Breaking News
Product News

ESC: Microchip simplifies access to embedded wireless

3/26/2010 05:00 PM EDT
9 comments
NO RATINGS
Page 1 / 2 Next >
More Related Links
View Comments: Newest First | Oldest First | Threaded View
Jon Adams
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
Jon Adams   4/7/2010 12:33:33 AM
NO RATINGS
IEEE 802.11, in any of its myriad and highly useful forms, is not a substitute for IEEE 802.15.4. There's a bunch of absolutely technically valid and market-driven reasons why the IEEE 802 group, while continuing to cook 802.11 and its variants, also created 802.15.4, and why there is such a tremendous, continuing and growing interest in adapting 15.4 to suit different uses. Long ago, the .11 folks decided that their future was speed and more of it. Likewise, the 15.4 folks decided that it was battery logevity and the appreciation for extremely low duty rate, sleeping devices in networks of varying topologies. These items alone represented a significant and fundamental fork between the two groups. That 15.4 was created "for no good reason"? Hogwash. As Mr Wright points out, 15.4 has a significantly lower overhead per frame than any version of .11. It was designed that way from the start. As Mr Gridmerge notes, ZigBee (the networking, security, application and certification "layers" atop 15.4) started out years ago with the assumption of the need for security and didn't have to suffer through something like WiFi's "WEP phase". The choice of data rate, spreading, and frame structure makes 15.4 have excellent link margins, way better than even the most emaciated 802.11b system. 15.4/ZigBee systems have significantly lower complexity, far better link margin in real world implementations, and can fit in much narrower channel bandwidths, allowing more independent systems per unit of spectrum. And spectrum is scarce. 15.4 has some great, native mechanisms to adjust latency, power consumption, provide for contention-based and contention-free channel access, to allow devices to go dead-to-the-world for long periods of time and still participate fully in a mesh network of varying topologies, with intermixes of all sorts of devices and functions. There are real-world 15.4/ZigBee systems deployed today with thousands of active devices in a single network. 802.11 can't do that, and would need a lot of rework to the basic spec to allow that. Look at what someone like Awarepoint is doing with tens of thousands of active, non-fixed (i.e., could be moving at pedestrian speeds) devices all in a single facility. Or the just-opened CityCenter in Las Vegas, with nearly 100,000 ZigBee enabled devices. Fanny, the IEEE 802.11 working group hasn't had much interest to date in addressing the problems that both the IEEE 802.15.4 task group and the ZigBee Alliance have solved. And I've been a participant and voter in all three groups. 15.4 latched on to these challenges a decade ago (and ZigBee beginning in 2002). They both have achieved great success, demonstrated by the market's growth and through adoption by other groups that see the technology as the way to go (Continua Health Alliance, IPSO Alliance, ISA100, RF4CE, etc.). Sounds like we're not the ones late to the market.

Gridmerge
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
Gridmerge   4/3/2010 4:51:13 AM
NO RATINGS
@Fanny: I don't know where you get your information from but your understanding about the security issue regarding ZigBee is wrong. ZigBee provides comprehensive authentication, confidentiality and integrity measures based on public and symmetric key cryptography both at a network-wide level and between two application peers in the network. ZigBee also hasn't had to suffer the pain of a transition from an initially broken specification. If WiFi is used for embedded applications with limited user interfaces it too will have to face up to more challenging credentialling requirements, some of which are used today in ZigBee through use of public-key cryptography providing certificate-based mutual authentication.

giuann
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
giuann   4/1/2010 4:32:03 PM
NO RATINGS
I think the solution is already here if the big brains would consider the '6lowpan' that has been adopted by the ISA society and is already used for medical monitoring in hospital. Too simple for the MBA mentality where simplicity is a dirty word. Big groups have spent a ton of money on the zigbee and they want their money back. good day gld

