Professor Jie Zhang Issue: Africa and the Middle East 2015
Article no.: 13
Topic: Delivering mobile data indoors and out
Author: Professor Jie Zhang
Title: Chief Scientist & co-founder
Organisation: Ranplan
PDF size: 337KB

About author

Professor Jie Zhang is Chief Scientist and co-founder at Ranplan, an innovative wireless technology company that has developed world leading software tools for outdoor/indoor wireless network planning, design and optimisation. Since 2011 Prof. Zhang has also held a Chair in Wireless Systems in the Department of Electronic and Electrical Engineering at the University of Sheffield. He and his students have pioneered research in femto/small cell and HetNets and published many widely cited papers.
Prior to Sheffield, he worked with Imperial College London and Oxford University and has been awarded over 20 research projects by the EPSRC, the European Commission FP6/FP7/H2020 and industry, including some of earliest projects on femtocell/HetNets.

Article abstract

Prof. Jie Zhang from Ranplan looks at the challenges for equipment designers and radio planners to deliver seamless mobile broadband coverage and Quality of Experience (QoE) 

Full Article

The original purpose of a mobile network was to provide voice coverage when users were away from their offices or homes. This was well served by outdoor macro networks which could largely be planned using straightforward outdoor RF coverage analysis.

In 2019, around 50% of mobile devises in Africa and the Middle East will be smartphones and 70% of mobile subscriptions will be either 4G or 3G. The regional growth in demand for mobile broadband has been unprecedented and will continue to escalate as the infrastructure continues to evolve. This demand has driven usage indoors and today this represents an estimated 80% of total mobile data consumption. The Middle East and African In-Building Wireless Market is estimated to grow to US$1,049.2million by 2019.

The default planning position adopted has been an outside-in approach, where outdoor macro coverage penetrates into buildings. But with subscriber densities increasing, the traditional ‘one size fits all’ outdoor macro network model is being supplemented with small cells, DAS (Distributed Antenna Systems) and Wi-Fi to both increase capacity and improve indoor coverage depth.

These systems must all work together in the ‘heterogeneous’ network to deliver good experiences wherever consumers are – and for an ever-widening range of different service types and applications. This requires careful RF (Radio Frequency) planning to model outdoor and indoor environments and their joint propagation patterns in order to understand how the different network components interact and interfere.

A holistic approach to network performance modelling is essential if we’re to take the guesswork out of understanding how these different types of infrastructure will work together to create the desired, seamless customer experience. Independent models of indoor and outdoor, RAN and backhaul, RF performance and service performance factors cannot take account of their overall interaction and will therefore be implicitly inaccurate.

This has the potential to greatly impact the profitability and success of network operators and their partners. In the early days of the voice coverage rollout, signal strength plots alone provided a good indicator of end user QoE. However, in today’s capacity-limited data paradigm, interference, loading and types of service must all be taken into account in order to understand what the end user experience will actually be. And although network density and complexity must scale to meet demand, revenues will not and so new and more cost effective processes must be introduced in order to maintain profitability.

The complex HetNet

The ‘heterogeneous’ part of the term HetNets can mean different things to different people. Perhaps the most common interpretation involves a network that’s comprised of a range of different cell sizes: macro, micro and femto or small cells. Capacity, or more specifically spectral efficiency, is maximised when the different sized cells all use the same carrier frequency, but this leads to a complex interference environment which demands co-ordination of technologies such as eICIC (enhanced Inter Cell Interference Co-ordination) and CoMP (Co-Ordinated Multi-Point).

This managed interaction implies that RF planners should simultaneously analyse the different components of the HetNet as well as any supporting co-ordination technologies.

Small cells deployed close to consumers, indoors or at street level, will increase capacity and coverage depth, but also introduce the new challenge of small cell backhaul. Identified by operators as one of the main barriers to deployment, small cell backhaul is very much a part of the HetNet and will need to be jointly designed and optimised alongside the radio access networks.

Looking outside in
The growing demand for mobile data is driving changes in the shape of radio access networks. Data services are more capacity hungry than voice and text and the majority of consumption today tends to be indoors rather than outside. It is predicted that wireless data traffic will grow at a CAGR (Compound Annual Growth Rate) of 57% from 2014 to 2019, reaching 24.3 exabytes per month by 2019. The Middle East and Africa will have the strongest mobile data traffic growth of any region with a 72% CAGR.

