Why the Rush to Voice over LTE?

FirstNet is hard at work developing a business plan and an architecture model for the first Nationwide Public Safety Broadband Network (NPSBN). When this network was first envisioned, it was to be a fully interoperable data and video network designed to augment, not replace, Public Safety voice over existing Land Mobile Radio (LMR) systems. Yet even before it becomes operational, some are pushing to add voice to the network. The question in my mind is why the rush to voice.

When Long Term Evolution (LTE) technology was chosen by the Public Safety Community then approved by the Federal Communications Commission (FCC), it was envisioned that the network would be used for data and video services and would be designed to provide access to and from first responders in the United States regardless of where they were. It was also envisioned that the network would be the foundation for interoperability on a nationwide basis, helping to fix the issues that have been hounding Public Safety for more than thirty years but only became known to the public because of the communications failures during the 9-11 attacks, Katrina, and several other major and widespread incidents.

During the long and often contentious discussions with Congress and the FCC that finally resulted in the assignment of the 700-MHz D Block to Public Safety, the formation of FirstNet and the initial funding for the system ($7 Billion), the use of voice on this network was somewhere in the future and not the primary reason Congress approved the re-assignment of this spectrum to Public Safety, nor was voice thought to be an integral part of the network for many years to come. Yet before we even get the network up and running, the idea that voice will be included from day one has gained momentum and now seems to be expected by many.

Today, commercial network operators are using their 2G and 3G systems for voice and LTE is being used for data and video. Yet Voice over LTE (VoLTE) is on all of the network operators’ drawing boards. However, dial-up voice for voice phone calls, what they consider an important voice service, is a far from being a requirement for the Public Safety community. One little known or acknowledged fact about the reallocation of the D Block to Public Safety is the requirement that Public Safety return the T-Band (470-512 MHz) to the FCC for repurposing. This requirement and the burden it is causing the eleven major metropolitan areas affected was placed on the Public Safety community because a single vendor in the wireless broadband business (not in the Land Mobile Radio business) was able to convince some congressional leaders that Voice over LTE for Public Safety is just around the corner and that those now using the T-Band could easily move down into the Public Safety LTE spectrum.

The Public Safety community has often stated that Voice over LTE is something that will happen sometime in the future but LMR systems need to be maintained and even expanded for the foreseeable future. APCO, NPSTC, and others have published documents about the importance of maintaining existing LMR systems, yet for whatever reason, there is more attention than ever being placed on adding voice to the NPSBN. Recently, the Open Mobile Alliance (OMA) proposed an open standard for push-to-talk (PTT) over LTE while MotorolaHarrisTwisted Pair (recently bought by Motorola), AT&T, Verizon, Sprint, and a handful of others are all talking about and demonstrating push-to-talk over LTE for Public Safety. Even in the upcoming Public Safety grade NPSTC report, push-to-talk over LTE is discussed and characteristics are being mapped. The Public Safety Communications Research (PSCR) group in Boulder has been very vocal with the standards organizations and has had some success placing a work item on the 3GPP standards agenda. It appears as though the ground swell for voice over LTE is growing in spite of the dangers of assuming that push-to-talk for Public Safety will, in fact, become a reality sooner rather than later over the Public Safety LTE network.

I would like to see everyone back off and wait until several things happen. First, let’s get the network in place for data and video services. Let’s see how much capacity is really needed by Public Safety, especially during incidents that are confined to a single cell sector (which I believe will be the majority of incidents). Let’s wait and see how much spectrum is available for adding partners that will assist FirstNet in funding the network build-out and then reselling excess capacity. At the moment, we don’t know how much, if any, excess capacity there will be in major metropolitan areas. Until Public Safety is up and running with data and video services, we won’t know how much capacity Public Safety will need either for day-to-day routine patrols or for incidents. Adding voice before we have some metrics to help determine usage patterns does not make sense to me. To simply assume that we will have capacity could result in instances of system blocking, which could impact not only data and video services but voice services as well.

Next there is the fact that, today, there is no VoLTE standard for push-to-talk. Permitting multiple and incompatible flavors of PTT on the Nationwide Public Safety Broadband Network does not make sense simply because it would set Public Safety back in terms of voice interoperability. Today there are still a number of Voice over LMR interoperability issues, but they are fewer and further between than they have been. Until and unless there is a common PTT standard, my vote is to not permit any form of voice communications on the Public Safety LTE network.

Adding dial-up voice to the Public Safety network does not make any sense to me either. It will radically change the system architecture and cost more money (we don’t have enough to build out the network as it is). Commercial operators provide robust dial-up-voice services on their 2G and 3G networks today. Moving forward, they will be adding VoLTE to their own LTE networks so they can refarm their 2G and then 3G spectrum, upgrading it to LTE to gain more capacity and provide better broadband services for their customers.

