IEEE Consumer Electronics Magazine - November/December 2020 - 59

packets are not received; and 3) in case of the primary coordinator failure, the backup coordinator
takes on the role of primary coordinator and
restore network functionality using the backed up
network information.

Multiple Backup Coordinator Methods
Replacement methods presented in the literature,11; 12 are based on multiple backup coordinators. These methods work on the following
principles: 1) each backup coordinator maintains
a live data connection with the primary coordinator; 2) the primary coordinator's failure is
inferred if ACKs for the transmitted packets are
not received; and 3) in case of the primary coordinator failure one of the backup coordinator is
selected as the primary coordinator (usually
backup coordinator's priority is used to select a
new primary coordinator).
Where Do Current Methods Fail
The state-of-the-art solutions to replace a network coordinator fail in a scenario where a user
wishes to change the HAS service provider. Suppose that a customer is using a HAS from company A, and the company uses one of the
backup coordinator schemes to replace the primary network coordinator in case that one fails.
However, due to some reason, a user needs to
switch to another HAS provided by company B.
In order to use Company B's HAS, the user
installs a new network coordinator from company B. As company A's coordinator shares its
network information with its own company's
backup coordinator, therefore, company B's
coordinator cannot obtain data from company
A's coordinator. This results in vendor lock-in.
Here, the network needs to be recommissioned,
which tends to be road-block for most users to
change to a new or more innovative technology.
Along with interoperability with HAS networking
protocols,7 this is one of the main roadblocks for
innovation in this space.
Existing network coordinators for HAS come
from a variety of companies, such as Amazon,
Google, Philips, Nest, and Samsung. These companies' systems do not have a backup mechanism to replace a faulty network coordinator
without requiring a network recommissioning.
For example, an existing HAS uses a Philips

November/December 2020

Figure 2. HAS architecture based on our HAS
proxy.

coordinator, if the user wants to replace the Philips coordinator with a coordinator that comes
from Nest, the user needs to recommission the
network using the Nest coordinator, even if Philips's HAS uses a coordinator backup method
(although this feature is currently not available
in any commercial product). In the following, we
present a proxy HAS that can eliminate the
stated problem.

HOME AUTOMATION SYSTEM PROXY

In this section, we present our concept for a
HAS proxy that can eliminate the stated issues
with network recommissioning. In general, the
following are distinguishing features of our
solution.
-

Ability to form a wireless personal area network (WPAN), and on-boards smart home
devices onto a network coordinator.
One-touch network recommissioning.
Incorporates method to replace faulty HAS
proxy.
It works with any existing networking protocol stacks for smart homes, such as ZigBee,
Z-Wave, Thread, and Bluetooth.

Figure 2 shows a HAS network architecture
based on our HAS proxy. Our HAS proxy is a
software component that can easily run over a
relatively low-end computing device, such as,
Raspberry Pi. To run our HAS proxy, a computing device should support relevant WPAN radio
and networking protocol stack. For example, for
a ZigBee-based HAS, it must include IEEE
802.15.4 radio support and implementation of

IEEE Consumer Electronics Magazine - November/December 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - November/December 2020