Don’t split a cell for less than 200 users. For 50 or less, use Wifi: Arieso
Network optimisation company Arieso has claimed that operators need to consider five key variables before they make decisions on whether to optimise their networks to deal with “extreme” users.
Arieso has previously said that 1% of users – termed extreme users – currently generate 50% of network data. Its studies have also shown that half of the data traffic generated in a typical mobile network is carried by the busiest 10% of cells. Arieso added that one-in-five of these busy cells have fewer than 200 unique data users. 2% of these busy cells have fewer than 50 data users.
“It’s clear that many busy cells have few users,” continued Flanagan. “This is critical intelligence for operators planning networks – heterogeneous or not. The practice of ‘cell splitting’ – the adding of a new cell to provide capacity relief – is at risk of being unjustified when there are too few users.”
Effectively planning hetnets demands that operators understand the complex interplay between previously known and five new variables:
1. The unique users per cell
2. The data they are consuming
3. The types and capabilities of devices they are using
4. Spatial distribution of the users within the cell
5. Temporal variations in demand
In a hetnet world, it’s impossible to provide hard and fast ‘rules’. Even allowing for some high ARPU subscribers, it is estimated that an operator would struggle to justify a macro cell split for less than 200 unique users. Below 50 unique users per cell, WiFi may well be the only cost effective solution – assuming the traffic is suitable for WiFi offload.
“Heterogeneous networks offer hope, but bring problems of their own,” said Dr. Michael Flanagan, CTO, Arieso. “Operators are not only faced with a dozen offload options, but also with complex subscriber behaviours, myriad charging models and extreme users. The result is a fiercely complicated business case with some brand new variables. The risk of getting it wrong, wasting money and still not solving the problem is significant.
“In a worst case scenario, where operators try and satisfy the demands of extreme users solely with macro sites, they will waste millions of dollars – $400m across the industry this year alone,” concluded Flanagan. “Further, a haphazard hetnet of differing offload solutions implemented without a deep knowledge of the variables in that location will cost inestimable billions in customer churn, regulator intervention and reputational damage.”
Sample justification for a macro cell having no business case if less than 200 subscribers (from Arieso):
- A macro has a capital cost of between US$100K and US$150K
- A high ARPU subscriber delivers US$20 of MARGIN to an operator
- 200 subscribers therefore deliver only US$48K per annum return to the operator giving a two-to-three year payback for a macro
- 50 subscribers deliver US$12K of returns, which should give a one-to-two year payback for a WiFi hotspot