Bluetooth Low Energy vs ANT+

Hey everyone,

Exactly my question, why does a solution  need two wireless technologies. Having two technologies on a single chip/single solution does not mean those have to be used.

I presume BLE was developed considering all the small devices needs (mainly power consumption, as casio is advertising a watch can run for 2 years on same battery),  if BLE serves the low energy needs why will some one need ANT+ and BLE.

I was reading about a solution in which a small device transmits data using ANT to a agent(a dongle or a small device nearby) and agent will transmit data to data collector using BLE. What I failed to understand was why does a solution need agent device.

regards,

Ashok

BLE is a new protocol and uses existing BT HW (RF and so on). There is no such thing as "backward compatibility" with regards

to BLE. If you have a BLE device, it can talk to either another BLE device (BLE single mode device) or a device that supports both classic (BR/EDR) and BLE.

(dual mode devices).

There are only handful of BLE dev kits out there. There are NO end user products available that uses BLE. 

Yes, but important thing to note is BLE can be run over existing Bluetooth radios, one just needs to upgrade the firmwares in the Bluetooth chip, Classic Bluetooth will be turned into BLE enabled dual mode chip. Which makes it adaptable to existing products too.

Another thing is some of the devices do not have any mechanism to update the firmware in Bluetooth radio chip.

No, it's not that simple. At the high level, existing BR/EDR devices were optimzied for BR/EDR only.

It would have flash/ROM (for firmware) and memory only for BR/EDR. For dual mode, it is a major

change to the firmware (LM, scheduler etc etc) thus requires larger flash or ROM and memory

for dualmode operation. I'm leaving out other changes that might require RF/BB parts for any

LE optimization.

So, you simply can't do firmware update to an existing BR/EDR device to support dual mode. 

Yes, agreed. Existing products in market have issue since most of them are optimized and it would difficult to upgrade those. If h/w is flexible like in most of the development kits one would be able to upgrade the s/w to dual mode devices.

Nogcas,

In order to answer your question I'd have to have more information about your product space and launch date, for now I'll just genearlize:

- If you're looking at the high level athlete (Ironman, Ultramarathon, distance cyclist, etc) and want to collect data to a pre-existing device such as a Garmin 310XT, then you'll want to use ANT+ to integrate into the existing product space of that market.

- If you're targeting the consumer athlete who collects data with their mobile then you will want to target BT (not LE) for a product launch in the next 6-8 months.

- However, if you are targeting a device launch which is greater than ~8 months from now, and you are either designing or partnering with a company who has BTLE in their device then you could use BTLE.

Of course another question that should be answered is - When do I need the data? Sometimes real time data transmission is not necessary when a simple event occurrence interrupt would be much more useful and efficient.

There are simply too many variables to give you a good reason without a concrete product usage story. If you provide more info I may be able to provide more insight and lead you to the correct solution.

John Amschler

Thanks for all replies,

I would like to continue discussing the battle between ANT+ and BLE. For what I have read, I think ANT does not have a good marketing strategy, since it has been 5 years and it still is not be integrated in any smart phone. And it seems like no one know about ANT (I mean non technical people). However, BL becomes a standard for smart phone. So, I am afraid that if I have to spend time to design a sensor network using ANT, and in the next few  years it still will not be integrated in any phones (or most of them, since next few months, the Xperia from Sony Ericsson will integrate ANT). Dont you think it will be wasted, but... anything should be standardized, right? If you develop non-standard tech, sooner or later you will have to change back to the standard one. 

Also, as John stated

john Amschler said:

However, if you are targeting a device launch which is greater than ~8 months from now, and you are either designing or partnering with a company who has BTLE in their device then you could use BTLE.

Regards,

Nogcas

I believe TI must have done tremendous business studies before they engaged resources in developing ANT chips. I  think there are at least the following reasons to answer your question.

