscriptions are a useful way for a search engine to maintain user engagement. However, generating and providing content updates regarding topics of interest based on knowledge repositories can be expensive. Further, updates generated in this manner can be too infrequent, providing an unsatisfactory user experience. This disclosure describes the automated analysis and use of fresh web content related to topics of interest to generate and deliver updates to subscribers. Fresh web content is analyzed, e.g., using a natural language classifier, to determine the topic and type of content. Updates are generated based on specific content types known to be …

Energy Efficient In-Line Encryption With Selective Round Masking, Anonymous

Defensive Publications Series

Side channel protection imposes substantial area, performance, and/or energy overhead, rendering such hardware unsuitable for devices that have space or battery constraints, e.g., head-mounted display (HMD) devices. To safeguard user data in such devices, energy efficient encryption techniques are necessary. This disclosure describes an in-line encryption engine that uses selective masking. The first and last rounds of encryption are accorded maximum protection by incorporating all masking features, while intermediate rounds are only partially masked at steps that are determined as likely vulnerable, or are left unmasked. Selective round masking in this manner provides data protection with low cost, and is …

Scalable And Efficient Real-Time Clock Maintenance In A Distributed Network, Venkat Reddy Ennu

Defensive Publications Series

Techniques are described herein for maintaining real-time software clocks in a distributed network. Generally, the techniques may approximate a synchronized timestamp for a delay request sent by a slave node based on a timestamp of a sync request received at the slave node. Then, a time stamping unit associated with the slave node may update the correction factor of the timestamp to adjust the synchronized timestamp appropriately. Notably, this avoids the need to read timestamps from first in, first out buffers of time stamping units and/or from field programmable gate arrays (FGPAs) and, thus, may significantly improve the scale of …

Order Independent Tiebreaker For Motion Search, Anonymous

Defensive Publications Series

A motion search integrated circuit (IC) is configured to determine the motion of objects in a video. In motion search, multiple candidate motion vectors may often have the same cost, such that the multiple candidates are each optimal and equally acceptable. The multiplicity of candidates leads to difficulties in verifying the IC design, e.g., in comparing the output of the IC to a reference (golden) model implemented in software. This disclosure describes tiebreaking techniques that enable the IC, as designed in a hardware description language, and the reference model in software to identify the same optimal motion vector for a …