Various arrangements are currently in use by governments and PPDR organizations for the procurement of PPDR communications services over dedicated narrowband PMR networks. The defining elements commonly used to characterize such arrangements include:
In European countries, the GO–GO and GO–CO models along with the designation of some amount of spectrum for dedicated PPDR use are the most common arrangements for current voice-centric PPDR networks. In these arrangements, the government, directly or via a controlled entity, procures equipment, deploys and operates a closed network dedicated to the PPDR users. These networks are typically shared by several PPDR organizations such as police, firefighters, medical services and civil protection of the same country or region. In many countries, the government creates a ‘pseudo-independent’ operator, often state-owned, in order to take responsibility for the build, management and financing of the network. Examples of GO–GO models are the ASTRID network in Belgium and VIRVE network in Finland. Examples of GO–CO networks are BDBOS in Germany and BOS in Austria. To a lesser extent, the CO–CO model is also deployed in some European countries. As in the case of GO–GO and GO–CO models, the designation of dedicated spectrum for PPDR is the prevailing approach in these CO–CO arrangements. Examples of CO–CO networks are Airwave in the United Kingdom and SINE in Austria.
In the United States, while the prevailing model is still largely based in disparate and separate analogue or P25 radiocommunications system owned and operated by individual PPDR agencies, many efforts have been conducted in the last years towards the realization of regional (multi-county) or state-wide shared networks that could provide service to multiple agencies. In this regard, different models are followed for the governance structure of these shared networks that range from direct contractual agreements among the participating PPDR agencies to governance models enacted by (state) legislative actions that usually also include funding mechanisms for construction [1]. A common denominator of these arrangements is that the network remains under the ownership of government entities. Nevertheless, another PPDR communications delivery model in which the services are provided by a private company (e.g. a telecommunications carrier) is also gaining momentum in the United States. This model is commonly referred to as carrier-hosted ‘P25 as a service’, which draws many similarities with the CO–CO concept previously discussed. This service model has been offered by Bell Canada in Canada for years, while its adoption in the United States is still at an incipient stage [2].
The adoption of the LTE standard brings up a major change in the way that PPDR services can be provisioned with respect to current narrowband voice-centric PMR systems. The use of LTE allows the service layer, which comprises the specific applications and service delivery platforms (SDPs) needed for PPDR, to be decoupled from the underlying network layer, whose main task becomes the delivery of IP-based connectivity to the service layer components. This approach, illustrated in Figure 5.1, follows the principles and general reference model for next-generation networks (NGNs) that establishes a functional division between [3]:
According to this model, the ‘PPDR service layer’ can be implemented mainly as application clients within the PPDR terminals and application servers within the PPDR SDPs and control room systems (CRS). This ‘PPDR service layer’ is then deployed over the LTE connectivity services, including individual EPS bearer services, MBMS bearer services and ProSe communications services. It is worth highlighting that such a decoupling does not exist in current narrowband network architectures, where the service functionality (e.g. the protocols and control capabilities used to support the voice basic and supplementary services) is embedded together with the transport functions in the same PMR network platform, as illustrated also in Figure 5.1.
There are important benefits associated with this decoupling. Remarkably, this allows service and network layers to be provisioned and offered separately and, importantly, to evolve independently (e.g. applications technology life cycle can be longer and not be affected by subsequent changes in the underlying communications technology). Moreover, this approach facilitates the provisioning of PPDR services independently of the network and access technologies (LTE/HSPA/Wi-Fi as well as fixed access) and of the operator of the IP-based connectivity network, which may be not the same as the operator in charge of the PPDR SDPs. Indeed, more than a single IP-based connectivity network, run either by the same or different operators, could be used to provide access to the PPDR SDPs. In particular, these IP-based connectivity networks could be dedicated networks, run by a PPDR operator or any other specialized player (e.g. an operator in charge of the dedicated communications infrastructure in an airport) as well as public access networks, run by MNOs.
An LTE network intended to support PPDR communications should be designed to resist failures, due to man-made or natural events, as much as possible. In order to establish the desired level of reliability and performance for a mission-critical mobile broadband PPDR solution, the National Public Safety Telecommunications Council (NPSTC) has defined the term public safety grade (PSG) and delivered a number of recommendations to provide guidance to the First Responder Network Authority (FirstNet) as it constructs and implements a US nationwide PPDR LTE network [4]. Qualitatively, PSG communications are defined simply as the effect of reliable and resilient characteristics of a communications system. On this basis, the NPSTC’s report provides measurable characteristics that would differentiate a mission-critical communications system from a standard or commercial-grade network. The report covers in detail environmental considerations, service-level agreements (SLAs), reliability and resiliency elements, coverage design details, push-to-talk (PTT) support, applications, site hardening, installation and operations and maintenance. A network or system intended to be considered PSG must address these topics during its design and later in its implementation.
In addition to the NPSTC recommendations, there are other documents that develop network-related requirements and identify the expected key characteristics for an LTE network to deliver mobile broadband PPDR communications [5–13]. Based on these, Table 5.1 gathers some of the key characteristics that illustrate the expected performance and capabilities of an LTE network designed for PPDR use.
Table 5.1 Key characteristics expected in an LTE network designed for PPDR use.
Areas | Characteristics |
Coverage and capacity | Geography-based coverage planning – as opposed to population coverage |
Symmetric UL/DL usage pattern – as opposed to DL-dominated traffic for commercial users | |
Strongly varying cell load, with limited possibilities to predict (stochastic) | |
Ubiquitous coverage, not just outdoors but inside buildings and other hard-to-reach locations (e.g. subways) | |
Off-network operation, to enable direct communication between PPDR users outside network coverage | |
Deployable systems, to secure wide area PPDR communication also when users are outside normal network reach | |
Support for air–ground–air (AGA) communication | |
Network availability and resilience | Robust network sites with enhanced physical protection and battery backup (site hardening) |
Geo-redundant network functions (intra-network redundancy) | |
Fallback to other networks at network failure (inter-network redundancy) | |
Multiple backup options | |
Hardening should not be a one-size-fits-all approach, but particularized to specific conditions of the sites and facilities locations | |
Network availability at least 99.99% of the time (i.e. <50 min/year of unplanned outages) | |
Security | User data encryption end to end |
Elevated link security, protecting both user plane and control plane | |
Elevated O&M security, securing node and network configurations and stored user data | |
Enhanced identity management, provisioning users with ‘right to use’ of network resources | |
Mutual authentication of infrastructure and terminals | |
Methods for temporarily and permanently disabling terminals and smart cards | |
Functions to detect and compensate for jamming at the air interface | |
Priority control | Differentiated priority classes so that high-priority communications are never blocked |
Dynamic control (real time) of priority and resource management. Also applicable in a visited network | |
Emergency calls | |
Accountability and service assurance | Real-time KPI monitoring to enforce service-level agreements between PPDR users and network operators (PPDR operators and/or MNO) |
Quality-of-service (QoS)-dependent charging to enable charging of PPDR services based on priority and QoS | |
Functionality and performance | Data and voice services and features (e.g. group communications, PTT) in need for PPDR operations (initial deployment may be only for data services) |
Reliability and adequate QoS unaffected by location (fast call set-up, reliable delivery of data) | |
Interoperability | Ability of communications equipment to comply to the same technical standard and to operate within the tuning range of the frequency bands used by these systems (this is especially relevant for European PPDR organizations) |
Seamless operation of broadband equipment across borders in Europe, including ad hoc deployments of extra capacity coverage when and where needed | |
Hierarchical control levels in a multi-jurisdictional shared network | Ability of operations centres at differed jurisdictional levels (e.g. national/regional/local) to exercise control on the network use |
Ability of PPDR agencies to have local network control to secure capacity and ensure their public safety priorities are met |
The definition of the right deployment scenario and associated business model for the delivery of mobile broadband PPDR communications is currently a matter of strong interest for PPDR agencies and governments around the world. As described in the previous section, the ownership of the new infrastructure, the management model for such infrastructure, the eligible users to be served and the potential designation of some amount of dedicated spectrum are among the central defining elements that have to be fixed. In addition, the formulation of a service delivery model for mobile broadband communications based on the LTE technology has to take into consideration the following elements:
Bearing in mind the above considerations, Table 5.2 compiles an overview of different service delivery options for mobile broadband PPDR communications that have been analysed so far [14–20].
Table 5.2 Studies that cover the analysis of delivery options for mobile broadband PPDR networks and services.
Study | Delivery options for mobile broadband PPDR networks and servicesa |
TCCA, ‘A Review of Delivery Options for Delivering Mission Critical Solutions’ [14] |
|
CEPT ECC Report 218 [15] |
|
Simon SCF Associates, ‘Study on Use of Commercial Mobile Networks and Equipment for “Mission-Critical” High-Speed Broadband Communications in Specific Sectors’ [16] |
|
German Federal Ministry of the Interior, ‘On the Future Architecture of Mission Critical Mobile Broadband PPDR Networks’ [17] |
|
APT Report on ‘PPDR Applications Using IMT-based Technologies and Networks’ [18] |
|
a The tagging of the options shown in the table does not obey to any order or preference expressed in the reference studies but mainly added here for the sake of clarity.
Regardless of some slight differences in terminology and focus of the different studies, the potential delivery models under consideration can be broadly categorized as follows:
A description of the different options, with an analysis of its advantages, limitations and specific factors to be considered, is addressed in the following sections.
A dedicated mobile broadband network for PPDR can be specifically designed and built to meet PPDR users’ requirements. This approach is often regarded as the best solution from the PPDR users’ point of view. The primary advantage of a dedicated network is that the government and PPDR users deploy and manage their own broadband network, having full control and guaranteed network access in mission-critical situations and without commercial users to take capacity during an incident. This approach would be the logical continuation of the present scenario with dedicated TETRA and TETRAPOL networks in Europe. Obviously, the main disadvantage of this approach is the high cost (capital and operational) of coping with such a new dedicated infrastructure. Moreover, the procurement process and network buildout could typically take 3–5 years to deliver the first service to users, thereby delaying the start of broadband services for the users of that network. The designation of some dedicated spectrum is a valuable asset that the government can bring up as part of the funding scheme for a dedicated network. However, the designation and clearance of new dedicated spectrum for mobile broadband PPDR can also take some years in those countries that have not initiated this process yet.
Within the dedicated network option, two main alternatives regarding the network’s ownership can be distinguished:
On the basis that the high costs and funding schemes are likely to be the biggest hurdles to overcome in any form of dedicated network, cost-efficient network footprints specifically tailored to fulfil PPDR particular needs are needed (i.e. the design of a dedicated network may not necessarily need to mimic that of a commercial network). In addition, enlarging the user base served through the dedicated infrastructure and finding ways of monetizing the excess bandwidth are some of the key dimensions that deserve careful consideration. These two aspects are further discussed in the following subsections.
Traffic demand in PPDR wireless communications networks is much less predictable, both geographically and temporally, than in commercial networks. Major incidents can happen anytime and anywhere. When such incidents do arise, communications needs are substantial and tend to be concentrated around a relatively small area. It would be impractical, both in economic and engineering terms, to plan a conventional wireless network based on such eventualities, since much of the capacity would never be used.
A more cost-efficient approach is to plan a network to provide a basic minimum level of wireless connectivity at all locations that can be rapidly expanded on an ad hoc basis to provide additional capacity to cater for unforeseen incidents. The rationale of this approach lays on the fact that two types of traffic can be distinguished in a PPDR network: the light traffic induced by routine activities, such as patrols and surveillance, and the heavier traffic due to a large number of PPDR personnel at a major incident scene. The network architecture must be designed to meet the peak capacity requirement due to the sum of both types of traffic.
In the conventional architecture adopted in commercial networks, the radio access infrastructure mainly consists of stationary base stations (BSs) connected spread at fixed locations over the territory. If such conventional architecture is adopted to deploy a wide area PPDR network, the placement of BS needs to be dense enough to meet the peak demand. An alternative approach is to design the radio access infrastructure based on a reduced number of more sparsely deployed stationary BSs for supporting light routine traffic complemented with a distributed set of mobile BSs ready to be quickly deployed to any incident scene by vehicle or helicopter [21]. A premise of this lighter architecture is that a mobile BS can be dispatched to the incident scene and can be set up and operational in a very short time. This imposes a requirement on the density and placement of mobile BSs, as well as on the technologies used to link and integrate the mobile BSs with the operation of the fixed infrastructure (e.g. fast settable wireless backhaul solutions, self-organizing features to automatically configure the BS settings). The analysis conducted in Ref. [21] shows that an architecture partially based on mobile BSs can potentially offer over 75% reduction in terms of the total number of fixed BSs that are necessary to cope with the same traffic demand. The analysis of a range of traditional, cellular-based solutions all the way down to a solution more dependent on mobile deployables is also under consideration by FirstNet in the United States [22]. Cellular designs using 35 000, 24 000 and 14 000 sites have been considered, the latter being a hybrid design with a very thin network and much more heavily reliance on transportable systems that can allow PPDR personnel to bring the network with them for those events that occur in areas where it doesn’t make sense to have a site around the clock. Solutions based on all-deployable public safety LTE networks are also being trialled in the United States [23].
Another compelling approach is the use of extended cell range solutions, also known as boomer cells, to reduce the permanent footprint of a dedicated PPDR mobile broadband network [24]. The conventional wisdom is that PPDR LTE could make economic sense in metropolitan, urban and even some suburban areas, where the capacity demand justifies the establishment of many cell sites located relatively close together, with coverage typically extending less than 4–5 km from the site. However, the case of rural areas raises concern in purely economic terms. For this reason, the use of cells with much larger cell radius for coverage in rural and remote areas can be a relevant capability to consider in the design of a dedicated PPDR network.
Extended cell range or boomer cells have been used for years in various cellular technologies to provide wide area coverage. Testing and evaluation of boomer cells able to extend an LTE cell radius to range of up to tenths of kilometres have been conducted since 2014 by the Public Safety Communications Research (PSCR) in the United States [25]. In particular, initial trials considered an LTE eNB antenna deployed at 85 m height. The transmit power of the eNB was set to 40 W, which is a typical transmit power for a macrocell. PRACH Preamble1 Format 1 was used to allow extended cell range up to 77 km (48 miles). On the terminal side, an LTE vehicular modem with 200 mW (23 dBm) transmit power (i.e. the same as a commercial cell phone) and two external antennas were tested, one omnidirectional car roof antenna with a gain of 3 dB as a practical application for mobile use and another 16 dB gain directional antenna as a practical application for fixed situations. In the case of the rooftop antenna, coverage ranges of around 40 km were demonstrated. In the case of the directional antenna, transmission at a distance of 77 km was possible, achieving a data rate of a few megabits per second. Additional experiments are planned for 2015 considering the eNB antenna placed at 280 m height and a PRACH Preamble Format 3, targeting the extension of the coverage up to 100 km. Besides the increased coverage, raising the antenna height is expected to improve the performance at a fixed location. This is particularly beneficial for the uplink, which is typically the limiting factor in the link budget. The use of high-power devices, which have already been specified by 3GPP for LTE Band 14, is especially relevant in this scenario.
Another big challenge for PPDR communications that could importantly affect the placement and density of the network sites is the capability to provide in-building coverage in large structures, such as office buildings, apartment buildings, warehouses, parking structures, tunnels and basements. In these scenarios, distributed antenna systems (DAS) and signal boosters are typically used to enhance coverage in commercial mobile communications. Deploying this type of solutions to improve indoor coverage in some critical locations could ultimately reduce the number of outdoor macrocells. Regulation could be a key factor to favour this case by issuing rules mandating, for example, the support of in-building DAS for public safety in new constructions. Indeed, installing an in-building system for cellular users to have good connectivity is almost a necessity for building owners in today’s environment. The relatively low cost associated with adding public safety support to the cellular support that the market demands and the maturity of the DAS technology for 700/800 MHz make this a compelling approach [26, 27]. However, special attention should be paid to the requirements on the PPDR DAS (e.g. UPS requirements, percentage of coverage in critical areas, components enclosures requirements, certification and periodical testing). Much more stringent requirements for PPDR DAS compared to carrier-grade DAS could be a deterrent for the adoption of a common solution for both domains [28, 29].
Finally, yet importantly, the adoption of small cells is also expected to have a huge impact on the architecture of next-generation mobile broadband PPDR networks [30]. As mobile operators roll out small cells to enhance LTE coverage, the technology is likely to expand to mission-critical broadband networks as well. ‘Small cells’ is an umbrella term for operator-controlled, low-powered radio access nodes, including those that operate in licensed spectrum and unlicensed carrier-grade Wi-Fi. Small cells typically have a range from 10 to several hundred metres [31]. Types of small cells include femtocells, picocells and microcells – broadly increasing in size from femtocells (the smallest) to microcells (the largest). These small cells, along with the mainstay macrocellular tower and rooftop antennas, make up what is known as a heterogeneous network (HetNet). Placing small cells in key locations such as police stations, fire stations and major government buildings will allow public safety an opportunity to cost-effectively add the coverage and capacity they need. Furthermore, advances in small cell technologies will also undoubtedly benefit the development of the mobile BS solutions discussed previously in this subsection.
Enlarging the user base of a dedicated infrastructure can greatly contribute to its economical sustainability. An expanded user base beyond the PPDR responders group would help to achieve higher economies of scale and monetize the excess capacity that a private LTE network only serving PPDR traffic may have. On a day-to-day basis, not all of the broadband capacity available on a dedicated network would be certainly used. Therefore, leveraging the priority features supported in the LTE standard, part of the network capacity could be shared with non-first-responder users to generate revenue that could be used to fund the operations and upgrades of the broadband network.
Besides the PPDR community, a dedicated network purposely built to support mission-critical communications can also become a compelling business proposition for other users who depend on efficient mobile communications to carry out their job and whose daily activities can also be fundamental to the health, safety and well-being of the citizens [20]. They comprise critical infrastructure services and other public and private entities such as [14, 16]:
A detailed compilation of the requirements, operational procedures and functional and safety needs for the utility and transportation sectors can be found in Ref. [16], together with a description of the wireless equipment and networks currently in use in these sectors.
There are strong arguments for these potential users having access to the same reliable network as PPDR:
Nevertheless, there are also some challenges that could prevent the sharing of a common dedicated infrastructure for various critical communications if these issues are not properly addressed:
In the United States, the possibility of collaborating with utilities and transportation on the buildout of the FirstNet broadband network is being considered [33, 34]. Indeed, in the law enacted in February 2012 that created FirstNet, language was included that allows a broader definition of the first responders, opening the door to embrace critical infrastructure users. Just for illustrative purposes, US commercial carriers like Verizon and AT&T are able to spread the capital costs of their nationwide networks across more than 100M subscribers. In contrast, the dedicated broadband system being pursued by FirstNet, which is expected to be more reliable and with greater coverage than the commercial counterparts, could count with a user base roughly estimated to be as high as 5M subscribers if only first responders are allowed access [33].
In a similar matter, the Utilities Telecom Council (UTC) in Canada has requested Industry Canada for the reclassification of some utilities with regard to their eligibility and priority in using the PPDR spectrum that has been allocated for the deployment of a dedicated network. In Canada, a public safety hierarchy has been defined with three categories. Category one users are the traditional first responders such as police, fire and medical services. Category two users include forestry, public works, public transit, hazardous materials clean-up, border protection and ‘other agencies contributing to public safety’. Category three users include ‘other government agencies and certain non-governmental agencies or entities’. Currently, hydro and gas utilities are classified as category three users so that they could only get access to this capacity in emergencies. According to UTC Canada, for utilities to see the worth of the network, they should be given day-to-day access to the spectrum for certain mission-critical applications.
Besides enlarging the user base of a dedicated network with critical communications users, there are other more controversial propositions that even consider supporting some citizen’s communications over the dedicated network. One of these proposals is the so-called Dynamic Spectrum Arbitrage (DSA) concept, proposed by a private company in the United States as a way to enable real-time auctions for bandwidth at any given time or location to monetize the excess throughput in the FirstNet’s network [35]. The rationale behind this approach is that the FirstNet system is expected to have a significant amount of excess bandwidth available during routine times that can be sold to operators to generate revenue that can be used to fund further FirstNet’s deployment and operation. Under the DSA concept, a DSA provider would finance the buildout of the PSG broadband network and would be given the right to sell excess bandwidth to operators, on the basis that first responders would get all of the bandwidth during an emergency response. A Dynamic Spectrum Arbitrage Tiered Priority Access (DSATPA) engine has been designed (and patented). The engine is intended to allow network operators to bid for access to unused broadband capacity on the FirstNet system on a nearly real-time basis, similar to the approach used by utilities in the energy markets. The use of this capacity is expected to be enforced through real-time modifications in the network operators’ subscriber databases, adapting setting such as subscribed quality-of-service (QoS) profiles, access restrictions for roaming and packet data networks to which the wireless devices are allowed to connect. In this way, MNOs’ subscribers can be properly steered between the networks. This flexibility in the management of the excess capacity is expected to be valuable for commercial services that are delay tolerant and can be accommodated easily over time such as software upgrades, M2M traffic, massive data backups, etc.
Another controversial proposal is to use the PPDR dedicated infrastructure also to support emergency communications with the citizens, in particular for the deployment of reliable alerting systems [36]. In this way, every commercial device could be mandated to support the frequency band allocated for PPDR to have such an alerting capability. According to Ref. [36], this approach could, on the one hand, make the chipsets supporting the PPDR spectrum band to become less expensive because of the economies of scale and, on the other hand, justify an additional charge to consumers and business users for the financing of this alerting capability (e.g. if a fee of 50 cents per month were established in the United States across 300M commercial devices, this could generate almost $2B funds each year for the sustainability of the FirstNet network).
As long as commercial broadband networks are becoming an important part of a society’s infrastructure, there is also an increasing consensus that these infrastructures undoubtedly will play a role in the delivery of critical communications solutions. However, there are important differences between the PPDR and commercial network models that should be taken into account when assessing the specific role that commercial networks could have in the overall delivery solution for mobile broadband PPDR communications. A compilation of some key differences is provided in Table 5.3 [16], which shows divergences in many fundamental aspects such as goals, capacity and coverage planning approaches, availability standards, types of communications services in demand, control over subscriber information, etc.
Table 5.3 Main differences between the PPDR network and commercial network models.
From Ref. [16].
Issues | Commercial network operator model | PPDR network model |
Goals | Maximize revenue and profit | Protect life, property and the state |
Capacity | Defined by ‘busy hour’ on a typical day | Defined by ‘worst-case’ scenario |
Coverage | Population density | Territorial, focused on whatever may need protection across a country geography |
Availability | Outages undesirable (revenue loss/customer loss) | Outages unacceptable (live lost or threatened) |
Communications | One to one | Dynamic groups, one to many, field crews/control centre |
Broadband data traffic | Internet access (mainly downloads) | Traffic mainly within agency (more uploads than downloads) |
Subscriber information | Owned by carrier | Owned by agency |
Prioritization | Minimal differentiation, by subscription level or application | Significant differentiation, by role and incident level (dynamic) |
Authentication | Carrier controlled, device authentication only | Agency controlled, user authentication |
Preferred charging method | Per minute for voice, per GB for data, per message for SMS | Quarterly or annual subscription with unmetered use |
The delivery of PPDR mobile broadband services over commercial networks has to be necessarily formulated via contractual agreements between the individual PPDR agencies, or a public entity on behalf of a group of PPDR agencies, and one or more MNOs. In some countries, it may be possible to have a national roaming agreement in place, allowing the PPDR users to have a single subscription with a particular MNO but to get services from other networks if that particular network fails or if the user moves out of the coverage of the currently used network.
