Report: 5G Technologies Workshop by IEEE

Workshop schedule [PDF]

Why is 5G needed?

The “Internet of Things” (IoT), increasing number of telephone connections, increasing use of data by mobile devices, and an exponential increase in the diversity of wireless applications, all require a backbone network in the form of 5G. At the same time, network operators need a sound business model. According to a white paper released by Cisco, the Visual Networking Index: Global Mobile Data Traffic Forecast Update 2014 [PDF], the number of connected devices will go up from 7 billion in 2014 to 12 billion in 2020. These include smartphones, mobile phones that are not smartphones, laptops, tablets, wearables, devices that support machine to machine communication (telemetry, automotive, smart grid and transaction devices), and other portable devices. Mobile data traffic is projected to have a compound annual growth rate (CAGR) of 61% from 2013 to 2018, and the number of connected wearable devices are projected to increase with a CAGR of 52% in the same duration.

5G is projected to completely roll out by the year 2020.

What are the use cases of 5G?

Driverless automobiles, vehicle-to-vehicle communication for road safety, smart cameras in cities, virtual reality gaming, remotely operated robots, tactile internet applications, applications in the cloud, and networks of sensors placed in home, industry and office environments are some of the use cases of 5G.

5G is expected to contribute to massive connectivity and ubiquitous coverage required for smart manufacturing, smart devices in the consumer electronics and personal electronics domains, smart healthcare, smart retail, smart transportation and so on.

Is there a definition of 5G yet?

There is consensus on the following new requirements for wireless communication systems. Most of these will mature beyond 2020.

• Speeds of up to 10 gigabytes per second: 100 times faster than 4G LTE and 10 times faster than LTE-Advanced.
• Very low latency, which will support augmented reality and tactile Internet.
• Very high mobility: Gigabit Ethernet is a form of LAN (local area network) technology that supports data transfer rates of approximately 1 gigabit per second. 5G would support Gigabit Ethernet (also known as Gigabit Everywhere). The user can experience high levels of mobility.
• Low energy consumption in networked mobile devices compared with the current rates of energy consumption in order to make mobility sustainable.
• Machine-to-machine (M2M) communication among a large number of devices to support the “Internet of Things”.

Challenges in implementing 5G in India/ Ways in which massive growth in the number of users, connected devices, and network traffic can be handled

5G is expected to provide numerous high quality services to millions of users and to provide connectivity to many heterogeneous networked devices, which may contain multiple multiband radios.

• While capacity and data throughput need to increase, the amount of radio spectrum available is limited. New spectrum bands could be released to deal with this additional demand. (Refer “Emerging spectrum licensing options” below.)
• Additional base transceiver stations will need be set up.
• 5G will exist alongside multiple and heterogeneous networks (3G, 4G, WLAN, NFC, and beyond with macro/femto/pico cells) while supporting numerous services all the time, many of which will be data-intensive will severely reduce the battery life of mobile devices and hence devices with greater battery life and very efficient power management will be required.
• Also, combining various technologies (LTE with WiFi and/or legacy infrastructure with WiFi) and implementing heterogeneous networks will cater to increased demands. LTE will remain the baseline technology for wide area broadband in the 5G era. Interoperability with 4G will be critical to the adoption of 5G.
• The backhaul will need upgrading in order to maximise speed and bandwidth. The National Optical Fibre Backbone is an infrastructural project that could work towards meeting this need.
• Spectral efficiency needs to be improved but currently spectral efficiency is as optimum as it can get. In light of this, the solution lies usability of currently available technologies, for instance, LTE Release 8 has not been deployed yet. The concept of Small Cell has been defined in LTE Release 12, which has been optimised as much as technologically possible for the current bands. A potential enhancement being discussed for Release 13 is to make LTE operable with unlicensed spectrum bands as well.
• Some amount of traffic can be offloaded using free spectrum (WiFi offloading) as well as the existing cellular spectrum (device to device offloading). [Refer: Licensed-assisted Access below]

--- Offloading can happen via cellular-to-WiFi hotspots (LTE and Advanced LTE)

--- Via Cellular small cells and relays (LTE/Advanced LTE, 3G)

--- Via co-located cellular and WiFi (Advanced LTE)

--- Via proximity services and D2D communication (Advanced LTE)

A unified solution lies in the end user enjoying ubiquitous connectivity and consistent user experience, and the telecom operator being able to operate efficiently with the gradual rollout of 5G and with close integration between heterogeneous technologies, with the use of the cloud and software defined networking (SDN) as underlying principles.

