See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/337794180

Performance Evaluation of E32 Long Range Radio Frequency 915 MHz based on
Internet of Things and Micro Sensors Data

Article  in  International Journal of Advanced Computer Science and Applications · January 2019

DOI: 10.14569/IJACSA.2019.0101106

CITATIONS READS

15 5,114

2 authors:

Puput Dani Prasetyo Adi Akio Kitagawa

BRIN Kanazawa University
38 PUBLICATIONS   286 CITATIONS    94 PUBLICATIONS   772 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Real-time and continuous output power monitoring of photovoltaic (PV) systems View project

Analysis performance of triangle microstrip antenna for basic construction of circularly polarized-synthetic aperture radar application View project

All content following this page was uploaded by Puput Dani Prasetyo Adi on 18 December 2019.

The user has requested enhancement of the downloaded file.

 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

Performance Evaluation of E32 Long Range Radio

Frequency 915 MHz based on Internet of
Things and Micro Sensors Data
Puput Dani Prasetyo Adi1, Akio Kitagawa2
Micro Electronics Research Laboratory (MeRL)
Kanazawa University, Kanazawa
Ishikawa, Japan

Abstract—This research discusses how to build and analyze a Thingspeak. Fig. 1 shows the comparison between Frequency
915 MHz Long Range (LoRa) E32 Frequency-based Node Sensor radio devices e.g, WiFi, Bluetooth, Bluetooth LE, Cellular and
network with a Micro Sensor with 3 sensor outputs produced i.e, LoRaWAN. Using Cellular and GSM technology e.g., in
Temperature (DegC), Air Pressure (hPa), and Humidity (%). research [5] studies on how to detect water distances using
therefore, This research succeeded in making a sensor node using Ultrasonic sensors and GSM SIM 900A. Furthermore, still, on
the LoRa E32 915 MHz using a mini type ATmega 328p the application of Radio Frequency, namely in research [6],
microcontroller with a 3.7 volt, 1000 mAh battery. The display LoRa is used in Street Light monitoring.
on the receiver uses an 8X2 LCD which will output 3 sensor data
outputs. furthermore, the result and analysis of this research are This comparison uses two parameters, i.e, the ratio of
how to analysis of the LoRa Chirp Signal, furthermore, LoRa Power Consumption to distance. With a distance of> = 15 km
Chirp Signal obtained from the Textronix Spectrum analyzer in on Free Space, placing LoRa or LoRaWAN in the best
realtime, Quality of Service (QoS), Receive Signal Strength ranking compared to other devices. LoRaWAN can be paired
Indicator (RSSI) (-dBm), uplink and downlink data on the with Cellular Devices in terms of Range but Cellular cannot
Internet Server. Furthermore, The Micro Sensor Graph Output match LoRaWAN in terms of Power Consumption. Therefore,
will be displayed on the application server with a sensor data LoRa and LoRaWAN only use battery power sources with the
graph. In this research Application Server used is Thingspeak ability to survive> = 2 years. nevertheless, LoRaWAN is the
from Mathworks. most ideal device that can be used as a sensor node based on
Keywords—Long range; microcontroller; internet of things;
Low Power and with the best data transmission range.
quality of service; micro sensor LoRa can> 30 km, in the condition of Free Space, of
course, with the right antenna position and an adequate
I. INTRODUCTION transmitter Antenna height. In research [7], packet loss is
The development of the Internet of Thing is currently calculated for different coverage of SF12 use in indoor and
growing rapidly, Sensor Nodes are developed with very low outdoor conditions. In research [8] very detailed in testing the
power consumption with long node durability, supported by reliability of LoRa signal and LoRa transmission data (bit rate,
the right environment and the right battery usage, in research data rate), Receive Signal Strength Indicator (RSSI), Error Bit
[1] and [2], sensor nodes are used in delivery of Internet-based Rate (EBR (%)) and EPR (Error Packet Rate (%) ), Packet
Pulse Sensors and Blood Pressure sensors with IoT devices Delivery Ratio (PDR(%) juga dilakukan menggunakan
with ZigBee sending devices. The advantages of ZigBee are perbandingan simulasi menggunakan LoRaSIM [9].
Low Power, with a data rate of 250 kbps, but weak in the
distance (m), a maximum of 120 meters on a regular type
ZigBee, and 1 km on a ZigBee Pro. So the development is
done by changing the radio frequency device with Long
Range (LoRa). LoRa center frequency e.g, 433 MHz, 868
MHz, and 915 MHz. for Asia, e.g, Japan and Indonesia use
Frequency Long Range (LoRa) 920-923 MHz (Center 915
MHz). The type of signal produced by LoRa is Chirp Signal as
in research [3]. In research [4], using Bluetooth type RN-42 in
sending LM35 temperature sensor data, and using Raspberry
Pi 3 Model B, the resulting analysis is Quality of Services
(QoS) on Bluetooth RN42 Master-Slave communication and
data management process on Raspberry Pi 3 using Python and
MySQL or MariaDB databases. Research [2] was developed
with subsequent research using GUI (Graphical User
Interface) using JavaScript Object Notation (JASON) and Fig. 1. Radio Frequency Devices with Power Consumption and Range
Application Server, e.g, The Things Networks (TTN) and Comparison.