One of the key advantages of a PPDR service delivery model based on commercial mobile broadband networks is the fact that 3G and 4G data networks are already deployed and becoming increasingly ubiquitous. Therefore, PPDR agencies can implement mobile broadband solutions early and explore some of the possibilities that broadband data offers without significant upfront investment. While early implementations may be limited to data services, including video delivery, it may be possible to obtain new capabilities as soon as the specific features that 3GPP is developing for critical communications are released and deployed by the MNOs (e.g. ProSe, group communications system enablers). Clearly, the use of commercial networks can result in a very cost-effective solution, since there is limited infrastructure CAPEX from the PPDR side (mainly limited to specific PPDR SDPs), and since the network is supporting many more commercial users than PPDR users, there are significant economies of scale. Needless to say, the applications considered as non-critical (such as administrative applications) can always be well handled via a commercial infrastructure.
On the downside, one of the main disadvantages compared to a dedicated network for PPDR is the less control that the PPDR agencies may have on the footprint and performance of the network (e.g. coverage of low populated areas, response times of public operators in the case of network disturbance, prioritization of PPDR users, security, control of subscribers’ profiles, etc.). Remarkably, special consideration should be given to the lower network availability standards typically in place in the commercial networks. In this regard, it is a fact that during emergencies, public mobile networks are more prone to suffer from congestion and even suffer important shutdowns during disasters, power outages and other events. On the contrary, many of today’s PPDR narrowband networks have been built to provide service availability close to 99.999%, which means less than 5 min of downtime per year [4]. Various system design elements are used to create this performance standard in current PPDR network, including redundant radio BSs, the use of self-healing backhaul transmission networks, backup power supplies and automatic failover to redundant critical components. Obviously, the achievement of this service availability level has a direct translation into increased network costs, which may not be considered as economically viable or desirable by a commercial network operator. Consequently, the fear that the network might fail due to congestion during a crisis is often cited as the most compelling reason for emergency responders not to rely on commercial networks for their communications needs [16]. The 2013 annual report on major communications network outages by the European Union Agency for Network and Information Security (ENISA) found that ‘Overload was the cause affecting by far most user connections, more than 9M connections on average per incident’. Even if the commercial network does not fail, unless PPDR users have been able to negotiate guaranteed capacity on the network, there is the danger that access will become restricted, or lost altogether, during times of peak demand. Experience shows that such peaks often occur during major incidents when critical communications users often need the service most [14].
These disadvantages can be modulated to some extent depending on the nature of the contract established between the PPDR entities and MNOs. This view leads to define two major types of arrangements for the delivery of mobile broadband services to PPDR users:
Regarding the specific PPDR contracts, hardening and disaster recovery plans for improved resilience are likely to be among the most critical extra costs that arise. In this respect, government paying for improved resilience is an option to be considered, though this may be classified as state aid as long as it may turn into a competitive advantage for a given MNO. Another critical extra cost can be associated with coverage extension requirements, which may also include specific areas such as tunnels or underground facilities. This can also be achieved through public funding or by establishing coverage obligations on the MNOs for the land-mass geographic coverage, data speed and building penetration.
Even considering the added cost of resilience and improved coverage, a recent study [16] on the use of commercial mobile networks and equipment for mission-critical high-speed broadband communications found that the PPDR communications delivery model based on commercial LTE operation was the cheapest option in simple financial terms. However, the same study pointed out that the main issue with this option is not the technology challenge of building a resilient network, but the regulatory, legal and contractual context.
In this regard, the next subsection provides some considerations on the organizational and contractual aspects between the PPDR users and MNOs for the delivery of mission-critical communications over the commercial networks. After that, some key conclusions from the study [16] are echoed with respect to the conditions and regulatory changes that would help in removing the obstacles of using commercial mobile broadband networks for mission-critical purposes. Finally, some insight into the operation and status of existing prioritization systems supported in some commercial networks and currently used by PPDR users for privileged access to voice communications services is provided in the last subsection.
The establishment of a central public entity as the contracting authority with MNOs rather than the individual PPDR agencies can help in leveraging the buying power of the entire PPDR sector by pooling the demand in one or more contracts and executing long-term service procurement with the MNOs.
This central role can be played by a new entity established for that purpose. This entity could be a kind of ‘delivery partner’ or ‘programme manager’ as envisioned in the UK contractual structure for the delivery of the ECS (described in Chapter 3). Another approach is to establish an MVNO as the entity responsible for the delivery of mobile broadband PPDR services, managing some network and service aspects and taking care of the arrangements with the MNOs and system integrators. An example of this approach is the Belgian network operator ASTRID with the Blue Light Mobile MVNO service for data communications (see Chapter 3 for further details on ASTRID’s approach and Section 5.5.2 in this chapter for a detailed description of the applicability of the MVNO model for PPDR communications).
Establishing contracts with multiple MNOs would also bring in some advantages. The most obvious is an increased availability level, as there is no a single network being the point of failure. In addition, the ability to roam across multiple MNOs would enable PPDR users to get connection through the most appropriate network at any situation. Hence, PPDR terminals would be connected to, for example, the nearest BS among those from different service providers, thus exploiting the diversity of BS sites and frequencies (multi-band and multi-operator site diversity). This ability can be expected to turn in better spectrum efficiency and achievable data rates for PPDR users. Moreover, involving multiple MNOs can also allow a selective hardening of sites across the multiple operators to ensure the coverage as needed in each location. This could reduce the costs of hardening considerably since only a fraction of the MNO’s sites would be hardened. This approach would also satisfy state aid rules by providing equal advantage to all the MNOs.
When using commercial networks for mission-critical purposes, the governments are expected to write specific legal requirements on the operator down in a contract. These contracts may include clauses on everything from government step-in upon failure, SLAs on minimum availability, contract transfer limits, limits on force majeure claims and so on. While these conditions are essential for a proper service assurance, there is the risk that the demand of stringent SLAs (e.g. very high availability levels, penalties for SLA infringements with liability for damages due to service interruptions) may ultimately hinder the incentives for mobile operators to offer such specialized services to the mission-critical sectors, especially for the PPDR sector. Examples of legal requirements imposed on the operator by the governmental authority are provided in Ref. [15] and captured in Table 5.4. Therefore, the ability to draw up balanced contractual and/or financial measurements is essential for both the PPDR community and MNOs. Anyway, contracts are not always 100% solid in comparison to the guarantees associated with a government-controlled network (e.g. in an extreme case, the operator business may go bankrupt or be sold).
Table 5.4 Examples of legal requirements that might be contractually imposed on the operator by the governmental authority.
From Ref. [15].
Level | Legal requirements that might be imposed on the MNO by the governmental authority in the contract for PPDR services |
Most stringent | Parent company guarantee (PCG)/performance bond (PB): (i) PCG – The supplier’s parent company agrees to meet the supplier’s financial and/or performance obligations should the supplier fail to do so. (ii) PB – The supplier provides the government customer with a PB usually valued at between 5 and 10% of the contract price. The government customer can redeem the bond if the supplier fails to meet its contractual obligations (even if the financial costs of the failure are lower than the value of the bond). Both are standard in government customer contracts |
Intellectual property: The government customer is to own the rights in any new, project-specific intellectual property developed by the supplier. Often, a non-negotiable requirement of the customer. This is a standard approach | |
Liability: Government customer contracts may specifically exclude a waiver of consequential and indirect damages, and the overall liability cap may be in excess of contract value. Losses for breach of the confidentiality provisions may be uncapped altogether. Light breaches will in some countries result in liabilities in the order of €500M and in other cases be fully uncapped. Standard provision in many jurisdictions | |
Open book accounting: The government customer has access to the supplier’s financial records in order to see any reduction in the supplier’s costs in performing the contract. If costs have reduced, the supplier and government customer will split the ‘profit’. Sometimes, the split is 50/50 although it is not uncommon for the government customer to receive the majority of any such profit. Provision is becoming more common | |
Most favoured customer: The government customer must have the best price. The supplier cannot sell the same (or similar) products and services to another customer at prices lower than those paid by the government customer. Reasonably frequent, but not a standard provision | |
Moderately stringent | Step-in: The government customer has the right to take over the performance of the contract in certain circumstances. For example, where the supplier suffers an insolvency event (e.g. insolvency, arrangement with creditors, etc.) or commits a material breach of the contract. The supplier is not paid during step-in and may also have to meet the government customer’s additional costs associated with step-in. The government customer may hand back the services to the supplier or terminate the contract. Provision is becoming more common |
Termination: The government customer has extensive rights to terminate, often including termination for convenience. Whereas the supplier will only be permitted to terminate in very limited circumstances (e.g. protracted failure to pay undisputed fees). Standard provision | |
Change of control: A change of control of the supplier will be subject to the government customer’s approval, which often may be withheld at the customer’s absolute discretion. In some instances, changes of control are prohibited altogether. Standard provision | |
Financial strength: The supplier is required to show financial strength on a regular, ongoing basis. If the supplier’s financial strength diminishes, the government customer may terminate the contract. Provision is becoming more common | |
Control over performance: This is typically very stringent in government contracts – the government customer takes a more involved role than is usual in other contracts, for example, in testing and acceptance procedures. In our experience, government customers are slower and less inclined to approve and accept elements of a project than commercial customers, resulting in delayed payment to the supplier. Standard approach | |
Liquidated damages/service credits (LD/SC): Although they can vary from contract to contract, LD/SC regimes are, in our experience, often more onerous with government contracts as the government cannot allow or afford the project to fail or be delayed or services to be compromised. Also see comments regarding force majeure events. Standard provision | |
Force majeure: Force majeure events are often defined much more narrowly than in commercial contracts. For example, industrial action is usually excluded from government customer contracts (although sometimes permitted if it is nationwide or industry-wide). Force majeure clauses may also include a proviso that, due to the very purpose of the contract (i.e. public safety), a circumstance will not be considered a force majeure event if the party invoking that event reasonably ought to have taken into account when the contract was signed. Standard provision | |
Export control: A US company is responsible for ensuring that its products are not exported to prohibited countries. This responsibility extends to onward sale by the company’s customers. Government customers will not accept such restrictions imposed on it by another government. Accordingly, standard export control provisions are routinely excluded from government customer contracts. Standard provision | |
Source code escrow: The supplier must place the system source code into escrow with a third party, at the supplier’s cost. The source code can be released to the government customer in specified events such as insolvency of the supplier, breach of contract by supplier, etc. Standard provision | |
Least stringent | Assignment: The supplier is not usually allowed to assign the contract without the government customer’s prior consent, which may be withheld at the customer’s discretion. Standard provision |
Security clearance: The government customer may require certain supplier employees (e.g. those who have access to certain customer sites) to undergo national security clearance. Standard provision | |
Data: Recording and retention obligations for data processed under government customer contracts may be subject to specific data protection legislation, which can be quite burdensome. Provision is becoming more common | |
Continuous improvement: The supplier must improve the operation of the system over time at no additional cost to the government customer. Reasonably frequent, but not a standard provision | |
Taxes: Contractual obligation on the supplier to regularly pay its taxes. Failure to do so would amount to breach of contract by the supplier. Standard provision | |
Confidentiality: Government customers are usually reluctant to agree to standard confidentiality provision, preferring to use their own. Standard provision |
While it is a fact that public mobile networks are not yet engineered for resilience and advanced voice features that critical communications users require, the challenge for MNOs is to decide whether they can recoup or not the additional investment that would be required from what is a niche market in comparison to their consumer customer base. In the context of an increasing pressure on public budgets and the introduction of a new and more efficient generation of mobile broadband technology, European research project HELP [37] considered strengthening the role and commitment of commercial wireless infrastructures in the provision of PPDR communications by exploiting network and spectrum sharing concepts (more details on the PPDR communications framework developed in this project are given later on in Section 5.6.7). More recently, the European Commission requested an in-depth independent study on the costs and benefits of using commercial networks for mission-critical purposes [16]. The study explored the communications requirements and options in three sectors – PPDR, utilities and transportation – and conducted a cost analysis of five network deployment scenarios. Among the analysed options (listed in Table 5.2), the use of commercial LTE networks was found to be the most attractive in terms of its value for money. The study concluded that it could be possible for commercial mobile broadband networks to be used for mission-critical purposes, though only if the following five conditions were fully met:
A crucial barrier pointed out by Forge et al. [16] for the use of commercial networks is the current MNO’s mass-market business model, which needs to be suitably amended to provide the appropriate levels of service to priority clients with special needs. In this regard, the study concludes that specific regulatory measures may be needed to reassure the three sectors (particularly those with regulatory obligations on continuity of service, such as the utilities) and ensure that the MNO’s performance levels are maintained over decades. These measures are necessary to gain the trust of the user communities that MNO commercial behaviour will never disrupt mission-critical services. In particular, measures needed to build the confidence of these users in the MNOs would be:
In this regard, a number of measures that give NRAs specific new powers to cope with this situation on behalf of mission-critical services are proposed [16]:
The ability to prioritize users is a key component in any model for the delivery of mission-critical communications services over commercial networks. Prioritization may also entail the ability to pre-empt or degrade some users or services during times of emergencies.
