The world is evolving rapidly. It seemed only yesterday that the Internet became available at home and the smartphone entered our everyday lives. On the horizon we see a completely connected world, aiming to make our lives more autonomous. This vision consists of smart cities with self-driving vehicles, delivery drones, smart homes and connected products. Today, cellular network providers are working towards this vision by introducing 5G.
What is 5G?
5G is the next generation wireless communication technology standard for cellular networks. 4G was mainly an introduction of mobile broadband communication, leading i.a. to mobile video applications. Its main goal was to provide broadband Internet to people everywhere. 5G however, is all about machines.
Figure 1: 5G frequency spectrum (source: NI)
5G is designed with the creation of a mobile ecosystem in mind, providing the means to fulfill new opportunities in a broad range of industries. The challenges are versatile. There are much more machines than humans. Furthermore, they can respond faster and energy consumption is a main issue. To tackle these requirements, the design know-how and required leap from 4G to 5G is very large and unprecedented.
5G frequency spectrum
While the focus in this perspective is not on providing detailed technical information about 5G, one major change has a key influence on communication behavior and product design: the frequency spectrum. 3 bands have been allocated globally and are auctioned in many regions:
- Low Frequency below 2GHz. Currently the focus is on 600-700 MHz (band 71 and 82). It’s the best possible spectrum for fast coverage penetration. However, as very precious and limited spectrum resource, they can only offer limited connections. Therefore, we cannot expect full-fledged 5G operating in this band.
- Medium Frequency range 2-6 GHz (sub 6 GHz). A new 5G spectrum which is the best frequency range to offer both coverage and capacity. This means that while penetration into buildings is possible, frequency reuse can be well planned.
- mmWave Frequency range (>26 GHz). The spectrum for denser populated areas with a higher capacity, which offers a greater freedom and more range to maneuver, in addition to less background noise. It’s excellent for providing capacity and offering extremely high data rates to specific use cases.
As each of these bands has a specific RF communication behavior, there’s a big influence on usage and design considerations to be taken into account.
Figure 2: Allocated spectrum per region (source: EDN)
Is 5G just a faster version of 4G?
The enormous amount of spectral range allocated to 5G points to an increase in communication bandwidth. Is 5G essentially just a faster version of 4G? The answer is no. The main benefits of 5G are to be found in multiple domains.
Figure 3: 4G vs. 5G (source: Western Digital)
The benefits of 5G can mainly be found in a higher bandwidth, a much lower latency with peak rates below 1ms, a much lower energy consumption from which battery driven mobile devices will benefit, and communication up to 1M devices per square km.
mmWaves you say?
The 5G specifications pose many challenges, of which the use of mmWaves is completely new. mmWave communication is a type of RF communication where line of sight (LoS) becomes a major factor. This means that the waves have a limited penetration depth, so buildings, trees and even rain decrease the link budget significantly. Base stations will likely offer up to 500 meter coverage, taking into account obstacles and foliage. This is obviously not a huge area, but this doesn’t mean it’s useless. Next to a broadband point-to-point connection, which mmWaves are ideally suited for, the above 26GHz range will probably only be used in urban centers or other places where a maximum number of consumers can be covered in a small space.
4G dead and buried?
5G is a new technology, its properties completely comprising the benefits of 4G. However, this doesn’t mean that 4G is dead and buried. 4G stays a valid alternative in case a broadband communication link is needed, as it doesn’t cope with the 5G challenges. Furthermore, it will stay for the years to come. 5G is just an extra, powerful possibility in the wireless communication technology selection that needs to be made during product design.
5G as a market enabler
As we explained earlier, 5G is all about machines. It mostly targets Internet of Things (IoT) as new market segment. The main benefits of 5G can be found in areas where other cellular technologies can’t follow, i.e. low latency in combination with high bandwidth.
Figure 4: 5G as a market enabler (source: Atlantik Elektronik)
Once 5G becomes widely adopted, we could see progress in areas that either consume or generate lots of data in real-time. These applications often combine 5G with Artificial Intelligence. Possible applications are in autonomous vehicles and robotics, augmented and virtual reality, Internet of context, predictive modeling, smart cities, factories and homes … It’s believed that 5G will play a major role in this.
5G roll-out
It was stated earlier that the leap from 4G to 5G is very large and unprecedented. This means that during the 5G roll-out, network operators are using a stepped approach.
Figure 5: Estimated roll-out scheme per region (source: Qualcomm)
The first step is adapting the existing 4G network to facilitate 5G in a non-standalone system (NSA). The advantage is that the construction speed is very fast. This allows users to experience the advantages of 5G network as soon as possible. This means that most of the base stations that are currently deployed by operators are NSA base stations.
Figure 6: Standalone vs. non-standalone 5G system (source:Gearbest)
For a complete standalone system (SA), operators need to rebuild 5G base stations and upgrade the equipment of the 5G core network. This is required to fully realize all the characteristics and functions of 5G network. But because all base stations and infrastructure need to be rebuilt, the cost of construction is quite high. Moreover, it takes a long time to migrate from 4G network to 5G core network. Reports predict that only in late 2021 SA networks will become available on prime locations in Europe.
Some people may ask if there’ll be a difference in terms of user experience, given the cost of NSA- and SA-systems. There won’t be much difference between the 2 in terms of speed and latency, NSA will be the 5G mainstream solution for the coming years.
No free lunch
For a large group of products or new product ideas, 5G has very attractive benefits. But there’s no such thing as a free lunch. Where the network operators need to bridge the gap towards 5G, for the product designer integrating this communication technology, 5G also poses many challenges. Verhaert’s EmbeddedLab aims to smooth things out for you. In the next parts, we provide more info on topics requiring extra attention.
