Digital Commons Network^™

Peer Node Eviction In Wireless Mesh Networks, Javier Cardona

Defensive Publications Series

Wireless mesh networks that include large numbers of peer nodes pose problems for routing frameworks. Multicast messages used for peer discovery can get dropped and synchronization of data between peer nodes does not scale to large numbers of peer nodes. A peer eviction technique for selective eviction of peer nodes is described. The technique can be used to limit the number of nodes for a wireless mesh network. Peer nodes are selected for eviction based on node-specific metrics. The technique ensures stability of the network by restricting the eviction rate and by adjustments to signal thresholds used to evaluate nodes …

Person Identification And Displaying Ar Banners, Anonymous

Defensive Publications Series

A technique for identifying a person and displaying a banner using Augmented Reality (AR) is described. The person is identified by comparing data gathered from orientation sensors such as gyroscope, accelerometer, etc., present in a smartphone of the person with estimated sensor data of the person (e.g. approximate readings of orientation sensors of the smartphone are estimated basis person’s motion/movement, which is detected by analyzing images of the person). Finally, AR is used to display an AR banner or an AR bubble (a 3D object) on top of the identified person. The AR banner or the AR bubble may be …

A Wearable Emotion Detection And Video/Image Recording System, Anonymous

Defensive Publications Series

A system for sensing an important event/occasion and automatic capturing of video/image is disclosed. The system detects important event/occasion by analyzing physiological signals and derives corresponding emotional state of a user. The system includes a wearable computing device having an emotion recognition system, a sensor array and a capturing device. Alternatively, the sensor array may be attached to the arm of the user when it is not integrated with the wearable computing device. The capturing device starts recording a video/image of surroundings if an event/occasion is detected. The user presets settings of the capturing device, which includes shutter speed, aperture, …

Verifying Authenticity Of Digital Images Using Digital Signatures, Neil Dhillon, Tanmay Wadhwa

Defensive Publications Series

This disclosure describes techniques to verify the authenticity of digital images using digital signatures. A secure element on a client device generates a public-private key pair and the public key is uploaded to a server. When a new image is captured by the client device, the device hashes the image and signs the hashed result using the private key to form a digital signature. Both the original image and digital signature are uploaded to the server. The server hashes the original image and compares it to the digital signature that has been decrypted using the appropriate stored public key. A …

Performance Delay Measurement And Fault Detection Of Sr / Srv6 Te Policies In Software Defined Networks, Clarence Filsfils, Rakesh Gandhi, Tarek Saad, Patrick Khordoc, Sagar Soni

Defensive Publications Series

Mechanisms are provided for Performance Measurement (PM) Delay Measurement (DM) as well as Fault Detection (FD) for Traffic Engineering (TE) Segment Routing (SR) policies (applicable to both SR-MPLS and SRv6 data-planes) in Software Defined Networks (SDN). For an SR policy, end-to-end atomic paths are "dynamically" computed using adjacency SIDs (or prefix SIDs in some cases) of all its ECMP paths by the path computation process running locally or on a controller. Key of the atomic path based on IP address of each hop is used for Telemetry, histogram and Client notification. The PM probe messages (or FD messages) are injected …

Visual Match Of Emails Or Landing Pages To Detect Phishing, Kuntal Sengupta, Vijay Eranti

Defensive Publications Series

In a phishing attack, a perpetrator attempts to obtain the online credentials of a user by impersonating a trusted entity such as a bank, email service provider, etc. Sophisticated phishers attempt to deceive spam filters by structuring the visual look-and-feel of their fake emails to be nearly but not precisely identical to emails sent by a trusted entity, such that the spam filter allows the fake email to reach a user’s inbox.

This disclosure applies machine-learning based techniques to assess the visual similarity of genuine and phished emails (or landing pages) for a given brand. The techniques detect visual near-duplicates …

Using Machine Learning To Improve Image Storage And Recall, Mary Witkowski, Kristina Bohl, Raja Ratna Murthy Ayyagar

Defensive Publications Series

Photographs taken using mobile devices are uploaded to image storage and sharing services at a resolution, server location, and upload priority that is blind to the importance of the photo to the user. This is sub-optimal, as there are images of great importance to a user (e.g., wedding photos) that are better uploaded at high priority and resolution. Similarly, there are images (e.g., screenshots, photos of receipts, etc.) that are not very important and can be uploaded at lower priority and resolution.