While prioritization capabilities are part of the LTE standard (see Chapter 4 for a detailed description), they have not been tested much in a real-world environment to date. In the context of the dedicated network being built in the United States by FirstNet, testing activities are being addressed by the PSCR programme, a joint effort of the Department of Homeland Security and NTIA/FirstNet, which is targeted to advance on public safety communications interoperability. The goal is to verify that QoS, priority and pre-emption features function correctly before first responders use them in a real situation [38]. This provides an opportunity for network equipment and UE vendors to debug such features and verify that pieces of equipment from various vendors interoperate properly. Early reports [39] indicate that the PSCR testing has already validated some basic features such as priority pre-emption of bearers, ARP and QCI configuration, admission control and packet scheduling. In addition, further development and testing are still required for advanced features such as Multimedia Priority Service (MPS), RRC Establishment Cause support and SIP priority. While the work at PSCR is focused on the support of priority and pre-emption on a dedicated LTE network, the LTE features tested could also be applicable to commercial networks. Besides the validation of the technical capabilities, the deployment of priority and pre-emption in LTE commercial networks requires the establishment and validation of clear policies to ensure that such actions do not have negative unintended consequences. An example that is often cited when it comes to the applicability of pre-emption in a commercial network is how to deal with the case of an ‘ordinary’ call between a medical specialist and somebody providing advice for a cardiac arrest that is just happening in the same area where there is a sudden increase of higher-priority PPDR traffic.
Thus far, schemes for privileged access to commercial mobile networks have only been adopted for voice communications in some countries. The current situation is that there is not a global prioritization system, though some efforts have been conducted at international organizations such as ITU-T by developing recommendations to facilitate the use of public telecommunications services during emergency, disaster relief and mitigation operations (e.g. ITU-T Rec. E.106 [40]).
The United States has been a pioneer in the development of this type of systems, initially implementing the prioritization of voice calls in the wireline network, known as Government Emergency Telecommunications Service (GETS) [41], followed by its extension to commercial mobile networks, known as Wireless Priority Service (WPS) [42]. This system gives priority to calls made by phones subscribed to WPS after dialling a specific prefix. Canada and Australia have also implemented a system based on the US WPS [43].
In the EU, only the United Kingdom possesses a prioritization system, known as Mobile Telecommunication Privileged Access Scheme (MTPAS), which substituted the old Access Overload Control (ACCOLC) system [44]. Upon the Police Incident Commander request, MTPAS restricts the access to the cell sites in the area where an emergency situation has been declared. In turn, Sweden is developing standards for the implementation of priority services but lacks of a system at the moment [45].
Following a brief discussion about the technical foundations of priority support in commercial networks for voice services, the next subsections cover the scope and capabilities of the WPS used in the United States and the MTPAS used in the United Kingdom as the two more representative initiatives for priority services on public mobile telephone networks. Further information on other initiatives related to the delivery of priority services on commercial networks can be found in Ref. [46]. Recently, some patents to enable prioritization and ruthless pre-emption to authorized users have been filled [47].
Priority Service [48] was first introduced in 3GPP Release 6 specifications and is applicable to voice calls offered through both GERAN and UTRAN. A user of the Priority Service is assigned access class(es) in the range 11–15 to receive priority access to the network. In addition to access attempt priority provided by access control mechanisms, a Priority Service call receives end-to-end priority treatment, including priority access to traffic channels, priority call progression and call completion. Priority Service is activated on a per-call basis using specific dialling procedures: the user dials a given service code (SC) after the destination number. It is worth noting that Priority Service supported by the 3GPP systems is one element in the ability to deliver calls of a high-priority nature from mobile-to-mobile networks, mobile-to-fixed networks and fixed-to-mobile networks. Hence, in order to effectively provide an ‘end-to-end’ solution, all Priority Service providers (mobile networks, transit networks, fixed networks) should adhere to uniform, nationwide operating access procedures.
3GPP technical report TR 22.950 [48] establishes the high-level requirements for Priority Service, and TR 22.952 [49] provides a ‘guide’ that describes how existing 3GPP specifications support such requirements. In fact, Priority Service is supported by relying on a service known as Enhanced Multi-Level Precedence and Pre-emption (eMLPP) that was already specified in GSM Release 98 and updated within GSM/UMTS Release 6 to be compatible with the Priority Service requirements. eMLPP service is specified in Refs [50–52].
The eMLPP service is offered to a subscriber as a network operator’s option for all basic services subscribed to and for which eMLPP applies. The eMLPP service supports two capabilities: precedence and pre-emption. Precedence involves the assignment of a priority level to a call. eMLPP supports a maximum of seven priority levels as defined in Ref. [50]. The highest level (A) is reserved for network internal use. The second highest level (B) may be used for network internal use or, optionally, depending on regional requirements, for subscription. These two levels (A and B) may only be used locally, that is, in the domain of one mobile switching centre (MSC). The other five priority levels (0–4) are offered for subscription and may be applied globally, for example, on inter-switch trunks, if supported by all related network elements, and also for interworking with ISDN networks providing the MLPP service. Levels A and B shall be mapped to level 0 for priority treatment outside of the MSC area in which they are applied. For each of the seven priority levels, the network operator can administer the parameters that control the treatment of that priority within its domain. This treatment includes the selection of a target call set-up time and whether pre-emption is allowed for each priority level or not. An example for an eMLPP configuration is given in Ref. [50]. 3GPP specifications do not define specific mechanisms to achieve the target set-up times as defined by the service provider. The use of pre-emption is an operator choice. In case pre-emption is not provided, priority levels may be associated with different queuing priorities for call establishment.
The eMLPP priority level for a given call depends on the calling subscriber. The maximum precedence level for each subscriber is set at subscription time. Information concerning the maximum priority level that a subscriber is entitled to use at call establishment is stored in the Home Subscriber Server (HSS). On the UE side, SIM/Universal Subscriber Identity Module (USIM) also stores eMLPP subscribed levels and related data (e.g. fast call set-up conditions). The priority level can be selected by the user on a per-call basis (up to and including their maximum authorized precedence level). The selection of the priority level is done via man–machine interface (MMI) at the UE in the case of an eMLPP subscription including priority levels above 4. For mobile-terminated calls, the priority level is established based on the priority of the calling party, and it is applied at the terminating end (presuming the call’s priority is passed via signalling between the originating and terminating networks). Interworking with ISDN MLPP is required. The eMLPP service also applies to roaming scenarios, if eMLPP is supported by the related networks.
eMLPP also supports automatic answering or called party pre-emption. Hence, if the called mobile subscriber is busy and automatic answering applies for calls with a sufficient priority level, the existing call may be released (if pre-emption applies) or may be placed on hold in order to accept an incoming call of higher priority.
With all the above mechanisms defined, no further development of the eMLPP service is envisaged in ongoing 3GPP releases.
WPS [42] is a method of improving connection capabilities for a limited number of authorized national security and emergency preparedness (NS/EP) cell phone users on commercial wireless networks. WPS was mandated by the FCC in the United States. In the event of congestion in the commercial network, an emergency call using WPS will wait in queue for the next available channel. WPS calls do not pre-empt calls in progress or deny the general public’s use of the radio spectrum. WPS is available to key NS/EP personnel with leadership responsibilities at the federal, state, local and tribal levels of government and in critical private industries (e.g. finance, telecommunications, energy, transportation, etc.). These users can range from senior members of the presidential administration to emergency managers and fire and police chiefs at the local level and to critical technicians in wireline and wireless carriers, banking, nuclear facilities and other vital national infrastructures.
The implementation of WPS in commercial 3GPP networks is based on the Priority Service described in the previous section. WPS actually uses five priority levels, and a Priority Service call is invoked by dialling *SC + destination number, with an SC of ‘272’. Within US networks, a WPS user is assigned one or more access classes in the range 12–14 to receive priority access to the network, in addition to an assigned access class in the range 0–9. Pre-emption capability is not required by WPS. Additional details of the implementation of the WPS in the US networks can be found in Ref. [49]. The WPS system places a limit on the number of voice channels that can be used for priority calls at any given time (25% of voice channels on each cell site [53]). This ensures commercial users not being completely starved of access to the network.
MTPAS [44] is a service launched in the United Kingdom in September 2009 that aims to provide privileged access to the mobile networks by installing a special SIM card in the telephone handset.
The service is only available to some categories of emergency responders who are entitled to get privileged access SIMs in their staff’s mobile phones. The activation of privileged access in SIMs does not necessarily require the physical change or acquisition of a SIM. Mobile operators can change the access of a SIM (from normal public access to privileged access and vice versa) remotely using SIM over-the-air (OTA) technology.
The MTPAS is activated upon request in an emergency situation. Thus, using an agreed protocol, the Police Gold Commander, in charge of the response to a major incident, notifies all MNOs that a major incident has been declared. In such situation, if networks become congested, handsets installed with a privileged access SIM will stand a much higher likelihood of being able to connect to their network and make calls than other customers.
MTPAS supersedes ACCOLC, the former scheme for managing mobile privileged access. SIMs issued under the old scheme will continue to work under the MTPAS.
Hybrid solutions, used here as an umbrella term to categorize those PPDR service delivery options that are based on a combination to different extent of dedicated and commercial mobile broadband network infrastructures, can help to strike the right balance between the approaches presented in the previous sections. In some cases, hybrid solutions can be considered as interim solutions in the medium run that could pave the way for smoothly transition towards the desirable long-term solution. For example, commercial networks can be exploited until a dedicated network become available. In other cases, hybrid solutions can be considered as the most valuable option from a cost–benefit perspective even for the long term.
Hybrid solutions can be formulated around one or a combination of the following approaches:
A further insight into the above hybrid solutions is provided in the following subsections.
Roaming is defined as the use of mobile services from another operator, which is not the home operator. The most well-known form of roaming is international roaming, which allows users to use their mobile devices when abroad. Nevertheless, national roaming is also quite common. National roaming is concerned with networks of operators within the same country. In some countries, national roaming is imposed by the NRAs with the objective to promote and stimulate competition by facilitating the entrance of new actors in the market. Sometimes, national roaming is implemented on a voluntary basis between providers (as part of business agreements), without the intervention or request from the NRA. Different types of existing national roaming schemes are the following [55]:
National roaming for PPDR is needed to extend the coverage of the PPDR services to the areas not covered by the dedicated network. National roaming with multiple MNOs enables higher coverage and capacity as well as improved resilience, since most areas are covered by BSs of several operators.
