I believe the speed increases are due to better beamforming algorithms, which use multiple fixed antennas to behave like one or more directional antennas at the same time. The idea being that the antenna only sends power in a small cone to your device and only hears responses from inside that small cone. This lets you fit many users onto the same frequency space, since they can't hear each other and each have the full slice of spectrum and time to themselves. It's space multiplexing, as opposed to time and code multiplexing that's currently widely used.
(The problem is that this requires really good digital signal processing. Normally we make directional antennas by putting bits of metal around the driven element that pick up the signal and re-radiate it out of phase. These work really well, and you probably have on on your roof to receive TV signals. But beamforming does this all in software, using a number of actual antennas to radiate out-of-phase to increase power in a certain direction. The reverse is true for receive; correlating phase information to "hear" a signal only from a certain direction. One direction is easy, but 100 directions is hard. That's where better DSP equates to more "bandwidth" for mobile phones.)
802.11n contains rudimentary beamforming, but the state of the art for software signal processing on $5 commodity chips is not amazing, and of course, the 2.4GHz ISM band is basically unusable because there is so much interference from non-beamforming (etc.) devices. 802.11ac improves this incrementally with better hardware. So it makes sense that at carrier grade budgets and with dedicated spectrum, good beamforming is possible. If so, and that's what this is, I totally believe 300MBps.
802.11ac has a 256QAM mode that can do almost 1Gbps. Of course, on the 2.4GHz band, you'll never hear 16 levels of signal over all the noise. But with dedicated spectrum, things change.
(This is mostly secondhand knowledge, so take it with a grain of salt. My cubemate at work is a WiFi expert :) I'm just a ham who is happy making QSOs at 31.25baud with PSK31.
(The problem is that this requires really good digital signal processing. Normally we make directional antennas by putting bits of metal around the driven element that pick up the signal and re-radiate it out of phase. These work really well, and you probably have on on your roof to receive TV signals. But beamforming does this all in software, using a number of actual antennas to radiate out-of-phase to increase power in a certain direction. The reverse is true for receive; correlating phase information to "hear" a signal only from a certain direction. One direction is easy, but 100 directions is hard. That's where better DSP equates to more "bandwidth" for mobile phones.)
802.11n contains rudimentary beamforming, but the state of the art for software signal processing on $5 commodity chips is not amazing, and of course, the 2.4GHz ISM band is basically unusable because there is so much interference from non-beamforming (etc.) devices. 802.11ac improves this incrementally with better hardware. So it makes sense that at carrier grade budgets and with dedicated spectrum, good beamforming is possible. If so, and that's what this is, I totally believe 300MBps.
802.11ac has a 256QAM mode that can do almost 1Gbps. Of course, on the 2.4GHz band, you'll never hear 16 levels of signal over all the noise. But with dedicated spectrum, things change.
(This is mostly secondhand knowledge, so take it with a grain of salt. My cubemate at work is a WiFi expert :) I'm just a ham who is happy making QSOs at 31.25baud with PSK31.