6.1.1 Global-Level Regulatory Framework

The international global regulatory framework is provided by the International Telecommunication Union (ITU), a specialized United Nations (UN) agency. Within the ITU, the ITU Radiocommunication Sector (ITU-R) plays the central role in the global management of the radio-frequency spectrum. ITU-R seeks to ensure the rational, equitable, efficient and economical use of the radio-frequency spectrum by all radiocommunication services. The main priorities of the ITU’s regulation of radio spectrum are:

• To protect against harmful interference
• To allocate radio services to the various radio-frequency bands in the radio spectrum (including globally harmonized allocations for systems used in international air and sea travel), taking into account sharing and compatibility studies
• To promote the effective use of the spectrum

To do so, the most important ITU legal instrument is the Radio Regulations (RR) treaty [3], which is ratified by the ITU member states. The ITU RR is the global agreement on how the airwaves are defined, allocated and used without harmful interference between the various wireless services around the world. ITU RR are binding for the states, not for individuals or operators. The compliance with ITU RR therefore presupposes that each state will take the measures required (legislation, regulations, clauses in licences and authorizations) to implement domestically those obligations to other spectrum users (operators, administrations, individuals, etc.). As such, the ITU RR provide an international framework for effective spectrum management that is primarily structured by the need for global harmonization in various domains (satellite communication, maritime, civil aviation, scientific research, etc.), coexistence capability between different types of radio communication networks and physical properties of frequency bands. It has major implications for the industry in terms of economies of scale and therefore for the design of radio products.

The regulatory and policy functions of ITU-R are carried out through World and Regional Radiocommunication Conferences and Radiocommunication Assemblies. World Radiocommunication Conferences (WRCs) are regularly held every 4–5 years, and the decisions adopted are incorporated in the ITU RR. Ahead of the celebration of each WRC, there are many ITU-R study groups and working parties [4], with representatives providing support, research and background for ITU RR. Contributors to this work include representatives from state’s regulatory authorities as well as from regional spectrum management organizations that represent the common interests of a group of states. The last WRC was held in January–February 2012 (WRC-2012), and the next one is planned for November 2015 (WRC-2015).

The ITU RR (Article 5 of the RR) contains the so-called Table of Frequency Allocations (TFA) that allocates frequency bands for the purpose of their use by one or more terrestrial or space radiocommunication services or the radio astronomy service under specified conditions. A radiocommunication service is defined as the transmission, emission and/or reception of radio waves for specific telecommunication purposes. Terrestrial services and space services can themselves be subdivided in several different types of services (fixed, mobile, broadcasting, etc.). Each frequency band can be allocated with one or more radio services (typically two to four). Frequency bands are allocated to radiocommunication services on a primary or secondary basis. Stations of a secondary service shall not cause harmful interference to stations of primary services and cannot claim protection against harmful interference from stations of a primary service. In order to recognize certain regional differences that may be necessary in the frequency allocations, the ITU has split the world in three geographical regions (1, 2 and 3), so that the TFA is particularized for each region.

In addition to the frequency allocations to radio services, the ITU RR also contain administrative provisions on the registration (and protection) of specific frequency assignments as well as on the use of radio frequencies. The principle underpinning most of the provisions of ITU RR stipulates that any new assignment (i.e. any new authorization to operate a radio station) must be made in such a way as to avoid causing harmful interference to services rendered by stations using frequencies assigned in accordance with the agreed TFA and the other provisions of the ITU RR, the characteristics of which are recorded in the Master International Frequency Register (MIFR). In particular, a new assignment can only be recorded in the MIFR after completion of a procedure aimed at ensuring that it will not cause harmful interference to assignments made in accordance with the RR and previously recorded systems.

6.1.2 Regional-Level Regulatory Framework

Regional spectrum management organizations also play a major role in the management of the radio spectrum resource. There are six main regional organizations in the world:

1. Inter-American Telecommunications Commission (CITEL)
2. European Conference of Postal and Telecommunications Administrations (CEPT)
3. Asia-Pacific Telecommunity (APT)
4. African Telecommunications Union (ATU)
5. Arab Spectrum Management Group (ASMG)
6. Regional Commonwealth in the Field of Communications (RCC)

These organizations seek the harmonization and coordination of spectrum use in their regions. These organizations also represent within global forums, such as ITU-R WRCs, the interests of their member countries and their NRAs along with telecommunications providers and the regional industry. In particular, regional spectrum organizations have a WRC preparatory function: administrations in each region submit draft proposals to the regional spectrum organizations, the regional organization adopts common proposals before the WRC in accordance with their own procedures, and the regional proposals are submitted to the WRC on behalf of all of their members.

At the European level, the Electronic Communications Committee (ECC) within the CEPT brings together 48 countries to develop common policies and regulations in electronic communications and related applications for almost the entire geographical area of Europe. The prime objective of the ECC is to develop harmonized European regulations for the use of radio frequencies. Permanent negotiation on conditions of use of spectrum is critical over Europe as it enables adapting spectrum use conditions to industry requirements and national situations. The CEPT ECC takes an active role at the international level, preparing common European proposals to represent European interests in the ITU and other international organizations. In order to achieve these objectives, the CEPT ECC endorsed the principle of adopting a harmonized European Table of Frequency Allocations and Applications (known as European Common Allocation (ECA) and published within the European Communications Office Frequency Information System (ECO EFIS)) to establish a strategic framework for the utilization of the radio spectrum in Europe [5]. The CEPT ECC delivers ECC decisions (measures that NRAs are strongly urged to follow), ECC recommendations (measures that NRAs are encouraged to apply) and ECC reports (result of studies conducted by the ECC) [6]. The implementation of ECC decisions and ECC recommendations by CEPT member NRAs is made on a voluntary basis since CEPT deliverables are not obligatory legislative documents. However, they are normally implemented by many CEPT administrations [6].

Legal certainty on the availability of the identified spectrum for a given usage and under specified conditions is actually accomplished among the European countries that form part of the EU. The European Commission (EC), the executive body of the EU, is responsible to undertake the EU radio spectrum policy aimed to coordinate spectrum management approaches across the EU. This radio spectrum policy was launched by means of the Radio Spectrum Decision (676/2002/EC) [7] in year 2002. Two complementary bodies were set up following the Radio Spectrum Decision to facilitate consultation and to develop and support radio spectrum policy across the EU member states:

1. The Radio Spectrum Policy Group (RSPG), a group of high-level representatives that advises on broad policy in the area. Comprising the EC and the spectrum authorities of the EU member states, the RSPG provides the EC with advice on high-level policy matters in relation to spectrum. Representatives of the European Economic Area (EEA) countries, the European Parliament and the regional and international bodies may attend as observers. Before being transmitted to the commission, the RSPG’s expert opinions are submitted to public consultations of all spectrum users, both commercial and non-commercial, as well as any other interested stakeholders. Therefore, the RSPG constitutes a unique platform for member states, the EU and all relevant stakeholders to discuss and coordinate regulation of radio spectrum.
2. The Radio Spectrum Committee (RSC), which assists the EC in developing specific regulatory implementation measures and can send EC mandates¹ to the CEPT ECC. In this way, EC can request CEPT to provide frequency allocations in support of EU policies, and CEPT outputs can be then codified into EC decisions, which are legally binding. An MoU between the EC and CEPT ECC defines the basis of their cooperation. In addition, an important element bridging the EC and the ECC regulatory frameworks is the ECO EFIS managed by ECC and that the EC decided to be ‘the European Common Spectrum Information Portal’ (EU member states are obliged to publish their frequency information in EFIS).

Therefore, harmonized conditions can be established on a European-wide basis either through an EC decision² (which is mandatory for EU member states to implement) or by implementing the aforementioned CEPT ECC decisions or recommendations.³ Deviations from EC harmonization measures are possible but necessitates that the EU member states request derogation to be granted by the EC, expectedly for a limited duration. In addition, as a result of a recent update of the regulatory framework for electronic communications services (ECS), the EC is now allowed to submit legislative proposals to the European Parliament and Council for establishing multi-annual Radio Spectrum Policy Programmes (RSPP) [8]. The first RSPP, approved in 2012, created a comprehensive roadmap contributing to the internal market for wireless technologies and services, particularly in line with the Europe 2020 initiative and the Digital Agenda for Europe. The RSPP sets general principles and calls for concrete actions to meet the objectives of EU policies. Among the specific actions set out by the RSPP is making available sufficient harmonized spectrum for the development of the internal market for wireless safety services and civil protection. In the wider field of telecommunications regulation, the EU has also established a harmonized regulatory framework for rights of use in the context of electronic communications networks and services (ECN&S) [9]. The ECN&S regulation also establishes some provisions in relation with the management of radio frequencies for ECS. Among them, regulation mandates that member states shall ensure that spectrum allocation used for ECS and issuing general authorizations or individual rights of use of such radio frequencies by competent national authorities are based on objective, transparent, non-discriminatory and proportionate criteria.

Complementing the work of CEPT ECC and EC, the European Telecommunications Standards Institute (ETSI), an industry-led organization, also plays an important role in the European regulatory environment for radio equipment and spectrum. ETSI is officially recognized by the EU as a European Standards Organization. ETSI produces globally applicable standards for information and communications technologies (ICT), including fixed, mobile, radio, converged, broadcast and Internet technologies. In the context of spectrum regulation, among other things, ETSI has the faculty to develop the so-called system reference documents (SRDoc) that provide technical, legal and economic background on new radio systems under standardization. SRDoc are used to inform the ECC so that harmonization measures or other kind of actions might be started to support the new radio system or application.

6.1.3 National-Level Regulatory Framework

At the national level, radio spectrum is managed by a designated competent government department or agency, which is commonly referred to as the NRA on spectrum matters. Examples of NRA are the Federal Communications Commission (FCC) in the United States, Ofcom in the United Kingdom and the Secretary of State for Telecommunications and Information Society in Spain.

Within the framework of action left by the observance of the regional and global agreements, NRAs must consider many factors, such as each country’s socio-economic benefit. Therefore, NRAs will typically consult with all interested parties and seek to release as much spectrum as possible to allow the country to benefit from global economies of scale, interoperability, interference minimization (including with neighbour countries), international roaming and alignment with regional and global agreements. Many trade-offs are likely to be faced by NRA decisions when assigning spectrum. An example would be the conjugation of the direct economic benefits arisen from the assignment of frequencies to commercial mobile communications compared with the less direct economic purposes that nevertheless benefit society in many indirect ways, such as the designation of spectrum for PPDR communications.

Figure 6.1 illustrates the general structure of the national legislation that is commonly in place in most countries to regulate the use of radio spectrum [10]. As depicted in the figure, the elaboration of a National Table of Frequency Allocations (NTFA) constitutes the first step and the basis of radio-frequency spectrum management, being the main instrument of the national legislation to govern the access to frequency bands. An NTFA primarily specifies the radio services authorized by an individual administration along the frequency bands (referred to as ‘allocations’) and the entities that might have access to these allocations (e.g. government use, private use). An NFTA can also include specific frequency assignments to individual users (e.g. reserved bands for the Ministry of Defence) or installations at particular locations (e.g. protection of specific facilities from potential radio interference). Moreover, an NTFA typically shows the internationally agreed spectrum allocations of the ITU RR.

In a second step, specific national frequency assignments allow the fine management of frequency bands in accordance with the rules set in NTFAs, particularly in bands shared by different type of users as well as to ensure the proper coexistence of the spectrum users in adjacent bands (e.g. channel bandwidths, guard bands, transmission power settings).

Finally, users have to be authorized to be able to transmit on specific frequencies (i.e. users are granted rights of use of the radio spectrum, which is property of the state). This is in accordance with Article 18 of the ITU RR that stipulates: ‘no transmitting station may be established or operated by a private person or by any enterprise without a license issued in an appropriate form and in conformity with the provisions of these Regulations by or on behalf of the government of the country to which the station in question is subject’. The term ‘licence’ is understood here in its broad acceptance, basically establishing that the use of spectrum must be explicitly permitted.

The way that authorization is performed differs between the two possible domains of use: ‘governmental’ and ‘non-governmental’.

‘Governmental’ use, or assimilated, covers various sectors (defence, civil aviation, maritime and waterways, public safety, meteorology, science, etc.). In this case, the rights of use are commonly limited to the rights described in the NTFA, and no additional legal acts are conducted to explicitly grant rights of use [11]. Access to spectrum resources by governmental users should be subject to regular review by the NRA.

In the case of ‘non-governmental’ uses of the spectrum, a public legal act issued by the NRA is necessary for the purpose of delivering spectrum usage rights to private entities or citizens. Under the EU regulatory framework, this legal act is referred to as ‘authorization’ (EU Authorization Directive [12]). On this basis, a legal differentiation is established between ‘individual authorization’ and ‘general authorization’ to reflect the obligation or not for the user to be granted individual rights of use before transmitting. For instance, radio applications that do not need individual frequency planning and coordination could be exempted from individual authorization and should be subject to a general authorization.

The different mechanisms in use nowadays to grant (individual or general) authorizations for the use of spectrum can be classified under three main approaches, referred to as ‘licencing regimes’. These three main approaches are individual licencing (also denoted as ‘traditional licencing’), light licencing and licence exempt (also referred to as ‘unlicenced’). This classification, shown in Table 6.1, is established in CEPT ECC Report 132 [13] as a result of a review of the various terminologies used across the CEPT countries. The distinction between the three licencing regimes uses as a first-level differentiation factor the legal basis given under the EU regulatory framework for ECN&S (i.e. individual or general authorization). In this regard, while there is a clear association between ‘individual licence’ and ‘individual authorization’ as well as between ‘licence exempt’ and ‘general authorization’, a ‘light-licencing’ regime can be based on either an individual or a general authorization. In either of these two options, a light-licencing regime is characterized by the existence of an obligation for the spectrum user to register or notify its use to the NRA. This obligation is of administrative nature and necessitates, as a prerequisite for use, that the user contacts the NRA. Such provision remain in the field of ‘general authorization’ as long as they are only meant to allow controlling the deployment and use of the application so as to avoid harmful interferences on radio services, but not to restrict it. Conversely, this provision falls under the ‘individual authorization’ umbrella if associated with possible limitation of the number of users and specific requirements for coordination prior to use.

It is worth noting that the licences that are issued as a result of a ‘light-licencing’ regime based on ‘individual authorizations’ are not different from ‘individual licences’ from a regulatory perspective. In both cases, the licences should contain technical conditions that are necessary and sufficient to avoid harmful interference to other systems and users. However, the process of issuing licences could vary significantly between the light-licencing and traditional licencing schemes. While in traditional licencing relatively complex administrative processes such as beauty contests or auctions could take place, licences under light-licencing regimes may basically involve the use of dedicated IT systems for automatic frequency planning and licence assignment. Also of note is that, as captured in Table 6.1, a licence-exempt regime is a general authorization regulatory regime where radio equipment operates under a well-defined set of regulator-imposed rules (e.g. constraints on maximum transmit power, spectral mask, duty cycle, etc.) and where no provision for registration and/or notification is required to the users of these devices.

6.1.4 Spectrum Management Models

Spectrum management should be designed and carried out with the main goal of ensuring the efficient use of spectrum. Efficient use implies that, given the state of technology, spectrum is channelled to its most productive uses. Importantly, efficient use has to be properly conjugated with the achievement of other non-market objectives such as national security, safety and equal access goals. As spectrum uses change over time, an efficient spectrum management regime also needs to minimize the transaction costs associated with these adjustments.

Historically, access to and use of radio spectrum has been highly regulated in order to prevent interference among users of adjacent frequencies or from neighbouring geographic areas, particularly for reasons of defence and security [2]. Fuelled by technological advances and the consolidation of liberalized regimes in the telecommunications market, there have been in the last years significant innovations in the theory of spectrum management along with gradual changes in practice of spectrum management and regulation. This gradual change follows a growing consensus that past and current regulatory practices originally intended to promote the public interest have in fact delayed, in some cases, the introduction and growth of a variety of beneficial technologies and services or increased the cost of the same through an artificial scarcity. In addition to these delays, the demand for spectrum has grown significantly, highlighting the need for efficient use of all available spectrum in order to avoid scarcity. Those factors are making policymakers and regulators worldwide focus on new spectrum regulation principles with an increasing emphasis on striking the best possible balance between the certainty required to ensure stable roll-out of services and flexibility (or light-handed regulation) leading to improvements in cost, services and the use of innovative technologies. In particular, a pioneering role in the introduction of many innovative spectrum management approaches should be credited to national authorities for radio-frequency management such as the FCC in the United States and Ofcom in the United Kingdom.

In practical terms, three main spectrum management models can be distinguished nowadays:

1. Command and control model. Under this model, individual users are granted rights to use spectrum on an administrative basis in the form of individual licences to private users or just spectrum assignments for governmental use. The usage is often set to be exclusive: each band is dedicated to a single user, thus maintaining interference-free communication. The command and control management model dates back to the initial days of wireless communications, when the technologies employed required interference-free mediums for achieving acceptable quality. Thus, it is often argued that the exclusive nature of the command and control approach is an artefact of outdated technologies. However, an apparent advantage of this model is that services related to public interest could be sustained. This is especially true with regard to government usage of spectrum (e.g. military, PPDR, transportation) as well as for licencing of spectrum used for maritime and aeronautical services or even for services such as over-the-air television, which may not be as attractive as other potential commercial uses in terms of profitability, but they can be retained by the administration as beneficial for the society. Even beyond such uses, however, some governments retain a belief that regulators are best suited to determine which operators should be granted licences, and in some cases, there may be insufficient demand for spectrum to warrant competitive bidding.
2. Market-based model. In this model, individual licences with exclusive spectrum rights of use in certain bands are acquired by market mechanisms such as auctions. Auctions remedy some of the flaws of administrative model, allowing the placing of a market value on spectrum. In addition, secondary markets for spectrum and spectrum licences can be established based on spectrum trading regulation under which both the ownership and use of the spectrum rights of use can change in the course of a licencee’s operation. Transfer control of spectrum licences may be subject to government or regulatory approval. This is a major step beyond the simple auctioning of licences without granting any real flexibility of use. It does, however, require the full specification of what rights associated with spectrum use can be traded and utilized. The trading of spectrum rights combined with flexible usage conditions could substantially benefit economic growth. Typical users of this model are commercial terrestrial operators and satellite operators. The terms of property rights or flexible rights of use are also often used to describe the market-based approach.⁴
3. Collective use of spectrum (CUS) model. The CUS model is actually an umbrella term to designate all spectrum management approaches allowing more than one user to occupy the same range of frequencies at the same time [15]. Examples of licencing regimes that fit within the CUS model are licence-exempt regimes (in this case, the CUS model is typically referred to as ‘spectrum commons’) as well as light-licencing regimes that allow either a restricted or unrestricted number of users to share a common band. Typical users of bands operated under the CUS model are individuals, though commercial telecommunications providers also rely on the use of these bands to wirelessly access to their networks or for traffic offloading. Hence, typical uses of these bands are Wi-Fi as well as many other low-power devices (e.g. remotes, garage openers, sensors). The proliferation of products and services based on Wi-Fi and other short-range radio technologies is a clear exponent of the benefits and economic value that can be brought by using a CUS model to manage some amount of spectrum.

Governments have generally been cautious in determining which approach to use. The result, across the world, is a combination of spectrum management regimes that mainly incorporate legacy command and control regimes for government services, auctions and bidding for many commercial licences and licence-exempt uses for low-power devices. The three spectrum management models have unique advantages and disadvantages, and no single approach is superior on all counts. The optimal spectrum policy should determine the right mix of these methods rather than adopting one model [16].

6.3.1 Spectrum Components

As outlined in Chapter 3 and further developed in Chapter 5, a multilayered communications approach is envisioned for future PPDR communications systems where complementary frequency bands and technologies can be used to deploy wide area (e.g. nationwide) coverage, with sufficient capacity to cater for routine, day-to-day communication requirements, together with the ability to extend coverage and capacity on an ad hoc basis to cope with challenging radio environments (e.g. tunnels, basements) or localized high-capacity demands. On this basis, spectrum needs related to different system components can be distinguished:

• Spectrum for terrestrial wide area networks (WAN). This is the spectrum to be used in the radio sites of the cellular network that delivers the PPDR communications services for day-to-day operations and most emergency scenarios. In this regard, considering the current situation in which voice communications are still supported over narrowband PMR technologies, spectrum estimations for WAN can distinguish between voice and broadband data communications.
• Spectrum for localized, ad hoc deployments. Radio equipment for large events and large disaster situations can be brought to the local area as required. This equipment may or may not be linked with the existing PPDR network infrastructure. Anyway, specific spectrum may be required for this use.
• Spectrum for backhauling. On the basis that additional local capacity can be deployed as required at the incident, it will be necessary to get data traffic to and from the locality, for example, to provide access to the Internet or other remote data sources and to maintain communications with the agency headquarters. Therefore, this may require the establishment of temporary fixed links using UHF or microwave frequencies or satellite links that may require some amount of spectrum to be available for such purpose. This is especially critical in the cases of unplanned events or major incidents, where there may not be any local alternative infrastructure available for backhauling.
• Spectrum for direct mode operation (DMO). DMO is currently an important means of communicating voice and narrowband data services when being out of network coverage. DMO operations could use spectrum designated for the permanent terrestrial WAN or need additional separate frequencies. In the case of supporting DMO also for broadband data services, the need of separate frequencies from the WAN would be dependent on the technical implementation of DMO into the future broadband PPDR solutions.
• Spectrum for air–ground–air (AGA) operations. In addition to the need for terrestrial operation, there may also be the need for AGA operation. AGA in this context means emergency services communication to or from low-flying (typically a few hundred meters over the ground level) airborne objects. These usually involve a video stream being relayed from a camera mounted on a helicopter or unmanned aerial vehicle (UAV) to a monitoring station on the ground. While this application could be regarded as a point-to-point link, it is difficult to deploy very directional antennas because of the aircraft movement. This turns into an increased risk of interference over a fairly wide area. In order to protect the land mobile infrastructure and to ease cross-border operation, there is a need to identify sub-bands or specific channels for airborne communications. As with the DMO case, AGA could use spectrum designated for the permanent terrestrial WAN or need separate frequencies.

6.3.2 Methodologies for the Computation of Spectrum Needs

In general terms, the methodologies used to determine the amount of spectrum required are based on (i) the characterization of the traffic demand in a given area or per incident, (ii) the estimation of the number of cell sites covering the area of interest and of any other radio equipment brought into (e.g. ad hoc network) and (iii) the characterization of the underlying wireless technologies mainly in terms of achievable spectrum efficiencies. This generic approach is applicable to the different spectrum components identified in the previous subsection, though specific assumptions and considerations have to be introduced for each of them. In the case of the estimation of spectrum needs for terrestrial WAN, the following aspects are considered:

• Traffic demand. This accounts for the amount of traffic generated by PPDR users when performing day-to-day tasks as well as when involved in different types of incident scenarios. It is typically represented in Erlangs⁵ for voice services and in kb/s or Mb/s for data services. The demand can be computed from estimations of the number of PPDR users within local/regional/national areas (e.g. based on PPDR population density analysis) and from the characterization of the services that are going to be used. Demand can also be characterized per incident (i.e. incident-based approach characterization), based on the number of effectives needed within an incident response. Demand can be specified per service or as aggregated values for those set of services delivered through the same type of communications resources.
• Number of cell sites covering the area. This information is relevant to determine what fraction of the demand can be served by the radio resources available in a single cell site. The cell size directly impacts on the spectrum estimate. There could be incidents spanning over several cells as well as cells serving the traffic of several incidents. In this latter case, considerations on the location of the incidents within the cell coverage are also relevant for the spectrum computation (e.g. whether the incident place is close or not to the cell edge).
• Characteristics of the technology providing the service. Radio link spectral efficiency is a key performance indicator of radio technologies. The spectral efficiency is given in bit/s/Hz, thus measuring the amount of bits per second that can be transferred per unit of radio-frequency spectrum (Hz). In modern wireless technologies, spectral efficiency is not fixed but dynamically adapted to the radio link conditions (e.g. the achievable spectral efficiency varies between terminals located close to the cell sites and terminals operating at the cell edge). In addition to the characterization of the achievable spectral efficiencies from a system-wide perspective, it is also important to characterize the capacity of the system as to frequency reuse in neighbouring sites/cells. Systems currently used for narrowband PPDR voice services require frequency reuse patterns as high as 12–21 (i.e. the amount of spectrum needed in the network equals to the amount of spectrum needed in a single site multiplied by the reuse factor). In contrast, LTE technology can support much lower reuse factors, being even possible to reach full frequency reuse at the expenses of some capacity and interference trade-offs. In the end, taking into consideration all the factors that reduce the capacity over the air interface (e.g. guard bands, co-channel and adjacent channel interference, channels assigned to other purposes within the band), an overall system spectral efficiency can be quantified in bit/s/Hz/cell.

With all of the above elements, a coarse-grained estimation amount of spectrum can be straightforwardly derived. For example, in the case of data services, a common approach is just to compute the amount of spectrum (in Hz) needed in a single site/cell by dividing the estimated demand (in bit/s) by the achievable spectral efficiency (either average efficiency or efficiency at the cell edge). Once the spectrum needed in a single cell is estimated, the total amount of spectrum can be obtained from multiplying the single cell estimate by the applicable frequency reuse factor. A more fine-grained estimation, if necessary, can be pursued by considering more specifics on the network deployment and traffic distribution as well as other factors such as the possibility to use multicast/broadcast delivery for some applications or the introduction of weighting factors to capture the correlation between environments in coincident busy hours. These fine-grained computations may require the use of computational tools such as the ones used in the commercial domain for radio network planning and optimization. In any case, it is worth highlighting that the outcomes of any spectrum computation are usually very sensitive to the considered inputs and assumptions, so that spectrum estimations can differ considerably if no reference models are established.

A generic methodology for PPDR spectrum calculation based on the aforementioned principles is reported in ITU-R Report M.2033 [18]. The model describes how to address the characterization of four fundamental variables that determine the amount of spectrum required (i.e. demand for a given area, number of sites/cells covering the area, spectral efficiency of the technology providing the service and amount the technology is able to reuse frequencies). The model provides a spectrum calculation for each type of PPDR service, considering both voice calls and data services. The basic equation employed in the model is as follows:

This fundamental equation is applied for the various service categories in both the uplink and downlink paths. In calculating the total demand per cell, the ITU-R model factors in the total seconds of usage per user, the average users per cell and the bit rate for the application in question. Protocol overheads (e.g. channel coding) and service-grade parameters are considered in the computation of the traffic per cell (e.g. introduction of multiplicative factors to account for blocking probabilities). On the other hand, the net system capacity is directly a computation of the system spectral efficiency where factors such as frequency reuse, guard bands and the use of some resources for signalling channels are considered.

Despite some differences in the computation of the model’s inputs and in the applicability of the previously mentioned formula per incident and per cell or accounting for the overall system efficiency, the ITU-R spectrum model calculator has been considered as the baseline approach in several other studies that provide estimations for broadband PPDR spectrum (e.g. CEPT ECC Report 199 [20], US NPSTC [21], Canada DRDC CSS [22]). As an example to illustrate the applicability of this type of methodology, the computation of spectrum needs for day-to-day operations scenarios is briefly explained in the following as addressed in ECC Report 199. The proposed methodology used for such scenarios consists of the following five steps:

• Step 1: Define the incidents (scenarios).
• Step 2: Estimate the total traffic requirement per incident including background traffic.
• Step 3: Calculate the link budgets and cell size.
• Step 4: Estimate the number of incidents that should be taken into account simultaneously per cell.
• Step 5: Estimate the total spectrum requirement based on assumptions on the number of incidents per cell, location of incidents within a cell and spectrum efficiency per incident.

With regard to the computation of the traffic demand, this methodology follows an incident-based approach where traffic is summed over several separate incidents and background traffic is then added. The outputs of each step are summarized in Table 6.4.

In addition to the earlier incident-based methodology, CEPT Report 199 also provides spectrum estimations based on a slightly different methodology that does not treat traffic separately from several incidents. This alternative methodology is based on the use of the Law Enforcement Working Party (LEWP)/ETSI Matrix described in Chapter 2. In particular, individual incidents and background traffic are combined together into the peak load in the busy hour in normal conditions and in emergency conditions. This is considered through a detailed characterization of the data requirements per application addressed within the LEWP/ETSI Matrix. Then, for each application, it is decided whether the application is ‘incident-centric’ or a background application spread out across the cell. This distinction is made to be able to consider either average or incident-specific spectrum efficiencies to be accounted for each application. In some way, this approach can be thought as a bottom-up approach where the estimated traffic from each individual user application and the number of users of that application are considered in creating the load. This is in contrast to the approach followed in the incident-based methodology to account for the background traffic load, which is completely separate from the incidents’ traffic load and estimated mainly based on the characteristics of the likely applications used in routine activities.

Methodologies such as the ones described previously can also be applied to some extent to the computation of the other spectrum components identified in Section 6.3.1. As a particular case, the basic methodology proposed in Ref. [18] has also been used to estimate the minimum bandwidth requirement for broadband disaster relief (BBDR) applications in localized areas using spectrum in the 5-GHz band [26]. However, the specifics of the localized area as well as the diversity of uses for that spectrum may prevent the ability to firmly determine specific spectrum needs. For example, in the case of spectrum for localized and ad hoc deployments, the applications can range from point-to-point, line-of-sight, broadband links with very high spectral efficiency to local area networking achieving much lower spectral efficiencies (e.g. Institute of Electrical and Electronics Engineers (IEEE) 802.11 type deployments with non-light-of-sight connectivity). The net spectral efficiency will then depend on the individual mix of applications. In addition, frequency reuse may also be highly variable. For example, if a region uses a given spectrum band for airborne operations on multiple assets, nearly the entire band can be exhausted by this application because of the difficulty in reusing frequencies.

6.3.3 Spectrum Estimates

Numerous studies have substantiated the spectrum needs for mobile broadband PPDR applications in different countries and regions across the world [20–26]. The main outcomes from a number of studies are summarized in Table 6.5.

There are many other similar studies addressed to compute the spectrum requirements. A study released in June 2013 considered eight Asian countries, namely, Australia, China, Indonesia, Malaysia, New Zealand, Singapore, South Korea and Thailand. The study supports that a minimum of 10 MHz for broadband PPDR is required on the basis of the opportunity cost argument [27]. In a separate study, China estimated between 30 and 40 MHz of spectrum required for broadband. The United Arab Emirates’ telecommunications regulatory authority (TRA) also conducted a PPDR spectrum study that concluded that PPDR use could in theory be supported in as little as 2 × 5 MHz of spectrum and that an allowance of 2 × 10 MHz would enable reasonable future growth [28].

Summing up, despite the differences across some of these estimates, reserving a minimum of 2 × 10 MHz for mobile broadband PPDR is becoming the prevailing option, though not excluding additional country allocations to meet specific needs.

