First, can R10 aggregate TD and FD LTE? I know that's not what you are really suggesting but I was curious about it.
I was wondering the same thing - WiWavelength over at srgru.com confirmed that you can aggregate unpaired spectrum to paired spectrum. You wouldn't be aggregating FD and TD but just taking spectrum and bonding it to the downstream to create a 5x25 setup, and since the uplink is typically limited by power limits on the handset, you wouldn't see the same performance limitations in terms of coverage that we saw with TD WiMAX deployment.
Which LTE bands are current or near future sprint devices able to support?
Next year I am sure they will have phones that support LTE on ESMR, PCS, and ERS/BRS bands. Right now current phones only support LTE on the PCS band. Don't forget though that as other devices in the future get multiple-band support, current devices on PCS should also see the benefit as there will be less load on PCS band. They can also convert (aggregate) other spectrum in the PCS band, and there is the H block auction coming up, which would allow sprint to have a nationwide 10x10 FD-LTE deployment in PCS band.
At the base of it, it is 30 MHz, 64QAM, 2x2 mimo vs 5 MHz, 64QAM, SISO, unless I've got one of those points wrong (is sprint doing SISO?) why would the LTE example be able to outperform the WiMAX example in the ideal case for each?
Sprint is deploying MIMO. At the base of it, it is 10MHz times 3 (30MHz total). Clear can't aggregate, so you are limited to what 10MHz can do. Sure, if both were 100% loaded up, WiMAX would outperform, but that isn't the case and as Sprint's LTE network starts to become loaded, they are moving to ESMR 5x5, more LTE in PCS band, and will "roam" on clearwire's TD-LTE network which will just move tonnage.
The gain of aggregate vs stacking is about 10% on HSPA for geographic sector throughput, and I don't know what it is for LTE or WiMAX (I read a single paper on this that focused on HSPA) but even at 20% loss in efficiency, and presuming attenuation loss vs 1.9 GHz at about 50%, each WiMAX tower is still going to be able to deliver more than each LTE tower in 30 MHz wimax markets, and given that it's TD it's possible that the majority of that 30 MHz gets used for forward link vs 5 MHz of forward link in LTE.
Aggregating leads to single user performance increases. Look at Sprint's EVDO network, EVDO is limited to 3.1mbit max - they can't aggregate it. You see a large performance increase when sprint adds another EVDO carrier but you still can't break the 3.1mbit max. With aggregation, it just adds up. In real world tests, on an unloaded network, I was able to hit 15.5mbit down with WiMAX. Today, I can almost certainly promise that you will see speeds on Sprint's 5x5 FD-LTE network that VASTLY exceed this (25-30+mbit).
In the simplified hypothetical town being supported by a single clear tower being outright switched to a single LTE tower those townsfolk will go from 30 MHz (with some efficiency loss because it is stacked, as you mention, but not that much) to 5 MHz, with no ability to aggregate until R10 deploys.
Sort of - First off, clearwire wholesales their network and offered home broadband plans etc. Second, look at Verizon. They've had LTE for 2 years and only 10% of their base is on it. Sprint doesn't need gobs of spectrum - they need a national deployment plan. 5x5 is only to start, not to futureproof their network. I definitely agree that all other things being equal, 30MHz of WiMAX = more total throughput than 5x5 of FD-LTE. The user experience will be different though - each user will see max speeds increase across the board because of less users, better coverage and a proper deployment.
Compared to verizon, it's also not as simple as halving the spectrum in LTE to give you half the delivered bandwidth, you get less than half of the payload bandwidth in half the spectrum, but I really wasn't intending to compare Verizon to Sprint LTE, I was intending to compare clear WiMAX to sprint's intended LTE deployment on PCS. There are other factors that complicate it, such as sprint intending to also use clear's TD LTE in the future to relieve some of the load and evolution to R10 - which Verizon will also do and in some markets, at least, verizon has a fair amount of spectrum they can deploy R10 in, as sprint evolves, Verizon and AT&T will evolve as well, and average use per UE will go up, but it will go up faster on sprint's unmetered network.
It will be curious to see how traffic on Sprint's network grows. I am guessing it is going to just skyrocket. In areas that Sprint has clearwire provide "fat pipes" of relief capacity, I suspect it is going to remain very fast. Even overall, I suspect with all the new QoS services Sprint will have, performance of the network will remain competitive.
From what I had last read, clear was intending or expected to deploy r10 in 60 MHz in the lower part of their spectrum.
Clear is going to have monster speeds. 20+20+20 or 20+20, whatever they do, it's going to be the fastest in the industry.
I suggest heading over to s4gru.com - you seem to be interested in this stuff (like I am). They are a great resource and I've learned almost everything from them. Very friendly group over there.
Facebook
Twitter
Instagram