The selection of the serving network would be controlled by the PPDR network operator. Indeed, in current commercial solutions, network selection relies on information that is directly stored in the SIM card of the mobile terminal:
Based on the previous information, network selection can be automatic or manual. In automatic mode, the mobile device scans the spectrum and finds all available networks. The UE then chooses a network in priority order based on the following list:
In this way, the UE requests first to the home network or equivalent as described in the first list. If none of them is found, the device selects a network in the next list and so on. In manual mode, the device researches and displays all available networks including the networks present in the forbidden networks list (in opposition to the automatic mode). The networks are presented to the user in the same priority order as do the automatic mode. Then, the user selects arbitrary a network from the networks list.
Solutions currently in use by MNOs or roaming hub operators to control the distribution of roamers in scenarios with multiple candidate networks (i.e. dynamic roaming steering) can find its applicability in the PPDR national roaming scenario. For instance, solutions for OTA provisioning of SIM card settings [56] can be used to organize the roaming priority lists within the PPDR SIMs and to specify a given roaming behaviour in a dynamic manner according to specific PPDR needs.
MVNO models have existed for more than 10 years in the commercial domain, and therefore, the principles are well established. Nevertheless, there are many different ways of implementing the MVNO model depending on how far the MVNO wants to control its services and how much control an MNO is willing to offer.
A standard definition of MVNO does not exist. Ofcom, the UK regulator, offers probably the most general definition of MVNO as ‘an organisation that provides a mobile (sometimes called wireless or cellular) service to its customers but does not have an allocation of spectrum’. Indeed, an MVNO is a business model that has emerged by the rupture of the traditional mobile value chain of an MNO. This has allowed new players to participate in the mobile value chain and extract value leveraging their valuable assets. The traditional mobile value chain can be separated into two main areas [57]:
It is in this second area of the value chain where other parties can participate by innovating, operating or selling mobile services. Both MNOs and MVNOs can take advantage of this business model. MNOs can exploit its network capacity, IT infrastructure and service and product portfolio to acquire untargeted segments, add a new revenue stream from wholesale business and reduce spare capacity and cost to serve per user. In turn, an MVNO can exploit its brand awareness, distribution channels and customer base to provide customized value proposition and complementary products and services to its customers. An MVNO venture brings multiple benefits to a company such as a new revenue stream, a low-cost entry strategy to the mobile market, a new vehicle to strengthen the value proposition and an opportunity to increase customer acquisition and/or retention.
Based on how the value chain is restructured between MNO and MVNO, the four main business models illustrated in Figure 5.2 have emerged in the MVNO market:
The success of the business model is dependent on creating a win–win situation between the MVNO and the MNO, where the MVNO has more knowledge than the MNO of a specific market segment that the MNO may not wish to develop itself. This may be due to particular market circumstances and/or difficulties in penetrating that market segment. The commercial relation between MVNO and MNO is based on relatively simple SLAs for the MVNO to obtain bulk access to network services at wholesale rates. Often, the MVNO bulk-buys minutes or data, based on coarse MVNO usage level predictions (for multiple months) and corresponding spare capacity predictions by the MNO. Typically, there is no differentiation between the MVNO and MNO users on the RAN. Thus far, there have been three main groups of players that have taken advantage of the MVNO business model:
In this context, the applicability of an MVNO model to cope with the market segment of ‘PPDR communications’, or more broadly that of ‘critical communications’, is gaining momentum. From an MNO’s standpoint, this market can be considered as a small, but very demanding, market in comparison to their mainstream large consumer markets. This is where an MVNO model is expected to fill the gap and bring value.
In the case of an MVNO model targeting the delivery of PPDR communications, the goal of the MVNO is to leverage the existing commercial mobile broadband radio infrastructure to create and operate dedicated services for the critical users. The MVNO stands between the user organization(s) and the MNOs, and it manages all the services for the users, such as provisioning, monitoring and managing all operational processes including incident, problem, change, configuration and release management that are needed to control the quality of the services. Therefore, out of the MVNO business models illustrated in Figure 5.2, a full MVNO is the most suitable approach to follow [58]. Alternatively, a light MVNO without core network equipment (i.e. P-GW and/or GGSN) but with control at least over the service platforms and subscriber management would also be feasible. Clearly, dedicated services can deliver added value including better availability, security, quality control and better customer care than can be delivered by the commercial MNOs individually.
The MVNO requires deep expertise in the relevant mobile communications technologies and a thorough understanding of the critical user requirements in order to make the best use of the existing commercial cellular radio infrastructures if it is to create and operate services dedicated to critical users. The MVNO has to negotiate special contractual arrangements with the MNO and a detailed SLA to reach its objectives. The MVNO will also manage the funding and financial aspects of the project. In order to spread the risk of network failure and exceptional traffic conditions as well as improving coverage, it may be advantageous to negotiate capacity with more than one MNO.
Better availability can be achieved by allowing roaming to any of the national MNOs. This also results in achieving the best coverage possible as well as improved resilience, provided the infrastructures are not shared. However, the implementation of national roaming could require (too) large investments by the providers if not national roaming practices are already in place for other purposes. As a way to address this, PPDR MVNO operators could initially leverage on existing international roaming infrastructure (i.e. use of international roaming hubs).
Reliability and service availability can be increased for PPDR users by granting higher access priority onto the commercial infrastructures based on special agreements with the MNOs. Security and confidentiality of the data flows can be managed by the MVNO (e.g. end-to-end security mechanisms between PPDR terminal equipment and MVNO’s service platforms). A schematic view of the MVNO model for PPDR mobile broadband communications is depicted in Figure 5.3.
The key advantage of an MVNO model is that it represents a relatively low-risk and limited investment option that would allow PPDR agencies to incorporate mobile broadband data into their daily operations in the short term, retaining the control on key aspects such as subscriber management, service profiles and service offerings. The implementation of an MVNO model could enable users to gain experience in the use of mobile broadband data and assist in the development of future plans. Additionally, an MVNO model can become very efficient especially looking for a ‘shared’ network/solution between different critical user organizations, as it enables them to reach an economy of scale through a common approach rather than each organization setting up its own MVNO with the MNOs. Indeed, a compelling reason for having a PPDR MVNO model is to negotiate and follow up the requirements and service commitments instead of having a situation where all the individual PPDR organizations make a contract with commercial operators themselves. A shared solution, through an MVNO, will also create an environment that will facilitate the exchange of information between user organizations when needed.
An MVNO can also function as a platform for PPDR application innovation from a very early stage and help to adapt services to the evolving requirements of the user’s mobile data applications. Critical communications users are now at the early stage of a new era regarding mobile data applications, and they will face growing and changing expectations towards the mobile data services. For existing PMR operators (i.e. TETRA, TETRAPOL, etc.), the MVNO model can provide a first step towards an eventual dedicated broadband network in the future while enabling the operator and users to better understand the benefits of mobile broadband. This gives also the advantage of a ‘1-stop’ service for different types of communications (i.e. PPDR users can go to the same entity for as both voice + data). The adoption of this model also represents a compelling opportunity for MNOs to build the confidence of the PPDR users in the MNOs as reliable and necessary players in the delivery of mission-critical data services.
On the downside, the disadvantages of this model mainly come from the limitations in terms of footprint and performance of the commercial MNO’s infrastructures, as discussed in Section 5.4. Hence, similarly to the case of standard commercial networks, the MVNO solution will only provide service where the operators have installed coverage and network resilience levels will only be those considered necessary by the commercial provider. However, as also discussed in Section 5.4, special arrangements can be established between the PPDR MNVO and the MNOs for increased levels of availability through increased robustness in the network design, extended coverage and priority access privileges for the PPDR users, especially in relation to emergencies and disaster events.
The adoption of an MNVO model for PPDR to use commercial capacity does not prevent the future deployment of private LTE networks. Indeed, the MVNO and dedicated network models are considered complementary [58, 59]. This model has been already put in place in Europe by the Belgian PPDR national network operator (ASTRID), which offers the Blue Light Mobile MVNO service for the delivery of data communications over the national commercial networks in Belgium. MSB in Sweden also plans to deploy a similar model to provide mobile broadband to their PPDR users, currently using a TETRA network for mission-critical voice. The MVNO model was also at the core of the PPDR communications framework developed in the European research project HELP [37], which is addressed in further detail in Section 5.6.7.
The deployment of a dedicated network may require a wide range of infrastructure and spectrum sharing mechanisms through public–public and public–private partnerships to be able to achieve its goals in a cost-efficient manner. Commercial MNOs are among those potential infrastructure and spectrum sharing partners. Commercial carriers can bring some key attributes to the table that other partners cannot, such as expertise in deploying a nationwide LTE network, back-office support systems and access to backhaul assets. Sharing agreements or partnerships with MNOs can minimize the initial investment required to launch service while providing the PPDR users with wide geographic service availability based on existing 3G/4G commercial cell sites. Indeed, it is a fact that many national TETRA and TETRAPOL networks in Europe already share some buildings, power, air condition and transmission with commercial operators [16, 13].
Network sharing can be viewed from two perspectives: geographical and technical [60]. From the geographical point of view, network sharing can take various aspects depending on the business model intended and the areas already covered by each involved mobile operator. The dimensions of network sharing can be characterized as:
From the technical perspective, the range of solutions encompass:
Apart from the network roaming and MVNO options, network sharing solutions today mostly focus on the access network, which is the most expensive part of an MNO’s network. The benefits for sharing core network elements are not as clearly identified as those for sharing the access network. While it is conceivable that there may be some cost reductions in operations and maintenance, the scale and practicality of these remain more uncertain. Sharing solutions might need the involvement of a trusted third party, which installs and operates the shared infrastructure and acts as a broker among the sharing operators. Such a solution comes at the cost of losing some of the operational control of the network operators and reorganization of the involved departments. A detailed list of the passive/active RAN and core network sharing solutions, together with their characteristics, is provided in Table 5.5. While there may be significant commercial and practical hurdles to overcome, there are no fundamental reasons why multiple operators cannot share networks. Agreements may concern individual sites, a number of sites or particular regions. Passive sharing and RAN sharing do not require a fully merged network architecture, and there are examples of unilateral, bilateral (mutual access) or multilateral agreements. Further considerations on business drivers, regulation and technical and environmental issues of network sharing can be found in Ref. [61].
Table 5.5 Network sharing solutions between MNOs.
