5G Has Arrived
After many years of hype and expectation, 2019 will be recognised as the year in which 5G services began to be rolled out globally. This article looks at the key technologies which will underpin 5G’s network performance and some of the challenges that lie ahead as operators around the world begin to roll out 5G networks.
5G: What and Why?
Emerging applications such as augmented reality, (AR), autonomous vehicles and the explosion of the Internet of Things, (IoT), are driving unprecedented demand for mobile bandwidth on a global scale. In the face of this huge growth in demand, current 4G/LTE networks will soon reach capacity but, at the same time, these technologies do not offer the levels of network performance – speed and latency – required by many of the new applications.
Recognising this trend, the International Telecommunications Union, (ITU), in 2015, defined the requirements specification for the next generation of mobile communications networks – 5G. Unlike previous evolutions between generations, such as from 3G to 4G, the IMT2020 specification represents a revolution in mobile communications systems, requiring a transformational approach to network design. 3GPP, the global standards body for the wireless industry has integrated 5G into its release schedule and is working to complete the process with release 16, which is scheduled to be completed in time for the ITU’s 2020 deadline.
The ITU’s requirements specification for 5G is contained in the document ITU-R IMT-2020 (5G), summarised in Figure 1. The IMT2020 specification demands a step change in network performance with peak data rates of up to 20 GB/s (10 to 20 times faster than 4G rates), ultra-low latencies of 1 mSec, (c.f. 30 – 50 mSec for 4G) and connection densities of 1million devices per square kilometre.
Figure 1: Selected Key Performance Indicators of 5G according to ITU-R
(Source: “5G for Connected Industries and Automation”, 2nd edition, White Paper, 5GACIA, November 2018)
IMT2020 identifies three major use cases for 5G (Table 1) and recognises that most applications will require combinations of the features of each case, as illustrated in Figure 2.
Enhanced Mobile Broadband (eMBB)
Provides extremely high data rates (of up to 20 Gb/s) and offers enhanced coverage, well beyond that of 4G
Massive Machine Type Communications (mMTC)
Designed to provide wide-area coverage and deep indoor penetration for hundreds of thousands of IoT devices per square kilometre. mMTC is also designed to provide ubiquitous connectivity for devices with limited processing capabilities. Also supports battery-saving low-energy operation.
Ultra-Reliable and Low Latency Communications (URLLC)
Facilitates highly critical applications with very demanding requirements in terms of end-to-end (E2E) latency (one millisecond or lower), reliability and availability.
Figure 2: 5G Applications and Use Cases
The Innovations at the Core of 5G
To meet the IMT2020 requirements, network designers have started from scratch developing a completely new radio interface – 5G NR, (New Radio). 5G NR’s improved performance is based on a number of key technologies:
5G transmitted waveforms are based on a version of Orthogonal Frequency Division Multiplexing, (OFDM), which makes extremely efficient use of radio spectrum, operates well with high data rates/ wide bandwidths and is compatible with the processing capabilities of current mobile handsets. Two further key technologies adopted by 5G NR are Beamforming and Multi-User MIMO – or MU-MIMO.
Beamforming (Figure 3) is a relatively new technology which enables the beam from the 5G base station to be directed towards the end-user mobile device, ensuring optimum transmission levels whilst minimising interference to other, nearby mobile devices.
Figure 3: Beamforming
Already used in many wireless systems, including Wi-Fi and 4G, MIMO, (multiple input multiple output) is a key enabler of 5G performance. The variant deployed within 5G NR is Multi-User- MIMO, (MU-MIMO) which uses a large number of antennas, (32 in 3GPP Release 32, rising to 64 or more in future releases), in the base station. MU-MIMO uses a complex algorithm and spatial information gained from a CSI-RS (Channel State Information Reference Signal) to enable the 5G Base Station to communicate with multiple devices concurrently and independently.
Together with Beamforming, MU MIMO, enables 5G to support an order of magnitude, (>1000) more connected devices than 4G, transmitting high-speed, low-latency data to many more users.
Effective use of radio spectrum is key to 5G performance and 5G signals will be transmitted at a wide range of frequencies, depending on the use case:
- Frequencies below 1GHz, the ‘coverage layer’, provide wide area and deep indoor coverage
- Between 1GHz-6GHz, the ‘coverage and capacity layer’; C-band spectrum around the 3.5GHz mark delivers the best compromise between capacity and coverage
- Above 6GHz - the ‘super data layer’, including frequencies above 30 GHz, the mm-Wave frequencies, which are needed to deliver the high data rates of the IMT2020 specification.
5G will offer a range of enhanced network management capabilities, including Network Slicing, which enables operators to tailor services according to the needs of the application. As discussed above, Smart City applications may predominantly require high connection density, whilst factory processes will need extremely fast, low latency connectivity to respond to events in real-time.
5G technology challenges
Although 5G services have begun to appear, the specification is still being finalised and several challenges must be overcome by operators as they roll out their 5G networks.
Most initial 5G deployments will be in the 1 – 6 GHZ range, where existing 4G/LTE infrastructure can be leveraged to provide 5G services. Operators can also save money by “re-farming” existing spectrum allocations as previous generation (e.g. 2G) services are retired. At some stage however additional spectrum will be required and most countries already have or will soon hold auctions of spectrum in the 1 – 6 GHz range, obliging operators to make further investments in spectrum. This region is already crowded, however, and will become more so as 5G traffic increases, and ultimately investments will have to be made in mmWave spectrum to enable the forecast growth in 5G. Figure 4 shows one analyst’s view of the timeframe for this in various countries.
Figure 4: Estimations of when spectrum capacity will run out
Transmission at mmWave frequencies will require a huge increase in base stations and, with MU- MIMO, 5G’s base stations will have up to 64 transmitters, requiring more power than 4G. Base-stations already account for 60% of operators’ total network power consumption and, with operators’ energy costs estimated to be 15% of opex, innovative efficiency measures must be incorporated in the design of 5G networks.
The RF Power Amplifier, (RFPA), is a key determinant of base-station energy performance and designers are grappling with the challenges of building RFPAs which can balance higher power with more efficiency, whilst operating across wider frequencies. Traditional LDMOS transistors are beginning to be replaced by GAN technologies and techniques such as envelope tracking, DC-DC conversion and ultra-wideband Doherty architectures are underpinning significant advances in RFPA efficiencies and performance.
2019 is the year when the 5G vision begins to become a reality, and most countries will see one or more flavours of 5G service launched this year. 5G network design incorporates a range of innovative technologies such as beamforming and MU-MIMO and is driving rapid advances in RFPA architectures as designers strive to increase power and efficiency levels.
Although the 5G journey has well and truly begun, the roadmap includes several technical and regulatory hurdles which must be overcome before the full, high bandwidth, low-latency promise is realised.