Tips and Tricks for Successful Network Capacity Planning
24 SEPTEMBER 2015
Benjamin Franklin once said, “By failing to prepare, you are preparing to fail.” But aside from relying on a crystal ball, how can network planners look into the future to know what RF network requirements will be one day, one week, or even one year from now? And how do you prioritize the network expansions you decide to invest in?
Poor network planning can lead to congestion problems that can quickly turn into a game of ‘whack-a-mole’. Solve one issue and another one pops up, and often, it isn’t until the customer complaints roll in that service providers even know that a problem exists. Equipment outages, congestion issues and radio interference are just a few of the things that can go wrong, and without the right tools in place, only your unhappy customers will know where the problems exist.
Keith Dyer, of The Mobile Network (TMN), recently pointed out that in many cases, less than 1% of a mobile network’s area can generate as much as half the data being consumed. These network bottlenecks and service problems create unhappy customers. But what can network planners do? Effective network capacity planning depends greatly on the degree that the traffic load is balanced and optimized over all cells, but reality isn’t quite that tidy.
The Trouble with Maps
Looking into the future is something most network planners do every day, it just comes with the job. The difference comes down to their accuracy. This often depends on the quality of the tools they have at their disposal. Typically, most network planners rely on outdated census maps with information on building density and whether an area is urban vs suburban. However this type of map data typically just characterizes areas by clusters and customer type, without regard to movement and trends. Shockingly enough, this information is often the extent of which many CSPs base their network forecasts upon. But mobile customers are – well – mobile. They don’t just use their devices in their offices and homes, and usage can increase suddenly when a new device or popular app or service is introduced. Map data only gives a hint of what future, and even current, network usage demands will require.
The Benefits of a Holistic View
Radio network planning is no longer a standalone business activity; there needs to be a way to connect to the wider OSS/BSS eco-system, and beyond, for maximum impact. Planners need to be able to see and interpret trends across a wide range of network equipment and subscriber activity so that bottlenecks do not emerge unexpectedly. There is a requirement for the system to be intelligent enough to alert where capacity trends are changing, estimating where and when limits will be breached.
Strategic planners will take a longer term view of how and when the network needs to expand. They’ll take into account how quickly new customers are signing up, the new services being introduced, and how much capacity is being used. Major investments in new technologies and the introduction of new services will also play a big part in their decisions. Major shifts in the industry, such as network sharing, partnerships, and mergers and acquisitions are possible events to expect in the future. Network capacity planning should utilize these inputs, plus market forecasts, performance management data, design capacity, and subscriber geolocation and social media data, two of TEOCO’s most recent offerings.
Geo-Locating the Future
By using real-time geo-located traffic data and social media activity, your customer’s exact location can be known, as well as the subsequent impact on network performance. Patterns and predictions can then be identified based on the busiest hours and places of the day. Today’s tools, like those from TEOCO, can help service providers plan their network capacity needs up to one year out by using predictive modeling techniques with sophisticated algorithms based on geo-located network traffic and actual network traffic data, with data dating back up to four years.
By pairing historical data with today’s data, which includes actual signal strength, throughput and quality measurements, planning engineers have the opportunity to gain an increasingly accurate view of network performance, especially when utilizing simulation models within the tool which can be tuned to more closely match reality. In addition, by understanding where static vs. moving and indoor vs. outdoor traffic is concentrated, planners can adjust to optimize even further. With this information at your fingertips, the ability to plan for the future becomes much easier; and network assets, including both spectrum and hardware, can be optimized to extend their overall performance and lifespan.