With user permission, the techniques of this disclosure use machine learning to predict the importance of an image …

Spatially Optimized Wireless Multicast Networks With Guaranteed Bit-Rate, Paul Heninwolf, Daniel Fernandes Barros

Defensive Publications Series

Multicasting is a mode of transmission in which the same information is relayed to a subset of users. Delivering the same information to multiple users of the subset via a series of one-to-one (unicast) transmissions quickly exhausts bandwidth. Delivering information in a series of one-to-all (broadcast) transmissions such that each subset receives its information in a round-robin manner causes latency, or equivalently, a reduction in bandwidth.

The techniques of this disclosure estimate the angular locations of each multicast subset using channel state information received across an antenna array. The locations are clustered using machine learning models. The available transmit power …

Remote Ranging Using Vibrations And Ultrasonic Transmission, N/A

Defensive Publications Series

This disclosure addresses the problem of ranging or imaging a remote environment, e.g., a remote conference room, using audio equipment present at the remote location. Ultrasonic waves are transmitted, e.g., via standard telecommunication channels, from the near end to the remote location, and their reflections off objects at the remote location are received at the near end. The reflections are used to reconstruct the remote environment and form an ultrasonic signature. The techniques of this disclosure can be used to secure a phone call, e.g., to verify the identity of the remote party; verify the remote location; differentiate humans from …

User Identification Across Online Providers, Anonymous

Defensive Publications Series

This disclosure describes construction of a user identification model based on confirmed user data obtained from different providers. The user identification model is usable by individual providers to provide missing pieces of user-specific data. An identification server maintains user data that includes globally unique user identifiers. The user data can be utilized to match user actions across different online providers and to perform analyses based on matched actions.

Smart Text Tool For Capturing And Sharing Visual Content

Defensive Publications Series

A system, method and computer readable media to facilitate capturing and sharing images through a device application is disclosed. The application includes smart and AR text tools for automatically placing the text in the image and choosing the color or the background of the text based on one or more rules. The rules may be based on analysis of the image such as facial recognition, contrast detection, color detection and content detection. The application additionally includes an augmented reality text filter for text placement on a capture screen. The AR text tool allows user to add expressive motion elements to …

Automated Augmented Reality Context Stickers

Defensive Publications Series

The disclosure relates to a method, system and computer readable media for automatically placing relevant augmented reality (AR) stickers in user’s field of view on opening a device application. The method places dynamic location, weather or time stickers on the ground plane or above the field of view, once the user opens the camera and also during the capture flow of moments. The user may either flick or dismiss them, or press-and-hold the stickers to move them to a desired position and then capture. The method brings the most used 2D post-capture sticker learnings to life automatically. The method gives …

Capture Modes, Filters And Frames For Visual Content

Defensive Publications Series

A system, method, and computer readable media are disclosed for performing swipe operations to edit visual content. The device application may include a smart feature to provide a plurality of subtle filters and/or frames for easy post-capture use with a swipe. The swipe filters and/or frames may be customized based on geographic location or nationality. Subtle filters with skin smoothing or improved lighting effects and boost filters may be used. Frames may be based on a location or city, or time - a day of week or time of day or relevant holiday frame may be used. The filters enhance …

Tool For Measuring End To End Latency, Michal Kielan, Jose M. Gonzalez, Michal Godek

Defensive Publications Series

A system, method, and computer readable media for measuring end to end (e2e) latency in a video calling system is disclosed. The latency measuring system includes a video calling system having at least two user devices that may be connected over a network, one or more external cameras, a device producing luminescence and a computing device. The camera is placed in front of each of the user devices and is configured to record the video call in both the devices. The luminescence device may be turned on for a short moment during the recording of the video call. The system …

Provisioning Day-Zero Configurations Through Passive Radio-Frequency Identification, Santosh Patil, Shyam Vaidyanathan, Prashant Kumar, Manoj Gupta

Defensive Publications Series

Embodiments presented herein provide a mechanism for setting the day-zero configuration of a network device without having to power on or unbox the device. Using an embedded passive radio-frequency identification (RFID) tag situated in a device, the device can be programmed at a distance using a mobile device.