38 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

II. RELATED WORKS The signal in Fig. 1 can be formulated with equation 1,
Research [10] analyzed the Parameter Packet Delivery while for Fig. 3 it can be expressed with equation 2. If it is
Ratio (PDR (%)) from direct measurements using Locations formulated with general function the signal for signal density
A, B, C, D, and E., In theory, these locations show a is like equation 3.
Spreading Factor (SF) parameter of 7 to 7 12. If the greater the
Spreading Factor, the smaller the Packet Delivery Ratio (PDR ( (( ) )) (1)
(%)). The closest distance is A is 650 m and the farthest E is
3400 m. furthermore, the analysis was also carried out by ( (( ) )) (2)
making a comparison of average throughput (bytes / s) with
Spreading Factor (SF). The greater the SF value, the smaller
the value of Throughput (bytes / s). In SF 7 the average value ( ) , *(( ) ) +- (3)
of 3 payloads is 51 bytes, 25 bytes and 1 byte is 3.96 bytes / s
and in Spreading Factor 12, from 3 payloads of 51 bytes, 25 while Phase Shift Keying (PSK) is a signal modulation
bytes, and 1 byte, the average throughput is 0.22 bytes / s. based on signal shifting, not high or low amplitude or signal
density. Furthermore, the value of signal modulation is 1 and
Research [3] conducted a detailed analysis of Chirp Signal 0, digital value 1 indicates ON and digital 0 states OFF. of
or CSS. chirp Signal Analysis, including Up chirp and down several types of signal modulation, LoRa uses the Chirp type
chirp. This research emphasizes more on the fundamental or called the Chirp Spread Spectrum (CSS), there are two
aspects of LoRa i.e. parameters, demodulation process, and types of Chirp namely Frequency Increases (Up-Chirp) and
Decoding Process. Furthermore, the CSS Packet Spectogram Frequency decrease (Down-Chirp) with time, as shown in
comparison between Frequency LoRa (125 kHz, 250 kHz, and Fig. 4.
500 kHz) against Time (ms) with Spreading Factor (SF)
differs from 7 to 12. And the results of the simulation are the
comparison between SF and SNR (-dB).
Research [11] analyzed in detail the Signal Interference
Ratio (SIR) by comparing the detailed Spreading Factor (SF)
Parameters e.g., Co-Spreading Factor Interference. LoRa
parameters are obtained with a detailed mathematical
approach that results in the Probability of Success (Ps) with
parameters e.g., P SNR, PSIR, and Simulation with a Comparison
of distances (km). the research compared the value of R = 6
km and R = 12 km.
III. METHODOLOGY AND DEVICES USED
A. Chirp LoRa Signal
Compressed High-Intensity Radar Pulse (chirp) is signal
modulation technique has long been used in the world of
commercial and Military RADAR Systems. therefore, There
are 3 types of signal modulation, e.g. Amplitude Shift Keying Fig. 2. Chirp with t = 10.
(ASK), Frequency Shift Keying (FSK) and Phase Shift
Keying (PSK). furthermore, Amplitude Shift Keying (ASK) is
a modulation signal based on the change in the value of the
amplitude is high or the low wave signal is generated, if the
amplitude of high-value and low-amplitude digital 1 digital is
worth 0. Frequency Shift Keying (FSK) is a modulation signal
which is based on the change in frequency indicated by the
Pulse Length (t) value or the resulting signal density as shown
in Fig. 2 and Fig. 3. In Fig. 2, the parameters are as follows.
t=0:0.05:10;A=1;omega=2;s_t
= t.^2/4;y_t=A*cos(omega*t + s_t)
t is the Pulse Length in Fig. 2 using the value t = 10 and
Fig. 3 the value of t = 40.
Like Fig. 2, Fig. 3 has the following parameters
t=0:0.05:40; A=1; omega=2; s_t = t.^2/4; y_t=A*cos(omega*t
+ s_t); Fig. 3. Chirp with t = 40.

39 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

The formula for calculating Atmospheric Pressure is in

accordance with the formula 2. Log10 P ≈ 5 - (h -15500),
where the value of P is Pressure (Pascal) and h is the height
(meters) Furthermore, the temperature has the default unit
Degree Celcius. the BME280 sensor, the unit used is Degree
Celsius and detects conditions in the environment and results
in precision. furthermore, to change to Kelvin (K), Reamur
(R) or Fahrenheit (F) C language is used in the Arduino IDE.
If the DegC value is known, to be changed to another
temperature unit to r = c * 4/5; f = (c * 9/5) +32; k = c +
273.16; furthermore, Humidity (%) is the amount of water
Fig. 4. Up and Down-Chirp Signal LoRa. vapor in the air that is not visible to the human eye, this
amount of water vapor will determine rainfall, dew or fog.
Furthermore, Fig. 6 describes the BME280 sensor diagram
block, There is a Pressure and Temperature Sensing element
that is converted from Analog to Digital using Analog to
Digital Converter (ADC) to Logic Gate and continues to 6
Output pins i.e.SDI, SDO, SCK, CSB, VDD and GND, and
just 4 pins used i.e. SDA, SCL, VDD (3.3 Volt) and GND.
Fig. 5. LoRa Signal with Encoded (data) on.
This sensor will be processed by the MCU (ATmega 328)
Fig. 5 contains 2 signal parts i.e., up and down chirp signal with BME280 pins on the Arduino Pro mini such as Fig. 3. In
on the LoRa signal prefix, consisting of 11 up chirp and 2 Fig. 3, three main components of a sensor node, i.e. MCU
down chirp. Next is the encoded signal. Fig. 8 is the realtime ATmega 328, FTDI and LoRa. The LoRa type used is SX1276
LoRa E32 signal that is the Chrip signal captured and a Universal Asynchronous Receiver Transmitter (UART) E32
analyzed by the Textronix Spectrum analyzer. 915T20D, this type of LoRa has the capability of up to 3 km.
The LoRa SX1276 UART E32 915T20D uses a default
In research [12], a decoding technique for LoRa signals frequency of 915 MHz (920-928 MHz), this Radio Frequency
was used using a decoding algorithm with two slightly can be used in Japan (920-928 MHz). A block diagram of the
desynchronized superposed LoRa Signals. furthermore, this BME280 can be seen in Fig. 7.
decoding Algorithm technique is used to increase the
effectiveness of sending data e.g, Throughput.
B. BME280 Sensor
The BME280 Sensor is a multi-sensory and applied in the
Health Care Application (e.g. Spirometry). in previous
research, Health Care Application using different sensors in
research [1], [2] i.e. Blood Pressure and Pulse Sensor Internet
of Things (IoT) based. BME280 sensor has a three types of
Sensors, i.e. Atmospheric Pressure (hPa), Temperature (DegC)
and Humidity (%). Atmospheric Pressure (hPa) is pressure at
any point in the Earth's atmosphere. Almost the same as the
hydrostatic pressure caused by the weight of the air above the
measurement point. Table I is the spesifications of the BME
280 Sensor. Fig. 6. Block Diagram of BME280 Sensor.

