Almost every major cellular network operator is looking towards network densification and the advantages it promises to bring for dense urban areas as part of their efforts to get ready for the demands of 2020. The technical benefits are clear: providing better service to more users in a smaller area, through the use of small cells mounted in unorthodox locations.
There are many challenges to doing so successfully, however.
First, the difficulty of physically installing the necessary network infrastructure. One advantage of traditional equipment installation areas such as cell towers is that they are well-understood from a deployment standpoint; the cellular carriers have relationships with the owners (or are the owners themselves) and work with teams that specialise in installing equipment on them.
These tower sites are also typically out of view of the majority of people, not interfering with their view of the city or cluttering up the street lights and corners with many pieces of equipment. These two factors combined make a powerful impediment to deploying widespread, densified urban wireless infrastructure: equipment must be nearly everywhere, but appear to be nowhere.
There are a few ways around some of the most common issues experienced by those wishing to deploy networks using this new model. First, a site like smallcellsite.com or a third-party company specialising in urban wireless network deployment can greatly speed-up the issues inherent in acquiring the right sites, such as reaching agreements with owners and identifying suitable sites.
Second, equipment must be used that conforms to a pre-agreed look and size, such that it can be mounted on street furniture. Cities and local groups have been known to object strongly to un-camouflaged, ugly wireless infrastructure deployed at ‘eye level’ – and it’s easy to see why looking at some deployments, which feature a bewildering array of antennas and cables hanging from boxes.
Any piece of network infrastructure equipment intended for eye-level urban deployment simply must be designed with non-obtrusive, sensible aesthetics in mind. Without this, it’s very possible that network equipment manufacturers and the system integrators they rely on will be told to go back to the drawing board, once a proposed deployment is rejected due to obtrusive visuals – costing both parties a not-insignificant amount of time and money.
Third, spectrum planning becomes more important in a densified network, due to the much higher number of cells operating in a given area. Coverage planning, though well-understood for macrocell deployments, begins to resemble more of the in-building WiFi networks seen in many enterprise locations, and deployment models must evolve accordingly.
A potential challenge occurs when the demand on the densified network begins to outgrow its current capacity, necessitating division of an existing cell into smaller cells. The knock-on effect of this on the spectrum plan can be severe, depending on the available frequency and number of cells in the area.
To support this difficulty, spectrum planning (and ongoing spectrum management) systems must be smart enough to balance different frequencies (such as licensed and opportunistic use of unlicensed) as well as firmly control the use of spectrum by each cell in the network, becoming ever-more critical as densification continues.
On this point, it’s worth noting that network densification is highly unlikely to be a single point in time – networks will not go from un-densified to densified, and that is that. Instead, it will be a gradual process, each network evolving slightly differently due to its own local differences in demand and user movement.
It is highly likely that a given network will experience a continuous, gradual flow of successive densification, rather than the single-event approach that seems likely at times when the topic is discussed. This has many implications for spectrum planning and management systems, who must now be able to cope with the network as a constantly changing array of ever-smaller cells, on a scale not seen previously.
Looking at the problem of network densification as one of continued gradual change, with the network diagram beginning to resemble cells dividing, changes how we can think of resolving it. By changing our thinking, tools and processes from a relatively static model to one expecting continuous change, our networks can be best adapted to meet their ever-changing needs.
That said, with these three challenges considered, the advantages of network densification still outweigh their negatives. What lies ahead for equipment manufacturers and system integrators is much learning on creating, installing and managing progressively densifying networks that must appear as invisible as possible, whilst still providing the ubiquitous, always-on connectivity desired.