What happens when carriers run out of spectrum? Earlier this month, the spectrum auction in Spain concluded with only three out of the four incumbent carriers securing 4G spectrum. Telefonica (O2), Orange and Vodafone each managed to secure 2x10 MHz license in the 800 MHz digital dividend band. However, the fourth operator, Yoigo, which is owned by TeliaSonera, did not secure any spectrum.

This was a repeat of the results of other auctions held in Sweden earlier this year and in Germany in 2010 where the largest of the four incumbents, T-Mobile, O2 and Vodafone, split up the 800 MHz "digital dividend" band leaving the smallest operator, KPN owned E-Plus, with none. Ongoing auctions in other European countries, such as the one currently underway in France, may well yield similar results because there just isn't enough "beachfront" spectrum in the 800 MHz band to accommodate four operators, particularly since LTE best operates in wide channel bandwidth such as 2x10 or 2x20 MHz.

Frank RayalFortunately, there are a number of available solutions, each with a price tag and benefits. Here, I address one such solution which works well to bridge the gap between operator's current and anticipated capacity: that of deploying outdoor compact base stations while continuing to evolve along the HSPA roadmap.

Compact base stations are recognized to provide double the capacity of large macro cells. The lower operating interference environment allows for higher signal quality in the cell coverage area leading consequently to higher throughput and user quality of experience. In this scenario, MNOs can continue to upgrade their macro cells along the 3G roadmap while solving specific capacity hot spots with compact base stations. This solution is economical because capacity constraints in a mobile network are typically not ubiquitous but are confined to certain geographic locations such as dense urban areas. In this case, the MNO avoids the expense of rolling out a market-wide LTE network in favor of resolving capacity constraints in a limited geographical area.

Evolving technology and fragmentation of the frequency band provides MNOs with more time before embarking on full-scale LTE deployments. In fact, many operators don't feel the rush yet to deploy LTE in urban areas, such as Vodafone in Germany, which is rolling out 800 MHz LTE to cover rural areas while urban area coverage is planned for later. MNOs can upgrade along the HSPA+ technology path and augment capacity where required by using compact base stations in the meantime until LTE is more mature and an orderly transition to LTE is planned and enacted.

Deploying outdoor compact base stations to extend the life of 3G investment is not without its cost. While compact base stations are available at relatively low cost and can be mounted inconspicuously on building sidewalls or public infrastructure assets such as light, telephone or utility poles, the business case becomes challenging when the cost of backhaul is factored in. The cost of fiber or wireline backhaul methods scales linearly with the number of compact base stations and can delay the deployment of compact base stations by several months due to the long installation cycle needed to secure the required permits. Microwave backhaul, on the other hand, requires line-of-sight connectivity which is difficult to achieve in the dense urban areas where compact base stations are needed.

Fortunately, a solution to this conundrum exists with non-line-of-sight (NLOS) wireless backhaul which uses frequency spectrum that falls between prime access spectrum and microwave backhaul spectrum, covering bands in 2 and 3 GHz. This solution is scalability to support relatively large number of small cells because of low marginal cost per link. It is quick to deploy and most importantly saves on either or both capital and operational expenditures of fiber or other wireline backhaul methods that makes such solutions prohibitively expensive.

Spectrum is one of the main cost drivers in wireless applications. NLOS backhaul spectrum in 2 and 3 GHz is available in abundance, especially in European markets where it has been planned for fixed wireless access services that have very limited applicability in dense urban areas where mobile broadband is highly desired. NLOS backhaul spectrum licenses can be obtained on a national or regional basis for a set number of years (usually 20 years) at a fixed price which is a fraction of what MNOs typically pay for access spectrum. For example, one MHz in 2 or 3 GHz band costs less than 2 cents per inhabitant versus over 50 cents per inhabitant for prime access spectrum (such as that in 700 or 800 MHz). This licensing model lends itself well to the small cell backhaul application since adding more links results in lower cost per link, making the business case scalable and more attractive to support a potentially large number of small cells.

The other aspect is related to ease of deployment. Small cells can become an expensive proposition if each cell needs to be planned in the same way as a macro cell and the leasehold has to be done cell by cell. What operators are looking for is simplicity and ease of planning, design and deployment that allows them to scale their deployment at low cost.

NLOS backhaul allows operators to design backhaul in a similar manner to designing their access network – a process they are well equipped for and used to. Line-of-sight microwave and E-band solutions, when the deployment allows, need to be planned and aligned individually, resulting in higher cost. For this reason, I view NLOS backhaul as a complement to fiber and microwave solutions which provide high capacity and connectivity to the core, while NLOS backhaul extends the reach of these solutions to the last few hundred meters into the urban clutter.

A further cost saving feature of NLOS backhaul is that it can operate in a multipoint configuration. Multi-point backhaul not only saves capital & operational expenditure because of fewer deployed hardware modules, but most importantly, limits the number of connections to the core network. This saves on additional backhaul expenses between the hub modules and the core network, resulting in higher savings the greater number of supported multi-points.

In conclusion, NLOS wireless backhaul provides a scalable and cost effective solution for compact base station backhaul. This allows MNOs to resolve capacity hotspots in the urban areas where traffic demands are highest and can therefore delay the deployment of LTE until additional spectrum is secured, or alternative methods are developed to migrate to LTE. 3G small cells are not meant to replace LTE deployments because LTE provides the platform for future capacity evolution. However, they do provide an interim solution that is cost effective and can serve the MNOs well by increasing the lifecycle of 3G investments until LTE is more mature, more devices are available and a long-term migration plan is devised.

Frank Rayal is vice president of Product Management at Blinq Networks.