Skip to Main Content

Strategic Business Insights (SBI) logo

Connected Homes February 2020 Viewpoints

Technology Analyst: Christian Feest

Connected Home over IP

Why is this topic significant?

Connected-home market leaders have announced the development of an open-connectivity standard for connected-home devices. This new standard—if it materializes—will significantly increase compatibility among connected-home brands and potentially reduce product-development costs for device makers.

Description

In December 2019, several prominent connected-home companies—including Amazon.com, Google, and Apple, as well as the Zigbee Alliance and Zigbee Alliance board members including Samsung Smarthings, IKEA, and Signify—announced the formation of a working group to develop an open-connectivity standard for connected-home devices: Connected Home over IP (CHIP). CHIP is based on the long-standing and pervasive internet protocol (IP), which should enable compatible devices and platforms to send and receive messages to each other across any IP-bearing network. The CHIP working group plans to release a draft specification for the CHIP standard in late 2020.

Implications

The support of many major players in the connected-home industry suggests CHIP possesses the potential to become the standard connectivity protocol for connected-home devices and not just another competing standard. However, disagreements between the working-group members or departures of significant players could delay CHIP or result in abandonment of the project altogether.

If CHIP comes to fruition, a standard protocol should put an end to compatibility issues that have long been a problem of the connected-home market. At present, consumers' existing devices (typically a smart speaker from Amazon, Apple, or Google) often limit which additional devices they can add to their connected-home network. Someone who has built his or her connected home around an Amazon smart speaker, for example, cannot integrate the Nest x Yale lock or the EveCam security camera, because these devices are incompatible with Alexa.

Current information does not describe the intended scope of the CHIP standard. At a minimum, CHIP will likely enable compatible devices to connect to any smart assistant—whether from Amazon, Google, Apple, or some other compatible brand. However, CHIP may also standardize a range of connected-home commands, such as turning lights on or off, locking and unlocking doors, and altering thermostat temperatures.

Impacts/Disruptions

Many device manufacturers have already attempted to overcome compatibility issues by supporting multiple connectivity standards. However, providing such support often represents additional development and manufacturing costs. A common and universal standard, such as CHIP, could enable connected-home-device manufacturers to reach a larger target market while also simplifying development and reducing costs. Further, the fact that CHIP is an open and royalty-free standard will likely mean additional cost reductions for device manufacturers. CHIP will also limit the ability of dominant market leaders to build exclusive ecosystems and exert monopolistic influence over device manufacturers—a direction the connected-home market at times has appeared to be heading in.

Scale of Impact

  • Low
  • Medium
  • High
The scale of impact for this topic is: High

Time of Impact

  • Now
  • 5 Years
  • 10 Years
  • 15 Years
The time of impact for this topic is: Now to 5 Years

Opportunities in the following industry areas:

Home-network equipment, lighting, security systems, heating, entertainment

Relevant to the following Explorer Technology Areas:

On-Device Processing Advances

Why is this topic significant?

Advances in artificial intelligence and machine learning enable even small devices to carry out sophisticated computation locally. Such on-device processing can potentially reduce system latency, increase privacy, and improve security within the connected home.

Description

Many connected-home technologies rely on artificial-intelligence processing from cloud servers. When a user issues a voice command to his or her smart assistant to turn on the lights, for example, the raw audio is typically transmitted to a remote server where artificial-intelligence algorithms interpret the meaning of the utterance and instruct the local device about the correct action to take.

At the Google I/O developer conference in May 2019, however, Google announced improvements to its virtual assistant that will enable it to operate using only the local hardware available to new Pixel smartphones. According to a blog post from Google, engineers at the company built on advances in neural networks to develop new language-understanding tools that require significantly less storage capacity. Previous language-processing models required 100 gigabytes of storage, but these new lightweight models apparently require less than 0.5 gigabyte of storage, enabling them to run on portable devices, such as smartphones. Facebook also uses on-device processing for the smart-camera and voice-boosting features of its Portal smart displays, as "Facebook's Second Generation of Smart Displays" in the October 2019 Viewpoints describes.

Implications

Although offloading processing tasks to remote servers enables more powerful computation, local processing functionality is advantageous for several reasons. First, on-device processing—in comparison with remote processing—can significantly reduce latency between input and output, because local processing does not require data transmission to and from servers hundreds of miles away. According to the above-mentioned Google blog, local processing reduces the time between making a request and receiving an answer by a factor of ten. A second benefit of local processing—and one that Facebook emphasizes for its Portal devices—is increased privacy. Instead of sending data such as voice requests or images from cameras to remote servers where companies can potentially misuse them, local processing keeps data within the user's home and within the user's control. Third, greater local processing abilities make connected homes more robust by reducing dependence on internet connectivity. Local processing enables devices to work offline, which is particularly important for security and safety features such as locks, alarms, and security cameras.

Impacts/Disruptions

Although Google announced on-device language processing as a feature of new Pixel phones, Google will likely implement the same technology in its next generation of smart speakers, too. If the software works as it should, it could pressure Google's competitors—Apple and Amazon—to develop similar technologies.

Beyond language processing, further advances in on-device processing could enable new features within the connected home. For example, cameras with advanced on-device processing could recognize individual users and automatically tailor connected-device outputs to their particular tastes. Although such functionality is already possible with remote processing, privacy concerns are likely to hinder adoption of such technologies. Devices that can recognize individuals without remote processing, however, may overcome many of these concerns.

Scale of Impact

  • Low
  • Medium
  • High
The scale of impact for this topic is: Medium

Time of Impact

  • Now
  • 5 Years
  • 10 Years
  • 15 Years
The time of impact for this topic is: Now to 5 Years

Opportunities in the following industry areas:

Artificial intelligence, software development, hardware development, voice assistants, security

Relevant to the following Explorer Technology Areas: