Mobile technology has gone through several radical changes or ‘generations’. The first generation from around 1982, now often known as 1G, is classical analogue transmission of voices. Its successor, 2G from 1991, was the first digital phone technology, and it allowed data transmission at a speed of up to 64 kilobits per second (Kbps), text messages, and later, MMS. 3G from 1998 was the first broadband technology for mobile phones, with transmission speeds of up to 2 megabit per second (Mbps), enough for limited use of the internet. Today, most smartphones use the 4G network, introduced in 2008, with typical transmission speeds of 100-200 Mbps. With a new generation every seven to ten years, it is about time for the next generation, 5G, which is in the pipeline, and expected to be ready in 2019 or 2020. Until then, we will undoubtedly see a number of bids for something that presents itself as 5G without actually being it. For instance, AT&T in the USA has launched something they call ‘5G Evolution’, which is really just 4G LTE with some increased speed.
There is not yet an established standard, but 5G is expected to make use of very high frequencies for the transmission of large amounts of data, a few blocks at a time, with speeds up to 20 gigabits per second (Gbps) – about a hundred times faster than 4G, with the latent period down to a millisecond. Maybe, 5G will make use of the same open frequencies as Wi-Fi without interfering with existing Wi-Fi networks. But everything comes at a price, and the high frequency means that its range will be shorter than that of existing networks, and obstacles such as walls, hills, or even rainy weather can more easily block the signals. Some will be familiar with this problem from Wi-Fi where the range of 5-GHz transmitters is shorter than that of 2,5 GHz. So, the 5G transmitter masts have to be closer to each other, maybe down to covering individual housing blocks. This probably means that 5G will typically be an urban phenomenon whereas rural areas will still have to make do with 4G for quite some time. It may also be difficult to access the 5G network in natural parks and recreational areas, such as forests, marshes, and beach meadows, so probably, most smartphones will use 4G as well as 5G networks, and switch between them according to the situation.
The devices that will be using 5G are expected to be cheaper and less energy-consuming than the 4G devices of today, and this will make it possible to connect more things to the net, in the ‘Internet of Things’, and it is expected that there will be more than 22 billion devices on the net in 2021. Today, most of the devices on the Internet of Things are connected to the net via the 4G phone network, or Wi-Fi connected to cable internet. A rise is also expected in the number of so-called Low Power Wide-Area (LPWA)-connected devices that cannot transfer data as fast as 4G+ and 5G connections, but use much less electricity, and have low connection costs, but a long range. This kind of connection is particularly well-suited for solutions that do not produce large sets of data, such as power, gas, and water metres, street lamps, tracking devices for pets, and personal assets.
Other devices, for example those of smart city systems, produce large amounts of data, and in such cases connections with poor bandwidth can become a limitation. Here, the 5G network can provide better bandwidth, and shorter latency, so larger amounts of data can be processed in real-time. This will also enable far more advanced smartphones, smartwatches, and wearables, because they won’t need much built-in processing power, but can access processing power in the cloud via 5G without noticeable delay. They will also be able to receive real-time data from sensors and other devices in the vicinity, relating to air quality, pollen, traffic, and much more. Future generations of smart devices, such as TV sets, laptops, and VR glasses, will probably also be directly connected to the phone network instead of using Wi-Fi and Bluetooth.
The semi-autonomous cars of today are connected to central computers via the 4G network, but we are not expected to derive full advantage of robot cars before they can communicate with each other, either directly from car to car, or via transmitter masts in lamp-posts and road-signs. In this area, the short latency period of the 5G network becomes important, for if the cars are moving at 60 mph, they can’t wait for very long for responses from each other. The same properties make the 5G network suitable for communication between robots in general, from flying drones delivering parcels and pizzas, to military combat robots in the field.
In October 2016, the Internet of Things was struck by a virus that infected and took over a large number of devices, and coordinated them in attacks on several major websites, such as GitHub, Twitter, Reddit, Netflix, and Airbnb. This is only one example that shows that the internet, as well as the phone networks, need to become safer than they are today. To this, we may add the increasingly massive surveillance of internet traffic and phone conversations that violates privacy. With the prospect of many more devices on the internet, and much more data processing redistributed to the cloud, 5G contributes to these problems – but may also become part of the solution.
With up to a hundred times more bandwidth than 4G, 5G allows the use of a much more thorough encryption than today. Even if solid encryption increases the amount of transmitted data by a factor of ten, the underlying data will be transferred more quickly than today, and with increasingly faster computers, encryption and decryption will delay communication less and less.
Today, encrypted websites are controlled by a small group of certificate authorities, and in some cases, these authorities have been hacked. It has been proposed that blockchains could be used instead, for the verification of encrypted connections in a decentralised fashion that will be less vulnerable to hacker attacks than centralised institutions.
When we access the internet on our computers, this is typically done through cable, either directly, or by a Wi-Fi network that establishes a wireless link to a cable connection. It has, however, become common to see people use smartphones as Wi-Fi hotspots, even if the data transmission limitation of a few gigabytes a month puts some natural constraints on this practise. With 5G, where a gigabyte can be transferred in seconds, the limits for data transmission will inevitably be raised or even disappear altogether, as will be the case for many internet subscriptions. This means that traditional cable internet may have a hard time competing with the telephone network. Historically, internet bandwidth has been increased by nearly 50 percent every year (an observation known as Nielsen’s Law), or about 50-60 times each decade.
As we have seen, the bandwidth of telephone networks grows about 100 times in a decade, or 60 percent a year. If this development continues, the telephone net will slowly but surely become better able to compete than the cable net, since new generations of cable networks typically require the burying of new cables, something much more complicated and cost-intensive than erecting transmitter masts. This prepares the ground for the kind of development we have seen in telephony where mobile phones have almost completely replaced landline telephones. This, however, requires that the telephone subscriptions keep up with the developments. In 2016, an average American household retrieved about 190 gigabytes of data from the net each month (chiefly video streaming), so if the telephone network is to replace the cable internet, a typical subscription must provide at least a couple of hundred gigabytes a month at a reasonable price.
As future generations of the wireless telephone network come to use shorter and shorter wavelengths that provide greater bandwidth at the cost of a shorter range, the transmitter masts will be closer and closer; so in time, it may not make much difference if our access to the net is obtained through cable or transmitter masts. Maybe, we will find an optimal balance between bandwidth and range for wireless connections where it makes no sense to increase bandwidth if this requires that the user has to be within a few feet’s distance of a transmitter mast. Cable connections are not subject to the same kind of limitations, so if the need for bandwidth keeps growing, cable internet and stationary computers may make a comeback sometime in the future. It is, however, not certain that ordinary people will need more bandwidth than the 5G network can supply, for everyday use. Today, 95 percent of data traffic on the net is video, and after all, there are limits to how high an image resolution people really need when watching movies and series, or playing computer games. Or are there?