Loading, scalability and cost

Data is greedy and will consume as much network capacity as it can take. The result is that networks today are ‘capacity limited’ and the user experience therefore depends on loading – the number of other users they are sharing with and the types of data service in use.

Scalability is becoming increasingly important for all aspects of the network lifecycle and this includes RF planning. SON (Self-organising Network) technologies are being incorporated into equipment but RF planning is still very much needed to design cost effective HetNets. A high degree of automation of RF planning tools is needed to find enough engineers to install an increasing number of small cells and keep service cost low. Although automation will never give ‘perfect’ results, it will need to be ‘good enough’ to quickly identify the amount of infrastructure needed to meet expected consumer demands across an area that comprises both outdoor and indoor environments.

The consumer’s apparent willingness to pay for mobile connectivity across Africa and the Middle East as elsewhere, is not keeping pace with their demands for richer services, increased data volumes and deeper coverage. Much has been said about mobile data creating a ‘scissor effect’ for the mobile industry, which assumes a scenario where the costs of meeting ‘exponentially’ increasing demands are rising faster than revenues, leading to a loss making situation.

However, equipment vendors and analysts have pointed out the flaws in this simplistic analysis. Demand growth is slowing, capacity will mainly be constrained by capital and there is still plenty of room for innovation and increased operational efficiency. RF planning will need newer and more integrated processes for all the different teams involved throughout the design, deploy, maintain and optimise lifecycle.

Equipment vendors need to understand and demonstrate how their technology meets network operators’ performance requirements and is within their cost constraints. Operators themselves need a vendor-agnostic comparison of the benefits of different technologies when applied to their network and subscriber base. They also need to perform detailed network planning to decide exactly where to site the equipment and how to backhaul it.

Service groups and OTT (Over-the-Top) players, who provide apps or services that bypass traditional distribution, need to understand the QoE that will be achieved over a given network. Finally, the network must be maintained, monitored, optimised and upgraded.

Performance modelling

At the centre of the lifecycle is a common performance modelling platform on which a detailed description of the HetNet technology, configuration and environment can be built. Each group has their area of specialist knowledge and associated data assets. Where a common platform is used, all groups can leverage the expertise and efforts of previous users, rather than starting from scratch at every stage. This conservation of effort reduces costs and brings value to all parts of the lifecycle.

The traditional roles of equipment vendors, network operators and assurance solutions providers are also changing in their scope. For example, major equipment vendors are increasingly managing infrastructure themselves, becoming providers of leased capacity to operators. Network operators in turn are collaborating with OTT players to manage the resulting end user QoE and exploit brands and social networking and content partners.

Deployment

The Small Cell Forum’s view of the deployment planning process is one that combines both RAN and backhaul in the planning stage, and also takes into account the existing macro network. Once a network design or upgrade has been planned, it needs to be transferred to the operations teams for deployment on the ground. Significant effort and costs can be saved if the tools used for planning can directly generate output in a format suitable for these teams.

The transfer of network designs from the designers to the deployers requires planning tools to be able to provide actionable instructions to teams on the ground. These outputs involve detailing installation and commissioning issues, such as the desired locations of infrastructure assets, the orientation and alignment of their antennas as well as power and backhaul details.

Mobile data connectivity has enabled a diverse range of different service types to be delivered to consumers on the go with widely varying requirements for the quality of the connection in terms of the data rates and latency. OTT players are entering the fray and developing robust service offerings that can tolerate the ‘best effort’ or ‘dumb pipe’ connectivity characteristic of the internet and IP. Both Operators and OTT players are also negotiating to arrange prioritised connectivity for monetiseable services, although some see this as a breach of net neutrality.

RF performance prediction tools can be used by service introduction and delivery groups to understand whether a service would be viably under given network and loading conditions. The tools must be able to model load in terms of varying user densities across both outdoor and indoor areas, as well as handle the types of services that users will be consuming.

In summary

The rise of mobile data is changing the shape of mobile networks – no more so than across the rapidly emerging Middle East and African markets. The need for more capacity is driving densification while the move to increased indoor consumption requires in-building systems including DAS, Wi-Fi and small cells technologies. All of these interfere with each other and the existing macrocell network coverage and so need to be co-ordinated using HetNet technologies such as eICIC. RF planning tools will continue to play an essential role in the design of cellular networks, but will increasingly need to offer greatly enhanced functionality, flexibility and cost effectiveness if they’re to meet the emergent needs of the HetNet era.