IEEE Systems, Man and Cybernetics Magazine - July 2023 - 39

are transferred in the given duration and the energy
consumption of all of the deployed nodes is equal. The
system cost of such a scenario may be reduced by using
different types of EDs based on the requirement of the
scenario. We address the following problem in this
work: How does one design an LCDT system to transmit
the huge size of sensory data of the smart building with
given energy and delay constraints? To solve this problem,
we present an LCDT system for smart buildings in a
noisy environment. The solution uses DL techniques for
the compression and effective transmission of sensory
data. The system uses the LoRa communication protocol
to transfer the compressed smart building data to
the EUs. Along with this, the key
contributions are as follows:
◆ We propose a compression-
decompression approach called
transmitter- and receiver-nets
for lowering the amount of
sensory data at the ED. The
approach employs deep neural
network (DNN) architectures
for compressing and decompressing
the sensory data. The
DNN designed for compressing
the data is lightweight and can
successfully run on low-processing
EDs.
◆ We employ a mixed-density network
architecture for the channel-net [21] to reduce
the noise effects between EDs and the LoRa gateway
(LG). The channel-net works on EDs after reducing
the size of the sensory data by using the proposed
compression DNN architecture.
◆ We present the analysis of the delay and energy required
for sensory data compression and communication. The
analysis considers the different types of devices with
unequal processing, energy, and storage capabilities.
◆ An optimization problem is formulated to minimize
the cost and energy consumption of the data transmission
system of the smart building. We also present a
low-time-complexity algorithm to solve the optimization
problem.
◆ Finally, the experimental results are presented to illustrate
the solution's effectiveness. The experiment's
parameters are defined based on the analysis of existing
hardware to make it practical.
The system uses
the long-range
communication
protocol to conserve
energy and enable
long-distance
communication.
The LCDT System
The LCDT system architecture
consists of SNs, EDs attached
with a LoRa node, an LG, a network
server (NS), an application
server (AS), and EUs, as shown in
Figure 1. The SNs attached with
the smart building collect the sensory
data in the form of the MTS
and forward it to the ED. The ED
is responsible for compressing the
received MTS and transmitting to
the LG. The LG receives the compressed
MTS and forwards the
same to the NS. The compressed
MTS is retrieved to the original form at the NS and forwarded
to the AS. The AS identifies the data and forwards
them to the respective user based on the
application. Finally, the EU receives the information collected
by the SNs.
Smart Building
With Sensor Nodes
Transmitter-Net
Channel-Net
Edge Device (LoRa Node)
Receiver-Net
LoRa
Gateway
Transmitter-Net
LoRa Communication
Channel-Net
Non-LoRa Communication
Figure 1. An illustration of the LCDT system components for smart building using LoRa. The transmitter-net and
receiver-net are the mirror image of DNNs.
July 2023 IEEE SYSTEMS, MAN, & CYBERNETICS MAGAZINE 39
Network Server and
Application Server
End Users

IEEE Systems, Man and Cybernetics Magazine - July 2023

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2023