TABLE. I. BME 280 SPESIFICATION

No Parameter Details
1 Dimension 2.5 mm x 2.5 mm x 0.93 mm metal Lid LGA
I2C (up to 3.4 MHz) and SPI (3 and 4 wire, up
2 Digital Interface
to 10 MHz)
VDD main supply voltage range 1.7 V to 3.6 V
3 Supply Voltage
VDDIO Interface voltage range : 1.2 V to 3.6 V
1.8 µA @ 1 Hz humidity and temperature
2.8 µA @ 1 Hz Pressure and temperature
Current
4 3.6 µA @ 1 Hz humidity, Pressure and
consumption
temperature
0.1 µA in sleep mode
-40 …+85o C, 100% rel.humidity, 300..1100
5 Operating range
hPa Fig. 7. Block Diagram of BME280 Sensor on MCU ATmega 328.

40 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

Battery Life Factor is the most important factor in the all the STAs in the Network. In research [14] conducted
LoRa or LoRaWAN network architecture, in research [13] research transmission data using LoRa to find out energy
reviewed how to make energy consumption on Sensor nodes efficiency over very long distance [15] by building
in LoRa or LoRaWAN networks can survive well with the communication 6 LoRa nodes which can cover 1.5 ha of
approach in Sleep mode. In addition to the Sleep mode network with sensor node specifications ie, 2 AA type
approach, the effectiveness of the sensor node and the batteries, data transmission every 5 seconds and level 80%
LoRaWAN network with a multi-node or multi hope is to use reliability. The Lora E32 or SX1276 / SX1278 Wireless
a method called the offset-CT Method, as in research [14]. Module has variants or versions, one of E32 915T20D LoRa
type shown in Fig. 5 and this type is used in this research,
The calculation approach to the node usage on the LoRa E32 868T20D (Permit at Japan 920-928 MHz ) this
Receiver is as follows, as in references [1], the calculation of Frequency radio is maximum 931 MHz, moreover the
the Power Consumption of Sensor node is based on the total E32915T20D uses a default frequency of 915 MHz, while the
load calculation component used, the Battery life calculation E32868T20D uses an 868 MHz frequency. e.g, for the
is in accordance with equation 4, furthermore, 0.7 is a value of Japanese region using E32915T20D, which works in the range
external factors which can affect battery life. A sensor nodes 920-928 MHz, therefore, the maximum frequency for
(Tx and Rx) can be seen in Fig. 8. Furthermore, The type of E32868T20D module is 930 MHz.
battery used in the sensor node is the 3.7 Volt 1000 mAh
Lithium Polymer Li-Po Rechargeable Battery as shown in The dimensions of the LoRa E32 915T20D Wireless
Fig. 9. Module have dimensions of 21x36 mm in Fig. 11. the voltage
used is 3.3 Volt DC (Tipycal), Transmit Power (dBm) is 20
1) Measurement of battery life (H) of receiver node: In dBm in Fig. 12, more complete can be seen in Table II.
measurements using the Ampere meter in Fig.10, the total
( )
value of the Power Consumption mote or node sensor on the [ ] (4)
( )
receiver (Rx) with LoRa E32 is 22 mA. so that the Battery Life
mote according to equation 4 is 1000 mAH / 22 mA = 45.45 x
0.7 = 31.8 hours. accordingly, This formula is the same as
finding power (P (watts) = V (Volts) x I (Amperes) = Psensor
= 3.7 x 0.022 = 0.0814 watts, so Pbattery = 3.7 x 1 Amperes =
3.7 watts, so the Power Required is 3.7 / 0.0814 = 45.45 x 0.7
= 31.8 Hours.
2) Measurement of battery life (H) of transmitter node: On
the transmitter, the total current sensor node is 119.5 mA, then
Battery Life mote = 1000 mAH / 119.5 mA = 8.36 x 0.7 = 5.8
hours. This formula is the same as finding power (P (watts) =
V (Volts) x I (Amperes) = Psensor = 3.7 x 0.1195 = 0.44215
watts, so Pbattery = 3.7 x 1 Amperes = 3.7 watts, so the Power
Required is 3.7 / 0.44 = 8.4 x 0.7 = 5.8 Hours. Furthermore, to Fig. 8. DIY Transmitter and Receiver LoRa E32 915 MHz Node Sensor.
reduce the Battery life (H) Sensor node it is necessary to do a
strategy on the sensor node by changing the C language
programming on the MCU to Sleep mode. With Sleep mode or
Reset mode, Battery Life (H) can be longer, in Sleep mode it
can be set how long the sensor node will sleep, and when the
sensor node will turn on again and send a signal.
C. LoRa 915 MHz (920-928 MHz)
The RF signal used in this research is two types, i.e. Fig. 9. Battery 3.7 Volt 1000 mAh Lithium Polymer Li-Po Battery
Dragino LoRa 915 MHz and LoRa SX1276. LoRa SX1276 Rechargeable.
uses the default frequency of 915 MHz (Japan: 920-928
MHz), this type of LoRa is intended for Arduino Uno Shield
and Arduino Mega. Complete data about the specifications of
the LoRa SX1276 915 MHz can be seen in Table I.
Furthermore, this research is divided into two analyzes, i.e. the
measurement of transmission devices (Transmitter and
Receiver) to get the RSSI (-dBm) value and analysis of the
915 MHz Internet Gateway e.g. uplink and downlink of LoRa
915 MHz Internet gateway. In research [8] the Framework of
the Distance-Ring Exponential Stations Generator (DRESG)
was introduced in handling and providing Multi-hop solutions,
e.g, Routing nodes, energy efficiency and consumption among Fig. 10. Total Load Measurement of Current (mA) on Rx dan Tx Sensor
Node.

