1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
|
\chapter{Discussion and usage scenarios} \label{chap:drysty}
\section{Comparison of discussed technology}
This section will comment on
\subsection{NB-IoT}
NB-IoT is best suitable for non-moving low speed low power devices where is possibility of conditions of very bad LTE coverage or placement in locations with high RF shielding such as basements or underground and every dB of signal is appreciated.
Mobile communication is possible but should be approached with cautious when dealing with critical application.
It is very comparable to LoRa and the choice between them depends large on other factors than purely radio properties.
Huge decisive factor are: size of anticipated area of service, coverage of cellular network by mobile operators, initial cost of modules and their yearly running cost, special needs and hacks that would need custom gateway.
\subsection{LTE-M}
LTE-M is optimally suited for moderate-speed, low-power embedded devices that require reliable mobile connectivity and higher data throughput than NB-IoT, but still benefit from extended battery life compared to traditional LTE. When in its signal range where it can safely communicated it should be preferred over NB-IoT because of its lower power demands and higher data speed.
It is effective in scenarios where devices might experience fluctuating signal conditions, high mobility or need to send larger data packets periodically.
\subsection{DECT NR+}
DECT NR+ waits for itself to matured and be used in application not clearly foreseen now. Its very low latency with ability to maintain connectivity of very large and dense network of devices would enable it to be game changer in massive IoT applications described below.
\section{Scenarios of IoT applications and recommended protocol}
This section will shift focus from particular protocol to possible scenarios in various industry sectors and evaluate best protocol from this point of view.
\subsection{Classic stationary IoT}
The most ubiquitous form of IoT are devices such as simple sensors such as remote environmental of river water level or remote reading of home water meters. This category includes even slightly more complex devices such as control of street lamps, control of different gates. For new application like these the choice of communication technology would have negligible impact on performance of the device. Only in places with limited coverage of cellular network some larger considerations will be needed.
\subsection{Mobile IoT}
For scenarios involving tracking, wearables, and mobile smart metering in areas such a smart agriculture or logistics LTE-M or possibly NB-IoT would be most suitable choices for their universal adaption and ability to communicate even while moving.
\subsection{Remote tracking}
There starts to emerge a new way of doing maintenance or at least diagnostics of different machines on remote using integrated IoT devices. This would save time and money of maintenance workers that would not need travel to diagnostics a trivial problem from their view. Suggestion of protocol is dependent on required data usage and LTE coverage for typical installation of such project.
\subsection{Massive critical industrial IoT}
Smart grid - A massive number of smart meters and controllers deployed across residential, commercial, and industrial sites, providing real-time data on energy consumption and grid conditions.
This could lead to possibility to optimize the power distribution to the point where residential user could access benefits of spot market both on consumption and on production side.
Machine Health Monitoring - Connection of virtually every machine in a large production line with sensors to monitor performance, vibration patterns, power consumption, and production output. This enables predictive maintenance, reduces unexpected breakdowns, and optimizes production schedules. Cost of the additional IoT devices would in most scenarios be far lesser that the possibility of machine prenamenet damage or production delay due to not foreseen maintenance. This would could lead to massive adaption of these sensors and need for more robust protocols. DECT NR+ is from the list of mentioned protocols ideal choice
\subsection{Low Earth orbit satellites}
Attenuation of frequency around 800 MHz to Low Eart Orbit around 400 - 1500km is about 144 dB to 154 dB. LoRa is already used in some of these satellites as cost effective way to communicate with a satellite without the need for expensive ground station. NB-IoT demonstrated to be able to manage attenuation of very similar magnitude. Use of NB-IoT to space and imagining nano satellites as new frontier of IoT devices is certainly very pleasing but any such attempt would need at least some support from at least one of established mobile operators. Until then established communication manners with occasional LoRa will prevail.
\section{Future work}
This section outlines some parts that would benefit from expansion of current data or repetition of existing experiment with slight tweaking.
\subsubsection{Direct comparison to LoRa}
Extension of LoRa measurements to comparable RF measurements with power consumption for fairer direct comparison of that and LTE protocols.
\subsubsection{Mobile usage with very low signal streght}
Check if the advantage of NB-IoT over LTE-M for weak signal decode is still present when the devices are moving and if speed has influence to it
|
