IEEE Consumer Electronics Magazine - January/February 2024 - 19

j keeps on trusting in i since its threshold is set
really low (0.1), which illustrates a malicious
device that easily trusts in any device so as to be
able to communicate with anyone.
During the operation of the network, it is necessary
to have mechanisms that allow the
adjustment of trust values. Hence, this is done
through the relative entropy component, which
adjusts the trust based on traffic behavior. Thus,
behaviors that deviate from the expected contribute
to the variation of the entropy, which
consequently impacts the trust value.
OPEN RESEARCH CHALLENGES
With human beings, the concept of trust is
related to the behaviors perceived during interactions
and experiences with other people (either in
the present or in the past). It is also related to the
behaviors and attitudes of the person itself who is
building trust in others. Thus, cultural aspects and
the environment have an influence on people's
behaviors, consequently influencing trust placed
in others when an interaction occurs.
Different from the human-centric concept, in
the IoT context, building trust from one device to
another may not consider the behavior and
actions of the device itself which is building trust,
but only information from other devices. For
instance, a compromised device will not consider
its malicious history to compute trust in another
device. In fact, it will always set its trust in others
as high as possible, given that it wants to transmit
to whichever device it can. The measure of trust
rules whether a device will accept the connection
from other devices or not. So, the trust for IoT
should be built only upon information from other
devices and not upon information from the
device itself that is building trust. Trust is asymmetric,
context-dependent, dynamic, and not
necessarily transitive, hence establishing trust
between IoT devices is challenging, as its definition
is also related to fundamental aspects, such
as belief, integrity, reliability, and verification,
among others6.
Based on previously presented security aspects
on consumer IoT electronics communication,
we bring some challenges and opportunities associated
with such context. In particular, trust is a
fundamental aspect to provide consumer IoT systems
with security,11,12 since the entities of such
January/February 2024
systems are inherently widely distributed and
dependent on reliable data. Many open issues are
still present on the development of consumer IoT
applications in terms of trust establishment,
mostly due to the characteristics of the respective
applications. For instance, the IoT feature of heterogeneity
in terms of device types, transmission
technologies, and systems involved, poses the
challenge to manage such features and make the
devices interoperate in a trustful manner. With a
huge amount of different technologies involved,
the challenge is to provide a comprehensive and
common trust metric that could be used by all different
devices in a consumer IoT application.
In addition, the scalability factor, with the
huge and continuously increasing amount of
connected devices, also contributes to hampering
trust establishment, which should consider
local and global trust metrics. Distributed solutions
tend to be more appropriate to inherently
distributed IoT systems. In sum, the integrity of
data and devices' identity assurance are still
challenging, with no convergence for any standard,
despite existing proposals (see the
" Related Work " and " Trust Modeling " sections).
Hence, we provided a trust definition suggestion
through our two-level approach trying to fill the
commented gaps and advance the development
of consumer IoT systems. We also encourage the
development of trust proposals that can deal
with the discussed challenges found in consumer
IoT applications.
CONCLUDING REMARKS AND
FUTURE WORK
In general, security aspects are not considered
from the beginning of the development of IoT consumer
systems. In particular, trust aspects are fundamental
to provide IoT devices with security, in
which such devices can trustfully establish communication.
We discuss strategies to assigning
trust in communication between IoT devices considering
the model presented in Macedo et al.'s
work.14 Nevertheless, we highlight some issues
that are still open and that are essential for the
development of trustworthy consumer IoT
systems.
A possible limitation of our approach is when
a device enforces a specific traffic behavior to
increase its reputation before others.
19
IEEE Consumer Electronics Magazine - January/February 2024

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2024