6.4.1 European Region

At the European level, the band 380–385/390–395 MHz is in widespread use for permanent narrowband⁶ PPDR systems (mostly TETRA and TETRAPOL national and regional networks). Indeed, this 5 + 5-MHz block is the only harmonized band at the European level for narrowband systems, as established in ECC Decision (08)05 [29] and reflected in ITU Resolution 646 as the preferred core harmonized band within the tuning range of 380–470 MHz. Certain channels within this band have also been identified at the European level for DMO and AGA purposes to protect the land mobile infrastructure and to ease cross-border operation (provisions established in decisions ERC/DEC/(01)19 and ECC/DEC/(06)05).

ECC Decision (08)05 also establishes that sufficient amount of spectrum shall be made available for wideband⁷ digital PPDR radio applications within the available parts of the frequency range 380–470 MHz. However, the reality is that these frequencies are being heavily used by non-PPDR applications (e.g. civil PMR systems) in most European countries, preventing the designation of additional spectrum for the use of wideband PPDR systems [23].

In addition to the frequency designations for narrowband and wideband PPDR systems, ECC Recommendation (08)04 [30] establishes that administrations should make available at least 50 MHz of spectrum for digital BBDR radio applications. In particular, two possible frequency bands are identified: 5150–5250 MHz (preferred option) and 4940–4990 MHz. Such amount of spectrum is expected to allow PPDR agencies to implement on-scene broadband wireless networks (e.g. ‘hot spot’ access points in localized areas, temporary incident command centres erected at an incident scene) as well as deploy temporary point-to-point radio links. Nevertheless, the implementation of this ECC recommendation so far is quite limited across European countries (according to Ref. [31], only one third of CEPT countries have actually implemented it).

It is worth noting that many European countries also have national frequency designations for PPDR in the VHF frequency range, which are not harmonized throughout Europe. Moreover, some countries have already reserved spectrum specifically for AGA use (e.g. the United Kingdom has aggregated approximately 42 MHz between 3.1 and 3.4 GHz for digital video from airborne vehicles; Germany has allocated spectrum for that purpose in the 2.3-GHz band). However, at the moment, there is no European harmonized frequency band to support this type of applications.

A summary of the main bands with PPDR spectrum designations across European countries is given in Table 6.6.

It is remarkable that no spectrum designations exist as of today in Europe for the deployment of WAN for mobile broadband PPDR services. Recognizing the need to provide the PPDR community with mobile broadband services and the benefits associated with European-wide harmonization of spectrum use for PPDR, the identification of new suitable spectrum bands for mobile broadband PPDR is on the agenda of the European institutions. Importantly, the RSPP approved in 2012 [8] adopted the following commitment in its Article 8.3:

The [European] Commission shall, in cooperation with the Member States, seek to ensure that sufficient spectrum is made available under harmonised conditions to support the development of safety services and the free circulation of related devices as well as the development of innovative interoperable solutions for public safety and protection, civil protection and disaster relief.

This commitment arrived after a series of efforts and initiatives fostered at different levels across the EU. In fact, back in 2009, the Justice and Home Affairs (JHA) Council of the EU approved a recommendation (Recommendation 10141/09, commonly referred to as the COMIX recommendation [33]) on improving radio communication between operational units in border areas. The COMIX recommendation concluded that law enforcement and public safety radio communication systems will need to support and to be able to exchange high-speed mobile data information beyond the capabilities of current networks, and a common standard operating in a harmonized frequency band will make this possible. Consequently, the COMIX recommendation suggested that CEPT/ECC should be tasked to study the possibility of obtaining sufficient additional frequency allocation below 1 GHz for the development of future law enforcement and public safety networks. Accordingly, the LEWP supporting the JHA Council activities established a Radio Communication Expert Group (RCEG) to work on relevant radio and spectrum matters, which asked CEPT/ECC to take into account the PPDR needs for a mission-critical broadband solution and for this purpose to designate harmonized frequencies [34].

It is worth noting that the current approach towards the harmonization of the broadband PPDR sector in Europe is not targeting a common frequency designation across Europe. Instead, a number of deployment options and spectrum bands for the delivery of broadband PPDR services in Europe are intended to be recognized [32]. In this way, national administrations could opt for the most suitable option or combination of options according to its particular national circumstances. At the end, this approach is believed to provide for the minimum harmonization needed to facilitate adequate interoperability and contribute to maximizing the benefits from the economies of scale for the delivered PPDR solutions.

The gross of the work towards the specification of this harmonized mobile PPDR communications framework in Europe is currently being conducted within the Frequency Management Project Team 49 (FM PT 49) at the CEPT/ECC level. This work is being addressed in cooperation with ETSI and other key organizations such as the RCEG/LEWP. It is expected that FM PT 49 work will eventually culminate in a CEPT/ECC deliverable by 2016 defining the harmonized framework for PPDR communications in Europe. More details on the expected deliverables and the roadmap established by FM PT 49 are given in Section 3.4.2.

Concerning the spectrum bands under consideration for this harmonized European PPDR framework, two candidate spectrum ranges have been identified within FM PT 49: 400 MHz (410–430 and 450–470 MHz) and 700 MHz (IMT-band, 694–790 MHz). The selection of the most suitable spectrum band for mobile broadband PPDR needs to assess and balance many different aspects such as:

• The potential availability of the band(s) for broadband PPDR and the associated timeframe
• The cost of the band’s refarming
• The available bandwidth and contiguity of the band
• The radio propagation conditions
• The risks and character of likely interference
• The potential for the development of business ecosystem in this band(s)
• The potential to achieve harmonization with band plans in other regions to benefit from economies of scale

Some main considerations in this regard concerning the two identified candidate spectrum ranges are given in Table 6.7. As previously mentioned, it is important to stress that the provisioning of broadband PPDR in Europe is not intended to be restricted to the use of 400- or 700-MHz spectrum. Instead, it is envisioned that PPDR capability can be enhanced by allowing mobile broadband PPDR services to be delivered across different bands (both below and above 1 GHz) and networks (both dedicated and commercial). Therefore, it is central that manufacturers produce multiple band integrated chipsets, including the designated PPDR ranges, using a common technology for PPDR user terminals ideally on a global or regional basis. In fact, leading chipmakers in the mobile industry have nowadays solutions that support access to over 12 spectrum bands, many of them under 1 GHz [39].

In addition to the harmonization of spectrum suitable for the deployment of wide area mobile broadband PPDR networks, CEPT is also conducting studies that may end up with the allocation of additional harmonized spectrum for other PPDR-specific uses [40]. In particular, ECC is developing CEPT Report 52 in response to a recent EC mandate to undertake studies on the harmonized technical conditions for the 1900–1920 and 2010–2025 MHz frequency bands (unpaired terrestrial 2-GHz bands) in the EU. The aim of the report is to assess and identify alternative uses of the underused unpaired terrestrial 2-GHz bands other than for the provision of mobile broadband services through terrestrial cellular networks, as well as the development of relevant least restrictive technical conditions for spectrum use. Potential harmonized uses of this band consider BB PPDR ad hoc communications (e.g. local video links) including, for some countries, PPDR broadband AGA applications (e.g. video links between terrestrial units and helicopters). In addition, spectrum at the 5-GHz frequency range (i.e. 4940–4990 MHz) identified for BBDR radio applications in ECC/REC/(08)04 may be subject to further harmonization activities within CEPT.

6.4.2 North America

In the United States, PPDR agencies can licence spectrum in seven separate frequency bands where the FCC has allocated spectrum for public safety use over the years [21]. Those bands, together with indications on the amount of spectrum available in each and its uses, are listed in Table 6.8. As shown in the table, there are several bands in use for narrowband voice and low-speed data systems. Each voice band has unique propagation characteristics, and each band is good or bad for different types of systems. The 30–50-MHz band is primarily used in some statewide systems to provide mobile coverage of highways. The VHF band is a good band for rural areas, while the 450- and 700/800-MHz bands are used in urban and suburban areas where good portable coverage is needed. The 700/800-MHz bands are best suited for trunking systems and increasingly are being used for large regional and statewide systems to provide improved communications and interoperability across multiple agencies and jurisdictions. Although not shown in the table, some PPDR entities in the United States currently utilize systems in some geographic areas in the 470–512-MHz band, known as the T-Band. However, the US Congress mandated in 2012 that public safety agencies with T-Band systems vacate the spectrum by 2021.

With regard to mobile broadband PPDR, the frequencies that have been allocated in the United States are those in the range 758–768/788–798 MHz. Indeed, this is the spectrum assigned to FirstNet for the deployment of the nationwide public safety network. It is also worth noting that the Spectrum Act passed in 2012 also allows a flexible use of broadband in the 700-MHz narrowband spectrum (769–775 and 799–805 MHz), but any move to do so would first need to consider the potential for interference between broadband and narrowband systems.

In addition, the 4.9-GHz band is available in the United States for short-range broadband data and point-to-point data links. The FCC reallocated this band (4.94–4.99 GHz, 50 MHz of spectrum) from federal government use in 2002 and adopted rules designating the band for public safety services in 2003. The band can be used for any terrestrial-based radio transmission including data, voice and video. All multipoint and temporary (<1 year) point-to-point links are primary users of the band. Permanent point-to-point links are secondary users and require separate site licences. The FCC has implemented a geographic licencing scheme for mobile applications. A licence grants a PPDR agency authorization to use all 50 MHz of spectrum within its legal jurisdiction whether that jurisdiction is a state, town, city or county. Licencees must share the spectrum and coordinate frequency use. However, the overall operation can be made very reliable because the general public is prohibited from using the band. This band is currently under review by the FCC to promote greater use of the band, including opening up a part of the band for commercial applications and finding ways to complement the nationwide interoperable LTE public safety broadband network currently in development [41].

The United States was the first country to allocate broadband PPDR spectrum in the 700-MHz band. Initially, 5 + 5-MHz spectrum was allocated, though early in 2012 the US Congress passed a law consolidating an additional 5 + 5 MHz (the so-called D-Block) that sat contiguous to the initial allocation. The channelling arrangement for the 700-MHz band in United States is illustrated in Figure 6.3. Canada is following a similar allocation.

For the deployment of LTE equipment in the US band plan, the 3GPP has designated four operating bands: 12, 13, 14 and 17 (Band 17 is a subset of Band 12). The spectrum already assigned for mobile broadband falls within Band 14 (758–768 and 788–798).

6.4.3 Asia-Pacific and Latin America

In the Asia-Pacific region, the 806–824-MHz segment is widely used for the provision of narrowband communications in support of PPDR applications. This allocation is harmonized through ITU Resolution 646.

