Coverage is also part of the equation, remember there original wireless network was planned using 850MHZ. So they will still have to place sites close to urban areas to achieve 2300 and 2600 coverage which is going to cost millions and the resultant outcry from Nimby's and councils.
Not necessarily. I'm sure Spark will have this 4G service across multiple bands including 700MHz which has even better reach/penetration than 850MHz. Deployment of 2.3/2.6GHz on even just their current tower locations would no doubt see a large percentage of urban customers in coverage. Those who aren't could likely be serviced by either the 1.8GHz or 700MHz network in a kind of 'infill' situation.
This may (necessarily) involve external antennas pointed at certain towers and/or locking modems to certain towers/frequencies but even still I don't imagine deployment is going to be too difficult.
To get the best results using 2300 and 2600 you have to be line of site to the nearest cellsite. And these bands will hardly penetrate walls let alone having the same coverage as the 850MHZ band. Perhaps somebody can explain to you signal propagation and the effects of using different frequency bands and coverage.
This article explains what I mean.
Yeah I know enough about signal propagation at higher frequencies to know it's not as easy as with the sub 1000MHz frequencies, hence my comments on the probable necessity of external antennas for service at these (2.3/2.6GHz) frequencies and the use of the lower 1.8GHz and (especially) 700MHz as 'infill'. It may be that it ends up less like the Skinny Broadband 'plug and play' model and more like the Vodafone RBI Wireless or Spark's own Rural Wireless Broadband with technician install, external antennas etc.
You must remember too that the Spark network has not been built out solely based on 850MHz, but on 1.8GHz and 2.1GHz as well, frequencies that have more in common with 2.3/2.6GHz than they do with 850MHz.
Anyway, not looking to argue on something that, (at this stage at least), is still just speculation on my part as to what Spark might be up to with this $9 million purchase.... guess we'll know by Christmas!
At this stage they are still trialling this product so time will tell.