Beyond dial-up voice, moving to on-network push-to-talk over LTE will have to be done in several steps and no one is sure of the timeframe. The initial step will be to offer non-Public Safety grade PTT services over LTE and tie the LTE network to the various Public Safety LMR networks in order to provide the first level of interoperability. This will require that the Public Safety community, FirstNet, and the vendors agree on a common and fully interoperable PTT standard over the NPSBN that will be presented as non-Public Safety grade PTT for administrative purposes and not for dispatch or mission-critical use.

Next will probably be what is considered as Public Safety grade PTT services using the LTE network, still interconnected to existing LMR networks. This version of PTT will also need to follow the standard chosen for the Public Safety system, and this type of PTT voice service for use for dispatching routine and incident traffic needs to be more robust and well thought out. Many of today’s PTT over LTE systems require that each device that is in PTT mode be assigned one or two resource blocks to keep them live. By design, LTE is based on the allocation of resource blocks on a per-user basis. A user will be assigned more resource blocks for streaming video than for simply using the network for data services, and a PTT session will need even fewer resource blocks. However, using most of the current PTT technologies, these blocks will need to be assigned to a device to keep it “alive” even when it is in listen mode. Therefore, given an incident within a single cell sector and the need for both uplink and downlink data and video services, a number of PTT users, all requiring resource block allocations, could have an impact on the amount of capacity left for data and video services.

It is not clear to me whether those working on solutions for when or if PTT is to be used for Public Safety grade dispatch and day-to-day operations truly understand that a single voice path in a given area will not be sufficient. Multiple talk paths will be needed so dispatch areas can be zoned, as is common in most major cities. That is, in the world of LMR, it is not unusual for a single precinct or group of precincts to have their own primary dispatch channel as well as some citywide channels. PTT over LTE will have to solve this problem, perhaps by the inclusion of talk groups as currently used with LMR trunked radio systems where each system supports a large number of talk groups as well as a large number of individual users within each talk group.

Over time, all of these issues will probably be addressed and solved, but my point here is that as of today, they remain issues and it is not wise to bet the farm on the fact that they will be solved and within what timeframe. Instead, the smart course of action is to place the NPSBN into operation as a data/video only network, learn how to manage the available network capacity, see what types of applications are needed, and which are used. From that, and with some careful gathering of metrics, we should then be able to determine if and when the network is capable of supporting voice and, if so, what type of voice services should be offered. One thought is to use PTT voice over LTE, interconnected to the LMR voice networks only for improving Public Safety voice interoperability, keeping true dispatch and field operation on existing LMR channels.

Off-Network Voice

Perhaps the least understood requirement for Public Safety voice services is the need for off-network one-to-one and one-to-many voice communications. This form of communications is used on a daily basis by many first responders. It is sometimes called simplex communications or talk-around, or perhaps in the IT world it could best be understood as peer-to-peer communications. We all know or should know that a cell phone out of range of a network is a useless device. It won’t communicate with anyone or anything until it reconnects with a network. Public Safety on the other hand, has the ability to talk unit-to-unit when out of network coverage or even when in network coverage but when working in groups near each other. Simplex communications is vital when out of network range but it is still extremely important even when devices are within range of a network.

Simplex communications takes the radio traffic off the network and confines it to local one-to-one and one-to-many channels. It is used to talk into areas not covered by a network (deep inside buildings, when out of network coverage, etc.) as well as to off-load traffic from the network in order to make the network available for other users and incidents. A typical scenario might look something like this: A fire is reported in a warehouse, multiple units are dispatched, EMS squads are dispatched, and law enforcement is notified and may respond to assist with crowd and traffic control. The first engine on the scene will report back over the network as to the condition of the incident. If it is a working fire, the incident will be assigned one or two simplex or tactical radio channels for fire ground use. These communications are intense, at least in the initial stages of the fire, and having all of the voice traffic on a voice dispatch channel would congest the network to the extent that the dispatcher might not be able to dispatch another incident as quickly.

During an incident, multiple simplex channels might be used. As an incident grows, especially larger fires or major law enforcement efforts, several groups, each with its own set of tasks, may be assigned to different simplex channels. In this way, each group can communicate with all of its group members without interfering with other groups assigned different tasks. Normally, only a few of these simplex channels are needed but during a major event such as a wildland fire, upwards of thirty to more than fifty simplex channels may be pressed into service.

Can off-network or simplex communications really be achieved over LTE? There are those who believe it can be done, but a number of obstacles must be overcome. The system must provide off-network communications both when units are out of network coverage and when within network coverage. Thus the first issue is how to design a device that can recognize that you want to have an off-network PTT conversation when the network is in control of the device, and what it will cost.

Next is the issue of the RF output power. A standard LTE device has an RF power output of about 250 milliwatts, and an internal antenna. The true output power of the device is more like 100 milliwatts. Most of the handheld Handie-Talkie devices  (HTs) have an external antenna and an RF power output of around 5 watts. In other words, the LMR PTT radio has 500 or more times the power of an LTE handheld device. Further, existing LTE devices provide battery life based on this RF power level. If the power level is raised, even when in simplex mode, the result will be reduced battery life that might not last a complete shift.