1. BLE is a late comer to the ultra low power market. It is at least 2-3 years in maturity behind its competitors like Zigbee, ANT, Nike+. While some companies would like to wait for BLE, many others won't. For example, the digitized sports and fitness market has been doubled every year in the past few years; a great chunk of the pie has gone to ANT

2. It is more and more of a common understanding that one technology can not fit all. Different technology has different trade-off to achieve ultra low power while to maintain certain minimum feature set. Today, Zigbee is focusing on smart energy, RF4CE  on remote control, ANT+ on sports and health, Nike+ on running sports, WirelessHart on industrial control... The market segmentation is a reality, but unity in foreseeable future is an utopia. As a semiconductor company, it would be rather an advantage offering multiple solutions to concentrating on a couple.

3. The PHY of ANT and Bluetooth are close enough if not the same. The difference is boiled down to almost firmware only. So one chip design can make both BLE and ANT.

@Ashok: The CC2567 is an ANT and Bluetooth combination, not a marriage of ANT and Bluetooth Low Energy. So the agent will transmit data to data collector using Bluetooth.

@Nogcas: Bluetooth is almost in all phones today, but these are v2.1, not the v4.0 which supports BLE. You will need to continue waiting for new phones launched with v4.0. Existing phones can not be retrofitted to receive BLE. So it is not your understanding of backward compatibility.

Regards,

Renzh

Renzh ilz said:

Bluetooth is almost in all phones today, but these are v2.1, not the v4.0 which supports BLE

I understand that, and my thought was BLE with dual mode can be used for older version of Bluetooth.

Renzh ilz said:

Existing phones can not be retrofitted to receive BLE

What do you mean by "retrofitted to receive BLE"? Do you mean the signal from BLE or the Low Energy tech from BLE? I think you mean the LE. If so, yes, I agree but at least I can design a device that can work for any version of Bluetooth in the mean time and wait for the BLE be integrated to new phone. In ANT case, if I want to use it for any smart phone, it will involve to build a dongle which has different USB port for each different phone... However in the case of BLE, I do not need since it can connect to older version of Bluetooth -> no USB port design for each particular phone -> Save money and time, am I correct? 

If you mean the signal from BLE, I understand that BLE can not talk directly to Classic Bluetooth (CB). That's why I need to design a device that can act like a HUB with BLE dual mode on that, something like : BLE (sensor) -> BLE (HUB) CB -> CB (phone). It should be better than: ANT (sensor) -> ANT-USB -> USB (phone), since for each particular phone, you should have particular USB port. By BLE-CB dual mode, for that I mean "backward compatible", am I right?

For the above design, in the power consumption perspective, it is worser than ANT since the HUB has to use more power for CB so LE is avoided. Anyway, it is just the solution in the mean time until all the phones are integrated with BLE. This comes to interesting part, will BLE become standard for all the phones (just phone market only)? or will ANT become standard? 

Regards,

Nogcas

All the major BT chipset vendors (TI,csr, broadcom, atheros) have qualified dualmode chipsets (check the bluetooth.org) and sooner or later they will land on Notebook, Phones and Tablets.And on other devices where it makes sense. It is inevitable. For OEMs, these chipsets will cost few cents or less more. It will be widespread around ~2012-2013 time frame. Whether there will be any BLE devices (peripherals) or not. No one will care. Highend devices will have dual mode support. It would be up to companies like MS, Google, Apple and Nokia to enable (BLE) use cases that make sense. I think there will be lot of AppStore development behind this if some interesting BLE devices become available.

We are in 2011 and no one has dualmode software stack in their platforms (Windows, iOS and Android). It will be end of 2011 or sometime in 2012 before we will see support for it. Until then, no one can do anything useful with any BLE peripherals unless chipset vendors release custom stack. 

Nogcas said:

Also, as John stated

However, if you are targeting a device launch which is greater than ~8 months from now, and you are either designing or partnering with a company who has BTLE in their device then you could use BTLE.

. Seems like you also tend to be with BLE :D.... And you predict in the next 8 month, BLE will become standard...

[/quote]

@Nogcas - from your statement above I think I was not clear, so let me clarify a bit further: I do not believe BTLE will be standard in the next 8 months

If you are going to make a sensor network that requires BTLE on a mobile I would make sure that you have a business relationship with a carrier and mobile device manufacturer to ensure that you have a supportive ecosystem.

 

Renzh ilz said:

I believe TI must have done tremendous business studies before they engaged resources in developing ANT chips.

