Spectrum sharing is becoming a necessity for the success of 5G

• Spectrum, a limited and finite resource, is crowded and becoming scarcer
• Sharing it can ease the tricky transition from 4G to 5G
• Dynamic spectrum sharing (DSS) is a pragmatic and sensible solution to a pressing problem
• The tricky part will be to convince (and/or pay) spectrum holders to partition their radio frequency allocation

As the demand for mobile wireless services continues unabated, the telecoms industry trade association 5G Americas, whose membership comprises leading telecoms operators, service providers, equipment manufacturers and software companies, has just released its latest briefing paper, ‘Spectrum Sharing: Challenges & Opportunities’. It provides in-depth coverage of the various spectrum-sharing models (such as licensed, unlicensed and shared) and their technical, regulatory and economic complexities, as well as the potential benefits they offer to the wireless industry in general and to the Americas in particular. 

It further, and importantly, looks at the specification and implementation of evolved spectrum access systems, including dynamic spectrum sharing (DSS). It is accepted that licensed spectrum is the gold standard for 5G networks, but spectrum is a limited and finite resource. As it becomes scarcer, particularly in the low- and mid-band ranges, a pragmatic solution would be to allow service providers to share what spectrum is available, and DSS is an antenna technology that permits 4G-LTE and 5G cellular wireless technologies to be used within the same frequency band. It allocates bandwidth based on the changing levels of usage whilst minimising the perennial problem of interference, and maximises spectral efficiency. It could well be a solution to the provision of commercial wireless services in the 3.1-3.45 GHz range.

Dedicated (and expensive) licensed spectrum allows mobile telephony providers to architect and dimension their networks, so that they adapt at speed to the varying demands of user traffic, which has increased massively as data-hungry apps and services have become globally popular to the point of indispensability. Regulators have responded by making more spectrum available – mainly by vacating spectra of other, usually redundant, often government-owned communication systems and networks, but demand continues to rise. Meanwhile, unlicensed spectrum has been used, in the main, to allow the provision of Wi-Fi services to enterprises and domestic premises. Again, demand has increased as the availability of new apps and services has surged.

Spectrum sharing is a technically complex operation as well as a regulatory nightmare. Historically, those service providers, desirous of sharing spectrum that has long been the sole preserve of incumbent operators, have had to accommodate themselves to their established regimes and requirements, the main one being that their operations do not interfere with the incumbent’s established performance and functional parameters. Unsurprisingly then, shared spectrum efficiency has usually been markedly lower than when a single operator has exclusive access to a particular slice of bandwidth.

The 5G Americas briefing paper lists the main problems that can affect shared spectrum systems. They include the overloading of the protocols that coordinate shared spectrum between different users. They are also subject to necessary transmission power limitations needed to minimise interference, which limits both the range of transmit signals and overall network capacity. It also increases costs because the degraded service requires the deployment of more transmitters across the network in an effort to provide the same levels of connectivity that would pertain in a network functioning at full power. 

Furthermore, networks by design and usage are dynamic environments, and spectrum has to be instantly available at all times. This can be problematical in shared-spectrum situations and can lead to reduced quality of service and concomitant customer dissatisfaction.

It is generally agreed that DSS is highly effective in a cognitive radio (CR) network by allowing the simultaneous use of 4G/LTE and 5G New Radio (NR) on the same frequency band by exploiting the flexibility of the physical layer as defined in today’s telecoms standards. In the past, the operation of different mobile wireless standards required a minimum of two antennas, whereas DSS allows 5G to share the same antennas, radio frequency (RF) units and 4G spectrum, and enhances spectral efficiency by allowing low-priority unlicensed users to share the bandwidth of high-priority licensed users.

This can ease the tricky transition from 4G to 5G because the technology obviates the need for lengthy and expensive spectrum reallocation exercises or acquiring separate 5G, which greatly helps operators to reduce their capital expenditure (capex), as several generations of wireless technologies can happily co-exist alongside one another in the same frequency band within the same geographic area. Such a solution is particularly valuable for remote and rural areas where standalone 5G would be extremely expensive and provides for the very efficient use of the network times 5G traffic is low.

It is worth noting that a CR network is, by definition, aware of itself, of what is happening within and around itself in the greater environment and can make decisions by mapping what it knows at any given point against a table of predefined objectives. The basic foundation of DSS are RF spectrum, software-defined radio (SDR) and network slicing optimised for specific types of services or different customers.

DSS also ideal for IoT

Another major benefit of DSS is that it has a major role to play in smart connectivity in internet of things (IoT) because its dynamic nature is particularly appropriate to the wide diversity of proliferating IoT systems and networks, as it allocates spectrum on the fly to ensure IoT devices have the bandwidth and reliable communication channels to work efficiently and guarantee the transmission of large amounts of batch or continuously provided data either in real time or at set timing intervals.

As 5G NR, LTE-M and narrowband IoT (NB-IoT) can all co-exist in the same spectrum, mobile network operators will be able to sweat their LTE-M and NB-IoT assets technologies for many years to come – estimates are that they will remain perfectly viable for at least another decade, and probably longer.

While there are varying definitions of DSS, such as those from 3GPP and the US National Telecommunications and Information Administration (NTIA), 5G Americas categorises it as “two or more disparate radio services (including incumbents and new entrants) adjusting their respective operations to achieve efficient utilisation of the same frequency range in time or space, such that they can each provide their intended service in a cost-effective manner with minimal performance degradation to their respective operations”, and that is certainly what DSS can facilitate. 

However, widespread uptake of DSS will depend on motivating those operators already holding spectrum to share it, and that will involve financial incentives. It will also require the difficult and potentially lengthy valuation of spectrum sharing and the need for investment in new technologies. Then there are the rather high administrative and technical costs associated with managing shared spectrum access.

So, as far as the widespread use of DSS is concerned, there are a lot of problems still to be solved, but the technology’s undoubted benefits make it a ‘must-have’ rather than a ‘would be nice to have’. Certainly, in an ideal world where spectrum is infinite, exclusive licensing would always be the preferred option, but in the real world, spectrum sharing is the next best thing. It will encourage new and non-traditional operators into the market, foster innovation, increase competition and, hopefully, provide better customer service. What’s more – spectrum sharing is ideal for private networks in a plethora of sectors, such as agriculture, automotive, energy, manufacturing and many others. Furthermore, DSS will be vital for mobile virtual network operators (MVNOs) in countries that do not have access to 3.3 – 3.5GHz for 5G services, including markets championed by 5G Americas.

The president of 5G Americas, Chris Pearson, summed things up nicely: “Dedicated exclusive-use radio spectrum is a critical resource affecting the wireless industry, economy and technology leadership. Yet, effective spectrum sharing could be a part of some solutions but will require innovative technologies and collaborative regulatory frameworks to ensure that both legacy users and new entrants can coexist without compromising performance.” 

It will happen, because there is an excellent solution to a deepening problem and necessity is the mother of invention.

- Martyn Warwick, Editor in Chief, TelecomTV