Cutting edge technology
At the moment of writing, multiple chip manufacturers already have 5G chip sets ready and are broadening their portfolio to keep up with the 5G network evolutions. The most important manufacturers are Huawei, Qualcomm, Samsung and Mediatek. The same can be told about module manufacturers. Among others, Quectel, Sierra Wireless and Telit currently have chip sets from the earlier mentioned manufacturers embedded on their modules. This means that it’s already possible to incorporate 5G in your product! However, using cutting edge technology has its drawbacks. Early technology adoption inevitably brings implementation challenges such as immature embedded software stacks, which have their influence on budget and timing. Still, this is often an acceptable cost when compared to a longer time-to-market.
RF engineering & antenna design
Each new wireless communication technology has its specific RF design challenges. Due to the large frequency spectrum and introduction of mmWaves, massive MIMO and beamsteering however, the hardware design complexity is much higher for 5G in comparison to other cellular technologies. This makes that the inputs from an RF engineer during the design of a 5G connected product are invaluable.
There are various aspects that need to be taken into account. One example is the influence of printed circuit board (PCB) material. FR4 is the board material used in almost every electronics design. It’s a low cost material, and performs well enough for everyday electronics. In the GHz range however, this material has a very high attenuation, making this design a challenge.
Figure 7: Board material attenuation vs. frequency (source: Intel)
So how to integrate a 5G module inside a product? Not only the 5G module itself, but also the implementation of the surrounding product becomes important. This is done by providing a good layout, shielding topology, resonance omitting measures, up to the mechanical design of the product itself. When it comes to designing a 5G compatible PCB, knowledge is key. Not necessarily the abundance of information, but having the right information counts.
For many UHF, SHF and EHF applications, an abundance of information is present. A broad range of specific topics are covered, but in contrast to UHF, little general layout guidelines are known for SHF and EHF. Information is detailed but widely dispersed amongst many different topics. Many of the guidelines applicable for UHF applications fail at higher frequencies such as the SHF and EHF bands as the wavelengths shortens and ordinary precautions prove inadequate. An example of such a design rule of thumb is via stitching between planes with an interval of 1/20 to 1/10 of the minimal on-board wavelength that can be expected. As the wavelength of e.g. 3.5 GHz is 43 mm on FR4, stitching vias should occur near the transceiver with an interval of 2.1 to 4.3 mm.
It’s clear that this frequency lies in the zone in which this rule of thumb stops to be practically usable. But this reasoning shows that the need for concise guidelines for frequencies above the UHF band exists and grows with the global release of 5G as the next generation of wireless communication. The insights of RF engineers will therefore be invaluable for the introduction of 5G in new products.
The same thing can be told about antenna design. Design of a broadband antenna is a specialty itself. Going to mmWaves however, this task becomes more challenging, as at these frequencies effects of the transceiver front-end come into play.
Figure 8: mmWave antenna (source: Taoglas)
Although the first 5G antennas are becoming available for sub-6-GHz as well as the mmWave range, it’s vital to take an extra look at the front-end combination as well as the positioning in the product. Not your cup of tea? Let’s guide you through this adventure.
EMC aspects
The complex integration of a 5G transceiver system into a product requires special attention. One of the critical effects to take into account is: self-interference. On one hand, there’s the computer system collaborating with such a high bandwidth communication system. This is expected to be an AI processing unit. To reach the required processing speed, a very fast transistor technology is used. Not only is the input capacitance much lower to be able to make the rise and fall in tens of picoseconds, providing an input bandwidth surpassing the sub-6 GHz 5G frequency band (e.g. PCI express). Furthermore, the supply voltage level of these chip sets is low, making them more susceptible to interference.
But no interference without receiving antenna. The 5G transmission frequency range is higher than in case of other commonly used wireless communication techniques. This means that a nearby transmitter is able to excite shorter structures. At this point, the mainly used dedicated 5G frequency is 3.5 GHz. This has a wavelength of 43 mm on FR4. This means that a trace of (multiples of) about 1 cm can become a great reception antenna. These are commonly used trace lengths on a regular PCB. And the susceptibility will further increase once the transmission frequency range broadens. A shielding approach needs to be defined.
And then there is the test problem. The radiated immunity test frequency range is currently defined by the product type and the operating frequency of the device-under-test. The standard radiated immunity test range is limited to 1 GHz. For higher frequency devices, an extension is made up to 6 GHz. This means that lots of devices have not been subjected to the higher cellular frequencies and their harmonics. This means that it’s unknown if unsafe situations can occur when a 5G transceiver is placed in the vicinity of a device. While it could be expected that a system not operating in the 5G operating range is immune against EMI, this is not the case. This means that to be able to bring a safe product on the market, the EMC test landscape will need to change in the coming years.
Interesting times
At the beginning of this perspective, a completely connected world was drawn on the horizon. Various manufacturers from different markets are working together to realize this. 3GPP and network operators are bringing the next building block: 5G, a wireless communication technology with the potential to open up new markets for anyone seeing the opportunity. And yes, each new technology comes with a few challenges that in hindsight will prove to be not as insurmountable as initially foretold. We’re living in interesting times, in which we, as Verhaert EmbeddedLab, are happy to participate.
Sources: raconteur.net – datamakespossible.westerndigital.com – verizonwireless.com – ni.com – androidauthority.com – tutorialspoint.com – qualcomm.com – taoglas.com – quectel.com – atlantikelektronik.de – edn.com – gearbest.com – intel.com – electronic.se
Download the perspective
Looking for solutions to innovate?
Leave us your email and get in contact with Lieven Claeys, Business Development Manager Services ‘Digital Innovation’, to help you with your innovation process.
