Single Cell Architectures vs. Scalable Small Cells

August 3, 2015

Now that SpiderCloud has proved that operators and enterprises are looking for an alternative to DAS in the medium to large enterprise market, quite a few companies are trying to enter this market. A number of them offer a solution that have a centralized baseband unit, connected to radio heads throughout the building over dedicated cabling. These companies then claim that these “cloud RAN” products are better than enterprise small cell products because they eliminate handovers between radio heads.

Is handover such a bad thing in a cellular network? Is it something that you want to eliminate?  To us, this sounds like a really strange idea. Inter-cell handover has been the basis of cellular networks for three decades, from the time the first commercial cellular system was launched in October 1983, till today. Before you finish reading this blog post, cellular systems around the world would have successfully done more than a billion handovers!

What handovers enable is capacity. Properly implemented and reliable handovers help cellular networks scale, serve more subscribers, deliver more minutes, bits, bytes, music, video, you name it. According to AT&T, in 1965, before the introduction of handover in the cellular network, 2,000 subscribers in New York City shared 12 channels (on a single cell), and typically waited 30 minutes to place a call.

So the real question is not whether a small cell system does handovers, but whether it does handovers reliably. SpiderCloud’s scalable small cell system is designed to do handover really, really well. We have published our KPIs, both for 3G and LTE, in the past. Operators who chose to trial our system can see it for themselves. See our published KPI information for 3G and 3G+4G.

Our competitors might look at the KPIs we have published and claim that they have found a smoking gun. They will say, “Look! SpiderCloud systems drop calls 0.5% of the time, we never do so”. They are right, even with SpiderCloud, handovers sometimes fail. Handovers, by the way, fail in every macro network. So, what should an operator do? Go back to a single transmitter serving New York City? As an operator, do you care about delivering capacity, or about eliminating low-probability handover failures?

Finally, if an operator really does not care about capacity, does not mind installing dedicated cabling, has a very large capital budget and just wants to deliver signal inside the building with no handover failure, well, why not use DAS – the original “single-cell” system?  One mobile service provider did a trial to answer exactly this question. They had a three floor office building that was covered with a DAS system (and a dedicated base station). They turned off the DAS on one floor, and replaced it with SpiderCloud’s LTE E-RAN system. You will be surprised to learn what they did… in the next blog post.

– Amit Jain, Vice President of Marketing & Product Management
Twitter: @SpiderCloud_Inc