Large enterprises and venues have traditionally installed Distributed Antenna Systems (DAS) for in-building cellular service. A new independent study commissioned by Airvana has determined that small cell systems offer a lower cost, even for multi-operator scenarios. We spoke with Josh Adelson, Airvana's Director of Product Marketing, for more insight.
What was the purpose of this study?
Enterprise Cloud RAN systems such as our OneCell are a new product category. They combine the benefits of high RF system performance from centralised baseband with simpler cabling using CAT5 Ethernet. .
This combination provides a set of attributes that make them a viable alternative to DAS, and that ought to have a cost advantage. We wanted to establish an independent view of the cost by taking a rigorous approach in comparing these two alternatives.
Why compare to DAS? Why not other small cells?
DAS is well established and yet because of its high cost it has low penetration in enterprises – less than 2% according to ABI Research. Traditional "standalone" small cells are more economical but they don't meet enterprise requirements. When densely deployed, they create multiple overlapping cells with interference at cell borders and frequent handovers. Our testing has shown this to cause choppy voice and up to 90% lower data rates. Standalone small cells also present challenges for RF planning and macro network interference, due to each small cell being its own cell in an enterprise. Some solutions use a "controller" but this controller simply assists with handovers and SON. It's a band-aid that doesn't address the root cause.
Enterprise Cloud-RAN systems solve the limitations of standalone small cells and therefore really must be compared to DAS.
Why did you choose Real Wireless?
We wanted the analysis to be comprehensive and independent. We selected Real Wireless for the task because they have consulted on many in-building systems including hundreds of real-world DAS designs. They understand the practical realities of deployment. They didn't just bring general benchmark costs but were able to fully specify and quote the design for different scenarios right down to the detailed parts lists and labour.
What was the scope of the report?
The report looked at three different technology scenarios for each of DAS and OneCell:
- 4G only (Greenfield)
- 3G and 4G (Greenfield)
- 4G Upgrade (existing 3G DAS already installed)
These scenarios were then applied to two sizes of building:
- 97,000 sq foot, 6 storey building
- 585,000 square foot, 32 storey tower block.
And considered for each of:
For each of these 24 scenarios, we took account of all costs including equipment, cabling, installation labour, design and project management.
What were the principal findings?
The study found that deploying OneCell was cheaper in all cases, both the initial CAPEX and ongoing OPEX.
CAPEX savings for the larger building was highest for the 4G single operator case at 69% less than DAS.
A key reason for these lower costs is the Wi-Fi-like deployment model of our C-RAN small cells. Standard Ethernet LAN CAT5 cabling was far less expensive than fibre and coaxial cables. A lower labour rate and shorter time is needed for installation, requiring less skilled expertise than for more complex DAS designs.
Don't DAS systems have an advantage where costs are split between multiple operators?
Until now, the general perception has been that DAS is more cost effective than small cells in large multi-operator deployments. But Cloud RAN systems such as OneCell are conducive to supporting multiple operators. It's true our architecture differs from DAS, requiring some of the active infrastructure to be dedicated to each operator.
However in spite of this, the OneCell system still came out as much as 44% less expensive because of the lower costs elsewhere in the system like cabling and labour as I mentioned.
Don't forget that DAS can also incur added costs for multiple operators. For example, a DAS may need extra or upgraded equipment support different or wider frequency bands.
What about the more recent DAS architectures/products now available?
Real Wireless determined the most cost effective design using the latest currently available technology for both DAS and OneCell. They chose not to use digital DAS because its up-front cost is higher. It's possible that this would be offset in some scenarios by savings during later upgrades in the lifecycle but OneCell is even more field-upgradeable, so we believe digital DAS would have shown an even larger advantage for OneCell.
The report claims that combining 3G-only DAS and 4G OneCell is still cheaper than a fully integrated 3G+4G DAS. Really?
The most challenging case is where a new building needs both 3G and LTE for multiple operators. The mobile industry is in a transition phase at the moment. We believe that in the long term LTE (including VoLTE) will dominate but we recognise that this will take some time and operators still need to support 3G meanwhile.
For greenfield sites, since OneCell only supports LTE, a separate DAS system is needed to provide the legacy 3G service. We weren't sure what the report would find, so were pleased to see this still came out as much as 18% less expensive than using a DAS which combined both. This has surprised a lot of people, and it shows that OneCell is applicable in all scenarios, even when a 3G-DAS is also required.
How future-proof are these deployments?
Installing any in-building system is an investment. It's not just the cost of the installation, but also any disruption during regular working hours of the business. You need it to last for many years with minimal physical hardware upgrades.
Both architectures support current LTE features such as MIMO and carrier aggregation. But if you project forward by two or three years, we can see growing demand for LTE-Advanced features such as joint transmission/ reception (CoMP) and higher order MIMO. Our C-RAN small cell architecture with coordinated radios can support these features whereas DAS, because it lacks intelligence in the endpoints, cannot. And with OneCell these features can be added through software alone, leading to longer lifetime and greater capabilities for equipment installed today
What are the key takeaways from the study?
This independent study by experts in real-world in-building wireless design validates our view that the emerging Enterprise C-RAN small cells, such as OneCell , are significantly cheaper than traditional DAS solutions.
We've proven this not just for single operator 4G greenfield deployments, but even where 3G is also required. The simpler and lower installation costs combined with the simpler solution architecture to make this very cost effective.
Add to this the superior future investment protection of C-RAN small cells, and we don't see any reason for enterprises to choose to invest in 4G DAS.
The report can be downloaded from the Airvana website [registration required]