Sharing solution | Type of sharing | Characteristics |
Site sharing | Passive | Operators collocate their own equipment. Very simple, it does not require operational coordination. Support equipment may or may not be shared |
Tower sharing | Passive | Operators share the same mast or rooftop. Operators have own antennas, huge CAPEX reduction and environment benefits |
Spectrum sharing | Active | Parts of the spectrum are leased by one operator to another. Improves spectrum efficiency and fights spectrum scarcity |
Antenna sharing | Active | Antenna and all related connections are shared. Passive site elements are shared too |
Base station sharing | Active | Operators maintain control over logical eNBs. Can operate on different frequencies, fully independent |
(Full) RAN sharing | Active | RAN resources are combined: antennas, cables, BS and transmission equipment. Separate logical networks and spectrum |
Core network element sharing | Active | Barely supported. Benefits are not as clearly defined as those for sharing the access network |
Transmission network sharing (backhaul, backbone) | Active | Transport is shared, for example, fibre access and backbone networks |
While all of the previous options for sharing are technically feasible, the specifics and complexities arisen when determining the most appropriate form of sharing could vary largely depending on many factors (e.g. network roadmaps of the involved operators, business incentives, costs, level of control required by the involved partners, physical security at the shared radio sites, hardening requirements). Importantly, sharing raises critical issues about the governance of the shared resources (who manages and controls) that need to be resolved in the form of win–win solutions. In any sharing model, the PPDR operator or government responsible entity for PPDR communications would presumably need to enter into an SLA and/or roaming agreement with the commercial partners. That business negotiation is critical to the success of any potential RAN sharing solution. Therefore, solutions may differ per region/country. In the context of FirstNet in the United States, in addition to the use of passive RAN sharing that entails less complexity in terms of integration, the following range of solutions and associated technical issues have been identified [63]:
In addition to commercial mobile operators, the deployment of a dedicated network may be required to forge partnerships with a wider variety of entities that will be open to sharing infrastructure. As an example, in the case of the CO dedicated network model, sharing opportunities for essential (and expensive) components of the LTE infrastructure can be seized depending on the consortium participants and contracts, as well on the possibility for the user organizations to lease or share their own existing network elements to the mobile broadband network provider. In this regard, the APCO Broadband Committee has delivered a white paper [64] that explores business tools, methods and techniques that will fuel and support the leveraging of public and commercial assets for use in building and operating the network by FirstNet in the United States. The document focuses upon asset standardization and valuation methodologies that have the potential to greatly simplify the negotiation, cost modelling, assessment and eventual use of resources needed to establish the network readiness, deployment and ongoing sustainment. This document presents recommendations, insights and open questions for potential LTE PPDR network stakeholders, such as state and local agencies looking to provide in-kind contributions to the network.
This hybrid model involves players who have the need and possibility to deploy local dedicated, redundant and extremely dense networks over certain sensitive zones. This could be the case of critical infrastructures such as airports and other transportation facilities. In addition, locations such as shopping malls, stadiums, concert venues and the like could also be considered as long as the specifics of the facilities require the deployment of dedicated professional networks or installations (e.g. DAS for coverage due to complex building infrastructures). In this model, these dedicated communications systems deployed in specific areas or facilities will be leveraged or integrated as part of a global, interoperable PPDR communications system, as depicted in Figure 5.5.
This model has been proposed by a major airport communications operator in France [54]. Airports are sensitive environments at the heart of which radio services are used to ensure the reliability of exchanges in real time, service performance and security of people. In both nominal mode and in crisis situations, airport operations require a secure and redundant radio network, a very high level of availability and a suitable coverage of buildings, technical galleries and runways. These different stakes are leading airport players to develop dedicated professional radio networks. In this context, [54] proposes a sharing service model by means of which the dedicated airport network could be integrated with the overall government solution for PPDR communications. The economic model proposed enables the absolute priority of government and airport PPDR uses over professional and potentially commercial uses. The model also enables much of the investment to be passed onto the specialist players. For its part, the government could keep the management of tactical networks and, more generally, the management of the overall service (networks, terminals, security applications). The interoperability and coordination of players are stressed as the main conditions for the success of this model. Commercial operators could also benefit from these dedicated networks, limiting the investment to be made to cover these zones and without competing with these specialist players.
This section describes some network architecture design and implementation aspects that are central for the realization of most of the previous described network delivery options for mobile broadband PPDR communications. To this end, the description of a reference architecture developed by ETSI that provides a high-level structure of a system intended to deliver mobile broadband mission-critical communications is firstly presented. On this basis, several constituent pieces are addressed in the subsequent sections, covering (i) interconnection options to get access to dedicated PPDR networks and SDPs from commercial networks, (ii) interworking solutions for the integration of broadband and narrowband legacy PMR services, (iii) the interconnection of deployable systems, (iv) the use of satellite communications and (v) the identification of the different types of connectivity services and frameworks for building the IP-based backbones and facilitating the interconnection of separate PPDR networks. Finally, and building on many of the previous constituent pieces, this section is concluded with the description of an MVNO-based delivery solution for mobile broadband PPDR communications, which is one of the approaches currently gaining more momentum as a viable short-term solution.
A reference model for the complete architecture of a critical communications system (CCS) is specified by ETSI in Ref. [65]. A CCS is defined as the whole system that provides critical communications services to users in several professional markets (e.g. PPDR, railway, utilities). As such, the CCS covers the complete communications chain, including terminal, access network, core network, control rooms and applications. The reference model developed by ETSI establishes various interfaces and reference points that comprise the overall architecture together with a brief outline of some of the most important services that the architecture supports. The ETSI reference model for a CCS is depicted in Figure 5.6.
The central element of the architecture is the critical communications application (CCA), which can be seen as the SDP providing professional communications services to critical communication users. The CCA includes capabilities on the terminal side (Mobile CCA) and on the infrastructure side (Infra CCA). The CCA contains both application-related services (e.g. registration of users, affiliation to groups, call services, etc.) and control functionality (e.g. ability to set up bearers with the required characteristics to communicate with the terminal units and control the levels of priority of the various bearers in the access subsystem where this control is available). The CCA provides services to additional (third-party) applications, for example, to provide group addressed services or prioritized access services. These applications can reside in or be distributed across both terminals and infrastructure (e.g. control room applications).
One CCA may be connected to more than one broadband IP network. The broadband networks may be of the same type (e.g. multiple 3GPP LTE networks). The broadband networks may also be of mixed network types, such as a mixture of 3GPP LTE and Wi-Fi networks that provide service to the same CCA. Multiple CCAs may also share the same broadband IP access network. Therefore, there can be a many-to-many relationship of CCAs and broadband IP access networks. The reference model also considers the support of direct mode of operations (DMO) between terminals. The interfaces illustrated in Figure 5.6 are:
The CCS reference model also considers the support of interworking with legacy PMR networks through the definition of an interface between the CCS and existing legacy systems (interface 8b shown in Figure 5.7). This interface is intended to support communications interworking between users operating on a CCS and on a legacy PMR system. This interface is anticipated to support at least a minimum set of features for the interconnection of TETRA, TETRAPOL and P25 systems: individual calls, group calls and short data services. These interfaces can be based on existing Inter-System Interfaces (ISI) already specified for some PMR technologies (e.g. ETSI TETRA ISI, P25 ISSI). Some further details on these ISI are provided in Section 5.6.3.
Further details on the CCS reference model can be found in Ref. [65]. In addition to this reference model, ETSI is also standardizing the architecture and protocols for interface (4) between the Mobile CCA and Infra CCA with the aim to support a generic mission-critical service equivalent to the existing narrowband technologies over LTE [67].
The interconnection of commercial networks to dedicated PPDR SDPs is a central element for the implementation of the hybrid solutions described in Section 5.5. This section describes the different interconnection options from a network architecture perspective, illustrating the main involved network elements and interfaces. Additionally, some considerations on the suitability of the different interconnection approaches to support critical communications are given. The interconnection options can be categorized under three distinct approaches depending on the services that the MNO provides through the interconnection.
The approach based on private connectivity services can leverage the solution in use nowadays for MNOs to offer private intranet services to business and corporate customers. With this solution, a (dedicated/private) PPDR SDP is reachable to mobile terminals gaining IP connectivity through a commercial network without traversing the public Internet. The delivery of private connectivity services is commonly done through the assignment and configuration of a private Access Point Name (APN). The APN is the parameter used in LTE (as well as in UMTS/GPRS) to identify the external network to which the UE is gaining IP connectivity access (see Section 4.2.2 for more details on the LTE service model). By using a private APN, all 3G/4G traffic leaving a UE can be routed to an IP endpoint at the private/corporate network (instead of the public Internet). Access to the private network is restricted to users who have a subscription for the corresponding APN. Nowadays, private APNs are common offerings from most mobile operators. An alternative approach to a private APN is the use of a virtual private network (VPN) solution over the public Internet (i.e. secure communications between a VPN client installed in the mobile device and a VPN server at the company/organization network). Nevertheless, there are several benefits to using a private APN with mobile cellular devices [68]:
In order to deploy a private APN, the interested organization should procure and provision an APN from a mobile operator and obtain SIM cards that are exclusively configured to use the private APN for their mobile data connection. Figure 5.8 shows a schematic of the network architecture based on the use of a private APN between an MNO’s infrastructure and a dedicated/private infrastructure of a PPDR service provider hosting a PPDR SDP and other elements needed for the delivery of PPDR communications services. The interconnection between the commercial MNO and the PPDR operator is based on dedicated resources (e.g. leased lines), thus preventing risks related to security threats and/or traffic congestion. End-to-end security services (e.g. end-to-end encryption) remain within the responsibility of the PPDR organizations that control the PPDR applications in the terminals and SDPs. Nevertheless, authentication and mobility management are under the full responsibility of the MNO, which owns and manages the HSS and SIM cards. Therefore, PPDR organizations have to trust the MNO with regard to network access security. On the plus side of this approach, and given that the MNO is who undertakes authentication and mobility management, is that the PPDR service provider can leverage any roaming agreement of the MNO and get access to the provisioned private APN service even when the PPDR terminals get connected through other mobile networks. This is especially relevant in the case of national roaming agreements for the PPDR service provider to be able to have multi-network access with the contract of a single private APN with a given MNO. Even in the case that national roaming is not in place, it would be possible for the PPDR service provider to contract the private APN service from an MNO in a foreign country or directly from an international roaming hub service provider that may hold international roaming agreement with all MNOs in the PPDR service provider’s home country.
In addition to the delivery of private connectivity, this interconnection solution can be extended with interfaces for user/subscriber management, QoS control and network monitoring by the PPDR service provider, as illustrated in Figure 5.8. In this regard, existing technical solutions and interfaces used by MNOs to cooperate with their resellers and sales partners can be leveraged for user subscription management. This would allow PPDR personnel in control rooms to check the status and, for example, disable or modify a subscription in case of theft or loss. Regarding QoS control, the MNO could provision access to the Rx interface of its Policy and Charging Control (PCC) platform (see Section 4.2.3) so that the PPDR SDPs could use the PCC capabilities to send service-related information, including resource requirements, to be enforced in the IP connectivity services (e.g. activation of dedicated EPS bearers with specific QoS parameters indicated by PPDR application servers). However, this is not a common offer by MNO nowadays so that special arrangements would be needed between the MNO and the PPDR service provider to deploy such interface. Similarly, the ability to monitor the MNO’s network for outages or other incidences is not a typical offering associated with private connectivity services. Therefore, a special arrangement with the MNO to get access to some information within its network monitoring system would also be required in this case.
Another possible interconnection approach is based on the delivery of roaming services. In contrast to the private APN solution, in this case, the PPDR service provider deploys some LTE core network elements within its infrastructure, as illustrated in Figure 5.9. In particular, a P-GW and an HSS are under the control of PPDR service provider, along with the SIM cards used in the PPDR terminals. In this approach, standardized LTE roaming interfaces can be used for the interconnection of the two infrastructures (i.e. S6a and S8 interfaces, as described in Chapter 4). With this solution, authentication and mobility management are performed collaboratively, though the MNO does not hold the permanent secret keys within the HSS and SIM cards. This solution facilitates the deployment of the user/subscriber management and QoS control interfaces, which reside within the PPDR service provider infrastructure. Nevertheless, as in the case of private APN services, the ability to monitor the visited network is not a usual arrangement in roaming-based interconnections. Therefore, if this were a mandatory requirement demanded by the PPDR organizations, a special arrangement with the MNO would be necessary. Furthermore, the PPDR service provider has to administer its own national and international roaming agreements since the roaming agreements of the MNO cannot be leveraged.
