Abstract
1- Introduction
2- Literature review
3- Methodology
4- Results
5- Policy implications
6- Conclusion
References
Abstract
Despite 5G still being embryonic in its development, there is already a quest for evidence to support decision-making in government and industry. Although there is still considerable technological, economic and behavioural uncertainty, exploration of how the potential rollout may take place both spatially and temporally is required for effective policy formulation. Consequently, the cost, coverage and rollout implications of 5G networks across Britain are explored by extrapolating 4G LTE and LTE-Advanced characteristics for the period 2020–2030. We focus on ubiquitous ultrafast broadband of 50 Mbps and test the impact of annual capital intensity, infrastructure sharing and reducing the end-user speed in rural areas to either 10 or 30 Mbps. For the business-as-usual scenario we find that 90% of the population is covered with 5G by 2027, but coverage is unlikely to reach the final 10% due to exponentially increasing costs. Moreover, varying annual capital intensity or deploying a shared small cell network can greatly influence the time taken to reach the 90% threshold, with these changes mostly benefiting rural areas. Importantly, simply by integrating new and existing spectrum, a network capable of achieving 10 Mbps per rural user is possible, which is comparable to the UK's current fixed broadband Universal Service Obligation. We contribute to the literature by quantifying the effectiveness of the spatial and temporal rollout of 5G under different policy options.
Introduction
Over the last decade, several factors have contributed to the cost-effective rollout of high-speed mobile broadband services, including the standardisation of LTE networks, enhanced spectral efficiency and the allocation of additional spectrum (Ghosh & Ratasuk, 2011; Holma & Toskala, 2012). While 4G is still reaching maturity different industrial, governmental and academic stakeholders are currently working together globally to develop the next generation of mobile networks known as ‘5G’ (5GPPP, 2016). A considerable number of papers have hypothesised the key characteristics of 5G networks and their potential capabilities (e.g. Rost et al., 2016; Akyildiz et al., 2016). There is an expectation in the engineering literature that 5G systems will provide peak data rates of 1 Gbps to mobile users and 10 Gbps to stationary users (Chih-Lin, Han, Xu, Sun, & Pan, 2016). The various use cases of 5G include enhanced mobile broadband, massive machine-type communications, and ultra-reliable and low-latency communications (see 5G NORMA, 2016; Tullberg, Fallgren, Kusume, & Hoglund, 2016; Mavromoustakis, Mastorakis, & Batalla, 2016; Hu, 2016). The current hype around 5G is pervasive in the telecommunications industry, with almost daily announcements being made by operators advertising either new test-bed commitments or advocating the potential benefits for vertical industries. Many operators are facing challenging times with static or declining revenues. While some see these technologies as being able to drive revenue, others are concerned about the economics of site densification. Indeed, adding new sites is a key strategy to increase wireless capacity, but the current deployment model is prohibitively expensive. Hence, for effective 5G rollout to take place, there needs to be major cost model changes.