Very quick for Broadcom to do this...and high level of integration...but space is a car is not limited, why bother integrating everything in one chip...would it be better to have a flexible architecture that can accomodate future wireless standards? after all the car will be driven for 10, perhaps 20 years
@krisi, true. But precisely because the car will be driven for 10 or 20 years, maybe it makes sense to look ahead and integrate Bluetooth Low Energy -- for which we may not see an immedate "apparent" market right now, but who knows, that may become a big market 10 years from now.
I agree Yunko, getting low power bluetooth makese sense...if I were a car maker I would stick as many IO standard as possible and have an option to configure them as needed...low cost Wifi, extra $ for Bluetooth, top dollars for everything etc...but not nessasarily in one chip, in one board maybe
Your point about space not being limited is well made. The other thing to consider is that Bluetooth LE is not the only "kid on the block" for short range, low power connections. Although, with the muscle of the Bluetooth SIG behind it, it may mow over some earlier technologies that already have a respectable product base.
That's true Junko. But, recongize that technologies such as ANT+ are built into some mobile phone platforms (android) and tablets and can be used on an iPhone (with a dongle). And, ANT+ has an impressive build out of devices. Check this out: http://www.thisisant.com/consumer/ant-101/ant-in-phones/. According to its website, ANT has over three hundred certified platforms and 60+million installed nodes. Just had to get that out there. Bluetooth tends to get the most airplay, and, since low-power wireless is one of my 'things,' I felt the need to share.
ANT is really strong in medical and sports related applications. People are much more familiar with BT in their automobiles. While I'd like to see ANT move into autos, I suspect it will be BT LE that dominates.
What intrgued me most in researching this story was when I've come to the following realization:
What connectivity technology choices carmakers make these days is totally driven by the mobile industry. In other words, the automotive industry no longer seems to have the luxury of thinking long and hard, and then coming up with automotive specific connectivity standards any more. They are truly more worried about keeping up with mobile devices drivers and passengers are bringing to their cars.
Who would have thought that this will be happening to carmakers 10 years ago?
True!! Mobile devices in the recent past (may be after the Apple & Android revolution?) has started driving the world, be it the automotive electronics, medical electronics, industrial electronics domain. As people will be playing with their mobile devices interfacing with the in-car infotainment devices wirelessly, automotive designer might have challenging time in designing the electronics system controlling the car, that takes care of the safety of the passengers, more rugged and EMI hardened.
If it takes you 5 years to bring an infotainment system to market, you're doing it wrong.
This suggests that people are thinking of wireless connectivity as part of the car. It's not, or shouldn't be. It's part of the infotainment module which rides inside the car, and it should be possible, through good interface design, for the customer to select an infotainment upgrade whose design phase extended well after the car was manufactured.
Even rear-view cameras shouldn't need to be deeply integrated into the automotive system, at least not electronically. They need mechanical integration but electrically just power feed and a coaxial video cable. Ditto for back-seat displays -- pick a standard like DisplayPort and let a brand-new multimedia system be plugged into it at the dealer.
Aftermarket radios are a long-standing tradition. Perhaps instead of trying to cut them out by limiting consumer options, the manufacturers should embrace the idea and provide their own upgrades. Using wireless technology that's 6 months, not 6 years, old.
@rbv, perhaps you are right. I get what you're saying.
But if you put yourself in the automotive industry's shoe, you do want to find a way to upgrade your cars with more features, rather than giving all that opportunities for added value to the mobile handset.
@junko, I'm not talking about all those features being in the handset.
But the vehicle can be designed with a infotainment connection interface, so that the dealer can install brand-new electronics (also sourced from the car manufacturer, which is in the profit interest of both manufacturer and dealer) into an automobile that spent 5 years in the manufacturing and design phase. Or have it installed at the factory, although that seems to needlessly reduce the dealer's opportunity to upsell the electronics.
What's important is to decouple electronic from mechanical design. A six-year design leadtime for infotainment systems just is untenable. Try to persist with that, and you WILL end up with all worthwhile features in the handset.
A side benefit is that consumers can bring old cars back to the dealer for an electronics upgrade. Yes I realize that aftermarket upgrades represent a loss of profit for the manufacturer and dealer -- but the first manufacturer to deal with that not by integrating the system so much that it can't be upgraded, but provide their own upgrade path, is going to have a huge hit with consumers.
This part being targeted towards the automotive industry: does this suggest the power draw is somewhat lower than a standard Access Point kind of product? Do you know what digital interface the chip is supporting?
Richard Barrett, Product Director, Wireless Connectivity, told us the following:
Broadcom's 5G WiFi for automotive is derived from the same mobility architectures that support smartphones and tablets. They are lower power by design relative to traditional access point architectures. The SDIO3 interface is used for mobility as it is lower power than current PCIe technologies used in traditional access point.
Blog Doing Math in FPGAs Tom Burke 2 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...