Long-term evolution (LTE) is the next step in mobile communications technologies from the 3GPP standards body. LTE, in its LTE–Advanced release, has been accepted as a 4G technology within the IMT-Advanced family of standards.
LTE provides a high-capacity IP mobile data network with low latency, suitable for many forms of multimedia delivery. It has a flatter architecture, giving a lower cost per byte for network operators. The use of orthogonal frequency division multiple access (OFDMA) gives good spectrum efficiency and is a very good technology to address multipath issues, which are a problem with mobile networks. Within the LTE standard, IP multimedia services (IMS) provide an architecture and framework for multimedia applications, which include IP telephony as well as multimedia services such as video conferencing, online games, instant messaging, all with quality of service (QoS) mechanisms.
LTE, therefore, is gaining momentum as a worldwide mobile broadband technology, and is starting to be deployed by network operators. It provides a flexible platform for modern communications.
At the same time, public-safety organisations see the need for new future networks, with increasing requirements for data communications. This could include data transmitted back in real time from an incident to inform the control room of what is happening, data sent to officers in real time to give them a situational awareness of what is happening, or non real-time data collection for evidential purposes. An Analysys Mason report for the TETRA Association (see http://www.analysysmason.com/About-Us/News/Press-releases/Analysys-Mason-study-into-Public-Safety-Mobile-Broadband-and-Spectrum-Needs/) reviewed and identified the requirements, and suggested four evolutionary paths, with three of the paths all needing significantly more mobile data capacity than that available within public-safety networks at present. Much of this data is mission-critical data, although there is some public-safety administrative data which is not mission critical. A key question is whether LTE technology can be deployed to meet this recognised data requirement, either using a public commercial network or a private LTE network designed specifically for public-safety use, and could such a network also provide suitable voice services for public-safety users (including particular requirements such as group calling), or whether a separate voice network is required.
It is worth considering the cardinal needs of the public-safety community for its mobile communications:
- Network must have high availability, with no single point of failure, a highly resilient architecture and guaranteed availability even in times of a major incident, when the network will be stressed. It must continue to operate through power outages; for some organisations this may be for five days without power for parts of the network.
- Coverage must be suitable for the users of the network. A police force network will need to cover all roads, and will need coverage into tunnels, shopping centres, etc. A utility will need to cover everywhere their service goes, even into remote rural areas.
- The public-safety organisation must have control of the network. It is not acceptable for a network to lose capacity when a major event is taking place or for maintenance or industrial disputes to result in a loss of service.
- The network must be secure, able to carry data without risk of interception and decryption, and protected against unauthorised access attempts.
- If the network is to be used for voice services, there are further requirements. These include a fast call set-up time of less than 300mS, the ability to replicate the group calling operational feature available in current networks, and a voice codec which works in high noise environments.
It is unlikely that a commercial network would be able to meet all these requirements. If it was designed to the high availability targets, the cost of implementation would be very high, and the network operator would be in an uncompetitive position (unless a government funded the hardening of the network, which is unlikely to be acceptable to other operators). Certainly, some of the administrative traffic can be carried on a commercial network, but it would be unsafe to rely on a commercial network for mission-critical data.
However, it would be feasible for public safety to deploy its own national LTE network, and for this network to be shared by all public safety responders. This is happening in the USA, where a block of paired spectrum from the 700MHz TV Digital Dividend was awarded to public safety for broadband use. The FCC has mandated LTE as the common air interface for interoperable mobile broadband networks deployed in 700MHz, and a pilot has started in San Francisco. Similar proposals are also being made in Australia.
So what is different about a public-safety LTE network, and what has to change?
Commercial networks are usually designed to cover a percentage of the population. A coverage target figure of 90% of the UK population allows large areas of Scotland and Wales, and even parts of England, to remain unserved. For public safety, there is no control as to where unplanned major incidents such as a train crash or shooting rampage will occur, so coverage has to be total, for instance covering all A and B roads. Capacity has to be either present, or able to be ramped up very quickly, to cover the incident. Coverage has to include tunnels, underpasses and metro systems, as well as sporting stadia and shopping centres. LTE lends itself to this sort of coverage, but spectrum below 1GHz is preferred, otherwise an excessive number of cells would be required for rural coverage.
The network has to be resilient and available. Much of this is how the implementation is done, with resilient links from the eNodeB to the enhanced packet core (EPC) and suitable power back-up. The eNodeB will need to have no single point of failure, which will need careful design of the LTE transmitter, and controller functionality will be duplicated. In the EPC, functions will be duplicated. There are already companies looking at providing public-safety standard LTE controllers. Network management must give the public-safety organisation visibility of the network and control of what is happening.
The security mechanisms within the LTE standard must be used, to protect both against interception, and intrusion, using authorisation processes.
If the LTE network is to carry voice traffic, changes may be required to the standard, or a way developed of making use of some the work already done within the standard.
LTE already has support for calls from the public to the emergency services (112/911 calls). These include pre-emption, location services and eCall. There is also access class barring, which gives priority to certain responders such as public safety. These give priority access to the network resources; they are not intended for everyday use, but for access control during a major incident.
LTE has provided enablers for the Open Mobile Alliance PTT over Cellular (OMA PoC) standard, which provides VoIP services and does have low latency, but more may be required to satisfy the public-safety group call function. Currently, the LTE VoIP services are unicast, where a separate path is established for each party in the group. This will be acceptable for occasional use, but once you have a major event such as New Year’s Eve in London, or the Grand Prix in Singapore, many hundreds of public-safety officers will be deployed, and a unicast mechanism will risk cell overload. What is required is a multicast solution, where the same transmission is heard by all the people in the group call (as is done in current PMR systems). Multicast exists in LTE to provide broadcast mobile TV services, but as yet not for VoIP.
Finally, there is great benefit in choosing a suitable public-safety codec. In a high-noise environment, such as a riot or a fire, the public safety professionals cannot find somewhere quiet to use their radios, and both the TETRA and Project 25 codecs were selected and optimised to work well in such high-noise environments, unlike cellular codecs which have different design targets. There is benefit to the public-safety community in continuing to use such codecs in the LTE network.
Whilst the choice of an LTE network specifically designed for public safety seems appropriate, it will have to meet exacting user requirements to be capable of carrying mission-critical communications, and will need preferably harmonised spectrum, ideally in sub-1GHz spectrum bands.
For more information, please contact David Taylor, Lead Consultant, at david.taylor@analysysmason.com