41 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

In general, Lora type E32915T20D has 7 core pins that are

used in the process of making sensor nodes, namely M0, M1,
RXD, TXD, AUX, Vcc and GND, in full about pins and
functions explained in Table III. There are 3 modes in sending
sensor data to Pins M0 and M1 are Wake up mode (mode 1),
if M1 and M0 = 1, then Power-saving mode (mode 2) if M1 =
1 and M0 = 0 and Sleep mode if M1 = 1 and M0 = 1.
A Zigbee using PAN ID (Personal Address Network)
Identifiers, e.g, 00 03 04 AA BB CC (00 03 = Address, and 04
= Channel) will send data to other LoRa Wireless Modules,
the receiver (Rx) LoRa Wireless Module have the same
address and channel, e.g, 00 03 04, if address 00 05 04,
message will not be sent or 00 07 04 (different address even
though the channel is the same, the message will not be sent)
while Broadcast Mode, FFFF will send all addresses and
channels such as FF FF 04 AA BB CC can send data to 00 03
04 and 00 05 04 (Different addresses but for channels must be
the same), e.g, channel 00 07 06, the data is not sent,
Fig. 11. LoRa E32 915T20D and Dimension. nevertheless the channel must be 4, same as Tx (Transmitter )
Lora Wireless Module. Writing the address module on the
broadcast is 0xFFFF or 0x0000 and channel 0x04, while for
Fix mode or point to point is 0x0003 or 0x0001.
In research [15], a study of the basic performance of LoRa,
LoRa Physical layer and comparison of Sigfox and LoRa in
terms of Bit Error Probability. In research [16] LoRa research
on Quality of Services (QoS) e.g, RSSI, SNR, Payload size
(bytes) and Spreading Factor. In research [17] LoRa
performance testing on image transmission, this is a new idea
Fig. 12. Circuit Diagram of LoRa E32 915T20D. because LoRa has the ability to transmit small data, in this
research the image transmission was successful, the camera
TABLE. II. LORA E32915T20D SPECIFICATION used has a SNR 45 dB specification and 38400 bps Boudrate,
Micro SD Card as saving data and LoRa shield with Arduino
No Parameter name Parameter detail
MEGA as a processor (MCU) and transmitter. In research [18]
Model Wireless
1
Module
E32 (915T20D) testing and study of the maximum or limit LoRaWAN
2 Frequency Min 900 MHz, Typ : 915 MHz : Max 931 MHz devices. LoRa test results in a comparison of Payload data of
3 Voltage Supply Min: 2.3 V DC, typ: 3.3 V DC, Min: 5.2 V DC
10, 30 and 50 bytes, and the maximum throughput of the
Min: 2.5 V DC, typ: 3.3 V DC, Min: 3.6 V DC
number of different end nodes, namely, 250, 500, 1000 and
Communication 5000 end devices.
4 If Communication level > Maximum Value,
Level
then Module may be damaged
5 TX Power Min :19.3 dBm, typ: 20 dBm, Max:20.6 dBm TABLE. III. LORA E32915T20D PINS
6 TX Current Min:110 mA, typ :120 mA, Max : 130 mA
Pin no. Pin Pin direction Application
7 RX Current Min:11 mA, typ : 14 mA, Max :15 mA
Input (weak Work with M1 and decide the four
Turn off Current 1 M0
8 Min:3 mA, typ : 4 μA Max :5 μA pull-up) operating modes
(Sleep mode)
9 Range Min: 2700 m, Typ : 3000 m, Max :3300 m Input (weak Work with M0 and decide the four
2 M1
10 Packing DIP pull-up) operating modes
11 Air Data rate (kbps) 2.4 kbps 3 RXD Input Connect to external TXD output pin
12 Antenna SMA-K 4 TXD Output Connect to external RXD input pin
Reception
13 -147 dbm 5 AUX Output To wake up the external MCU
Sensitivity
14 Air data rate 0.3 kbps – 19.2 kbps, default 2.4 kbps Voltage positive references of
6 VCC Input module; Power Supply 2.3 Volt –
15 Test distance 3000 m (3 km)
5.2 Volt DC
16 Antenna Gain 5 dBi
7 GND Input Ground
Communication UART, boud rate 1200 – 115200, default: 9600
17 Fixing
Interface bps 8 Fixing hole
hole
18 Transmitting Power 100 mW
Fixing
19 Parity 8N1 9 Fixing hole
hole
Default Parameter Fixing
20 C0 00 00 1A 0F 44 10 Fixing hole
Values hole