Emerging Spectrum Licensing Options

Static assignments and exclusive licenses of spectrum paved the foundation for reliable services and innovation on technology evolution. Cellular systems have now gained the ability to operate on frequencies of up to 5 GHz and to operate over system bandwidths up to 100 MHz.

To deal with the increased demand for spectrum, new spectrum bands could be released. These bands are likely to be in the higher frequencies with a carrier bandwidth of up to 1 GHz. Initial research shows that such high frequency bands might require the development of a new radio waveform, a new radio technology. It is not known yet if and when the standardisation of the new radio technology will be undertaken. Spectrum is a costly investment, so telcos tend to be very picky about it.

The prevalent scenario in India

• India is not seen as an early adopter of new wireless communication technologies.
• Indian telcos invest late in ‘new’ technologies, but they invest massive amounts of money and for periods of time comparatively much longer than in early adopter countries. Indian telcos are still starting and/ or expanding their 4G operations and these will stand for a long time to come.
• Call tariffs are among the lowest in the world.
• New spectrum is unlikely to be released in the near future. World Radio Conference 2019 (WRC 2019) is likely to be the earliest possible time of release.
• 5G system design is therefore likely to happen over two phrases: Evolution design (up to 5 GHz) by the year 2020 and Revolution design (in bands over 5 GHz) around 2023.

Licensed sharing/ Authorised sharing of spectrum

One way of dealing with the spectrum crunch, could be spectrum sharing. When the holder of spectrum is known to be underutilising it, and there is little possibility of changing the policies governing the quantum of utilisation, spectrum sharing is a preferred solution.

License-exempt use of spectrum with the implementation of policy guidelines could be practiced in cases such as apartment complexes and bus stations handling large amounts of traffic.

Traffic offloading: LTE over unlicensed spectrum

-- Licensed-Assisted Access (LAA)

LAA can be use to opportunistically boost data rate. It works by aggregating a primary cell, operating in licensed spectrum to deliver critical information and guaranteed Quality of Service, with a secondary cell operating in unlicensed spectrum. LAA can be implemented globally in the 5 GHz band. The secondary cell operating in unlicensed spectrum can be configured either as downlink-only cell or contain both uplink and downlink. This also facilitates some degree of co-existence between the operators of LTE and WiFi as well as among LTE operators.

How do the availability of network services and cost efficiency affect network performance?

– In general

Availability of network service is inversely proportional to bit-rate.
Costs pertaining to terminals and networking are inversely proportional to latency.
The length of the battery life of terminals (i.e. mobile devices) is inversely proportional to spectral efficiency.

– 2G

2G technology has an emphasis on voice and SMS with low bit-rate and low spectral efficiency. This leads to high mobility and high availability of the network service at the boundaries of the cell. (The cell boundaries are the furthest from the cell tower in terms of physical distance.)

– 2.5G

2.5G (EDGE and GPRS) provided higher bit-rates and hence lower spectral efficiency and lower availability of the network service at the boundaries of the cell.

– 3G

Targets for 3G performance were not comprehensively defined by the ITU-R in the IMT 2000 set of standards. The target defined peak bit-rates for a single user. Hence, in its early stages 3G did not meet expectations for data transfer speeds.

– 4G

Mobility continued to be low in 4G technology even as the ITU-R provided more comprehensive specifications by including spectral efficiency and latency targets. Like 3G, 4G focussed on single-user peak data rates.

– 5G

Attributes currently proposed for 5G:

• High connection density (approx 10^5 users per km^2)
• Low latency (less than 10 ms)
• High bit-rate (approx 10^8 bits per second)
• High capacity density (approx 10^3 bits per second per Hertz per km^2)
• High spectral efficiency (approx 3 bits per second per Hertz)

Which would result in:

• Very high terminal costs for the operator (between USD 100 and USD 1,000)
• Low availability of network service
• Low battery life of the user equipment,i.e., mobile devices and fixed devices utilised by the end user (less than 1 day)
• Low energy efficiency (approx 10^-6 joules per bit)
• Somewhat low mobility (less than 10 km per hour)

However, targets for the Internet-of-Things and for public safety conflict with the above attributes and their consequences, indicating the possible future emergence of more than one technical solution.

Key technologies for 5G wireless communication networks

– Massive MIMO (multiple-input multiple-output)

The number of receiver and transmitter elements as (also transceiver elements) are expected to increase to 100- 1000 low-power antennas per base transceiver station (BTS).

- New antenna technologies

Large scale antenna system (LSAS), 3D-MIMO, Steerable array antennas

Transmitted radio significantly reduces as the number of antenna elements is increases. Hundreds of thousands of antennas could be used together to improve the energy efficiency of wireless communications. Steerable arrays of antennas could be used for dynamic beam-forming patterns.

