Private 5G – Why not there yet?

stein@opensky

From time to time, I get triggered by something I read – and this time it is no different. This article from Mobile World Live on what is holding back 5G in the industry is one of those – as it refers to a panel discussion at the Hannover Messe 2025 just finishing last week. The panel concluded that private 5G is moving forward – although still too slowly. In my recent update from MWC25 in Barcelona, I also commented on a similar session on “Scaling private networks – What is holding us back?”. Without directly duplicating the Hannover article, some of my own views are commented below – often also already commented earlier in other articles.

Complexity

In the Hannover article above, the Chair of the 5G Alliance for Connected Industries and Automation (5G-ACIA) referred to the “perceived complexity and costs associated with the (5G) technology” as a barrier to adoption. This is an important perception to address.

In my earlier article from 2021 already on “Looking for a private 5G network?” I elaborated on considerations for an enterprise around implementing and operating a private 5G network. If the conclusion is that you really need a private mobile network, then you need to consider how to deploy it, operate it and upgrade it when needed - and an obvious question is whether this should be handled in-house or by a third party.

In the context of today I consider mainly a private network based on private spectrum acquired from the regulator – as a campus network to get improved local coverage and security and to run “Operational Technology (OT)”. This limits the use case to supporting industrial automation in a campus with wireless and secure technology – and with interoperability with the public 5G network mainly optional. It also means that a number of base stations with private spectrum will have to be installed – complemented with a private core controlled by the enterprise.

With reference to my earlier article, building and managing a (public) mobile network requires a lot of dedicated expertise, effort, investments and staff – which is something MNOs have been building up over decades – and in many cases, traditional telco vendors have also acquired this competence – as MNOs increasingly have outsourced deployment as well as operation of their networks to the telco vendors.

Most private enterprises don’t have this capability, however, building and operating a private 5G campus network is much simpler, as there will most likely be only a handful of base stations needed. It is also possible today to purchase a “private 5G core network in a box” from standard telco vendors like e.g. Ericsson – or even from smaller software companies based on COTS hardware (refer also my earlier article on “software telcos”). The job will thus mainly be to install and operate a small 5G core – plus a handful of base stations. Installation of this in a campus already owned by the enterprise should be quite feasible, however, operating and maintaining it (including bug fixes, upgrades, new features, etc) will require a certain expertise not readily available in the existing IT department – so outsourcing may be a very relevant consideration. A large enterprise might, however, have the capability to handle it in-house. Good contracts handling life cycle issues with your vendors will also be a pre-requisite. Buying a “private 5G network as a service” may be a good option as well.

On the more general perception of complexity, it is clear that the 5G architecture is very complex, however, this does not necessarily mean that installing, operating and maintaining a private 5G network is. It should be noted, however, that it will require special competence not necessarily available in a classical IT department - as private 5G must be seen as Operational Technology (OT) – just like any other legacy OT existing in the company.

IT, OT and internal turf issues

Most of my life, I have been working for mobile operators – and in those organizations the main technical challenge with associated Capex has always been to plan and roll out the mobile network with thousands of base stations nation-wide and a core network of switches etc. The IT side of this has mainly involved BSS (“Business Support Systems”) and OSS (“Operations Support Systems”), both planned for telco grade availability and highly linked to the vendors of the mobile network. The more standard “Office IT” area has been small in comparison. In a telco organization as above, there has therefore traditionally been a split between OT and IT, with a main focus on OT – and, if seen in context – which is not always the case, IT has often reported to the CTO with the main focus to build and run the mobile network.

In a small company, all technical areas often belong to the IT department, which are mostly occupied with Office IT – and there may be no OT as such. In a larger company with some form of industrial production and with a potential interest in a private 5G network, there will be a certain amount of OT solutions available – but these are (like also in a telco) typically very much linked to the vendors of the production equipment.

The bottom-line here (which is also commented in the article from Hannover Messe) is that IT and OT are not always handled within the same organization – and also that they are not used to (or want to) work together. To install and operate a private 5G network, working together in the process is most likely required.