42 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

In research [10], the LoRa test on the Packet Delivery

Ratio of the LoRa field test in SF 7, 9, 12 at 5 different
locations. Also obtained a comparison of the average
Spreading Factor to Throughput (bytes/s). in research [19]
data generated from research on LoRa are AWGN Channel
and Rayleigh channel, Loss of Received signal Power to
maintain PER = 10-1 as in AWGN channel when the end-
devices velocity is 30 km / h.
D. Fresnel Zone Approach
In this research, transmitting data from the 915 Mhz LoRa
Transmitter to the 915 MHz LoRa Receiver with a long
distance (Km). this affects the geographical location of the
earth in the form of an ellipse. So a theory about the Fresnel
Zone is formed. Fresnel Zone is an elliptical form on the
Direct Line of Sight condition between the transmitter and
Receiver. furthermore, The formula for calculating radius in
the Fresnel Zone is as r=8.657 x √(D/f), r = Fresnel Zone
radius (m), D = distance (km), f = frequency (GHz). so that
from equation 1 a table of comparison of Fresnel Zone can be
made radius (m) at a certain distance and certain Frequency,
for example at a distance of 500 m or 0.5 Km and Radio Wave
Frequency used is 915 MHz or 0.915 GHz then the Fresnel
radius of Zonen r = 8,657 x √ (0.5 / 0.915) = 6.3994 meters,
this condition is illustrated in Fig. 5. Conditions when sending
BME280 sensor node data from the transmitter to the receiver
node which can be seen from the Fresnel Zone Clearance
condition. The Fresnel zone is the ellipse part that shows in
Fig. 5(a), outside the ellipse part is not the Fresnel Zone. For
this reason, the Fresnel Zone is influenced by the transmitting
antenna height (HT) and the receiving antenna height (HR).
analysis of data transmission between Transmitter (T x) and
Receiver (Rx) LoRa node is determined by the Fresnel zone to
obtain the correct delivery results, in addition to the influence
of the Fresnel zone, other influences, an obstacle (buildings,
trees) and interference e.g.weather.
There are four states of the Fresnel Zone described in
Fig. 13 and Fig. 14. Fig. 13(a) is the Fresnel Zone Equation Fig. 13. Conditions of Fresnel Zone Theory.
Fig. 13(b) is the Condition Transmitter and Receiver without
interference or no obstacle in the Fresnel Zone, Fig. 13(c) is TABLE. IV. RESULT OF FRESNEL ZONE APPROACH WITH H
CONSIDERATION
Condition Transmitter and Receiver with a reflected signal
from the ground surface and Fig. 13(d) is Conditional No Distance (km) Height (m)
Transmitter and Receiver with considering factor H, H is 1 0.1 Negligible
2 0.5 Negligible
heigh from curvature of the earth and to calculate H = 1000 x
3 1 Negligible
D ^ 2 / (8 x Rearth), accordingly the theory, H is Height (or 4 2 Negligible
earth curvature allowance in m), D is the distance between the 5 5 0.4
end node and the gateway in km, and Rearth is earth, the 6 10 1.5
radius in km equal 8504 km. In full about the effect of Height 7 15 3.3
(m) on the results of the calculation approach of the Fresnel 8 20 5.9
Zone can be seen in Table IV. Moreover, Table IV is a 9 25 9.2
comparison between the distance (km) to the height (m) of the 10 30 13.2
arch of the earth, and this is an approach that is not significant Furthermore, the Fresnel Zone area is influenced by Ht and
right, considering the different areas when transmitting data. Hr, the higher the value of Ht and Hr, the greater the
The essence of adding Fresnel Zone parameters is to take percentage of Fresnel zone clear, as shown in Fig. 14.
into account the earth's curvature which affects Reflection, Then from the formula r = 8,657 x √ (D / f), it changes to r
Diffraction and Scattering wave signal propagation, in this = 8,657 x √ ((0.7 x D) / f), this is caused by adding a Fresnel
case, the position of Transmitter is important, tx should be zone value of 70% or 0.7 x D.
placed at a position far higher than the Fresnel zone.
Therefore, Fresnel Zone is used at distances between 5 km to Therefore, Fig. 15 shows the Comparation of Fresnel Zone
30 km. Clear Percentage (%) with the Fresnel Zone approach with a

