Page 150 - ITU Journal Future and evolving technologies – Volume 2 (2021), Issue 2

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

by operators to provide services.

Fig. 8 – Latency evolution as a function of the number of RRHs supported
by BBU resource and fronthaul distance

When the distance evolves the 100km with each BBU
shared between several RRHs, the latency time added in Fig. 9 – Evolution of latency as a function of deployment cost per site
the network is from 6 ms to 6.4 ms. Indeed, in the spe‑
ci ic case of the rural and isolated areas, the techniques 5. CONCLUSION
in fronthaul usable are the microwave, the high‑altitude
platform system, and the satellite because of the dif iculty In this paper, we focus on the evaluation of different in‑
of access. For the microwave, knowing that it is a ter‑ frastructure sharing techniques in mobile networks. For
restrial link, we can say that the latency introduced will this purpose, we model the total investment cost for mo‑
be about 6.4ms if we stay within 100Km of the fronthaul. bile network extension in different scenarios. The objec‑
For the high‑altitude platform system, given that they are tive is to ind the most advantageous scenario for connect‑
generally located between 18 and 20Km from the ground ing the unconnected and eliminating the digital divide.
[15], the latency introduced into the network will then be The main contributions of this paper are the formulation
of the order of 6,2ms for the round trip time. In summary, of a mathematical cost model for each infrastructure shar‑
active infrastructure sharing, even if dif icult to achieve, ing technique, the proposal of an optimal sharing model,
is possible with the cloud‑RAN. This option is, from an and the application of these models in the context and re‑
economic point of view, the best in the different scenar‑ alities of rural, poor and isolated areas. The presented
ios presented. Fig. 7 shows that as the level of sharing scenarios combine several infrastructure sharing tech‑
increases, the cost of deployment decreases, and Fig. 8 niques and propose a large‑scale deployment involving
shows that sharing can lead to signi icant additional la‑ several operators for a fast, ef icient and cost‑effective ex‑
tency in the network, which can have consequences on tension of ICT in general and mobile telephony and broad‑
performance. Indeed, to reduce the deployment cost sig‑ band, particularly in remote areas. After demonstrating
ni icantly, an operator can share more of its active infras‑ the bene its of these scenarios, we show the constraints
tructure. When we are interested in quantifying the re‑ that must be respected to limit the impact on network
lationship between the network deployment cost and the performance using latency as a parameter. The main lim‑
additional latency in the speci ic case of the selected sce‑ itation of this work is the consideration of network per‑
nario 5, we obtain the curve in Fig. 9. In our fronthaul formance indicators such as KPI and the feasibility and
simulation, we have considered a transmission by HAPS, sizing of the network in a real environment. The other
which imposes a round trip of about 40km. We vary the limitation of this paper is that it does not take into account
number of operators sharing the infrastructure to deter‑ spectrum sharing. The latter being a scarce and expensive
mine the deployment cost per site and calculate the cor‑ resource, our future work will focus on this point more
responding additional latency value in each case. This ig‑ precisely by considering that white space is important in
ure shows that the deployment cost can be considerably these regions. As a follow‑up to this study, we are also
reduced thanks to sharing; however, the network perfor‑ undertaking ield studies to evaluate the implementation
mance can decrease signi icantly by prioritizing the cost of the chosen scenario to suppress the digital gap in two
reduction. It will then be necessary to ind a good com‑ developing countries.
promise between cost and performance. The additional
latency introduced by the centralisation of BBU is signi i‑ ACKNOWLEDGEMENT
cant, especially for technologies such as 4G and 5G. How‑
ever, it will give a considerable boost to bridging the digi‑ We are grateful to the African Center of Excellence
tal divide. Therefore, it must be allowed by primary legis‑ in Mathematical Science, Informatics, and Applications,
lation, encouraged by regulators voluntarily and adopted IMSP, Dangbo, Republic of Benin, which supported this
work.

136 © International Telecommunication Union, 2021