Hello,
Actually, I think the headline is kind of strange, in that I don't really see BLE and Dash7 as competing much at all. BLE is an ultra-low power PAN with hubs in PCs and mobile device, Dash7 has its roots in active RFID, defence and asset tracking. So why compare these two?
The trigger was this blog post over at the Dash7 Alliance blog. Now, it's a bit difficult to know where to start, as the blog post is full of opinions and anectdotes, but short on facts. Rather than talking about how good Dash7 is, the author spends most of his time badmouthing Bluetooth and Bluetooth low energy in particular. Attributing the success of Bluetooth to the availability of blue LEDs gives you an idea of the type of "analysis" provided. Rather than trying to debate at this level, I would rather try to put together a more objective analysis based on facts (TI, by the way, offers both Dash7- and Bluetooth-compatible products, so we have no reason to promote one over the other).
To start, Bluetooth low energy's design goals where to create a wireless protocol that built on Bluetooth (and therefore could tie into the existing Bluetooth ecosystem), but optimized for devices that exchange quite small amounts of data relatively infrequently. The main idea was to win new applications which Bluetooth did not yet address. Therefore, the assession that if you cannot do it with Bluetooth, you cannot do it with BLE, is not correct. For example, because of the low power consumption of BLE, it can address many applications what Bluetooth cannot address because power consumption is too high. This is true for sports applications and proximity tags, for example. Another example is that BLE eliminates the 7-device slave limit in regular Bluetooth, which again would allow BLE to address sensor applications where more than 7 devices is an absolute requirement.
Another strange assertion is that WiFi is touted to be a dangerous competitor to BLE. This probably deserves a separate BLE vs WiFi blog post, but I do find it strange that a Dash7 representative is bad-mouthing some competing stardards while promoting others. The idea that BLE is a harmful "bad neighbor" to WiFi is a strange one, considering that BLE has a very small footprint in the 2.4 GHz spectrum (narrow bandwidth, very low duty cycles - a typical BLE device would be active <1ms per second ie. 0.1%). If there is an elephant in the china store, it is WiFi, where a single linked-channel WiFi device could take up 2/3rds of the total available 2.4 GHz spectrum.
OK, so I promised some analysis based on facts, let's first start with some facts:
There's no doubt that Dash7 gains some benefits by using 433 MHz rather than 2.4 GHz. Most importantly, assuming you have space for a decent antenna (a quarter-wavelength antenna is much bigger at 433 MHz than at 2.4 GHz), Dash7 will be a lot longer range, yes. This is both because the link budget is better (due to the lower data rate) and because range measured in m/feet increases with decreasing frequency (it would be constant if you measure range in wavelengths).For very space-constrained applications, the range at 433 MHz may not be better than 2.4 GHz, as the antenna efficiency drops dramatically when the antenna becomes very small compared to the wavelength. Incidently, these are the types of applications mainly targeted by BLE (body-worn sensor etc.). 2.4 GHz is a resonance frequency of water (this is why this spectrum is available world-wide in the first place), so 2.4 GHz is not suitable for applications where the radio signals must travel in water or through water.
Interference-wise, it is probably true that the 433 MHz band is generally much less crowded than 2.4 GHz today, but if 433 MHz devices are very successful, would that also be true in the future? I'm sure the pioneers in 2.4 GHz also thought they would have it easy, and therefore did not have to create robust protocols. Actually, the 433 MHz band would fill up much more quickly, as the band is very narrow (just a few MHz) and the systems operating on that frequency use lower data rates (and therefore stay on the air for a longer period of time).
When it comes to power consumption, it is a bit difficult to answer this as I do not right now have access to an apples-to-apples comparison. What I can say, is that we are seeing 30 uA average current for one packet per second with our CC2540. This is including the whole protocol stack. Now, the Dash7 protocol stack could be lighter, which would reduce the current consumption consumed by the software. On the other hand, I don't see a big difference in current consumption between our 2.4 GHz and 433 MHz radios (433 MHz is maybe 30% less). However, the lower data rate of Dash7 means that the devices will spend more time (5x or more) transmitting and receiving, which would easily more than negate the 30% peak current advantage. Depending on how significant any stack-related power savings are, Dash7 could be better than BLE or BLE might be better than Dash7, it's really hard to say.
At the systems level, Dash7 was originally designed as an active RFID system, and I am sure it is well suited to things like container tracking. In this application, range is obviously important (if you need to track containers throughout a huge port facility), tag size is not very critical (so you would be able to put in a good antenna) and connectivity with mobile phones is probably quite unimportant. I don't see BLE playing at all here. I'm sure they will also do well in military applications and things like irrigation control (again, I see the need for long range here).
On the opposite side, I don't see any reasons why Dash7 should be better suited for body-worn sensors (be they sports sensors, medical sensors or watches). The lack of mobile phone connectivity is likely to be a big issue (the absolutely last thing a mobile phone designer wants to deal with is yet another radio), and the bigger antenna becomes a real issue as these sensors need to be tiny to avoid being in the way.
Now, there may be some applications in the middle where there could potentially be some question which standard would be used. A couple of good examples might be automotive applications (Remote Keyless Entry - RKE, Tire Pressure Monitoring) and Access Control. Here we'll have to see how the market plays out. Bluetooth has good traction in the automotive sector already through handsfree and entertainment systems, Dash7 is not there yet but RKE is often 433 or 300 MHz for range reasons. I don't see any clear benefits one way or the other with TPS, so that one is difficult to call.
For Access Control, it probably depends what kind of system we are talking about. If you want users to be able to use their phones as access control, then it comes down to what is already in the phones - BLE or NFC. If the Access Control ties into Inventory management or something like that, Dash7 could stand a fair chance.
But please, let's have a discussion based on facts rather than hyperbole. Silly charts like this one isn't helping anyone (hey, come on, if Bluetooth doesn't have any major customers, how come the industry is shipping 1 billion Bluetooth devices a year?). Part of the reason for this may be that people are operating in very different industries (the defence industry and consumer electronics don't have much in common, for example) and people are making assumptions that industries work similarly when they really don't. Maybe Dash7 meant that Bluetooth didn't have any major container tracking customers? That would probably be true.
But then again, that is why there are so many different wireless standards. Different applications have vastly varying requirements, when it comes to data rate, power consumption, usage models, range and bandwidth requirements. One standard definitively does not fit all.
Best regards,
Karl