Overview

The function of the medium access layer (MAC) is to share the available radio resource among the many thousands of devices that might wish to access it. As discussed in Chapter 4, the starting point in understanding the Weightless MAC is the design options that were selected as a result of the underlying requirements and constraints. Those that concern the MAC include:

• the use of TDD

• long time duration frames

• super-frame structure with sleep mode.

There is never a clean distinction between MAC and PHY and the MAC must provide PHY-level information to the terminals such as the hopping sequence hence there will be some blurring between this and the next chapter.

Scheduled and contended access

There are two ways that a terminal can obtain resource for communications or that the network can communicate with the terminal. These are scheduled access and contended access. Scheduled access occurs when the network and the terminal have pre-arranged to communicate at a particular time. For example, when transmitting a meter reading, the network and terminal may agree the next time that a reading will be sent and this resource is reserved in advance. Contended access occurs when there is information to send that was not envisaged. For a terminal, contended access requires attempting to gain radio resource to signal a desire to communicate. For the network it requires the base station controller making space within a downlink frame to carry the traffic.

Introduction

Dr Antony Rix

Weightless could be suitable for a wide variety of applications, from smart metering to healthcare and even asset tracking, as noted in Section 1.3. In this chapter we consider which application requirements fit well with Weightless, explore what is required to build a system that integrates with Weightless, and then look in more detail at example applications in automotive, energy, healthcare and consumer markets. This chapter is not a complete guide on building end-to-end services – substantial organisations exist with this as their main focus – but is intended to provide an informative introduction to some important considerations.

AlthoughWeightless is intended to be reasonably application-agnostic, it is important to understand that design decisions made in defining the standard, and the unlicensed nature of white space spectrum, place important limits on which applications will work best with a Weightless service. For example, while half-hourly readings from a utility meter could easily be sent over the Weightless system, video streaming would likely overload it.

Conversely, the requirements of some applications and business models do need to be addressed within the system, even potentially in the physical layer. Generally these requirements will be common to several markets, so the solutions are of quite generic benefit. Security is one of these requirements, and something which cellular networks address in detail. A topical example of why this is important could be the following. Consider a Weightless-connected diabetes monitor used by a celebrity. Machine IDs like MAC addresses are often assigned to manufacturers in ranges, each range to be used by a particular product to provide traceability, and thus identify not only the vendor but even the type of device. If the machine ID is sent unencrypted over the air, it might be possible for an attacker with a channel monitor to deduce that the celebrity has diabetes, track the celebrity’s movements, or even guess when the device is used. Appropriate security precautions in the system can make this difficult or infeasible.

What defines a machine?

Wireless communications are widespread, but tend to be used predominantly by people. Mobile phones allow people to talk or send emails, Wi-Fi systems allow people to surf the Internet from a laptop and Bluetooth links allow people to use cordless headsets. There is an entirely different class of applications for devices that do not directly have users and whose communications are not instigated by people. A good example of this is a smart electricity meter. It might send meter readings to a database every hour. It has no direct linkage with any person – although indirectly it makes their life somewhat better by enabling smart grids and automating the meter-reading process.

There are so many different applications and machines that a clear definition of one is not possible. Broadly, these are devices where transmissions occur due to the function of the machine rather than any person. They send information not to another person but typically to a database within the network from where it can be processed by other machines. Of course, sooner or later someone benefits from the service provided, but typically not from the radio transmission itself. Applications include automotive engine management updates, healthcare monitoring, smart city sensors and actuators, smart grids, asset tracking, industrial automation, traffic control and much more.

Security in general

Security is important in many machine applications. Smart grids must return accurate meter readings that cannot be modified. Network communications to instruct meters to change demand must not be modified or sent by a terrorist organisation. Financial-related transactions must be secure and it must be certain that they were received. Almost all machine communications have some level of security requirements.

To some degree these could be layered over the top of the Weightless network at the application layer. So a smart meter could encrypt its data stream before passing it to the Weightless radio embedded within it. The data would pass through the Weightless network in secure form and be decoded by the client in their central IT system. Indeed, this can be done even when Weightless has its own security to provide extra protection.

There are some functions that cannot be achieved at the application layer. These include authentication of the network by the Weightless device. Itwould be possible for a terminal to be ‘captured’ by a rogue base station and stay attached to it indefinitely. Authentication mechanisms are needed to prevent this occurring. Because of this, and because it prevented all applications having to build in their own security, it was decided to provide the Weightless standard with cellular-grade security mechanisms. However, as will be explained, some of the differences in message type and threat type mean that the same security mechanisms as are used in cellular cannot just be copied.

This book has introduced the Weightless standard, setting out the need for a wide-area wireless machine communications network and explaining how the advent of white space spectrum has provided the last piece of the puzzle needed to make such a network reality. However, the mix of very particular requirements for machine communications and the need to operate under complex rules and challenging interference within white space has led to the need for a bespoke new technology. Chapters in this book, and the standard, set out in detail the design decisions for Weightless, the specifics of the network, the base stations, the medium access layer and the physical layer. Subsequent chapters demonstrate how Weightless could be deployed in a nationwide network and the coverage, capacity and business case that would result from such a deployment. Finally, we concluded by looking at some of the key applications for Weightless and how they could be served by the standard.

