Bluetooth 1.0 uses GFSK at 1 Msym/s, just like BLE / Bluetooth 4.0. So it's the same r/f modulation protocol (it's the higher layers that are incompatible.) Is it the actual GFSK parameters, eg. frequency shifts, that are different?
I know that Bluetooth 2.0 and 3.0 use different r/f modulation (π/4-QPSK and 8DPSK at 1 Msym/s), but why do people say Bluetooth 4.0 is so different when it seems to be the same as Bluetooth 1.0?
One difference is that BLE uses 40 channels in its adaptive frequency hopping scheme, from Wikipedia: "Instead of the Classic Bluetooth 79 1-MHz channels, Bluetooth Smart has 40 2-MHz channels"
Now that I write it out, I see that frequency hopping would count as a "higher layer," but I think it's still managed in hardware, which could be why there are many BLE-only chips.
It's possible the GFSK parameters are also different, and of course the layers above the link manager are vastly different (for some reason).
A major difference is that while original Bluetooth used time-based channel hopping BLE uses sequence-based channel hopping. This means you don't have to keep your local clock or receiver running to know what channel to listen to, rather you listen on a given channel just before you want to speak/receive something and once something shows up there you know where you are in the sequence. BLE-only chips can basically be entirely powered off most of the time without losing connection, whereas bluetooth needs a running clock or a large receive window to stay connected.
Bluetooth 1.0 uses GFSK at 1 Msym/s, just like BLE / Bluetooth 4.0. So it's the same r/f modulation protocol (it's the higher layers that are incompatible.) Is it the actual GFSK parameters, eg. frequency shifts, that are different?
I know that Bluetooth 2.0 and 3.0 use different r/f modulation (π/4-QPSK and 8DPSK at 1 Msym/s), but why do people say Bluetooth 4.0 is so different when it seems to be the same as Bluetooth 1.0?