43 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

distance of 5 km produces a value of Fresnel Zone radius In Fig. 16, 433 MHz Frequency of LoRa at a distance of 5
100% clear as far as 639.9 m, Fresnel Zone 70% clear as far as km has a value of Fresnel Zone as far as 778.3 m, 868 MHz as
535.4 m, Fresnel Zone 60% as far as 495.6 m, and Fresnel far as 549.7 m and 915 MHz as far as 535.4 m. so it was
zone 50% as far as 452.5 m. Furthermore, from Fig. 16 it can concluded that by transmitting LoRa data with Low Frequency
be concluded that the farther the distance (m) between the and with the same Fresnel zone clear percentage condition
transmitter (Tx) to the LoRa (Rx) receiver, the greater the (70%) has the furthest Fresnel zone (r) value. Please note that
value of r (Fresnel zone), so that it will be more detailed if the the H value is calculated when the distance is km 5 km. so the
addition of data is done using a comparison of Frequency The value of r ≥ 5 km equals r + H, for example, r + 0.4 at a
different LoRa are 433 MHz and 868 MHz in Fig. 16. distance of 5 km and r + 1.5 at a distance of 10 km and so on
as shown in Table IV.
E. Network Concept
Fig. 17 and Fig. 18 is a concept on this research, end
devices/end nodes or mote consists of more than one sensor
node connected to the LoRa Internet Gateway. Therefore, this
LoRa Internet Gateway functions to store and capture LoRa
signals from end devices to be forwarded to the e.g compatible
Application Server, The Things Network or Thingspeak and
then forwarded to Devices connected to Internet devices. The
Fig. 14. Fresnel Zone 70% Clear. analysis stage is the data rate (byte/s) of the Uplink and
Downlink process, continuous data transmission by the sensor
node will make the battery run out quickly and this is
undesirable in the LPWAN method.
In Fig. 17, by only using 1 LoRa Internet Gateway, packet
data or data bytes will be a buildup of data bits from n-number
end-nodes that send data to one source, consequently, this
multiplexing method results in increasing the bit rate of Error
Ratio (BER (%)) or Packet Error Ratio (PER (%)). therefore,
we need an approach to the number of gateways that can
dynamically manage packet data (bytes/s) coming in from
end-node or end-device. End-node needs to use a sleep mode
approach, therefore, it didn't continuously send packet data
(bytes/s) without stopping causing bottleneck packet data
(bytes) at the gateway. The addition of the gateway an answer,
furthermore, the end node can select the destination gateway
furthermore, that there no packet data buildup resulting in a
reduction in throughput.

Fig. 15. Comparation of Fresnel Zone Clear Percentage (%).

Fig. 17. Simple TT Network Concept.

Fig. 16. Comparation of 70% Clear (30% Blockage) Fresnel Zone with
different LoRa Frequency (433, 868 and 915 MHz).

44 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

Fig. 20. Interrupt Process.

1. Initialization Sleep mode and LoRa Library

#include <RH_RF95.h>
#include <ThingSpeak.h>
#include <SPI.h>
#include <LoRa.h>
#include <avr/sleep.h>
#define interruptPin 2
2. Initialization Sensor Library
#include <Wire.h>
1. Initialization Boud Rate or Speed Data
Rate(bps)
void setup() {
Fig. 18. Adaptive Data Rate (ADR) Algorithm TT Network Concept. Serial.begin(115200);
2. Initialization Input and Output
So we need an approach using an algorithm that can
regulate the sending of data (byte/s) continuously by the Serial.println("LoRa Sender");
if (!LoRa.begin(915E6)) //915MHz Freq
sensor node or end node/mote to the gateway. The core of the {
ADR algorithm is that sensor nodes can dynamically switch to Serial.println("Starting LoRa failed!");
off or sleep mode so that power consumption can be reduced, }
therefore the lifetime on sensor nodes is getting longer. pinMode(LED_BUILTIN,OUTPUT);
Therefore, the Adaptive Data Rate (ADR) algorithm is used to pinMode(interruptPin, INPUT_PULLUP);
digitalWrite(LED_BUILTIN,HIGH);
optimize data rates, airtime and energy consumption, the ADR }
Flowchart in Fig. 19. 3. Determine the Sleep Time
void loop() {
F. Sleep Mode delay(10000);
Sleep mode is one way to manage LoRa data transmission Going_To_Sleep();
}
using MCU, in this research the Arduino Pro mini MCU type 4. Command to Sleep
3.3 v, 8 MHz. in general the structure of the Sleep mode can void Going_To_Sleep(){
be seen in Fig. 20. rf95.sleep();
sleep_enable();
attachInterrupt(0, wakeUp, LOW);
set_sleep_mode(SLEEP_MODE_PWR_DOWN);
digitalWrite(LED_BUILTIN, LOW);
delay(1000);
sleep_cpu();
Serial.println("Stop Transmit data LoRa and
Sleep! ");
digitalWrite(LED_BUILTIN,HIGH);
}
5. Command to WakeUp
void wakeUp() {
Serial.println("Transmit data LoRa!");
sleep_disable();
detachInterrupt(0);
}
====Pseudocode 1. Sleep Mode Use Interrupt =====

IV. RESULT AND DISCUSSION

A. BME280 Sensor Output
Realtime data BME280 Sensor on the Serial monitor is an
example of the output generated from the end node transmitter
Fig. 19. Adaptive Data Rate (ADR) Flowchart. (Tx) to the Receiver (Rx). Shown in Fig. 21 and Fig. 22,

45 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

therefore, this output can be added to the analysis parameters,

namely Receiver Signal Strength Indicator (RSSI) as in
Fig. 23.
Fig. 22 is the Realtime Sensor node taken from the Serial
monitor with a 9600 bps baud rate which is indicated by
realtime time, temperature, humidity, and Air Pressure data.
Fig. 23 is the sending of Sensor data from the Transmitter
(Tx) and Receiver (Rx) accompanied by RSSI Parameters, in
addition to RSSI, SNR is also very necessary to determine the
strength of the LoRa signal.

Fig. 23. RSSI (-dBm) LoRa 915 MHz.