On a similar note, I once got a new colleague from the oil industry, to be responsible for procurement – and his initial statement was that “mobile network planning is the first stage of the procurement process”. Coming from outside into a mobile telco where the CTO was in charge of building and operating the mobile network and where procurement was simply tasked to negotiate a good deal for what had already been planned, this was a slight mismatch.

The truth is probably somewhere in the middle – but in all cases, the organizational structure and priorities depend on the industry and the company. There may also in many cases be internal turf issues that can hamper the take-up of private 5G in industries – and while industrial automation must be regarded as OT, resistance from the IT area might be seen.

Private 5G or WiFi

About five years ago, I wrote the article “WiFi6 and 5G – Is 6 better than 5?”, commenting on the tendency around a possible “technology war“ between 5G and WiFi6 at the time. In the meantime, WiFi7 is now also coming to the market, but many of the topics have not changed. For industrial use cases though, we could probably disregard anything in the referred article on roaming and B2C.

As commented in the article and in several other, 5G and WiFi will both be there going forward – and to some extent they are complementary technologies. For industrial use cases, however, some differences may be relevant to consider – and for an enterprise looking for industrial automation, there will be trade-offs between speed and ease of implementation and available competence in the organization, etc. Some issues to highlight for an industrial use case include, however:

• WiFi is a radio access technology only – so it can be easily installed using standard routers and a number of radio Access Points.

• WiFi operates in an unlicensed frequency band – so it may be exposed to interference from other WiFi devices in the area. 5G operates on dedicated frequencies managed by the enterprise and is not exposed to external interference.

• 5G is a complete system and needs a dedicated core and a number of base stations / antennas, however, the number of antennas might be 10 times fewer compared to WiFi due to longer range.

• 5G offers gigabit speeds and very low latency – and WiFi may provide similar.

• 5G is designed for telco grade availability – while WiFi quite often will have outages – which may be a concern for OT.

• Use of public WiFi is more and more discouraged for security reasons, but end-to-end security can be added. 5G has inherent security with in-built authentication and encryption.

• If interoperability with the public 5G network is of interest, this is inherently available for private 5G. For OT this may not be of interest, but for users operating the OT it could be relevant.

In terms of ease and speed of installation, I would assume that this would be similar between WiFi and 5G. With a “private 5G network in a box” and many fewer antennas, installation of 5G could even be faster. In terms of O&M, private 5G requires competence not necessarily available in the enterprise.

Cost aspects

In an internal office IT scenario, an enterprise will typically install WiFi with routers and Access Points purchased from a vendor and the IT Department will in many cases manage the WiFi network itself. In some cases, it might also outsource the basic management of the WiFi network. For a private 5G network, this might not be the applicable scenario – firstly because it is all about OT and secondly because 5G involves a lot more software and regular software upgrades – with associated software licenses.

The cost of a private 5G network will generally depend on the vendor chosen and on the operational model. What you will generally have to pay for, however, are the base stations / antenna equipment and the core network / baseband equipment – with associated software licenses and upgrades as required. It also depends on whether you buy and operate yourself or buy a private network as a service. On the device side, you will also need to purchase specific 5G devices to connect to your legacy OT equipment.

In case you would use WiFi as wireless connectivity supporting your OT systems you would still need to use specific WiFi devices applicable to your production environment. You would probably have to install around 10x more WiFi Access Points but probably pay fewer software licenses (as there is less SW). With WiFi you could also outsource or insource the O&M – but IT and OT should probably be separated in any case.

Availability of devices

I am so old that I have witnessed the introduction of most “G”s to the market. The acronym “GSM” was initially a French name for a standards group under CEPT called “Groupe Spécial Mobile” - but the name of the GSM (2G) system was later referred to as “Global System for Mobile communication”. Already at the time it was introduced into the markets in Europe in the early 1990s, however, there was a shortage of devices which clearly hampered its take-up commercially. For fun, GSM was then simply at the time referred to as “God Send Mobiles”.

The essence of it all is that you need not only available technical standards and network infrastructure for it to take off - but also commercially available devices at reasonable prices based on global volumes, so that people could actually use the network. For later “G”s you could probably also add availability of relevant use cases and services or applications. In the case of private 5G networks, there may be a need for specialized devices, probably different for different industrial applications or for different industries. They may be for indoor or rugged outdoor use – and the global scale is most likely at a different magnitude from commercial B2C mobile phones.

