Page 115 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 1

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 1

through BLE to 1. We achieved BLE raw transmissions REFERENCES
through the use of single BLE advertisements, hence the
packet losses. With the enhanced RODENT, we can see a [1] D. Ye, D. Gong, and W. Wang. Application of wire‐
better PDR for all nodes, close to the one obtained with less sensor networks in environmental monitoring.
only WiFi. In International Conference on Power Electronics and
Intelligent Transportation System (PEITS), 2009.
11.3 Energy consumption [2] B. Foubert and N. Mitton. Long‐range wireless radio
technologies: a survey. Future Internet, 12(1), 2020.
Physical measurement of the Pycom FiPy’s energy con‐
sumption is hazardous since it suffers from design prob‐ [3] B. Foubert and N. Mitton. Autonomous collaborative
lems which lead to erroneous measurements [24]. We wireless weather stations: a helping hand for farm‐
choose to stick to the energy ratings given in the compo‐ ers. ERCIM News, (119):37–38, 2019.
nents data sheets [17,21] to get a general idea; these are
shown in Fig. 11. Compared to WiFi, BLE needs approxi‐ [4] F. Bari and V. C. M. Leung. Multi‐attribute network
matelyhalf‐less current and LoRa a tenth. With the Pycom selection by iterative topsis for heterogeneous wire‐
FiPy’s CPU, WiFi and BLE offers the same bit rate. LoRa’s less access. In IEEE Consumer Communications and
bit rate is much slower leading to longer transmission for Networking Conference (CCNC), 2007.
the same amount of data. WiFi and BLE require a heav‐
ier traf ic control than LoRa does, which allows LoRa to [5] W. Zhang. Handover decision using fuzzy madm in
consume less energy. Thus, we can assume that RODENT heterogeneous networks. In IEEE Wireless Commu‐
enables signi icant energy savings. nications and Networking Conference (WCNC), 2004.

12. CONCLUSION [6] M. A. Senouci, M. S. Mushtaq, S. Hoceini, and A. Mel‐
louk. Topsis‐based dynamic approach for mobile
WSN deployments are limited by the coverage and perfor‐ network interface selection. Computer Networks,
mances of the devices’ RAT. The use of several RATs in an 107:304 – 314, 2016.
MTN allows these shortcomings to be overcome. In this
article, we introduced the novel Routing Over Different [7] E. Stevens‐Navarro and V. W. S. Wong. Compar‐
Existing Network Technologies protocol (RODENT). RO‐ ison between vertical handoff decision algorithms
DENT is based on a lightweight TOPSIS method that re‐ for heterogeneous wireless networks. In IEEE Ve‐
duces the complexity of the computations and eliminates hicular Technology Conference (VTC), 2006.
rank reversal issues, lessening computation time of about
38%. The resultingnetwork interfaceselection is still sim‐ [8] F. Bari and V. C. M. Leung. Automated network selec‐
ilar to the one obtained using classic TOPSIS in 82% of the tion in a heterogeneous wireless network environ‐
experiments. ment. IEEE Network, 21(1):34–40, 2007.
RODENT is designed for routing in MTN and enables the [9] L. H. Teixeira and A. Huszak. Preemptive network
use of multi‐technology routes. We demonstrate the fea‐ selection for v2v communication. In International
sibility and utility of MTN with a prototype network based Conference on Telecommunications and Signal Pro‐
on a custom implementation of RODENT. Results show cessing (TSP), 2019.
that RODENT increase lexibility, reliability, energy sav‐
ings and maintains a good PDR. [10] I. Bisio and A. Sciarrone. Fast multiattribute net‐
For future work, we plan to precisely measure energy sav‐ work selection technique for vertical handover in
ings and extend RODENT to support downlink communi‐ heterogeneous emergency communication systems.
cations. This will further increase nodes’ lexibility and Wireless Communications and Mobile Computing,
possibilities (e.g., irmware over the air upgrade). We plan 2019:1–17, 2019.
to conceive an ef icient link layer protocol for precise link
costs and performance assessment for multi‐RAT. In ad‐ [11] J. Famaey, R. Berkvens, G. Ergeerts, E. D. Poorter, F. V.
dition, we plan to combine RODENT with an ef icient data d. Abeele, T. Bolckmans, J. Hoebeke, and M. Weyn.
reduction scheme to reduce even more energy consump‐ Flexible multimodal sub‐gigahertz communication
tion. We intend to run a larger simulation and experimen‐ for heterogeneous internet of things applications.
tation, where end‐to‐end delay will be measured to assure IEEE Communications Magazine, 56(7):146–153,
that RODENT’s usage does not slow down the network, 2018.
and we will show how this satis ies QoS in terms of delay.
[12] J. Hoebeke, J. Haxhibeqiri, B. Moons, M. Van Eeghem,
ACKNOWLEDGMENTS J. Rossey, A. Karagaac, and J. Famaey. A cloud‐based
virtual network operator for managing multimodal
This work was partially supported by a grant from CPER lpwa networks and devices. In Cloudi ication of the
DATA, Sencrop, FEDER, I‐SITE and Lirima Agrinet project. Internet of Things (CIoT), 2018.

© International Telecommunication Union, 2021 99