B. Signal Analyze with Analyzer

Fig. 24 and Fig. 25 shows when the LoRa transmitter is on,
it will appear in the Analyzer a noise and signal that moves at
the amplitude and signal strength (-dB) at a frequency that
matches the radio Frequency LoRa used. (a) is the high LoRa
signal when sending sensor data, in this section expressed by
Fig. 21. Realtime Sensor Data from End Node.
the high signal that is the Amplitude signal, while (b) is the
Noise, which is part of the LoRa Frequency signal or radio
wave. Fig. 24 is the LoRa signal which consists of two parts,
the LoRa signal and the noise signal at the bottom. Therefore,
Power Noise can be determined from the strength of the LoRa
signal.
Fig. 26 is analog demodulation, i.e. IQ versus time,
Inphase and Quadrature (IQ) is two amplitude sinusoidal
waves with phase is one-quarter cycle (π / 2 radians). Inphase
and Quadrature signals are expressed in equation x. in realtime
the Inphase and Quadrature signal can be seen in Fig. 8. The
yellow color is the amplitude of the In-phase signal and the
Green color is the amplitude of the Quadrature signal.
Quadrature and phase show the sending of a LoRa signal.
(a) therefore, LoRa uses the FSK (Frequency Shift Keying)
method if it is shown in the wave signal amplitude in Fig. 27,
then the signal density shows the sending of LoRa data in
realtime and a low signal or low frequency indicates no LoRa
signal.

(b)
Fig. 22. Data Transmit (a) and Receive (b) on Serial Monitor with RSSI (-
dBm) value.
Fig. 24. LoRa Signal Analyze with SDR v.1.0.0.1700 RTL-SDR.

46 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

( ) C. Output Sensor on Application Server

( ) (4)
The sensor output is shown in the Fig. 28(a) and (b).
(5) Taken from the application server e.g, Thingspeak, TTn. This
data can be captured in realtime by smartphones, desktop
computers, mini PCs or other devices connected to the internet
.CR (6) network, so that data can be received easily and quickly,
dynamically and freely, wherever they are.
D. Realtime Chirp Sinyal
During data transmission, the LoRa signal called CSS
signal modulation (Chirps Spread Signal) can be seen in real-
time using a Signal analyzer. Chirps signals consist of 2 Up
Chirps and Down Chirps which are also called Preambles.
Fig. 25. LoRa Signal Analyze with Textronix RSA 3408B Analyzer. And when the data is received, the Signal Chirps tend to be
intermittent, because it contains LoRa data that is being
transmitted. furthermore, Fig. 29(a), (b), (c) and (d) are
examples of various Chirps LoRa signals.
Sensor node management is important to improve energy
savings in the battery. therefore a requirement for analysis of a
method using Arduino C Language to change the node to
sleep mode and from the Gateway or Server-side is to use the
Adaptive Data Rate (ADR) Algorithm method meaning the
sensor node can dynamically transmit data on Gateway-1,
Gateway-2 to Gateway-n, the adaptive means the data rate can
be dynamically controlled by each node, the important
function is for the effectiveness and management of the
battery and sending data. furthermore, this will also reduce the
percentage (%) of the Bit Error Rate (BER) and Packet Error
Rate (PER) values. Therefore, This sensor node management
pattern is to maintain Long Life or sensor node Lifetime in
Wireless Sensor Network (WSN) which has a large number of
nodes and by using LoRaWAN will enable Wireless Sensor
Network Architecture with a wide range and small bit rate for
monitoring data the sensor.

Fig. 26. Inphase and Quadrature (IQ).

(a) DHT 11 Sensor IoT based.

(b) BME_280 Sensor IoT based.

Fig. 27. Transmit Power LoRa Signal (dB). Fig. 28. (a) and (b), Output Sensor pada Application Server Thingspeak.

