Page 72 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 7 – Terahertz communications

ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 7 6. CONCLUSION AND FUTURE WORK

Number of awake nodes (100K fs) 2000 Rec-2-1200 This paper presents the effect of the sleeping mechanism
2500
Rec-1-2400
Rec-3-900
in a heterogeneous nano‑network. The new idea embeds
a density estimator algorithm (DEDeN) to automatically
1500
tune the node awaken interval (awakenDuration). The
estimator algorithm can be executed in different times
1000
(network deployment, or when a node needs to estimate
its number of neighbours). It shows its usefulness when
500
becomes clear that relying on average neighbour node
0
160
60
40
20
80
140
120
100
density has a positive in luence in preserving node
Number of awake neighbours 180 200 used in conjunction with the sleeping mechanism. It
resources (CPU, energy, memory ...) and decreases the
number of sent packets compared to results based on
Fig. 18 – Number of full‑awake nodes over different average density val‑
ues. static percentage of awaken duration.
sumption) ranging between 20 000 and 100 000 fs (diffe- Considerations have also been taken for the speci ic case
of the destination zone (packet loss by the destination
rent awaken duration for each node). Furthermore, for an
node because its sleep may coincide with the packet’s ar‑
awaken percentage of 90%, all the nodes in the network
(4500 nodes) will be awake along this duration (Fig. 19). rival at the destination zone). An algorithm has been pro‑
posed in case the application needs data packets to reach
a speci ic node in the destination zone.
5000
60 Awaken neighbours As future work, we would like to consider different al‑
4500 Awaken duration 90 %
gorithms and variations to handle what happens at the
4000 destination zone. Especially, a destination could be ex‑
Number of nodes 3000 pressed as “any node in the destination area that provides
3500
a given service” instead of all nodes in the zone or a single,
2500
speci ic node.
2000
1500
1000
500 7. ACKNOWLEDGMENTS
Ali Medlej has a grant from Islamic Center Association
0
0 10 20 30 40 50 60 70 80 90 100 for Guidance and Higher Education (Lebanon). This work
% Awaken duration has also been funded by Pays de Montbéliard
Agglomération (France).
Fig. 19 – Number of nodes depending on an awaken duration of 90%
and 60 neighbour nodes.
REFERENCES
The total number of sent packets is a metric we rely on to
compare between the methods used in applying the sleep
[1] Xin‑Wei Yao and Ye‑Chen‑Ge Wu and Wei Huang,
mechanism (awakenDuration vs awakenNodes). Based on
”Routing techniques in wireless nanonetworks: A
the results in the previous sections, setting the node
survey”, in Nano Communication Networks, pp. 1–
awaken duration based on node neighbour density (using
13, Sep, 2019, Elsevier.
DEDeN) will help in decreasing the number of packets sent
by 14% when compared to a ixed awaken percentage [2] Büther, Florian and Lau, Florian‑Lennert and
(awakenDuration). Stelzner, Marc and Ebers, Sebastian, ”A formal
To conclude, in heterogeneous networks, the SLR rou- inition for nanorobots and nanonetworks”, in
ting protocol is acting well when crossing areas of
Internet of Things, Smart Spaces, and Next Gene‑
several densities. It smoothly deals with nodes of ration Networks and Systems, pp. 214–226, Aug,
different awaken durations, and packets are successfully 2017, Springer.
delivered to their intended destination. Applying
DEDeN allows nodes to have different awaken durations [3] Akyildiz, Ian F and Jornet, Josep Miquel, ”The Inter‑
compared to a static awakenDuration. This leads to a net of nano‑things”, in IEEE Wireless Communica‑
decrease in packet transmission cost, and preservation tions, pp. 58–63, Dec, 2010.
of node re‑ sources (CPU, energy, memory), therefore
[4] Jornet, Josep Miquel and Akyildiz, Ian F ,
an increase in network lifetime. Moreover, applying the ”Femtosecond‑long pulse‑based modulation for
retransmission algorithm at the destination zone terahertz band communication in nanonetworks”,
increases the chances for the destination node to in IEEE Transactions on Communications, pp.
receive its intended packet. Therefore, this algorithm 1742–1754, Apr, 2014.
helps to increase the reliability of data arrival to the
destination node, which is an important parameter in [5] Tsioliaridou, Ageliki and Liaskos, Christos and Dedu,
some applications. Eugen and Ioannidis, Sotiris, ”Packet routing in 3D
60 © International Telecommunication Union, 2021