The deployment of 5G mobile networks does not only involve an evolution of the radio access segment (RAN). The impact on fronthaul/backhaul networks will be very important because new requirements are emerging (flows, latency, synchronisation, etc.). The fixed fibre access networks (FTTx) currently being deployed are in principle natural candidates to constitute this transport network (Fiber-to-the-Antenna, Fiber-to-the-Tower). Therefore, a number of issues need to be addressed now to optimise the overall costs of deployment and operation as well as the quality and resilience of networks. In particular, the question of the mutual consideration of these networks, 5G and FTTH, in the context of their deployment arises.


5G application areas

The network requirements may be different for different areas of application of 5G.

  • -eMBB (enhanced Mobile Broadband)
  • URLLC (Ultra Reliable Low Latency Communications)
  • mMTC (massive Machine Type Communications)

To date, operators who have introduced 5G services are positioning them primarily as a continuation of 4G, for improved speeds and user experience.

The expansion of the service range and the use of higher frequency bands (and therefore with less favourable propagation characteristics) will imply an increase in the number of antennas. In cities, the massive use of street furniture is envisaged.

Furthermore, depending on the application, the core functions of the network will be close to the radio segment, with particular constraints on latency.

Finally, the arrival of 5G goes hand in hand with the growing need for indoor coverage: multi-operator smallcells should become widespread, as well as other technical solutions (DAS and others), or the intervention of non-public operators.


1- Compatibility of the transport technology with the application

Mobile operators will make their choice of backhaul network based on a trade-off between:

  • the uplink and downlink rates available to the base station
  • The latency induced by the transport solution
  • Quality of service – Costs (CAPEX and OPEX / TCO)


Figure 1: Relative costs of network elements (IEEE 802 / ITU /Nokia)
Figure 1: Relative costs of network elements (IEEE 802 / ITU /Nokia)

These trade-offs will, a priori, differ according to the areas and applications proposed. For example, the relative cost of transport for a small cell is significantly higher than for a macro-cell.

The question is whether, and how, to optimise current investments in fibre network deployment for the shared optical local loop (SLLL) to ensure that they also serve 5G.

2- Timing and nature of needs

Already, the increase in traffic on 4G networks means that operators are looking to fibre as much as possible. However, it is clear that microwave (FH) is and will remain largely capable of meeting the throughput needs.

This trend is confirmed, for example, in the responses to the Arcep consultation on frequency requirements for microwave links1 , or more recently in a study conducted by the GSMA2 , which predicts that in Europe only 38.3% of backhaul links will be fibre, the rest being provided by FH.

1 Frequencies for fixed service point-to-point links (radio-relay systems): future needs and evolution perspectives – Arcep public consultation – April 2012 2 “Mobile backhaul options – Spectrum analysis and recommendations” – GSMA – Nov. 2018

Figure 2: Trend in backhaul in Europe (GSMA)
Figure 2: Trend in backhaul in Europe (GSMA)

Above all, it should be noted that when asked about their prospects, mobile operators only envisage the use of fibre in the context of dedicated links.

Most FTTx equipment manufacturers emphasise the ability of GPON systems to serve consumer and business customers and to connect antennas on the same PON tree, thanks to the low latency and the possibility of synchronising the clocks. However, the operation in the same network of customers of such different natures and which therefore require extremely distinct levels of quality of service represents a constraint which seems, to date, to be prohibitive for operators.

However, it should be noted that Free has obtained permission, in the context of a dispute settlement, to connect its mobile base stations via the supernumerary optical fibres of the FTTH network deployed by Orange and co-financed by Free, in less densely populated private initiative Mobile operators cannot afford to make their mobile access networks dependent on backhaul links whose quality of service would be degraded compared to current services.


The shared optical local loop (BLOM), since it is intended to connect all subscribers from the general public to professionals, is by its very nature an infrastructure on which numerous interventions will be carried out, whether for the connection of new customers or for operator migrations. In addition, the French model, which involves several commercial operators physically intervening on the network, creates specific operational constraints, increasing the risk of incidents. This increased risk has a potential impact on quality of service.

Mobile operators cannot afford to make their mobile access networks dependent on backhaul links whose quality of service would be degraded compared to current services.

3- Issues according to territories and places


In very dense areas, operators generally have their own FTTH network down to the foot of the building, with only the uplink part being shared. Also, since mobile and fixed operators are the same companies, they are able to size and adapt their rollouts to their future needs.

In less dense areas, a single infrastructure operator deploys the network (BLOM) from a PM located on the public domain, several hundred metres or even a few kilometres from the connection points. We note that the main BLOM operators are beginning to set up offers with mobile antenna connection boxes and the possibility of diverting the collection cable before the PM cabinet.

In the most rural areas, some stations will generate little traffic. This is the case, for example, with the New Deal Mobile stations, as the operators considered that the expected traffic levels did not even justify the deployment of a specific cell.

These differences in expected throughput levels are recalled in the table below, which shows that caution should be exercised in relation to the claims by some players of an imperative need for fibre.

Figure 3: Throughput requirement for collection (Ericsson 2017)
Figure 3: Throughput requirement for collection (Ericsson 2017)

4 The notions of “very dense areas” and “less dense areas” used are based on the definitions given by Arcep in its decision n° 2009-1106 of 22 December 2009


As explained above, forecasts predict a multiplication of the number of antennas, in particular for small cells. These antennae, operating at higher frequencies, are mainly intended to increase throughput capacity very locally, with the basic coverage being provided by the macro-cells.

In order to reduce deployment costs and the clutter of overhead equipment, it is conceivable that mobile operators could differentiate backhaul networks according to the type of base station, as the number of mobile users potentially affected by a cut-off of these small stations is limited.

Similarly, specific installations for in-building mobile coverage could rely on readily available collection solutions without additional deployments.

4- FTTx network design


Some BLOM network owners are legitimately asking themselves the question of how to take into account the future collection of 5G networks when they define the layout and sizing of fibre networks today. This consideration could be of different kinds: – Specific oversizing of optical cables

  • Creation of additional dedicated connection points (BRAM or PRSM depending on the operator)
  • Reserve dedicated modules in the cables (the same as for FTTE?)
  • When laying ducts, provision for the passage of collection cables.


Although passive offers (dark optical fiber) are starting to appear, it is unlikely that mobile operators will buy activated services on BLOM infrastructures because they want to have full control over their networks.

The questions will be more about the SLAs that infrastructure operators will be able to commit to.

When mobile operators use FTTH passive infrastructure, they can either connect these fibres directly to their backbone network or through their LTOs. The question then arises as to which PON activation technologies will be able to meet the requirements of mobile networks.

The constraints of synchronisation and latency mean that today equipment manufacturers are putting forward NG-PON, rather than GPON or XGPON, a technology that has not yet been implemented on the networks.

If the fibres are to be taken directly from the collection network, then gartering solutions at the NRO should be available.


The variety of technical solutions available for the collection of 5G base stations, and the questions that remain open about how these networks and the different types of cells will be operated, mean that it is not possible to say that 5G collection will necessarily rely on FTTH networks.

A dialogue with mobile operators to understand their needs in the different areas will enable BLOM operators and project owners to better anticipate these needs.

Whatever the intended use, mobile operators will only use these networks if they are able to guarantee high levels of service, by securing and guaranteeing the quality of the network’s points of flexibility (NROs, pooling points, connection points). Therefore, even before considering oversizing or specific deployments, BLOM infrastructure operators must demonstrate the ability to ensure reliable and sustainable operation.

Source : Credo


What's new at Skylane Optics

See all news April