It's kind of shocking to think that carmakers would not embrace Wi-Fi Bluetooth combo chips. As you say, Junko, nearly all smartphones use these chips. Why would automakers not fully embrace the joys of wireless connectivity?
As others have noted, designing chips for the automotive industry may be a big leap for some of the manufacturers who have traditionally been supplying parts for consumer devices like smartphones and tablets. A similar situation exists for the space industry, which Rajan Bedi recently discussed in a blog post - You guys are dinosaurs! - on EDN.
I want to point one thing; Bluetooth +HS mode (which is introduced at BT3.0) is actually WiFi. It is officially called PAL/AMP (Protocol Adaptation Layer / Alternative MAC and PHY), which is essentially "hijacking" WiFi chip by Bluetooth driver stack. So adapting BT+HS technology is essentially "switching Bluetooth to WiFi" as the result.
Bluetooth itself has its own "high speed" mode called EDR (Enhanced Data Rate), which was introduced at BT2.0, but EDR is only 3Mbps max.
The another subtle but big problem is supply issue. WiFi/Bluetooth combo chip is mainly desgined for cellphones and tablets, usually have product lifecycle 3-5 years. In order to be competitive (smaller footprint, lower power consumption, higher throughtput / latest standard), no chip manufacture want to stick with old generation process.
Of course car industry is very different. They want to secure parts supply at least for 10 years. For chip suppliers, that means they have to keep old generation process fab running for long time, or must have large dead stock - either way raising the cost, reducing their profit.
Unless they made sure about huge growth oppotunity in WiFi - for car market, or they suffers market satulation in consumer electoronics WiFi (PC, Phone, Tablet, etc), it is high risk (and perhaps not very profitable) challenge for chip suppliers to jump in to car industry, I think.
Chip making for Automotive industry is bit different than cell phone or tablets. Usually automotive system needs lot more reliability under extreme conditions and redundancy too. You certainly cannot afford to have wireless rear view camera image chopy or stuck due to some wireless noise around causing network congestion. And of couse you dont like nearby car causing interfearence in your car wireless system.
It will be interesting to see how we slowly moves from using wireless beyond just entertainment system.
I wonder if it means that current combo chip suppliers for phones/tablets are not interested in making "automotive-qualified" versions. Or, they are simply holding back, to see if car OEMs believe the combo is the way to go...
@Junko, my understanding from working with clients in the automotive market is that you don't just "enter the market." It is an establishment of trust and vendor longevity process that is, perhaps, more like the mil/aero market then the consumer one. It takes years to get "into cars" and with a burgeoning commercial market, many vendors probably don't think it is worth the learning curve.
It takes years to get "into cars" and with a burgeoning commercial market, many vendors probably don't think it is worth the learning curve.
@Janine, I totally get that. But I beg to differ on your second point. Believe me, every major wireless chip company I talk to is deeply interested in moving their consumer chips into the automotive market by getting their chips automotive qualified.
Most of the cell phone chip makers are watching for Automotive industry. I know Nvidia/Qualcomm/Broadcom and other such companies ar elooking into Automotive industry as well however none of those companies are right now actively pursuing these chips. My feeling is that WiFI + BT are not going to solve prolem in longer term, there need a whole new system to support the need. A system that can integrate all high data requirement needs within car + integrate system to cell phone netowrk and system that can easily connect them to home network as well.
This says WiGig transfers huge video files from tablet to car (not direct from the Internet?). A simple cheap USB or WiFi will do that quite nicely. Then the car sends to the user (using a $600 tablet/phone) such useless data as tire pressure and battery state; which probably require about 64 bytes of data, and that data may change from week to week. What big problem is being solved by this technology? How about telling me precisely what component on the engine is throwing the P0171 trouble code and whether it needs cleaning, replacement, or just a jiggle on the connector? The dealer would charge $150 for that one-time info.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...