Page 94 - ITUJournal Future and evolving technologies Volume 2 (2021), Issue 1
P. 94
ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 1
is [12] in which the authors used parallel ilters with only short‑frame OFDM communication system where sen‑
one Hamming code as an error correcting code. Their sors’ devices send only one OFDM symbol per frame on
scheme allows more ef icient protection when the num‑ the PLC channel. Each time slot is composed of an OFDM
ber of parallel ilters is large. symbol which represents a communication between the
Our contributions can be summarized as follows: master device and each slave‑sensor node device.
In Table 1, we denote our short‑frame OFDM parameters
• The design of this low‑powered short‑frame OFDM that are considered later in our MATLAB‑Simulink simu‑
communication system in terms of error correction lations:
is modelled by MATLAB‑Simulink tools.
Table 1 – Scenarios and short‑frame OFDM parameters used in our
• We analyse and choose the adapted Error Correcting MATLAB‑Simulink simulations
Codes (ECCs); such as Hamming and Reed‑Solomon
for this low‑powered short‑frame OFDM communi‑ Parameters Value
cation model.
: FFT size 64
• We discuss the trade‑off between the low implemen‑ : Used sub‑carriers 30/32
tation complexity and the high error correction ca‑ : Sampling frequency 1.6 MHz
pacity. Δ : Sub‑carrier spacing 25 KHz
: Total Symbol dura‑
• Moreover, we propose a new model of parallel Ham‑ tion 45
ming coding in order to increase the error correction : Modulation size 4 (QPSK)
capability of our model and we illustrate its perfor‑ : Packets of L bits 50 bits
mance in terms of Bit Error Rate (BER) vs. EB/N0. : Error correcting Hamming, Reed‑Solomon
codes
• Finally, we validate the performance of parallel Ham‑ : Coding rates = (4/7), = (11/15),
1
2
ming encoder/decoder in terms of complexity of im‑ = (23/31) , = (26/31),
3
4
plementation on an FPGA board using VHDL simula‑ = (57/63), = (56/64)
6
5
6
tions. : BER Perfor‑ = 10 ;
mance Analyser GUI
The remainder of this paper is organized as follows: First, = 10 10 ;
= 0 ∶ 15 dB
a brief description of the communication model in Sec‑ 0
tion 2. Then, we study the trade‑off between the two ECCs
(Hamming code and Reed‑Solomon code) which are more In this paper, we use MATLAB‑Simulink tools in order to
adapted to our communication model. We propose a new model this low‑powered short‑frame OFDM communica‑
design of a parallel Hamming coding in the case of a short‑ tion system as shown in Fig. 2.
frame OFDM sensor network in Section 3. Moreover, we
illustratethe performanceofthese ECCsin termsof Bit Er‑
ror Rate (BER) for different scenarios using BERTool ap‑
plication in Section 4. In Section 5, we validate the per‑
formance of parallel Hamming encoder/decoder in terms
of complexity of implementation on an FPGA board using
VHDL simulations. Finally, we conclude this paper in Sec‑
tion 6.
2. DESCRIPTION OF THE COMMUNICATION Fig. 2 – Short‑frame OFDM communication model using MATLAB‑
Simulink tools: General model
MODEL
Next, we will describe each block represented in Fig.
Here, we consider a sensor’s network which is composed 2 representing our general low‑powered short‑frame
of one master device and several slave‑sensor node de‑ OFDM communication model:
vices ( , , ..., ) using Time Division Multiple Access
1
20
2
(TDMA) to share the PLC channel illustrated in Fig. 1.
2.1 Random input generation
In this block, the input data is randomized by a ”Bernouilli
Binary” block in order to spread the energy over all the
bits before being encoded by the following block.
2.2 Error correcting code (ECC)
Fig. 1 – Shared bus architecture for our sensor’s network
The purpose of this block is to add enough redundancy
Since low‑power consumption is a critical factor in the to the data packets being sent, so that even if some of the
design of sensor’s network, we consider a low‑powered received data includes errors, there will be enough infor‑
78 © International Telecommunication Union, 2021