47 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

ACKNOWLEDGMENT
Thanks for Micro Electronics Research Laboratory
(MeRL) Kanazawa University, JAPAN for supporting through
a materials e.g., devices, analyzer. and to Professor Akio
Kitagawa who has guided the writer and has provided support
and role models.
REFERENCES
[1] Puput Dani Prasetyo Adi and Akio Kitagawa, “ZigBee Radio Frequency
(a) (RF) Performance on Raspberry Pi 3 for Internet of Things (IoT) based
Blood Pressure Sensors Monitoring” International Journal of Advanced
Computer Science and Applications(IJACSA), 10(5), 2019. DOI:
10.14569/IJACSA.2019.0100504.
[2] Puput Dani Prasetyo Adi and Akio Kitagawa, “Quality of Service and
Power Consumption Optimization on the IEEE 802.15.4 Pulse Sensor
Node based on Internet of Things” International Journal of Advanced
Computer Science and Applications(IJACSA), 10(5), 2019. DOI:
10.14569/IJACSA.2019.0100518.
[3] V. Fialhoab, F. Azevedoa, "Wireless Communication Based on Chirp
Signals for LoRa IoT Devices",i-ETC: ISEL Academic Journal of
(b) Electronics, Telecommunications and Computers, IoT-2018 Issue, Vol.
4, n. 1 (2018) ID-6.
[4] Puput Dani Prasetyo Adi and Akio Kitagawa, “Performance Evaluation
WPAN of RN-42 Bluetooth based (802.15.1) for Sending the Multi-
Sensor LM35 Data Temperature and RaspBerry Pi 3 Model B for the
Database and Internet Gateway” International Journal of Advanced
Computer Science and Applications(IJACSA), 9(12), 2018. DOI:
10.14569/IJACSA.2018.091285.
[5] Puput Dani Prasetyo Adi, Rahman Arifuddin ,”Design Of Tsunami
Detector Based Sort Message Service Using Arduino and SIM900A to
GSM/GPRS Module”, JEEMECS (Journal of Electrical Engineering,
Mechatronic and Computer Science) Volume 1, No.2, 2019, ISSN :
(c) 2614-4859, DOI:10.26905/jeemecs.v1i1.1982.
[6] Rudy Susanto, Wiedjaja, James Anthony, Jacksen, "Comparison of three
LoRa devices and its application on street light monitoring system",
2018 IOP Conf. Ser.: Earth Environ. Sci. 195 012066,
doi:10.1088/1755-1315/195/1/012066.
[7] Jetmir Haxhibeqiri, Eli De Poorter, Ingrid Moerman and Jeroen
Hoebeke, "A Survey of LoRaWAN for IoT: From Technology to
Application", Sensors 2018, 18, 3995; doi:10.3390/s18113995.
[8] Marco Cattani, Carlo Alberto Boano and Kay Römer,"An Experimental
Evaluation of the Reliability of LoRa Long-Range Low-Power Wireless
Communication", Institute for Technical Informatics, Graz University of
(d) Technology, Graz 8010, Austria; J. Sens. Actuator Netw. 2017, 6, 7;
doi:10.3390/jsan6020007.
Fig. 29. (a) Start Chirp Signal (Preamble) (b) Realtime Chirp up-Down Chirp
and Encoded Signal(c) Up and Down Chirp (d) Encoded Message Signal. [9] Muhammad Omer Farooq, Dirk Pesch "Extended LoRaSim to Simulate
Multiple IoT Applications in a LoRaWAN", International Conference on
Embedded Wireless, Systems and Networks (EWSN) 2018.
V. CONCLUSIONS
[10] Aloÿs Augustin, Jiazi Yi, Thomas Clausen and William Mark Townsley,
Transmission packet data (bytes/s) Long Range (LoRa) "A Study of LoRa: Long Range & Low Power Networks for the Internet
from End Devices or End Nodes using BME280 and DHT11 of Things", Sensors 2016, 16, 1466; doi:10.3390/s16091466.
sensors Temperature sensor has been successfully. and [11] Amir Mahmood, Emiliano Sisinni, Lakshmikanth Guntupalli, Raul
furthermore, the packet data (bytes/s) sent to the ThingSpeak Rondon, Syed Ali Hassan and Mikael Gidlund, “Scalability Analysis of
a LoRa Network under Imperfect Orthogonality”, IEEE Transactions on
application server and can be seen in realtime on devices that Industrial Informatics, Volume 15, Issue:3, March 2019. DOI.
have an internet connection. Furthermore, the LoRa signal 10.1109/TII.2018.2864681.
sending data realtime can be received by the Signal Analyzer [12] Alexandre Guitton, Megumi Kaneko, "Decoding Superposed LoRa
according to the LoRa Signal Radio Frequency used in this Signals", The 43rd IEEE Conference on Local Computer Networks
research and produce a Signal Chirp (CSS). In the next (IEEE LCN), October 1-4, 2018, Chicago, USA, DOI:
research, it is hoped that it can be further investigated, i.e., 10.1109/LCN.2018.8638253.
broadcast data transmission, therefore, that the weaknesses of [13] Taoufik Bouguera,Jean-François Diouris, Jean-Jacques Chaillout, Randa
Jaouadi and Guillaume Andrieux, "Energy Consumption Model for
the Gateway can be identified and what and what algorithm Sensor Nodes Based on LoRa and LoRaWAN", Sensors 2018, 18, 2104;
approaches can be used so that the sensor data analysis is doi:10.3390/s18072104.
detailed.

48 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 10, No. 11, 2019

[14] Chun-hao liao , Guibing zhu, Daiki kuwabara, Makoto suzuki, and [17] Akram H. Jebril, Aduwati Sali, Alyani Ismail and Mohd Fadlee A. Rasid
Hiroyuki Morikawa, "Multi-Hop LoRa Networks Enabled by ,"Overcoming Limitations of LoRa Physical Layer in Image
Concurrent Transmission", IEEE Access, DOI: Transmission", Sensors 2018, 18, 3257; doi:10.3390/s18103257.
10.1109/ACCESS.2017.2755858. [18] Ferran Adelantado, Xavier Vilajosana, Pere Tuset-Peiro, Borja
[15] Guillaume Ferré, Audrey Giremus. LoRa Physical Layer Principle and Martinez, Joan Melià-Seguí, Thomas Watteyne, "Understanding the
Performance Analysis. ICECS 2018 25th IEEE International Conference Limits of LoRaWAN", IEEE Communication magazines 2017,
on Electronics Circuits and Systems, Dec 2018, Bordeaux, France. hal- arXiv:1607.08011v2 [cs.NI] 13 Feb 2017.
01977497. [19] Xuan-Chien Le, Baptiste Vrigneau, Matthieu Gautier, Malo Mabon,
[16] Norbert Blenn and Fernando Kuipers, "LoRaWAN in the Wild: Olivier Berder. “Energy/Reliability Trade-off of LoRa Communications
Measurements from The Things Network",Delft University of over Fading Channels”. International Conference on
Technology, Mekelweg 4, 2628 CD Delft, The Netherlands, Telecommunication, Jun 2018, Saint-Malo, France. hal-01816574.
arXiv:1706.03086v1 [cs.NI] 9 Jun 2017.

49 | P a g e
www.ijacsa.thesai.org
View publication stats