Identifying Ai Participants In Calls And Escalating Ai-Only Calls Into An Optimized Machine Communication Mode, Denis Savostiyanov, Rob Hanton

Defensive Publications Series

The embodiments presented herein enable two human-emulating artificial intelligence (AI) speech agents to identify each other in situations where the two are engaged in a conversation. Once two speech agents recognize each other as artificially intelligent, they may transition to a more efficient method of communication in order to complete tasks more quickly, accurately, and inexpensively.

Device History Recall Optimization With Virtual, Locally-Stored Acquisition Databases Using Previously Encountered Communication Access Points And Services, Jibing Wang, Qin Zhang, Srinivasa Prasad Vangaru

Defensive Publications Series

A mobile communication device of an end user (user equipment or “UE”) can use more than one radio access technology (“RAT”) such as GSM, WCDMA, TDS-CDMA, LTE, and NR. Searching for information needed by the UE to use a specific RAT can consume large amounts of power and time. Locally stored acquisition databases (“PLMN Info DBs”) of previously encountered communication access points and services can speed up searches, but can be limited by the storage capacity of the UE. Also a locally stored PLMN Info DB may be invalided when a UE moves to a location that is distant from …

Crowdsourced Market Research And Survey Completion Using Intelligent Conversational Assistants, Reena Singhal Lee

Defensive Publications Series

Insights into customer behavior are important for making business decisions. Yet current market research methods, e.g., surveys, multiple-choice questionnaires, etc., are long or complex, and have little or no incentive for customer participation. The quality and coverage of data that results from traditional surveys is sometimes found to be insufficient.

This disclosure leverages conversational assistants to conduct market research. At appropriate times, and with user permission, a conversational assistant asks a user questions relating to their everyday purchases, behaviors, or preferences. The questions may be longitudinal, audio-based, location-based, etc. For users that provide permission, user responses can be studied, without …

Lightfield Compression Using Commodity Hardware Video Codecs, Sai Deng, Pavel Krajcevski, Matthew Pharr, Stuart Abercrombie

Defensive Publications Series

A lightfield video is a series of lightfield stills, which include a number of reference views and residual data describing differences between the reference views and various viewpoints. The set of reference views or images, though different enough to provide the required basis for predicting intermediate views in the dataset, are correlated enough to approximate a traditional video stream. Additionally, reference images from a single time point may be compressed as a collage of smaller images that make up a single frame of a traditional video. Temporal correlation between reference images at successive time points is exploited, similar to standard …

Micro Touchpad For Wearables, Chun Yat Li, Guangzhe Cui, Debanjan Mukherjee

Defensive Publications Series

Users interact with electronic devices via a variety of input devices, such as buttons, touchpads, touchscreens, keyboards, styluses, and mice. Touchpads allow users to provide the electronic devices with a variety of commands and instructions because of their flexibility in receiving many types of input, including touches, swipes, gestures, taps, and holds from one or more fingers or other input devices. Touchpads also allow for robust construction and ergonomic variability in environments in which other input devices may not function as effectively. In particular, a micro touchpad sized for wearable devices, such as watches, activity trackers, or headphones, that retains …

Cognitive Cloud Collaboration, M. David Hanes, Sebastian Jeuk, Gonzalo Salgueiro, Chuck Byers

Defensive Publications Series

Techniques are described for using a cognitive learning instance to learn a user's preferences across a wide range of cloud collaboration services and provide a customized user experience. This customized user experience is then leveraged on the back-end for optimizing cloud resources.

Mechanical Controller For A Smartphone, Gökhan Bakir, Joseph Lange

Defensive Publications Series

This disclosure describes a mechanically operated controller for a mobile device. The controller includes a grid of black and white squares whose pattern can be mechanically changed. Particular patterns are read by a mobile device, e.g., via its camera, and interpreted as different commands. The mechanical controller can be used as an interface between a mechanical device, e.g., a musical instrument, and the mobile device.

Asymmetric Pcie, Nishant Patil

Defensive Publications Series

Many devices such as accelerators, storage appliances, video transcoding accelerators, etc. have asymmetric bandwidth requirements. For example, the ingress bandwidth for a machine learning accelerator can be more than ten times the egress bandwidth. However, this asymmetry in bandwidth is not reflected in the interconnect. For example, PCI-express typically has as many lanes for host-to-device communications as it does for device-to-host communications. This disclosure presents techniques for asymmetric links between host and device. By reflecting the relative magnitudes of to-and-fro traffic more accurately, the asymmetric PCIe link achieves greater efficiency of communication.

Software Service For Facilitating Unbiased Ratings, Tuna Toksoz, John Dukellis

Defensive Publications Series

Online application stores, e.g., that enable users to download software applications to a user device, include rating systems that allow users to provide ratings and reviews. Some application developers attempt to obtain favorable reviews for their apps by first asking users if they are enjoying the app, and requesting a review only from the users that answer affirmatively to provide a review. As a result, the ratings and reviews for such applications on the store may be skewed.

This disclosure describes a service for obtaining unbiased ratings directly from users. An application store that utilizes the techniques sends a ratings …

Machine Learning To Optimize Controlled Experiments, Neil Hoyne

Defensive Publications Series

Businesses use pre-sales questionnaires to quantify the value of incoming leads. Results from such questionnaires can improve allocations of sales resources and optimize advertising campaigns. A given questionnaire may ask for dozens of pieces of information from a prospective client. However, it is well understood that the questions are not always predictive of the likelihood that a client completes a purchase.

This disclosure provides machine-learning techniques that discover the predictive ability of questions on a questionnaire. The techniques enable a business to select or design questions or experiments that correlate with the closure of a sale. The business benefits by …

Conversion Detection For Improved Online Ad Selection, Stephen Hsu, Eric J. Kolb, Steven T. Dieter

Defensive Publications Series

Online ads sometimes continue unchanged even after the ad had resulted in a conversion, e.g., after a user has purchased the advertised product. Many users find such continued presence of the ad annoying. Also, not refreshing the ad with a different product represents lost revenue for the website publisher and the ad network.

This disclosure describes techniques to detect if the user has purchased a given product by analyzing, with user consent and permission, user data such as emails and other data. Upon detecting conversion, display of ads relating to the product or similar products is discontinued.

Machine-Learned Alt Text For Images And Videos, Sandro Feuz, Mohammadamin Barekatain

Defensive Publications Series

HTML has a provision for adding a textual description, known as alt text, to media elements in a webpage, such as images and videos. This is useful for improving accessibility of web pages, for display environments that do not support a particular media item, for content searching, etc. The alt text is currently added manually, e.g., by the web page developer or publisher. This is a labor intensive activity and does not work for dynamic media. This disclosure utilizes machine learning techniques to automatically add alt text to media content.

Context-Dependent Output Volume For Voice-Controlled Virtual Assistant, Victor Cărbune, Thomas Deselaers

Defensive Publications Series

Voice-controlled virtual assistants typically maintain a standard output volume level. In certain circumstances, this leads to a response from the virtual assistant that is either too loud or too low. For example, if the last-used context of the virtual assistant was a party with high ambient noise, then the virtual assistant continues responding too loudly even after the party has ended and the ambient noise level has dropped.

This disclosure describes techniques to automatically adapt the output volume level of a virtual assistant based on current context, user preferences, user feedback, etc. A machine learning model predicts an optimum volume …

Fast Wifi Scanning, Jiwoong Lee

Defensive Publications Series

In some countries, unlicensed spectrum used by WiFi networks is shared with radar systems. WiFi networks that operate in such bands employ radar detection and avoidance. A WiFi transceiver spends a certain channel sojourn (dwell) time scanning a channel prior to starting communication on it. A full scan takes about seven seconds, a user-perceptible delay.

This disclosure describes techniques to reduce scan time by leveraging radar detection procedures that have already been completed by nearby access points. If a nearby access point has recently transmitted a frame or emits a wide-band transmission, it is an indication that the nearby access …

Task Completion By Seeking Input From Multiple Parties, Matthew Sharifi

Defensive Publications Series

People frequently perform online tasks that require input from others, typically requiring multiple context switches to consult with multiple parties until a consensus is reached. The process can be time consuming, requiring the user to pause until others provide input. A user must also remember to return to the task and complete it after the input sought from others is received. This disclosure describes a technique to obtain input from others regarding an online task. At the point when completing the task requires input from another party, the techniques solicit such input, with user permission. Conversational responses received from other …