Finally, the third interconnection approach, illustrated in Figure 5.10, is based on the exploitation of RAN sharing services. It assumes that the PPDR service provider has deployed a full EPC and use the LTE RAN capacity deployed by one or several MNOs. Specifically, the interconnection solution depicted in Figure 5.10 corresponds to the MOCN solution supported in the 3GPP specifications, as described in Section 4.8. In this case, the interconnection is based on the deployment of the LTE S1 interface between the two infrastructures (i.e. S1-U for the user plane and S1-MME for the data plane, properly secured with NDS/IP). With this solution, authentication and mobility management are entirely handled by the PPDR service provider within its own EPC. As in the previous case, the PPDR service provider has to administer its own roaming agreements. On the other hand, this interconnection solution would allow leveraging the RAN sharing enhancements being introduced in the 3GPP specifications (see Section 4.8). In particular, selective access to operations, administration and maintenance (OAM) functions could be provided by the hosting RAN provider to the participating operators (i.e. the PPDR service provider among them) for network monitoring, together with other enhanced RAN sharing functions such as the flexible allocation of shared RAN resources and on-demand capacity negotiation (e.g. the peak capacity delivered to the PPDR service provider could be adjusted to better fit the needs if a particular emergency response).
Current PMR systems are expected to continue being the central platform for mission-critical voice for many years. As illustrated in Figure 5.7, the interconnection approach for communications interworking under consideration in the ETSI reference model for CCSs is based on the deployment of an interface between the new PPDR SDPs and the core networks of the legacy infrastructures. This approach allows leveraging the existing interfaces within the legacy technologies, especially those already developed for the interconnection between multiple independent legacy systems.
One of these interfaces is the TETRA Inter-System Interface (TETRA ISI). Back in the 1990s, the ETSI started the standardization process for the TETRA ISI to interconnect independent TETRA networks. The ETSI TETRA ISI standard [69] relies on the QSIG/PSS1 protocol (i.e. an ISDN-based private signalling standard) for the signalling plane to exchange a set of upper layer protocols, known as Additional Network Features (ANFs), that provide different sets of functions (e.g. there is an ANF set for supporting individual calls, another for group calls, etc.). On the user plane, the TETRA ISI uses 64 kb/s E1 channels for the transmission of TETRA-coded user voice and data information. However, the fact that this is actually a circuit-switched technology is seen as an important hurdle to overcome. Indeed, the TETRA ISI standard is employed nowadays only by a few TETRA vendors for limited functionalities (i.e. basic registration scenarios, individual call, short data and telephone).
In this context, the European research project ISITEP [70] is developing a new ISI for the interconnection of TETRA networks based on the adaptation of the current TETRA ISI so that it can be deployed over an IP transport network. The new interface is referred to as IP ISI and, in addition to enable the interconnection of TETRA networks, is intended to be used also for the interconnection between TETRA and TETRAPOL networks and between TETRAPOL networks. While the ultimate goal of the ISITEP project by pursuing the development of the IP ISI is to facilitate the interconnection of the different national TETRA and TETRAPOL PPDR networks deployed across European countries, the consolidation of such IP ISI is expected to facilitate the integration of current narrowband PPDR networks with the forthcoming all-IP PPDR SDPs as well. The new IP ISI protocol is based on the Session Initiation Protocol (SIP), which is the current de facto standard for Voice over IP (VoIP) communications. The approach adopted in ISITEP is to allow that the already standardized ETSI ISI ANFs can be exchanged through SIP messages and use the Real-time Transport Protocol (RTP) for the voice traffic encoded with the corresponding codecs. A simplified view of the IP ISI protocol stack is depicted in Figure 5.11.
A similar approach is envisioned for P25 systems. The P25 standard has already defined an ISI known as Inter-RF Subsystem Interface (ISSI). It aims to connect disparate P25 networks, regardless of vendor. Its first commercial implementation dates back to 2010. P25 ISSI protocol is already based on IP, also relying on the IETF SIP for the control plane and the RTP to convey P25 voice frames [72]. Besides the connection between legacy P25 systems and new IP-based PPDR SDPs, another applicability of the P25 ISSI is found in the development of a standard for delivering P25 services directly to LTE terminals. This standard is being specified by the TIA Mobile and Personal Private Radio Standards Committee (TIA/TR-8) in the United States, and it is commonly referred to as P25 PTT over LTE (P25 PPToLTE). A preliminary architecture of this solution is depicted in Figure 5.12 [73]. It comprises a P25 PTToLTE server located at application level, within an SDP, reachable as an external network from a (commercial or dedicated) LTE network. The access point into the P25 network is via the standard ISSI protocol between an ISSI gateway (GW) in the P25 network and the P25 PTToLTE server. Then, a new protocol named Subscriber Client Interface (SCI) is used to connect the P25 PTToLTE server to a P25 PTToLTE client in the mobile terminal through the LTE network. The overall solution also incorporates the already standardized P25 Console Subsystem Interface (CSSI) protocol for the interconnection with control room dispatch consoles. Voice transmission relies on the standard P25 vocoders (voice coder/decoder) and allows up to AES 256-bit encryption services to be used at the LTE terminal. In this way, the P25 end-to-end encryption is preserved even when terminated in the LTE terminal.
As discussed in Section 5.3.1, deployable systems can be central for achieving cost-efficient network footprints. Furthermore, the use of deployables is instrumental for network restoration, network extension and remote incident response.
Deployable systems can be classified under two categories: cell on wheels (COWs) and system on wheels (SOWs). On the one hand, COWs typically include a BS (e.g. LTE eNB) along with one or more backhaul transports (such as microwave or satellite). COWs require connectivity to the core network where the LTE EPC and SDPs are located to be able to provide services to the mobile terminals under the coverage of the COWs. On the other hand, SOWs are fully functional systems that can act without backhaul connectivity. Indeed, SOWs are more complex systems (and consequently more expensive) than COWs. As a general approach, SOWs are more appropriate in rural environments and in disaster areas with heavy volumes of local traffic, where broadband backhaul connectivity is an issue. In turn, the use of COWs is envisaged in smaller incidents in urban and suburban environments, where better connectivity to the core network can be guaranteed. Both COWs and SOWs can adopt different form factors depending on their capabilities and design constrains (e.g. size, functionality, power supply, deployment time, environmental protection, etc.) and can be installed in trailers, mobile vehicles or transportable rack-mounted chassis.
Figure 5.13 depicts the main functional communications-related components integrated within COWs and SOWs together with their remote connectivity needs. As shown in the figure, SOWs include an eNB and a light implementation of the LTE EPC (i.e. HSS, P-/S-GW, MME, PCRF functions) to provide IP connectivity services to LTE devices. Additionally, SOWs include a local SDP to be able to offer services without the need to have wide area connectivity to remote SDPs. Optionally, SOWs could be supplemented with elements such as local Wi-Fi connectivity and PMR radio bridges to enable the communication with Wi-Fi-enabled devices (e.g. tablets/laptops) and legacy PMR terminals (e.g. TETRA/P25 handheld radios) on the scene. In the case of COWs, the deployable typically consists of the eNB functionality alone. Therefore, wide area connectivity to the remote EPC and SDP is necessary for its operation.
Solutions for COWs and SOWs face important design challenges. One of these challenges is the ‘smooth’ integration and coordination of the radio settings of the deployed equipment with those used in neighbouring LTE sites (which may overlap in coverage) as well as other in other deployables that may operate in the local radio vicinity. This challenge is especially relevant if deployables are expected to use the same frequencies used in the WAN (e.g. use of a single PPDR dedicated 10 + 10 MHz allocation). In this regard, advanced inter-cell interference management capabilities of LTE, together with other features such as self-organization and handover management, are fundamental to support this scenario [74, 75]. Another important technical challenge is the integration of the different tiers of connectivity services and applications when SOWs are used. The target here is to achieve a consistent and interoperable operation of the local EPCs and SDPs within the deployables and the remote EPCs and SDPs so that all the components together work as a single wide area distributed system. In this context, distributed peer-to-multi-peer architectures for the application layer are being tested in FirstNet trial systems in the United States [76].
Satellite communications provide a unique and important method for PPDR to plan around the hazards of earth-based infrastructures that can be susceptible to all manners of natural and man-made catastrophes. Satellite communications can also be used in areas where existing terrestrial infrastructure is insufficient or overloaded. This turns satellite communications into an important component within the complete PPDR communications toolkit [77].
One primary application of satellite communications is backhauling of COWs and SOWs. Indeed, solutions for cellular backhauling over satellite have been around for many years to backhaul 2G BS traffic in some remote locations as well as used in PMR deployables (e.g. TETRA BS with satellite connectivity). However, its adoption has been very limited mainly due to the high operational expenditures associated with satellite connectivity, relegating satellite cellular backhauling to a sort of last-resort or worst-case transport solution only used when microwave links or wired alternatives are not available. Nevertheless, the advent of high-throughput satellite (HTS) technology and the use of IP-based TDMA satellite interfaces instead of single channel per carrier (SCPC) are changing the economics of satellite backhaul [78]. HTS operating in the Ka and Ku bands are able to exploit frequency reuse technologies along with multiple tightly focused spot beams, yielding greatly improved spectral and power efficiencies that are bringing down the cost per megabit per second. On top of this, IP-based TDMA satellite interfaces allow the sharing of the satellite bandwidth among many sites (i.e. capacity aggregation with trunking gains), allocating bandwidth on demand based on the real-time requirements at each site (in contrast to SCPC links where fixed capacity was allocated per site).
Additionally, innovation in the terrestrial infrastructure with regard to how satellite connectivity services are provisioned, customized and managed is anticipated to contribute to further driving down the costs. In particular, the advance towards a more efficient integration of satellite and terrestrial components can be fuelled by the use of technologies such as network functions virtualization (NFV) and software-defined networking (SDN). Enabling NFV into the satellite communications domain provides the operators with the tools and interfaces to establish end-to-end fully operable virtualized satellite networks over shared satellite communications platforms, which can be customized and offered to third-party operators/service providers (e.g. a PPDR operator could manage ‘its’ virtual satellite communications platform for the interconnection of its deployables). In turn, enabling SDN-based federated resource management between the terrestrial and satellite domains paves the way for a unified control plane that allow operators to jointly manage and optimize the operation of the overall communications chain encompassing the satellite and terrestrial equipment. In this respect, the European research project VITAL is working towards these objectives [79].
Figure 5.14 illustrates the applicability of satellite communications for the interconnection of COWs. The eNB is connected with the remote EPC through the satellite connection, which is used to support the S1 interface (see Chapter 4). The fact that the S1 interface is an IP-based interface enables the use of different bandwidth optimization techniques like header compression and performance-enhancing proxies (PEPs). Satellite terminals with transmit power less than 10 W and antenna dishes of 75 cm in diameter are common settings. State-of-the-art satellite interfaces are DVB-S2 and its upcoming evolution DVB-Sx for the forward link (from the hub/GW to the satellite terminal) and DVB-RCS2 for the return link. Typical achievable speeds can be up to 12 Mb/s downlink and 3 Mb/s uplink, being also possible to reach 50 Mb/s downlink and 20 Mb/s uplink in more demanding settings [78]. Figure 5.14 also shows that the satellite capacity can be shared among a number of COWs and SOWs as well as used in fixed sites, which is an example where the provisioned satellite connectivity can be used either for traffic overflow of congested sites or for improved resilience of some critical sites.
A comprehensive generic architecture considering satellites as a transport solution in emergency communications has been developed by ETSI under the concept of Emergency Communication Cell over Satellite (ECCS) [80]. ECCS is a concept understood as a temporary emergency communication ‘cell’ supporting one or more terrestrial wireless standard(s) (e.g. cellular telephony, PSTN cordless, Internet access via Wi-Fi, PMR services), which are linked/backhauled to a permanent infrastructure by means of bidirectional satellite links. An ECCS system is intended to be a quasi-autonomous communications infrastructure in the field that enables communications between users inside and outside the disaster area using different sorts of communications devices. Indeed, the ECCS concept is not specific to communications between PPDR organizations but also applicable to solve the communications needs to the affected persons, victims or any other kind of involved people.