– Cloud technologies for flexible Radio Access Networks (RAN)

Cloud-based network architecture would include a centralised base station with numerous radio units distributed over the cell and ideally connected by optic fibre in order to reduce latency.

– Advanced Interference Management

New air interfaces under consideration include:

• UFMC: Universal Filtered Multi-Carrier
• FBMC: Filter-Bank Multi-Carrier
• GFDM: Generalized Frequency Division Multiplexing
• SCMA: Sparse Code Multiple Access
• NOMA: Non-Orthogonal Multiple Access

– Network Densification

The proposed 5G requires new network architecture -- HetNet and Small Cell

– Millimetre wave band

New channels models will need to be developed to deal with different propagation conditions. 5G will work on higher frequencies, that is, in the millimetre wave band which has shorter wavelength (10GHz to 50GHz, 60 GHz, and possibly 70 GHz to 80 GHz) and wider bandwidths (500MHz to 3GHz). Millimetre wave (mmWave) MIMO requires dynamic beamforming at the transmitter and receiver.

More technical assumptions:
-- 5G will require significantly higher albeit low cost backhaul capacity (about 400 Gb/s).
-- 5G will have very low round-trip latency requirements.
– Higher frequencies and higher densities will dictate small cells.

Over-The-Top (OTT) Applications on Mobile User Equipment and IoT Devices

Reproduced with permission from Rohde and Schwarz

Green Communications Using 5G

The joint optimisation of Energy efficiency and spectral efficiency is critical for 5G research. There is still a long way to go to develop a unified framework and a comprehensive understanding of the tradeoff between energy efficiency and spectral efficiency. The latter has been pursued for decades as the top design priority of all major wireless standards, ranging from cellular networks to local and personal area networks. The cellular data rate has been improved from kilobits per second in 2G to gigabits per second in 4G. Spectral efficiency-oriented designs, however, have overlooked the issues of infrastructural power consumption.

The total power consumption of India’s mobile telephony infrastructure was 11.16 billion KWh in 2010. About 15% cell sites in India are either not connected to the electricity grid or receive power for less than eight hours a day; only about 10% receive more than 20 hours of power (Source: Intelligent Energy Limited). Two billion liters of diesel is consumed per year to power these cell sites, contributing to CO2 emission levels and massively adding to the operational costs of telcos.

The Indian Institute of Information Technology-Bangalore has proposed two power saving mechanisms.

1. Power-Saving Semi-Persistent Scheduler (PS-SPS) for VoLTE traffic in LTE-Advanced: This is a new scheduling algorithm implemented for VoLTE traffic in the downlink of LTE-Advanced cells in order to reduce power consumption in the link level. This proposed solution submitted to the IEEE is claimed to save power without affecting the network’s ability to support a large of VoLTE calls.
2. Random Access strategies for IoT devices in LTE-Advanced network: This scheme reduced the number of what is called “retransmissions” by Internet-of-Things devices so that these devices may complete the data transmission procedure [Random Access Channel procedure] in a comparatively short time. This scheme does not need the use of additional spectrum or barring mechanisms for IoT devices.

Glossary

Beamforming or spatial filtering is a signal processing technique used in sensor arrays for directional signal transmission or reception.  -- Van Veen, B.D.; Buckley, K.M. (1988). "Beamforming: A versatile approach to spatial filtering" (PDF). IEEE ASSP Magazine 5 (2): 4. doi:10.1109/53.665.

Latency is a measure of the time delay that occurs when data packets travel from one networked point to another.

Multiple-input and multiple-output, or MIMO, is a method for multiplying the capacity of a radio link using multiple transmit and receive antennas to exploit multipath propagation.  -- Lipfert, Hermann (August 2007). MIMO OFDM Space Time Coding – Spatial Multiplexing, Increasing Performance and Spectral Efficiency in Wireless Systems, Part I Technical Basis (Technical report). Institut für Rundfunktechnik.

Spectral efficiency is a measure of the performance of channel coding methods. It refers to the ability of a given channel encoding method to utilize bandwidth efficiently. It is defined as the average number of bits per unit of time (bit-rate) that can be transmitted per unit of bandwidth (bits per second per Hertz). – Taylor and Francis, Encyclopedia of Wireless and Mobile Communications, http://www.tandfonline.com/doi/pdf/10.1081/E-EWMC-120043448

Throughput is the amount of data transferred from one place to another or processed in a specified amount of time. Data transfer rates for disk drives and networks are measured in terms of throughput. Typically, throughputs are measured in kbps, Mbps and Gbps. – Webopedia, http://www.webopedia.com/TERM/T/throughput.html