If we limit the use case to industrial automation with high speeds and low latency, the industry could most likely get to a reasonable scale, be it for indoor or outdoor use – but at another magnitude than for B2C equipment.

SMEs or large enterprises

A year or two ago, I took part in a “5G Forum” organized by Digital Norway with the main ambition of enhancing and supporting the take-up of (mostly private) 5G in Norway. This was attended for the most by mobile operators, some large international telecom companies with affiliates in Norway (like Siemens, Ericsson, Huawei etc) – and large Norwegian industries with a potential interest in implementing private 5G networks. SMEs were not really in the picture.

It is clear that the most relevant enterprises using private 5G networks for industrial automation would be large enterprises as above. However, also SMEs could have an interest in private 5G – although not necessarily for industrial automation. The only challenge is that SMEs may know even less about 5G than large enterprises. For large as well as small enterprises, there is therefore a high need in society for awareness about 5G and to educate them about what 5G can do for them.

Summary

A private 5G network can in principle be used by any enterprise, but it is probably most relevant for enterprises with a certain production and for industrial automation (OT).

So why is private 5G struggling to take off? …  and what are the challenges?

• Perceived complexity: The 5G architecture is complex and also most familiar to mobile operators and telco vendors (and less to the general IT community). However, it should be quite feasible to install a private 5G network. To operate and maintain it needs some specific competence though.

• Cost: What you get with a private 5G network cannot be compared to an internal IT network using WiFi – and the costs associated may not be comparable. With 5G you would get a telco-grade performance and also software updates – which you would have to pay for. In the context of OT and continuity of production, this would normally be a requirement.

• Internal organizational issues: IT and OT are often handled in separate organizations – and they may not be used to (or want to) work together. While IT units are likely to favour WiFi as a wireless technology, it may not provide the required performance or availability. OT or production units may have other priorities though.

• Awareness and competence on 5G: 5G (as well as any other G) have historically been technologies only available to mobile operators and telco vendors – and enterprises have purely bought mobile subscriptions from operators. In a context where 5G will be used for OT and industry automation, enterprises will need to understand more and secure sufficient control of their own production environments.

What can be done?

• Availability of devices: Devices for private 5G have been a challenge so far but are now coming to market (at least for low latency broadband communication). These will be at another cost level than regular mobile phones, but in an OT perspective, this might be acceptable.

• Private 5G is about OT: A private 5G network for industrial automation is about OT and must thus be seen in the context of production and business continuity.

• IT, OT, WiFi and internal organizational issues: While the legacy production infrastructure in an enterprise typically would be connected via cables and the internal IT organization often would be more inclined towards WiFi should wireless technology be considered, a private 5G network is often unknown territory. To bridge the IT/OT gap, it is therefore recommended that a cross-functional team between IT and OT units is set up in the preparation phase.

• How to organize for running private 5G: As internal competence on telco-specific technology (like 5G) may be scarce in the enterprise, it may be wise to start off with buying “private 5G as a service” with everything from installation and detailed configuration to Operations & Maintenance. A large industrial enterprise may have the capabilities to handle it internally – but outsourcing the job (at least from the start) may be a good place to start. An in-house takeover can of course be considered with time.

• Communication and awareness on 5G: Not only for large enterprises with industry production but also for SMEs and enterprises in general, there is a need to educate enterprises and industries about how 5G works and what it can do for you. Operators have so far focused on what they can sell to enterprises themselves, however, in a private 5G perspective operators may not even need to be involved – so there is a lack in the industry for more education on 5G. It would be good if players independent from the operators could take on this challenge.

For any further reading about 5G, I have previously written a number of articles on 5G (and 6G), the most recent earlier this year on “Making 5G a success – or waiting for 6G?”. Earlier ones from 5-6 years ago include also “5G or 6G? The story all over again?”, “5G will save the world – or will it?”, “5G is all around us – but who will eventually benefit?” and “What will really happen with 5G?”). These are all 4-5 years old now, but surprisingly mostly still true.