The 800-MHz band is also considered by some countries (e.g. Australia, Singapore) for the designation of new spectrum for mobile broadband PPDR, while other national administrations consider the 700-MHz band (South Korea, Mexico, United Arab Emirates) [42]. In the case of Australia, the NRA (ACMA) is undertaking a number of initiatives to improve spectrum provisions for public safety. The most important are:

• Making provision for 10 MHz of spectrum from the 800-MHz band for the specific purpose of realizing a nationally interoperable a PPDR mobile broadband cellular 4G data capability. This band supports 4G (LTE) systems, and as such, it is considered to be ‘beach front’ spectrum by carriers and PPDR agencies alike. The actual frequencies to be provided within the 800-MHz band will be determined later in the context of the ACMA’s review of the 803–960-MHz band.
• Enabling 50 MHz of spectrum from the 4.9-GHz band for PPDR agencies. This spectrum is recognized internationally as a PPDR band in Regions 2 and 3, capable of extremely high-capacity, short-range, deployable data and video communications (including supplementary capacity for a WAN in locations of very high demand).
• Implementing critical reforms in the 400-MHz band, where spectrum has been identified for the exclusive use of government, primarily to support national security, law enforcement and emergency services, is ongoing.

The channelling arrangements for the 700 and 800 MHz are illustrated in Figure 6.4, showing some of the considered allocations for mobile broadband PPDR.

The segmentation of the 700 MHz follows the harmonized APT band plan, commonly known as APT700 band plan. This band plan has been standardized by the 3GPP into two operating bands: Band 28 for FDD operation and Band 44 for TDD operation. The FDD option (Band 28) has attracted the most support from the industry so far. Indeed, many countries have already allocated, committed to or recommend allocating APT700 FDD (Band 28) spectrum for LTE deployments. It is worth noting that the APT700 band plan is not compatible with the 700-MHz arrangement adopted in the United States, since both rely on different channel bandwidths and channel locations within the 700-MHz band.

In a further boost to the dominance of the 700-MHz band for PPDR LTE systems, it was recently announced that Brazilian public safety, national defence and critical infrastructure services will be allocated spectrum in this band. Significantly, in a departure from the common South American practice of adopting the US spectrum planning arrangements, the announcement stated that the APT 700-MHz band (Band 28) would be used [43]. In a related effort, the CITEL recommended to its member states across North, Central and South America that PPDR broadband spectrum allocations should be made along the 700-MHz band, considering either the APT 700-MHz band (Band 28) or the US band (Band 14).

6.5.1 Spectrum Sharing Models

Spectrum sharing may be achieved by numerous methods such as coordinating time usage, geographic separation, frequency separation, directive antennas and so on. In the past, the employment of spectrum sharing mechanisms has typically been on a static, pre-planned basis. For example, the simplest means of spectrum sharing is the operation of systems in the same frequency band but in different geographical areas. Nevertheless, spectrum sharing can be more complex and allow sharing frequencies in the same geographical area. For instance, cognitive radio (CR) technologies (e.g. geo-location databases (GLDBs), spectrum sensing) can be utilized to adapt, in a dynamic and flexibly manner, the spectrum used by PPDR communications equipment considering the presence and activity of other users of the shared band.

A general classification (not specific to PPDR) of spectrum sharing models is based on two defining features [49]:

1. Whether the spectrum sharing arrangement comprises primary–secondary sharing or sharing among equals. In the former case, some systems have the right to operate as a primary spectrum user, and policy mandates that secondary devices are not allowed to cause harmful interference to a primary system.⁹ In the latter case, all devices have equal rights, and typically there is more flexibility about how to behave in the presence of peers.
2. Whether sharing is based on cooperation or coexistence. In a model based on cooperation, systems or devices sharing the band must communicate and cooperate with each other to avoid mutual interference. With a coexistence model, devices try to avoid interference without explicit signalling (at most, devices sense each other’s presence as interference and apply ‘good-neighbour’ sharing practices to use the common resource).

Based on these two above-mentioned features and on the types of licencing regimes and spectrum management models discussed in Section 6.1, three categories of spectrum sharing models for PPDR use can be distinguished [50]:

1. Models based on the dynamic transfer or coordination of individual spectrum rights of use between sharers so that, at a given time and location, there is only one user authorized to use the spectrum.
2. Models based on primary–secondary sharing, where there is a primary user that holds individual rights of use for a given spectrum band but multiple secondary users are allowed to access the spectrum in an opportunistic manner whenever the primary user is not affected. Two variants of this model are distinguished based on the use of either cooperation (through coordination mechanisms) or coexistence approaches between the primary and secondary users.
3. Models based on the CUS of a shared band, where multiple distinct users with equal rights operate in the same range of frequencies at the same time and in a particular geographic area under a well-defined set of sharing conditions. Two variants of this model are also distinguished based on the use of either cooperation (through coordination mechanisms) or coexistence approach between the sharers.

The feasibility of each sharing model for PPDR communications primarily depends on the type of users involved in the sharing framework, which may involve users across the commercial, military and PPDR domains. The adoption of a given sharing model may require changes to the organizational structures and relationships among these users. In some cases, these changes are just an extension of existing agreements (e.g. joint procedures for disaster management between military and public safety entities in large natural disasters). In other cases, new agreements (e.g. sharing rules or conditions among users) must be put in place. Moreover, the amount of changes required in the existing infrastructures managed by these different communities is another important aspect to consider. Any proposed sharing model should minimize the changes to the existing infrastructures.

The adoption of a sharing model is also dependent on the development of suitable technologies and regulatory frameworks. Different sharing models may require more or less complex modifications to existing standards and undertake different technical challenges. In some cases, the technical requirements for specific functions (e.g. spectrum sensing case of models built upon CR technologies) may be difficult to implement with existing technological capabilities (e.g. computing/processing power). Furthermore, international and national spectrum regulations must be modified to permit the deployment of some of the sharing models.

Based on earlier observations, Tables 6.9, 6.10 and 6.11 describe the principles, applicability and some examples for the three identified categories of spectrum sharing models. The tables also provide a discussion on the suitability of each model considering organizational and operational aspects of the involved users as well as relevant technical and regulatory initiatives that can contribute to pave the way towards their adoption.

6.5.2 Shared Use of Spectrum Based on LSA

LSA is a new spectrum regulatory approach that facilitates the introduction of new users in a frequency band while maintaining incumbent services that may exhibit low or localized utilization in the band. LSA fits under an individual licencing regime, so that LSA licences are to be granted to the new users. LSA aims to ensure a certain level of guarantee in terms of spectrum access and protection against harmful interference for both the incumbent(s) and LSA licencees, thus allowing them to provide a predictable quality of service (QoS) (i.e. each user would have exclusive individual access to a portion of spectrum at a given location and time). Therefore, LSA excludes concepts such as ‘opportunistic spectrum access’, ‘secondary use’ or ‘secondary service’ where the new user has no protection from primary user(s). Remarkably, LSA applies only when the incumbent user(s) and the LSA licencees are of different nature (e.g. governmental vs. commercial users) and operate different types of applications, being subject to different regulatory constraints. Hence, an LSA sharing framework is expected to have limited impact – likely no impact – on the market regulation policy objectives since incumbent and LSA licencees belong to two different vertical markets. From the incumbent’s perspective, LSA could be an alternative to spectrum refarming processes that might be too costly to be implemented, not possible in a reasonable timeframe or simply not desirable.

The LSA¹⁰ framework was proposed by the EC’s RSPG in 2011 [61]. The RSPG Opinion on LSA [62] approved in November 2013 provides the following definition:

A regulatory approach aiming to facilitate the introduction of radiocommunication systems operated by a limited number of licensees under an individual licensing regime in a frequency band already assigned or expected to be assigned to one or more incumbent users. Under the Licensed Shared Access (LSA) approach, the additional users are authorised to use the spectrum (or part of the spectrum) in accordance with sharing rules included in their rights of use of spectrum, thereby allowing all the authorised users, including incumbents, to provide a certain Quality of Service (QoS).

It is worth noting that the LSA concept embraces the so-called Authorised Shared Access (ASA) concept, which was firstly proposed by an industry consortium (involving Qualcomm and Nokia) to provide shared access to IMT spectrum under a licensing regime in order to offer services with a certain quality of service [63]. In particular, it could be said that both LSA and ASA concepts are equivalent excepting the fact that ASA was proposed in the context of IMT bands, while the LSA concept has been defined with a broader scope of applicability, not limited to IMT bands. Therefore, the terms LSA and ASA terms are often used interchangeably, and even sometimes some documents use the notation ‘LSA/ASA’ model [64].

A report on LSA (ECC Report 205 [10]) was released by CEPT ECC in February 2014. ECC Report 205 establishes the scope and components of the LSA regulatory approach and provides detailed considerations towards the potential implementation of LSA in the 2300–2400-MHz (‘2.3-GHz’) frequency band in the EU to provide access to additional spectrum for mobile broadband services (referred to them as mobile/fixed communications networks (MFCN)). In this regard, technical harmonized conditions for this band to support wireless broadband ECS are being developed by CEPT in response to a mandate by the EC [65]. In particular, ECC Report 55 [66] and related ECC Decision 14(02) [67] set out the technical conditions to allow coexistence between wireless broadband applications in the 2.3-GHz band and to ensure coexistence with the services and applications above 2400 MHz (e.g. Wi-Fi networks). The ECC Report 55 also contains guidelines on the additional conditions to be applied, on a case-by-case basis, for coexistence between wireless broadband and services below 2.3 GHz. This band is rather different from others where similar initiatives have been applied, because of a range of important incumbent services in limited but specific parts of Europe. In this context, the ECC intends to develop further guidelines to aid administrations in developing an appropriate sharing framework for coexistence between communications networks and incumbent services and applications at the national level (e.g. see [68]).

In parallel, and in close cooperation with regulatory efforts, the ETSI technical committee on RRS is addressing the standardization of an LSA technical framework. In this regard, an SRDoc proposing the adoption of LSA usage in 2.3–2.4 GHz was issued in 2013 [69], and LSA system requirements were released in October 2014 [70]. Work is currently ongoing towards the specification of a system architecture and high-level procedures (to be reported in ETSI TS 103 235 [71]).

A description of the regulatory and technical frameworks of the LSA model is provided in the next two subsections. Then, the use of a potential applicability of the LSA framework for PPDR communications is discussed.

6.5.2.1 Regulatory Framework

The central piece for the implementation of LSA is the establishment of the so-called sharing framework to define the spectrum that can be made available for alternative usage under an LSA model.

The sharing framework can be understood as a set of sharing rules or sharing conditions, which have to be established under the responsibility of the NRA. In particular, the sharing framework has:

• To materialize the change, if any, in the spectrum rights of the incumbent(s) (e.g. delimitate the geographical areas where incumbent spectrum rights are actually necessary)
• To define the spectrum, with corresponding technical and operational conditions, that can be made available for alternative usage under LSA while guaranteeing the protection of the incumbents’ services

The definition of an effective sharing framework requires the involvement of all relevant stakeholders (illustrated in Figure 6.5):

• The Administration/NRA
• The incumbent(s) (most likely to be governmental bodies)
• The prospective LSA licencee(s) (e.g. mobile network operators (MNOs) in the case of the 2.3-GHz band intended to be open for mobile broadband services across Europe)

The Administration/NRA should identify the relevant parties to be involved in the development of the sharing framework. After this, a dialogue between Administration/NRA, incumbent(s) and prospective LSA licencees should be initiated, with the aim of determining the terms of the sharing framework:

• The incumbent should report on the conditions under which LSA can be facilitated. These should include its statistical current and future spectrum requirements in order to operate its services in the band. In particular, it may report frequency band, predefined time, frequency use by geographical area and statistical use of the band as well as other technical conditions such as pre-emption conditions, in case of urgency, where the incumbent may retrieve use of the spectrum.
• The prospective LSA licencees should provide some indication of the minimum duration of the sharing framework required to enable an adequate return on investment. It may also be useful for the LSA prospective licencees to report on the frequencies, locations and times where spectrum is most acutely required. These conditions are needed to ensure the proper spectrum usage by both the incumbent and the LSA licencee in adjacent time/space/frequency domain(s).
• The Administration/NRA should determine the relevant conditions in particular to ensure operations of the incumbent services to be protected. Based on these conditions, the Administration would set the sharing framework, which can be eventually referenced under the NTFA. The administration may also need to modify the incumbent authorization accordingly.

Then, in accordance with the established sharing framework, the Administration/NRA would set the authorization process with a view to delivering, in a fair, transparent and non-discriminatory manner, individual rights of use of spectrum to LSA licencees. LSA does not prejudge the modalities of the authorization process to be set by NRA taking into account national circumstances and market demand (e.g. traditional licencing, light licencing with individual authorization). Granting LSA individual rights of use can also be associated with a number of obligations, as commonly done in traditional (exclusive rights of use) licences.

Depending on the dynamic nature of the spectrum use by the incumbent(s), the LSA licencee may need to be provided (e.g. through a database) with information on the area(s)/time of availability of the spectrum. If this information remains constant over time, it can be provided when the LSA licencee applies for its LSA authorization. Should the incumbent needs to have access to (a part of) the band used by the LSA licencee in accordance with the conditions defined in its LSA authorization, the LSA licencee has to be informed by agreed means and has to modify its use.

The concept of ‘sharing framework’ associated with LSA should not be mixed with a conventional sharing arrangement that is applied for, for example, fixed services such as microwave links or PMR-like services. In such cases, there is no ‘incumbent’ having priority or exclusive spectrum access across a territory, and new systems are typically introduced on a first-come/first-served basis by applying appropriate coordination measures (e.g. geographic frequency separation measures).

6.5.2.2 Technical Framework

ETSI is standardizing a technical framework for the implementation of an ‘LSA system’ in the context of enabling access for MFCNs to the 2.3-GHz band. As such, an ‘LSA system’ comprises one or more incumbents, one or more MFCNs (LSA licencees) and the means to enable the coordination between the incumbents and the LSA licencees, such that the latter may deploy their networks without harmful interference.

The LSA system is designed considering that sharing may in general be dynamic (i.e. the requirements of the incumbent may be such that some portions of the spectrum are not permanently available to the LSA licencee in any given location). In this regard, the set of practical details for sharing a given LSA spectrum resource is referred to as ‘sharing arrangement’, which may be subject to change but should always remain consistent with the ‘sharing framework’ defined by the Administration/NRA. For instance, a particular spectrum sharing arrangement between an incumbent and an LSA licencee may include constraints on the potential variations of resource availability to, for example, facilitate the implementation and operability of the sharing system. Examples of such constraints are:

• Changes in the spectrum resource availability may only occur at preset times (e.g. periodic).
• There may be minimum allowed intervals between successive changes (in general or affecting a given area).
• The availability of spectrum resources may be preconfigured (only a finite set of possible combinations in space/frequency is allowed).
• Changes may not be allowed if they violate certain statistical criteria (e.g. overall availability of a certain resource in a given time frame).

On this basis, the proposed LSA system architecture at the time of writing (work is still in progress) is shown in Figure 6.6. It consists of two functional entities:

1. LSA repository (LR). The LR supports the entry and storage of information describing incumbent’s usage and protection requirements. It is able to propagate this information to authorized LSA controllers (LC) and is also able to receive and store acknowledgement information received from LC. The LR also provides means for the NRA to monitor the operation of the LSA system and to provide the LSA system with information on the sharing framework and LSA licence details. The LR enforces the sharing framework and the licencing regime and may in addition realize any non-regulatory details of the sharing arrangement.
2. LC. The LC is located within the LSA licencee’s domain and enables the LSA licencee to receive or request LSA spectrum resource availability information from the LR and to provide acknowledgment information to the LR. The LC interacts with the licencee’s network in order to convey availability information and support the mapping of this information into appropriate radio transmitter configurations and receive the respective confirmations from the LSA licencee’s network.

Three reference points are defined:

• LSA1: Reference point between LR and LC
• LSA2: Reference point for Administration/NRA interaction with the LR
• LSA3: Reference point for Incumbent interaction with the LR

An illustrative scenario implementing this LSA system architecture is shown in Figure 6.7. The scenario considers the case of an MNO who holds two (traditional) licences to operate some amount of spectrum (referred to as licenced band/carrier A and licenced band/carrier B) over the entire coverage area where its network is deployed. In addition, it is considered that the MNO also holds an LSA licence to use some additional spectrum that is only available in a smaller restricted area due to the presence of incumbents in the rest of the network coverage (referred to as LSA licenced band/carrier C). In this situation, the MNO is responsible for ensuring that only the appropriate base stations (i.e. those out of the grey area depicted in Figure 6.7) can actually operate in the LSA spectrum, either using this spectrum in a stand-alone manner or combining its use with the other licenced bands if carrier aggregation capabilities are in place. To this end, an LC would form part of, or interact with, the network management system (NMS) of the MNO. The LC will receive information on LSA spectrum resource availability over the LSA₁ interface from the LR that governs the access to this band. On this basis, the NMS then would translate the information on spectrum availability obtained by the LC into the appropriate radio resource management commands to the base stations in the operator’s network. Hence, a user terminal located in the area where the LSA band is available can have access to either of the licenced bands and the LSA bandusi or to all of them if it has the appropriate carrier aggregation capabilities. A user terminal located in the area where the LSA band is not available can only use the licenced bands.

From MNO’s point of view, LSA introduces a variability in the amount of spectrum that can be utilized for its normal operation, and in principle, the mechanisms required to utilize the LSA band are similar to the inter-RAT or inter-band load balancing means that already exist in the NMS and are widely deployed in today’s cellular systems. Hence, using the LSA spectrum will require the following steps:

• At the appropriate time indicated by the LC, the MNO’s NMS instructs the relevant base stations to enable transmission in the LSA band.
• If needed, reconfigurations and system information updates of the other networks operating in the underlying band are performed.
• Existing load balancing algorithms in the radio access network (RAN) will make use of the newly available resources and transfer devices to the new band as needed.
• Transferring the devices can be achieved using different techniques such as:
  • Reselection procedures: User terminals in idle mode that migrate in and out of the coverage area of the LSA frequency may reselect on such frequency.
  • Inter-frequency handover procedures: The RAN initiates handover procedures to transfer user terminals in connected mode from the underlying band towards the LSA band.
  • Carrier aggregation procedures: The RAN reconfigures appropriate user terminals (i.e. terminals supporting carrier aggregation between the underlying band and the LSA band) to start operating in a carrier aggregation mode.

Vacating the LSA spectrum will require the following steps:

• When the granted time period for the operation by LC in the LSA band expires or when due to emergency situations, the incumbent requires its spectrum back (e.g. public safety or other incumbents that require that stipulation), the existing load balancing algorithms in RAN will ensure that devices are transferred back to the underlying band.
• Transferring the devices back can be done via the same aforementioned reselection, inter-frequency handover and carrier aggregation procedures.

6.5.3 Shared Use of Spectrum Based on Secondary Access to TVWS

A white space can be defined as ‘a part of the spectrum, which is available for a radiocommunication application (service, system) at a given time in a given geographical area on a non-interfering/non-protected basis with regard to other services with a higher priority on a national basis’ [74]. The concept of allowing additional transmissions in white spaces is a technique to improve spectrum utilization as well as to unlock some spectrum for new uses, provided that the risk of harmful interference to the existing licenced users of the spectrum can be appropriately managed. Most efforts currently underway related to the exploitation of white spaces are focused on the parts of the VHF and UHF frequency bands used for TV broadcasting (i.e. TVWS). The existence of a significant amount of unused parts of spectrum within these bands has been reported in some studies (e.g. see [75]). However, the actual availability of white space spectrum differs significantly among countries (depending on the relevance that terrestrial TV broadcasting service plays in front of other TV delivery platforms such as cable TV or satellite TV) as well as among areas within the same country (e.g. urban and rural areas).

The favourable propagation of radio waves in the VHF/UHF bands and their ability to penetrate deep inside buildings make this frequency range quite valuable. Indeed, many and diverse potential use cases for the exploitation of unused TVWS have been identified such as broadband Internet access in rural areas, wide area machine-to-machine communications, wireless backhauling, rapid deployed networks, in-home networking, etc. [76].

In this context, consideration should be given to the use of TVWS by PPDR equipment to leverage on the long-reaching and highly penetrative signal capabilities of this band, especially in hard-to-reach areas (e.g. tunnels, building basements) as well as in low or very low populated areas where an important part of this spectrum might remain underutilized.

In the regulatory domain, some of the world’s most influential regulators, including FCC in the United States, Ofcom in the United Kingdom and the European CEPT/ECC, Japanese Ministry of Internal Affairs and Communications (MIC) and Singapore’s Info-Communications Development Authority (IDA), are at the forefront of developing the rules for the use of TVWS [58]. In the standardization and industrial domains, the ETSI, IEEE, Internet Engineering Task Force (IETF) and ECMA International have also started several activities to shape future solutions for the so-called white space devices (WSD) or TV band devices (TVBD).

The regulatory and technical frameworks that set the playing field for the potential exploitation of TVWS by PPDR applications are described in the following two subsections. On this basis, a functional architecture of a technical solution for the control and use of TVWS in a LTE-based PPDR network is presented in the last subsection, together with the identification of some regulatory actions that would further contribute to increase the level of dependability that PPDR users could have on this kind of spectrum sharing solution.

6.5.3.1 Regulatory Framework

A first regulation on the use of TVWS was released by the FCC in the United States in 2010 [77] and went officially into effect in early 2011. Since then, some corrections to the rules have been introduced [78], and the operation of TVBDs for commercial use is now a reality in the United States in the TV channels frequencies across the VHF and UHF bands [79].

In Europe, leading efforts are taking place in the United Kingdom towards the allowance of WSDs in the spectrum range between 470 and 790 MHz. While this spectrum is mainly used for digital terrestrial television (DTT), other services such as programme making and special event (PMSE) systems (e.g. use of wireless microphones, talkback systems and in-ear monitors in concerts, sport events and others) have also to be accounted as incumbents to be protected. Regulatory requirements have been already drafted by Ofcom [80], and white space technology is currently being piloted in the United Kingdom [81].

Furthermore, possible harmonization measures for the use of WSDs in the band 470–790 MHz across Europe are also under consideration within CEPT/ECC, which established the Frequency Management Project Team 53 (FM PT 53) in September 2012 [82] to study the potential use of TVWS in Europe. Among others goals, FM PT 53 is intended to provide a master set of the overall requirements for CEPT countries and develop, if appropriate, harmonized regulatory measures to complement the related standardization activities in ETSI with the aim of enabling the development and deployment of WSD while ensuring protection to incumbent services.¹³ Thus far, some technical studies have been performed within CEPT on the technical and operational requirements for the possible operation of (so-called) cognitive radio systems (CRS) in the frequency band 470–790 MHz (ECC Reports 159, 185 and 186 [74, 83, 84]). Based on these CEPT/ECC studies, ETSI produced a harmonized standard that went into effect in September 2014 [85]. The ETSI harmonized standard covers the essential requirements that must be met by WSDs to be conformant with the R&TTE Directive that regulates the placement of radio and telecommunications terminal equipment in the EU market.

Other countries that are moving forwards in the regulation of TVWS are Canada, Singapore and South Africa.

All these regulatory frameworks advocate for a licence-exempt regime for spectrum authorization where WSDs learn about the available WS channels that can be used at a given time and location from a GLDB. The main components of the regulatory framework enabling the operation of WSDs are depicted in Figure 6.8 and briefly described in the following.

The existing regulations consider both fixed and portable WSDs. Fixed WSDs are intended to be deployed at a fixed location with a fixed antenna height. Fixed WSDs typically must store their location, which can either be entered by a professional installer or self-determined using geo-location technologies. On the other hand, portable WSDs must be able to self-geo-locate whenever they move more than a specific distance (typically, it could be 50 m, which is related to the accuracy of the geo-location technology in use). The issues inherent with self-geo-location (e.g. latency, accuracy and time to fix) mean that the use cases for these devices tend to be more limited than those for fixed devices.

WSDs provide their location to a given GLDB, which returns information on the available frequencies and permitted transmission settings (e.g. a list of channels on which they may operate, maximum transmit power, etc.). This approach shifts the complexity of spectrum policy conformance out of the device and into the database. This approach also simplifies the adoption of policy changes, limiting updates to the GLDB, rather than numerous devices. It also opens the door for innovations in spectrum management that can incorporate a variety of parameters. In the current approaches, WSD queries typically provide device location, device information (e.g. type, serial number, certification ID, etc.), antenna height (for fixed) and additional identifying information (e.g. device owner). In the future, GLDB can include other parameters, such as user priority, signal type and power, spectrum supply and demand, payment or micro-auction bidding and more.

GLDBs are likely to be provided by third parties authorized or contracted by regulatory agencies. There is also the case that a regulatory agency decides to manage the GLDB on its own, much like an online licencing system. Regulatory agencies can opt to have single or multiple GLDB providers. Competition among multiple database providers can be beneficial to end users, as it is likely to drive innovation and give users greater choice. Regulatory agencies will collect the details of the incumbents’ usage (e.g. DTT, PMSE systems). Incumbent information can range from very detailed planning information gathered from the incumbent user (e.g. detailed planning of the DTT broadcast network) to the information gathered from the registration processes used to protect entities that are eligible to receive interference protection in TV spectrum (e.g. online registration of licenced wireless microphones to operate in a given location). This information can be delivered to the GLDB providers for them to conduct the calculations to apply protection criteria to these systems (e.g. computation of exclusion zones for specific TV channels and sometimes for the adjacent channel as well). Alternatively, a regulatory agency may decide to carry out the bulk of the calculations in-house and deliver to the database very specific information datasets such as the maximum equivalent isotropic radiated power (EIRP) values that a WSD can use at all locations for a combination of parameters. These two approaches are not exclusive and can be combined (e.g. a regulatory agency must decide to keep all the computations for the DTT service and task the GLDB providers with the computations concerning the protection of PMSE equipment). The protection criteria of the methodologies for the protection calculations are commonly established through regulatory proceedings allowing feedback from all stakeholders.

It is worth mentioning that spectrum sensing can also be used to determine the availability of unused channels, though today’s regulations mainly consider spectrum sensing as an optional feature that complements the mainstream geo-location approach. With spectrum sensing, WSDs would try to detect the presence of the protected incumbent services in each of the potentially available channels. Spectrum sensing essentially involves conducting a measurement within a candidate channel, to determine whether any protected service is present. Ideally, spectrum sensing could be used as a stand-alone technique to determine the usability or not of a given channel, not requiring any existing local infrastructure such as connection to a database. However, the fact that measurements are taken only at a given location and the low-power levels that some incumbent signals might have prevent the full reliance on spectrum sensing. The emergence of cooperative sensing, in which devices share their findings, may bring in the future the potential to improve sensing reliability and reinforce the role of the sensing approach into the overall solution [74].

Based on the previous description, Table 6.12 outlines and contrast some of the main characteristics of the US and UK regulations.

	FCC regulation [77–79]	Ofcom draft regulation [80]
Types of TVWS devices	Fixed devices. Operate at fixed positions and obtain channel availability from databases	Two main, non-exclusive, dimensions to classify devices: Fixed or mobile/portable devices, according to whether the location of the device is permanent or can change Master or slave. A master device directly communicates with a database, while a slave device can only obtain operating parameters from a master device
	Portable/mobile devices. The location of these devices might change. Two different subtypes: Mode I. Do not use an internal geo-location capability and do not access to databases. Channel availability is obtained from Mode II or fixed devices Mode II. Use an internal geo-location capability and access to databases for channel availability
Information provided to the GLDB	Unique device identifier	Unique device identifier
	Antenna geographical coordinates and their accuracy (50 m)	Antenna geographical coordinates and their accuracy (50 m)
	Device type (fixed, Mode I, Mode II)	Technology identifier
	Device’s antenna height above ground level (meters)	Fixed or portable/mobile nature
	Contact information of the owner of the device and of the person responsible for the device’s operation	Lower and upper frequency boundaries and maximum EIRP spectral densities of in-block emissions
		Indoor or outdoor nature (optional)
		Antennas’ characteristics (optional)
Power limits	4 W EIRP for fixed devices	Variable maximum transmit power per location. Power limits are specified in terms of maximum permitted EIRP spectral density (in units of dBm/0.2 MHz). The power limit is a function of the quality of the DTT coverage in the geographical area where the DTT receiver is located. Spatial resolution of the power values that is based on 100 × 100 m geographic pixels. The area of the United Kingdom is covered by over 20M pixels
	100-mW EIRP for personal/portable devices. This is further reduced to 40-mW EIRP in case of operation of these devices in channels adjacent to occupied TV bands channels
	50-mW EIRP for sensing-only devices
	Additional limits on power spectral density (PSD) are defined considering a uniform distribution of transmit power limits over the channel frequencies. Adjacent channel emission limits are also specified	Additional limits are to be set out in terms of through minimum adjacent channel leakage ratio (ACLR) settings
Specification of authorized frequency	A list of permitted TV channels	A list of lower and upper frequency boundaries (not restricted to TV channel boundaries) with a resolution of 100 kHz. Maximum permitted EIRP spectral density for each frequency boundary pair is indicated
Authorization principle	Licence-exempt regime based on geo-location and database access for spectrum authorization	Licence-exempt regime based on geo-location and database access for spectrum authorization
	Spectrum sensing is not mandatory but optional feature	Spectrum sensing is not mandatory but optional feature
	Sensing-only devices also permitted (these devices have additional approval requirements and lower power limits)	No provisions for sensing-only devices
Authorization validity and rechecking	Database must be checked at least once per day. If channel list cannot be refreshed, it times out next day at 11:59 p.m.	Time validity of parameters (in minutes) is to be provided by the database as part of the channel availability response
	Mode II device needs to check its location at least once every 60 s, except in powered-down modes. Rechecking database must be done if location changes by more than 50 m with respect to previous consultation	The database can send instruction for a WSD to cease transmission in 60 s (i.e. the so-called ‘kill-switch’ feature)
	Mode I devices must either receive a special signal from the Mode II or fixed device that provided the list of available channels to verify that it is still in reception range of that device or contact a Mode II or fixed device at least once every 60 s to re-verify/re-establish channel availability
Database awareness of the operation of WSDs	Fixed devices shall register their operation in the database	Registration is not mandated
	No provisions for any kind of reporting back to databases from WSDs once channel availability information has been provided	Requirement to have an acknowledgement of receipt of information from WSDs on used channels and EIRPs

It is also important to note that TVWS operations face some challenges. For example, geo-locating portable WSDs is problematic when trying to support indoor operation or full mobility at high speed. Geo-location accuracy, latency and time-to-fix limitations may not easily support the requirement to re-query the GLDB whenever the device moves more than, for example, 50 m. In addition, the WSD out-of-band emission (OOBE) masks are more stringent than for most unlicenced devices in other bands in order to protect operation close to TV receivers. This OOBE limit has proven difficult to achieve at a cost point low enough to make WSDs further broadly attractive.

Moreover, regulatory uncertainty is also hindering the uptake of TVWS operations. For instance, in the United States, there is an ongoing incentive auction proceeding by the FCC in the 600-MHz band aimed to repack and repurpose some spectrum now used for TV stations and auction it off to mobile licenced services. This poses uncertainties on how much spectrum will be available for TVWS following that auction. Similarly, the future spectrum availability for TVWSD within the TV UHF band in Europe is also unclear [86, 87]. In this regard, co-primary allocations for mobile and broadcasting services between 694 and 790 MHz will be a fact for the EU after WRC-15, so that this band could be used for mobile services in some countries in the near future (around 2020, with some countries like Germany and France already planning the auctioning of this spectrum along 2015), importantly reducing the availability of TVWS. In addition to the 700-MHz band, the possible need and point in time for a further co-primary allocation for broadcasting and mobile services below 700 MHz, that is, 470–694 MHz, is also under discussion for the longer term (beyond 2030) in Europe [38]. However, no consensus has been reached in this case since the sustainability of the current European audiovisual model is highly dependent on the availability of this core spectrum.

In spite of these technological challenges and regulatory uncertainties, the capability of database-enabled devices to operate in vacant TV spectrum without causing interference is already proven (there are hundreds of registrations of database-controlled fixed TVBDs in the United States), paving also the way to apply this concept in other spectrum bands other than the TV bands.

6.5.3.3 Applicability to PPDR Communications

The access to TVWS by PPDR users can bring additional spectrum for PPDR communications. The long-reaching and highly penetrative signal capabilities of this spectrum make it especially valuable in hard-to-reach areas (e.g. tunnels, building basements) as well as in low or very low populated areas where an important part of this spectrum might remain underutilized.

Based on the prevailing regulatory approach being adopted for the operation of WSDs (described in Section 6.5.3.1), the following challenges shall be considered for the potential exploitation of TVWS by PPDR applications:

• PPDR systems intended to exploit TVWS will have to deal with temporary unavailability of specific channels or groups of channels, due to coexistence decisions or use of the channels by primary incumbents (e.g. licenced wireless microphones).
• PPDR systems intended to exploit TVWS may need to deal with interference from other secondary users in the same band or find mechanisms to coexist with those secondary users so that the bandwidth is still used efficiently by the PPDR systems in the presence of these secondary systems.
• Coverage range can be significantly limited by the relatively low authorized transmit power values so that externally mounted antennas (EMA) solutions may be necessary.

A functional architecture for the exploitation of TVWS for PPDR was developed within the EU research Project HELP [72]. This functional architecture, reproduced in Figure 6.9, is intended to enable and control the use of TVWS spectrum in a PPDR network. LTE is assumed to be the technology in use. The solution encompasses the following functional components:

• Geo-location Database (GLDB). As described in Section 6.5.3.1, this is the entity that contains the information about spectrum availability at any given location and time, as well as other types of relevant information related to the white space spectrum. The database can be operated by the spectrum regulator or a third-party entity (e.g. authorized TV band database manager).
• Network spectrum manager (NSM). This functional element is a central control point allocated within the NMS of the PPDR network and used to control the access of a number of LTE eNBs to TVWS in a coordinated way. The NSM directly interacts with the WSD database to acquire the information of TVWS spectrum usage status in a given area. Hence, the NSM serves as a CM (i.e. a sort of SC according to the ETSI framework [91]) so that channel availability information obtained can be used to decide on the most appropriate spectrum allocation of the TVWS resources among eNBs. This NSM can also support priority schemes for the allocation of the available spectrum to the eNBs over specific periods of time. Besides information received from the WSD database, the NSM may be able to collect sensing results from eNBs and terminals and produce a Radio Environment Map (REM) to enhance decision-making. Through the NMS, the PPDR network operator will be able to control and oversee the proper configuration, activation and deactivation of this additional capacity according to PPDR operational needs in an incident area.
• Base station spectrum manager (BSSM). This entity is allocated within eNBs to manage the spectrum use of any individual eNB. Two main modes of operation are envisioned for this entity. In one of these operation modes, this entity mainly interacts with the NSM and is in charge of enforcing decisions coming from this central control point. In this case, whenever some triggering situation occurs (e.g. a temporary base station to be switched on, congestion threshold reached in licenced spectrum, etc.), the BSSM contacts the NSM to get the usable TVWS frequencies. In the other operation mode, this entity provides full autonomy to the eNB for TVWS access. This mode can be exploited in case that the NSM is not deployed or unreachable for whatever reasons, providing an inherent level of redundancy. In this autonomous mode, the BSSM can directly contact the GLDB to get the information of TVWS spectrum usage status and decide on spectrum utilization locally. Spectrum sensing functionalities in eNBs can also be used to improve decision-making. In any case, regardless of both possible operation modes, the BS is always assumed to control all RF parameters of its attached terminals (frequency, EIRP, modulation, etc.) in a ‘master–slave’ relationship.
• UE spectrum manager (UESM). This entity is mainly needed to support UE-to-UE communications over TVWS when there is no BSSM to control the allocation of this spectrum. This functionality allows a UE to behave as a master node for the allocation of RF channels for UE-to-UE communications for, for example, ad hoc network deployment in cases that the PPDR network infrastructure is missing. Hence, UEs embedding this functionality are able to access the WSD database on their own by means other than PPDR network access (e.g. through commercial networks or satellite Internet connection). The UESM can also support relying or proxying functions for other master UEs that cannot reach the database by themselves.

Regarding the communication interfaces depicted in Figure 6.9, interface (A) is used for GLDB access and could be based on the IETF PAWS protocol. Interface (B) forms part of the protocols used by the NMS to remotely access to network elements and manage them, thus leveraging/extending the protocols used for network management. Interfaces (C) and (D) are technology dependent (e.g. LTE interface adapted for use in TV bands, IEEE 802.11af interface) as they actually support the radio transmissions. Interface (E) is needed for the registration of entities that are eligible to receive interference protection in TV spectrum. Web-based interfaces can be a solution for this interface. Finally, interface (F) represents the Application Programming Interfaces (APIs) that are reachable in control room systems (CRS) to develop applications enabling tactical and operational PPDR managers to have control on the capacity and coverage of the network, including the use of TVWS spectrum (e.g. the use of TVWS within the network can be enabled or disabled from control room positions).

The implementation of this spectrum sharing solution requires that the configuration of LTE hardware installed in eNBs can be adjusted to operate in TV UHF bands. There are certain aspects that need to be considered for LTE to operate in TVWS:

• The allowable modes of operation for LTE. Most likely, deployments of LTE in TVWS will be either TDD mode (either in TVWS stand-alone operation or through carrier aggregation with the primary carrier provided via non-TVWS frequencies) or FDD mode in the context of carrier aggregation where the primary carrier is provided via non-TVWS frequencies. In the FDD case, both DL and UL operation should be supported in TVWS via DL component carriers or UL component carriers.
• Robustness of control signalling and data transmission. The operation in unlicenced bands will naturally require a system to function (either temporarily or for a long period of time) under some levels of interference that are caused by other secondary users. Control signalling and data transmission of any system need to take this into account. Silencing gaps in the LTE transmissions can be employed for coexistence purposes with other secondary systems [100].

Indeed, the implementation of this spectrum sharing solution could leverage the technological enhancements being targeted under 3GPP Release 13 for the operation of LTE in unlicenced spectrum [101], commonly referred to as LTE-U. These enhancements are expected to introduce a so-called Licenced-Assisted Aggregation (LAA) operation mode in LTE to aggregate a primary carrier (using licenced spectrum, to deliver critical information and guaranteed QoS) and a secondary carrier (using unlicenced spectrum, to opportunistically increase capacity and data rates). The secondary carrier operating in unlicenced spectrum could be configured either as downlink-only carrier or contain both uplink and downlink. While current work in LTE Release 13 is on the use of LTE in the 5-GHz band, the underlying principles and technologies could be directly extended to other unlicenced bands (e.g. 2.4 and 3.5 GHz, TVWS).

The opportunistic nature of the usage of the TVWS does not fit well with the deployment of services that require provisioning of QoS as there is uncertainty about the availability of spectrum resources and there is no protection from harmful interference from other WSDs. However, different measures could be considered by the regulatory authority to increase the degree of reliability and QoS guarantees on the exploitation of TVWS for PPDR:

• Allowing higher transmit powers for PPDR devices. This approach would be aligned with the ECC/REC/(08)04 recommendation for the allocation of a BBDR band at 5150–5250 MHz, where PPDR devices are expected to coexist with other devices (e.g. Wi-Fi devices) though higher transmit power limits are established for the operation of PPDR devices. The more controlled nature of PPDR equipment and its operation may be turned into less stringent incumbent protection levels requirements for those devices (and so higher allowed transmit powers) as compared to commercial devices.
• Supporting priority access for PPDR WSDs. In this case, if access to radio resources is requested by several WSDs concurrently, the GLDB could treat PPDR WSD devices with a higher priority (i.e. higher precedence than ordinary devices but always remaining below the precedence of the current incumbents).
• Reserving a number of TV channels for PPDR use when an emergency situation is declared by a competent authority. In this situation, GLDB could be mandated to guarantee that a fraction of the available capacity in a given region is excluded from the channel availability information provided to conventional WSDs and only advertised to PPDR WSDs.
• Registering PPDR base stations in the TV bands database and establishing protection criteria to prevent interference to these stations when an emergency situation is declared by a competent authority. In this case, PPDR equipment would form part of the incumbent radio services/systems authorized for operation with a regulatory priority (e.g. PPDR devices will be registered similarly as it could be done for PMSE equipment). Another option could be to create a second, high-priority tier of WSD above the general authorization tier. This would require that the regulator defines rules for the coexistence of WSD so that higher tier devices do not get interfered. Access to the higher-priority tier could be through a licence. Indeed, the realization of a three-tiered shared access system is currently being pursued by the FCC in the United States to create a new Citizens Broadband Service (CBS) in the 3550–3650-MHz band managed by an SAS based on the use of geo-location-based opportunistic access technology [48]. In this case, the first tier (denoted as incumbent access) includes authorized federal users and grandfathered fixed satellite service licencees. These incumbents are afforded protection from all other users in the 3.5-GHz Band. The second tier (denoted as priority access) includes critical use facilities, such as hospitals, utilities, government facilities, and public safety entities that are afforded quality-assured access to a portion of the 3.5-GHz band in certain designated locations. The third tier (denoted as general authorized access) includes all other users, including the general public, that have the ability to operate in the 3.5-GHz band subject to protections for incumbent access and protected access users and can use the spectrum when incumbent and priority access users are not using it.