IEEE Consumer Electronics Magazine - January/February 2020 - 17

Figure 1. Evolution of different home appliances.

This work, therefore, proposes an intelligent
home appliance system. The proposed system is
designed as a vision-based interaction system.
For data privacy purposes, the proposed system
conveys basic security messages, without sending pictures or video to server. The proposed
system is then able to prevent accidents or
potentially dangerous situations involving home
appliances, allowing parents to ensure the safety
of their children at all times. In addition, many
advanced techniques are used, such as intelligent
vision processing and the automatic identification of events. To make a real-time system, an
embedded system is designed which includes
Raspberry Pi with Bluetooth 4.0 transmission in
order to interact with remote users.

BACKGROUND
Network Topologies in IoT
Communication bandwidth is always a bottleneck in IoT scenarios. ZigBee is a useful protocol
for data transmission.1 In ZigBee, the indoor
line-of-sight transmission distance can be about
10-100 meters and maximum frequency up to
250 kbps. In the paper by Zhang et al.,2 it can manage and control the home furnishing centralized with a short-range communication protocol.
Bluetooth protocol was proposed with the purpose of higher frequency and transmission distance. It is a high performance and stable
protocol, compared with ZigBee. Since its power
consumption is not high, it is widely used in cell
phones, automotive, and home appliances.
On the other hand, multimedia, such as audio,
image, and video, is very useful and should be
adopted in IoT applications. However, this will
give rise to a bandwidth issue since ZigBee and

January/February 2020

Bluetooth are not able to transmit large-bandwidth data. If users want to transmit multimedia
data, traditional wireless communication techniques are used, such as Wi-Fi. In Wi-Fi, frequency
can be up to 150 Mbps, which can transmit video
easily. With the popularity of smartphones, Wi-Fi
has become ubiquitous in recent years. However,
it was not built for IoT environments with implementation complexity and cost concerns.

Sensors
In IoT-based scenarios, sensors play significant roles, along with other components in systems. Some conventional applications utilize
such sensors as light detecting sensor, smoke
detecting sensor, and gyro sensor. Light sensors
detect the presence of a person in a room and
control light switching accordingly. Some human
movements and hand gesture recognition can be
detected by gyro sensors, which can tell the status and position of human movement. Different
sensors can communicate with each other to
predict what next step may occur and what
appropriate action to take.3
Vision-Based Techniques
Vision-based sensors in IoT are indispensable in a smart home scenario. Technological
advancements have meant that vision-based sensors can automatically monitor various areas and
identify potentially dangerous situations. This
surveillance system has undergone significant
development since its production and was put
into practical application in the last decade.
Infrared (IR) sensors have also been combined with some video techniques. One of the
most well-known applications is Kinect. This
powerful platform can be applied in many

IEEE Consumer Electronics Magazine - January/February 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2020