Weightless is a unique technology. It could enable smart energy, smart cities, healthcare solutions to allow people to live at home longer and much more. It can generate excellent returns for those that make and deploy the technology and those that deliver applications using it. And, by providing the enablers to address climate change and the aging population, it might just save the world.

Defining white space

Most of the spectrum below 10GHz across most of the world is allocated for particular applications and assigned to certain users. These include broadcasters, mobile phone network operators, defence departments and many, many more. The net result is that there is little obviously spare spectrum, particularly in the preferred frequency bands between around 300MHz and 3GHz where propagation is favourable but antennas are conveniently small. However, measurements of the actual utilisation of this assigned spectrum suggest that it is typically only used in around 20% of the locations. Such measurements undoubtedly underestimate usage but nevertheless it is clear that there is some potential for more efficient use of the spectrum.

One way to visualise where there might be underused spectrum is to plot on a map the strength of signal from the licensed user of that band. If colours are used to represent signal strength then those parts with no coverage at a given frequency will be uncoloured and appear white on a black and white map. Hence the term ‘white space’ for areas where there is potential for others to use the spectrum.

Machine communications does not yet have the necessary standard

The observation, noted in the last chapter, that there was massive potential in machine communications is not a new one. Over the last few decades many have noted that the installation of a wireless connection into myriad devices would bring a range of benefits. However, the market for machine communications to-date has been weak. There are some cars with embedded cellular modems and some relatively high-value items such as vending machines are equipped with cellular packet-data modems. But the market today is only a tiny fraction of the size it has long been predicted to grow to. This is predominantly due to the lack of a ubiquitous wireless standard that meets the needs of the vast majority of the machine market as set out in Chapter 1. There is no current wireless system that comes close to meeting all of these requirements.

Cellular technologies do provide sufficiently good coverage for some applications but the hardware costs can be $20 or more depending on the generation of cellular used and the subscription costs are often closer to $10 per month than $10 per year. Battery life cannot be extended much beyond a few months. Cellular networks are often ill-suited to the short message sizes in machine communications resulting in massive overheads associated with signalling in order to move terminals from passive to active states, report on status and more. So while cellular can capture a small percentage of the market which can tolerate the high costs and where devices have external power, it will not be able to meet the requirements of the 50 billion device market. Indeed, if it could, it would have done so already and there would be no further debate about the need for new standards.

Introduction

It would be possible to write a book about Weightless without much discussion about the network. This is partly because the network is not included within the specification and partly because the network is relatively simple.

The Weightless specification covers the interface from the base station into the network. However, unlike cellular networks there are no defined interfaces within the network. This is because it is envisaged that all the network functions can be run on a virtual machine within the cloud. The interface from this machine to the base stations is well defined. The interface from this machine to the client networks will likely be bespoke for each client network and therefore not appropriate for specification. Hence, there are no further interfaces to standardise.

It might seem rather incredible that the core network of a cellular system consisting of many large and expensive computing platforms can be collapsed into a single virtual machine within Weightless when a machine network is likely to have many more subscribers than a cellular network and further much of the intelligence typically found in a base station in a cellular network has been transferred to the core.

The key requirements

Weightless is designed to enable machine communications in white space spectrum. This leads to two sets of requirements – those related to machine applications and those related to operating in white space. The machine requirements were introduced in Chapter 1 and are summarised below:

• Support of a large number of terminals. A typical cell might have between 100 000 and 1 million devices within it and a national network could easily contain 1 billion devices.

• Long battery life. Ten year lifetimes from one battery are needed in many cases.

• Mobility. A subset of applications has moving terminals which need to be supported as they move, potentially across national borders.

• Low-cost equipment. Costs of $2 per chip or less would appear to be necessary.

• Low-cost service. Network costs must be low and the marginal cost of each terminal very low.

• Global availability. Some applications will require global roaming. Others, like automotive, will require that one solution can be fitted into all vehicles regardless of their country of destination.

• Ubiquity. Excellent coverage, including within buildings, is needed.

• Guaranteed delivery. Some applications require certainty that messages have been delivered. This may also require strong authentication and encryption.

• Broadcast messages.

• Efficient transmission of small bursts of data. Most machines send data packets of the order of 50 bytes.

• Accommodating sub-optimal terminals. In many cases terminals will be small and low cost and will have a poor-quality antenna and limited power supplies.

• Event-stimulated loading peaks. The network needs to be able to accommodate and control the resultant peak in loading.

As discussed in the previous chapter, white space operation leads to the following requirements:

• Very low levels of out-of-band emissions. This minimises interference caused to licensed users and so maximises spectrum availability.

• Avoid interference caused by other unlicensed users. This can be random and sporadic.

• Reduce the impact of interference where it cannot be avoided. Where interference cannot be avoided the system needs to be able to continue to operate.

• Reduce power where there are few white space channels available. It is often possible to increase availability by transmitting with lower power and hence causing less interference.

Some of these requirements have immediate design implications as discussed in the section below.