@Renzh - This may be the case, but I like to thing that there is a high likelihood that since the PHYs are so close, one of the firmware engineers took it upon themselves to add ANT+ to the chip ;)   I just bring this up because I have lots of experience doing this cost free experiment on high volume products, and I encourage others to do this to accelerate innovation.

 

Back on track -

@Nogcas - ANT+ has not been targeting mobile phones for the last five years, because many mobile phones back then weren't used for personal data collection. Only after the iPhone came onto the market did Phone Manufacturers in the US start getting more freedom to design consumer technology into their devices. Mobile phones may be classified as a consumer device, but in the US the customer is actually the Carrier, and unless the Carrier can monetize a feature, they are less likely to subsidize it.

Having said all that, if you are targeting mobile phones to be the hub of your sensor system, you may want to figure out how the Carriers can monetize your products, thus they will market it for you, and when they market a product, it becomes a top seller.

 

John

john@jxaengineering.com

 

Nogcas,

First of all, BLE is not compatible with Classic Bluetooth and you will have to wait at least one generation of Smartphones before BLE is integrated into any real product. Second, there is no guarantee that BLE will be adopted into all phones in the future. As was accurately noted in some of the responses above, ANT and BLE both share the same PHY, so at least from a technical point of view, any phone running BLE is also capable of running ANT. Whether or not a particular protocol will be adopted into a cell phone will be up to the cell providers, cellphone manufacturers and individual semiconductor companies – however ultimately the market will decide which protocol it wants to use. While BLE has an enormous marketing machine behind it, it has no real products in the market. ANT has an install base of over 13 million devices and is prevalent in many consumer devices already, especially in the sports and fitness arena. It also beat BLE to the smartphone with the introduction of the Sony Ericcson.

One way to connect to Classical Bluetooth (and consequently existing phones) is by utilizing the TI CC2567 chip, which has both Bluetooth Classic and ANT. This part can be used as a bridge to receive data from ultra low power sensors and subsequently re-broadcast over BT Classic to a cell phone. The advantage of this approach is that it links existing low power sensors with existing cell phones.

Which protocol you decide to use will ultimately depend on your particular application. However, the risks you need to weigh are choosing the wrong protocol when vs. delaying your product development waiting for the answer to that question. 

Regards,

Sebastian

Hello everyone,

Good to see a lot of spirited debate on this topic! While it is fun to speculate about the future, I would like to summarize the facts and what is publicly known:

• Bluetooth low energy is not backwards compatible with "classic" Bluetooth in the sense that existing Bluetooth devices can communicate with BLE devices, nor can existing Bluetooth devices generally be updated to be compatible with BLE (due to BLE PHY being somewhat different from classic Bluetooth and low-level code usually being in ROM or hard-coded in HW gates).
• However, most Bluetooth chip vendors (including TI) have announced that their future Bluetooth devices will include support for Bluetooth low energy
• Devices will need a full dual-mode stack to support Bluetooth low energy, not just hardware support. A lot is happening behind the scenes here, what is currently publicly know is that work is being done on adding BLE support in BlueZ (popular open-source Bluetooth stack used in many Linux environments). TI is also planning to come out with a dual-mode Bluetooth+BLE product, including stack support.
• Bluetooth low energy is quite new (being ratified last year). At CES last week, two BLE-based prototype products were shown. On the semiconductor side, the CC2540 is in mass production and TI is selling development kits and chips to any interested party. Other semiconductor vendors have announced BLE products, but to our knowledge are not shipping in volume as of the time of writing.
• ANT has been in the market for quite a few years, and has an installed base of 13 million devices (according to Dynastream). Previously, there was just one source of silicon for ANT, but with the introduction of CC2570/71 and CC2567, TI is also supplying ANT silicon.
• SonyEricsson just announced support for ANT in their Experia series of phones, enabled by TI silicon. Hence, ANT did beat BLE to market in mobile phones. TI is predicting that the first BLE-enabled phones will appear around the middle of this year.

Based on these facts, you are free to draw your own conclusions.

TI sees value in both ANT and BLE, and is supplying silicon for both protocols.