Next is the issue of how to communicate via simplex or peer-to-peer communications with an LTE device on 700 MHz. The device will be transmitting in the 788 MHz to 798 MHz portion of the band and will be listening on the 758 MHz to 768 MHz portion. That is a difference of 30 MHz and in order to accomplish simplex communications, each device will probably require a second transmitter or a second receiver. This is not in keeping with the goal of employing LTE to take advantage of commercial LTE wireless technology and standards. LMR radios are normally designed, from the beginning, to operate over a wide portion of the spectrum and to be able to move between two different transmit channels, especially since they are in PTT mode, which means they transmit, then receive. LTE devices operate in full-duplex, transmitting and receiving at the same time. They normally use a common antenna and filters and what are known as duplexers to filter out the transmitter signal from the receiver signal before passing it to the radio, enabling a single antenna to transmit and receive simultaneously. These filters and duplexers would also have to be bypassed in order to operate in simplex mode.

There are a number of other issues that will need to be addressed before simplex over LTE can become a reality. I have touched on the two major ones. I am not saying they are insurmountable, only that the solutions are not available today. In talking with a lot of very smart LTE engineers, not one has been able to give me a date when these issues and others will be resolved. I am told that given time they PROBABLY can all be resolved, but the key phrase here is “given time.”

My final point before I conclude this article is that if those within the Federal Government who believe broadband is the only future of wireless come to believe, as some are saying, that Voice over LTE for Public Safety is only a few years away, they are liable to pass additional laws that will require Public Safety to give back even more LMR spectrum. Giving back the T-Band will create enough anguish. During the Boston Marathon bombings, spectrum in the T-Band provided the interoperable voice communications that enabled the massive communications efforts. NYPD, FDNY, and many others in the New York area rely on T-band spectrum for their primary Public Safety voice communications as do Chicago, LA, and every other area where the T-Band is widely used. The T-band is also used by volunteer fire departments in the Philadelphia suburbs, and by many business and industrial users in the eleven cities where it is authorized.

What if, because of false expectations, the Public Safety community is required to give up even more of its spectrum? California Fire services are some of the most interoperable in the United States. Each mobile radio is capable of at least 320 voice radio channels and each handheld device is capable of at least 160 channels. Most or all make use of VHF radio spectrum (150-174 MHz including 170-174 MHz channels used by Federal Government Forrest service and other Federal Fire agencies). What if they had to give up their VHF spectrum? What happens if LTE voice services are not ready for prime time and cannot provide all of the voice services needed by these agencies?

Wireless technology is advancing at an amazing pace but those passing laws and making decisions about the future of wireless only know about their handheld devices, tablets, and perhaps laptops connected to commercial broadband wireless networks. They don’t know that voice communications is the lifeline of the first responder community and MUST be available no matter what. If first responders cannot communicate via voice, even if they have interoperable data and video, they are putting their lives and the lives of all citizens at risk.

To those who believe voice over LTE will be able to replace LMR Public Safety voice anytime in the near future, I ask you, “Will you bet your life on it?

Andrew M. Seybold

print

3 Comments on "Why the Rush to Voice over LTE?"

  1. I couldn’t agree more Andrew! As usual you’re talking a huge amount of good and common sense here. In particular I think a lot of people on your side of ‘The Pond’ should really ponder long and hard on your final paragraph. I for one would not like to see any Public Safety or other critical user’s life left at the mercy of voice over LTE anytime soon!

  2. Excellent article Andy, thanks for highlighting the bigger issues before anyone can seriously address VoLTE for PS.

    By the way, great to see you again after so many years after our time at IBM.

  3. I too agree that we should not rush into VoLTE for PTT but we need to lay the groundwork for such an offering today. The LTE VoLTE standard needs to be set by FirstNet. No other platforms for PTT should be allowed.( Their license, their kingdom!) I believe that you will see RF and RoIP portable and mobile radios within the next 5 years. If you would have made to comment to me 5 years ago,I would have taken you by the hand and introduced you to a good psychologist.

    Rf communications will never go away. There will always be a need to talk directly between users-without any infrastructure. And even furthermore a need for radio networks with total interoperability (maybe even to a greater extent that P25 II). I do believe that RoIP is a cost effective way top deliver interop communications to public safety entities with very limited budgets. Some volunteer fire companies could deplete their entire yearly budget with the purchase of just 6 P25 radios.

    VoLTE could be a cost effective solution to bridge the gap with inexpensive PTT handsets.

    The last thing we need on FirstNet is 20 different VoLTE operating systems out there, each one talking in its own language. We all need to stop at the fork in the road and carefully choose the right path!

Leave a comment