The ECCS architecture is shown in Figure 5.15. Despite the multitude of technical solutions that could be used to implement an ECCS system, the architecture is technology agnostic and mainly identifies the constituent logical blocks. The overall transmission path involves a number of domains, which compose an ECCS communications chain. The domains represent network elements involved in the ECCS communications chain, playing logically neighbouring functionalities in this chain or jointly enabling the provision of a given functionality (e.g. local access). A central role in the architecture is played by the ECCS servers placed on remote locations and the ECCS terminals on the field. These equipment together provide the backhauling of the different wireless services via satellite as well as, if necessary, the interconnectivity and interoperability between different services by means of GW or any other specific equipment (e.g. connectivity in the field between two PSTN cordless terminals can be achieved via a private branch exchange as part of the ECCS terminal). Indeed, different classes of ECCS terminals are likely to coexist due to different design constraints. Some examples are portable yet basic ECCS systems that can be easily packaged in an airborne-cabin-format suitcase, providing voice and data access via satellite; transportable and more powerful ECCS systems packaged in an airborne-container format or multiple man-carried containers; and mobile ECCS systems with ‘on-the-move’ access. It is worth noting that the ECCS architecture assumes that an ECCS service provider operates the overall ECCS system and interacts with other actors (e.g. satellite capacity providers, PPDR operators, MNOs, etc.). The ECCS operator acts as a kind of ‘concentrator’ for a complete and tailored service provisioning, in terms of communications services, content and infrastructure, to the system users and should be their main/single direct interface. Figure 5.15 depicts the four main interfaces identified in the ECCS architecture:
Thus far in the previous discussion, the role of satellite communications for PPDR communications has only been addressed as a backhaul solution. However, mobile satellite service (MSS) systems can also form part of the toolkit of PPDR communications means. The fact that MSS systems provide very large coverage areas compared to terrestrial-based systems is a relevant feature that can be exploited in the context of PPDR communications. MSS systems currently in operation (e.g. Iridium, Inmarsat, Thuraya) are able to provide voice and data radiocommunications and access to the Internet. Further, these systems can facilitate access to public and private networks external to the MSS system. In terms of data transmission capabilities, current offerings provide data rates of up to 500 kb/s with flat antennas of the size of a laptop. These MSS terminals can even be carried by an individual. There are other MSS terminals that can be mounted on a ship, an airplane, a truck or an automobile and can deliver Communications-On-The-Move (COTM). An illustrative description of uses and examples of different MSS systems for emergency communications is provided in Report ITU-R M.2149-1 [81]. Recently, solutions that integrate the functionality of a satellite MSS terminal into smartphones have also been launched (e.g. Thuraya SatSleeve for a set of Samsung and iPhone models). The SatSleeve comes in the form of an adaptor that is plugged into the smartphone. Nevertheless, data capabilities of this solution are limited to short messaging, and data services with data speeds are up to 60 kb/s for downloads and up to 15 kb/s for uploads.
Finally, some consideration should also be given to MSS satellite networks that introduce a complementary ground component (CGC). The CGC consists of a core terrestrial network that is to be operated in an integrated fashion with a satellite segment of the network using the MSS frequency band. This network configuration combines the geographical coverage benefits of the satellite with the high throughput of terrestrial mobile communications networks. In Europe, the European Commission granted two European satellite operators the right to operate 2 × 30.0 MHz (2170–2200/1980–2010) of MSS spectrum (the so-called S-band) using a CGC network. The use of this spectrum asset was granted in 2009 for a period of 18 years and covers all EU member states. Indeed, an S-band satellite segment was launched in 2009 and could already support service launch for a portion of the EU member states. Additionally, the MSS spectrum band is set to become 3GPP compliant and hence can be deployed in Europe leveraging the economies of scale of the LTE ecosystem globally. Therefore, a satellite–terrestrial network on the S-band spectrum is an option to the PPDR sector to implement a dedicated network approach relatively quickly [82]. However, this option is not currently receiving widespread support across the PPDR and other critical communications users across, which is fundamental to move forward such a project [14]. Moreover, the current satellite system is limited in capacity and could not meet the entire demands of the European critical communications sector immediately. The development of the market on a pan-European basis would require a new high-performance satellite to be manufactured and launched.
The interconnection of the multiple and diverse components of the overall PPDR communications system (e.g. radio sites, data centres hosting the EPC and SDPs, MNO’s data centres with the commercial network equipment, PPDR deployables, emergency control centres and PSAPs, interconnection of regional/national PPDR networks, etc.) requires the deployment of a vast backbone network infrastructure. This backbone network or combination of networks is likely to combine different transmission technologies, from fibre to satellite links, and designed and operated with the sufficient redundancy and protection mechanisms to guarantee the high availability expected in this kind of critical infrastructures. This required interconnection infrastructure can be fully or partly owned by the government or public entities established for the delivery of PPDR communications, though it can also rely on the use of interconnection services and facilities provisioned by private carriers.
In this context, fibre has become the predominant technology for backbone networks, though some niche applications for microwave and satellite are in place [83]. Whereas in the 1990s backbone networks used dedicated technologies, such as Asynchronous Transfer Mode (ATM) and Synchronous Optical Networking (SONET) and Synchronous Digital Hierarchy (SDH), most of today’s (Internet) backbone networks are built on the Ethernet suite of standards, which was originally designed for offices and data centres. The speeds of 1 and 10 Gb/s are now those most commonly used, with 100 Gb/s becoming increasingly available. Ethernet has become dominant because the high volumes used in data centres created a high-volume market that overshadowed the demands generated by the traditional telecom voice market. The Ethernet standard was expanded to support the requirements of carrier networks. SONET/SDH still remains central in backbone and interconnection for telephony services. In addition, modern fibre networks offer an additional level of flexibility through optical routing. Optical routing allows the network to route wavelengths irrespective of the content in the wavelengths.
Connectivity services offered nowadays by commercial carriers are typically categorized according to the OSI layer at which the interconnection service provider’s systems interchange reachability information with customer sites:
When it comes to interconnection solutions that may involve several carriers and service providers, the GSM Association (GSMA) has specified a managed IP network solution named IP exchange (IPX). IPX is a global, private, managed IP network solution that supports end-to-end QoS and the principle of cascading interconnect payments (i.e. all carriers in the interworking value chain are expected to receive a fair commercial return). While one of the motivations behind the specification of the IPX framework was the interconnection of mobile networks, the IPX service is not restricted to MNOs but intended to accommodate the connection to and between other communities such as fixed network operators (FNOs), Internet service providers (ISPs) and application service providers (ASPs). Collectively, all these types of potential IPX users are simply denoted as service providers within the IPX model.
IPX may be used to interconnect any IP service, either standardized (e.g. VoLTE services) or not. The IPX offers three standard modes of interconnection, which service providers are free to choose on a per-service basis: (1) bilateral transport only, where the IPX only provides transport at a guaranteed QoS between two service providers; (2) bilateral service transit, where the IPX provides QoS-based transport as well as service-aware functions; and (3) multilateral hub service, where the IPX provides QoS transport and service-aware functions to a number of interconnect partners via a single agreement between the service provider and IPX.
The IPX network model is illustrated in Figure 5.16. The IPX is formed from separate and competing IPX providers (or IPX carriers). Service providers are connected to their selected IPX provider(s) using a local tail interface (e.g. EPL). Service providers may be connected to more than one IPX provider. IPX providers connect to each other via peering interfaces. All parties involved in the transport of a service (up to the terminating service provider border GW/firewall) are bound by end-to-end SLAs. A common Domain Name Service (DNS) root database supports domain name resolution within the private network and E.164 NUmber Mapping (ENUM) capability to assist with the translation of telephone numbers to IP-based addressing schemes. This root database may be used by all IPX parties. IPX proxy elements within IPX provider networks can be used to support service awareness and interworking of specified IP services and make it possible to use cascading interconnect billing and a multilateral interconnect model (i.e. hub functionality within IPX proxies). Common rules regarding IP addressing, security, end-to-end QoS and other guidelines needed to ensure interoperability among service providers connected to the IPX backbone are described in the technical specification IR.34 [87]. Security guidelines established in IR.34 are complemented by another document, IR.77 [88], which concentrates on providing the detailed security requirements for IPX providers as well as service providers connecting to the IPX backbone.
In practice, a number of international carriers already provide IPX services (e.g. BT Wholesale, Telefonica International Wholesale Services, etc.) although some of them are in a limited form (e.g. support for bilateral transport only). In the context of PPDR communications, the use of a third-party IPX solution to achieve interconnectivity between regional networks was proposed in response to a FCC Notice of Inquiry (NoI) regarding the implementation options of the broadband PPDR network being built in the United States [89]. Currently, the IPX model is being considered, among other options, within the European research project ISITEP for the interconnection of national TETRA and TETRAPOL networks using the new IP ISI described in Section 5.6.3 [90].
As discussed in Section 5.5.2, a mobile broadband PPDR communications delivery solution based on the MVNO model is the approach currently under consideration as a viable short-term solution in some European countries. Building on many of the network design and implementation aspects described in the previous subsections, this last subsection outlines the main features of an MVNO-based delivery solution developed within the EU research project HELP [37, 58]. A high-level view of the overall system architecture with the key network elements and interfaces is illustrated in Figure 5.17 and described in the following.
The system architecture defined in Project HELP considers a PPDR operator’s core infrastructure that consists of IMS functions, application servers and EPC network components (i.e. HSS, PCRF and P-GW), all interconnected by means of a private IP network.
This core infrastructure is used to provide PPDR services to users in the field equipped with LTE-enabled PPDR terminals through a number of commercial LTE networks interconnected by means of standardized 3GPP interfaces (e.g. S8 for data transfer and S6a for signalling transfer, as depicted in Figure 5.17). National roaming agreements are assumed to be in place between the PPDR operator and multiple commercial MNOs. In addition to the use of the commercial capacity, it is considered that the PPDR operator has also deployed dedicated LTE access capacity in some specific areas (represented as the dedicated LTE-based PPDR network in Figure 5.17). Hosting the P-GW within the core infrastructure provides a secure access to this critical infrastructure and allows the PPDR operator to fully manage the IP connectivity service (e.g. private APN and IP address allocation). Besides, mobility between multiple networks (commercial or dedicated) without service disruption is facilitated since the P-GW serves as a mobility anchor point for all PPDR traffic. On top of the IP connectivity service, commercial or PPDR-customized mobile VPN solutions would be used to add an additional security layer between the client and server application endpoints within the SDP.
The PPDR operator’s core infrastructure also contains a GW for communications interworking with a legacy PMR network infrastructure. In the context of Project HELP, legacy TETRAPOL networks were under consideration, though the proposed solution approach is not specific to the TETRAPOL technology. The GW converts TETRAPOL protocols to SIP and RTP and vice versa, in order to keep the same group call service on both sides.
The PPDR operator controls the procurement and provisioning of the USIM cards used by PPDR subscribers. This approach avoids PPDR users ending up with a number of separate subscriptions to different commercial operators (i.e. handling multiple USIMs and using terminals with multi-SIM support) as well as provides independence from commercial MNOs through the ability to switch among them or support a number of them without changing PPDR users’ USIM cards.
The network management system (NMS) illustrated in Figure 5.17 represents the collection of technical and operational management tools used by the PPDR operator to control and monitor the operation of the core infrastructure, terminals, USIMs, legacy PMR and any dedicated LTE access network infrastructure that may be deployed.
The PPDR operator’s core infrastructure is connected through, for example, Application Programming Interfaces (APIs) to CRS used by PPDR users for tactical and operational management. CRS can include dispatch applications to access the PPDR services (e.g. dispatch positions to communicate with users in the field) as well as control and monitor applications to deal with administrative and operational issues of the provided PPDR services. In particular, the CRS include: