Device Fingerprinting Across Home Networks

by Parks Associates | May. 14, 2019

Content Submitted by Universal Electronics Inc.

You can imagine that a device can be assigned a unique fingerprint based on a combination of attributes. Device fingerprinting is the technique where relevant data and features about a device are identified, extracted and profiled in order to uniquely identify the device within the home. This can be done across multiple networks within the home; and if paired with a knowledge graph of devices, can be used to automatically retrieve their capabilities and properties. This technology has today enabled many interesting use cases such as device recognition, more intelligent network scanning, intrusion detection or even personalized user experiences.

There are many different networks in the home including IP networks across Wi-Fi and Ethernet; HDMI wired networks connecting entertainment equipment together; and a range of low power wireless networks used in smart home applications such as Zigbee, Zigbee rf4ce, Z-Wave and Bluetooth networks. In traditional device fingerprinting techniques used in IP-based networks, feature extraction and profiling is done at different OSI layers of IP networks to achieve the desired application goals. These are typically focused on networking technologies to identify the device characteristics which limits their reach to other mediums that are available in the home. Another point to note is that as the number and diversity of connected devices within the home continue to rise, it’s typical for manufacturers and providers to make regular firmware updates to them. These updates may sometimes change the device fingerprints and such real-time updates must also be addressed by the device fingerprinting approach chosen. If not addressed, these gaps in the fingerprinting techniques may limit their ability to enable many important use cases that are primarily driven by daily user experiences.

User Experience Improvements 

Consumers purchasing smart home devices are looking for solutions that simplify their user experience right from the time when they open the box, starting with set-up and installation of their devices. The initial setup and configuration can be automated utilizing device fingerprinting to retrieve the necessary information and provide an easy out of the box experience. 

A good user experience addresses both the out of the box device setup as well as a smooth daily interaction with the devices. Features such as user-friendly names that can be easily recognized by consumers on the Smart TV UI as opposed to "HDMI 1", "HDMI 2"; or proper device types and names instead of generic MAC addresses on a wireless router management page, both address real usability issues for the consumer. At the core of both features is the need to dynamically discover and recognize devices across different networks.

Better user experiences across connected devices, from Smart TVs to Wireless Routers, both out of the box and in the daily use, enabled by device fingerprinting techniques.

Security

According to the Client Computing Group at Intel, the world is expected to have 50 Billion connected devices by 2020! With such a growing number of devices, security remains to be one of the top concerns for the consumers. There are different applications available in the market that are using device fingerprinting techniques to provide security by acting as the primary gatekeeper in the home network. A robust security application can only be achieved with proper device fingerprinting, to uniquely identify and track the devices and subsequently grant access to their services. These applications and services provide necessary monitoring and analytics about the devices through their intuitive user interface, generate alerts on unusual activities and even enable remote access in events of troubleshooting.

50 Billion connected devices by 2020, device fingerprinting and AI-based behavior profiling can protect consumers.
 
Remote Technical Support

According to Parks and Associates, 50% of Smart home consumers face set-up problems and 32% face technical issues resulting in poor device performance. That being said, remote technical support and system diagnosis have gained immense popularity among the smart homeowners today. According to Customer and Product Experience 360, in the U.S., the average consumer spends as much as 2.5 hours between self-help and technical support to fix the problems with their connected devices. For a technical support to be able to properly address the consumer concerns, they need to have a full understanding of the piles of connected devices within the home.

Let's take a look at a simple yet popular tech support case where a customer is unable to set up the device and ends up making a support call. With the technology at hand, technicians can access the home network to identify the correct device and guide the client with necessary steps. A simple and quick fix of their problems.

Comprehensive information about devices such as their brand name, model number, series number, manufacturer, current capabilities and offerings, their physical and software characteristics and will lead to a correct diagnosis of consumers’ problems. Device Fingerprinting can provide proper information and access to the consumers’ devices for an efficient real-time troubleshooting. Remote diagnostics will also help the technicians to understand the real problems in consumer homes and better prepare for their customer visits.

Customer issues are being diagnosed and resolved remotely by access to the home network through device identification and troubleshooting.

Personalization

Smart home technologies are entering the consumers' homes at a faster pace than ever before, and companies are working towards creating user experiences that are exclusive and revolve around their usage pattern. Device fingerprinting provides a complete understanding of devices in an ecosystem by uniquely identifying them and their properties. This wide dataset of information can help paint a true picture of consumers’ likes, interests and their interaction pattern with the devices. This information can further enable interesting use cases for personalized user experiences and monetization from customized advertisements. It is important to note, the proper balance between personalization and privacy should always be kept in mind, and the technology chosen should take privacy into consideration from the beginning.

Devices within the home along with the consumers' behavior and preferences are bringing about customized user experiences.

Device fingerprinting with QuickSet

QuickSet Discovery and Predictive engines detect device signatures across multiple home networks including HDMI, IP, and Zigbee rf4ce; calculate fingerprints and traverse the largest knowledge graph of devices to uniquely identify a device. With its predictive modeling, QuickSet intelligently identifies devices and their additional attributes. The results obtained across the different networks are merged to provide a single device object definition towards the host application. QuickSet, using its device fingerprinting technique and leveraging the constantly growing knowledge graph of devices, enables support for brand new devices as well as existing devices with changing behavior within the home. When it comes to the control information, QuickSet again goes beyond a single protocol approach for interacting with multiple devices.

QuickSet’s technology today enables many different use cases, primarily driven by the user experiences in the home. By implementing the whole-home discovery and control capabilities embodied in QuickSet, companies and application developers make it possible for consumers to automatically discover and interact with all points of access to content and applications in the home through a single point of control. 



Next: Device Fingerprinting capabilities now available to a wider audience!
Previous: Interoperability: The Cornerstone Of Connected Home

Comments

    Be the first to leave a comment.

Post a Comment

Have a comment? Login or create an account to start a discussion.


PA_Staff_Quotes_Jennifer_v1_600x60.jpg

PA_Staff_Quotes_Jennifer_v1_600x60.jpg