MWright
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
MWright   3/29/2010 10:14:14 PM
NO RATINGS
Security remains at the risk of the weakest link. This may be the inappropriate use of a particular protocol, an improper selection of available security mechanisms, an inappropriate conversation between people, or an unlocked door. It is the responsibility of the system designer and every user. The use of various networks will allow good decisions to be made in terms of power consumption, ease of use, and ease of integration; but it comes with the responsibility of tying the various components together so that appropriate security measures apply where they are used. Less and less items are "standalone" in the new "internet of things" paradigm. Even contemplating security just because something is supposed to be "standalone" is dangerous -- keep in mind the military datafeeds from UAV's. An appropriate level of security must always be considered for the environment, and use of an embedded application. This in itself, may force a decision for a particular technology. Thus it is one of the important selection criteria.

MWright
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
MWright   3/29/2010 10:02:55 PM
NO RATINGS
@Fanny, You are correct, strong security is part of Wi-Fi. Also, it is of little use to have new standards re-inventing the wheel. But, Wi-Fi will not get as low power as ZigBee on a same transmit power comparison because the ZigBee protocol is simpler and made for lower bandwidth uses. So they are not apples to apples. The alliance will prove valuable if it can facilitate simpler connection between networks of both kinds, not replacing one with the other.

patrick.mannion
User Rank
Staff
re: ESC: Microchip simplifies access to embedded wireless
patrick.mannion   3/26/2010 9:11:55 PM
NO RATINGS
The Wi-Fi security scares of the early years have subsided, for sure. However, with embedded Wi-Fi-enabled devices now possibly connecting to potentially less secure ZigBee (and cellular) networks, combined with the Google-enabled Internet access to all, does security now loom large again, or is it just not an issue? It's a new space for embedded systems designers used to 'standalone' systems.

Fanny0
User Rank
Blogger
re: ESC: Microchip simplifies access to embedded wireless
Fanny0   3/26/2010 8:41:08 PM
NO RATINGS
My understanding is that security issue has been solved for Wi-Fi, but not for ZigBee and that ZigBee is about 10 years behind Wi-Fi on introducing authentication and crypto features. So, I‚??m not clear on what the problem is and why embedded Wi-Fi would be any different from regular Wi-Fi when it comes to security. Or am I missing something? Fundamentally, there is no technical reason why Wi-Fi could not operate at the same low power consumption as ZigBee. The technology is not fundamentally different. All you have to do is reduce the Wi-Fi transmit power and data rate and ‚?? bingo ‚?? you have a low power wireless device based on mass-produced technology. The world doesn‚??t need yet another standard such as ZigBee trying to solve the problems that have already been solved by Wi-Fi. New standards occasionally created for no good reason other than that the key players want to invent their own rules when they‚??re late to market, rather than play by the rules already established by someone else. Sometimes there is a good reason for a new standard, but often you could just as easily adapt the already existing technology.

MWright
User Rank
Rookie
re: ESC: Microchip simplifies access to embedded wireless
MWright   3/26/2010 7:25:17 PM
NO RATINGS
Wi-Fi has evolved very well allowing varying levels of security. There is a choice of none, right up to triple-A grade enterprise security. The question on any of these for an embedded device is how to add the device to the network and enter these credentials. Not an easy task for something that may not have a keyboard and display. Fortunately, work being done by Stuart Cheshire (Apple) and Google are addressing both getting ‚??things‚?Ě to get on easily, and also for the secure communications to be seamless to the end user. Designers have the ability now to develop WiFi products for the embedded space that are easy to use and secure. The details of the underlying apps will be discussed at 2010 ESC in April.

patrick.mannion
User Rank
Staff
re: ESC: Microchip simplifies access to embedded wireless
patrick.mannion   3/26/2010 5:51:31 PM
NO RATINGS
The opportunities for embedded wireless are manifold, but so too are the security issues associated with it. For many applications, it's not a concern, but with Google now in the monitoring game and Internet access ubiquitous, should designers really start to get concerned? Where's the cut-off between caution and paranoia -- and what should we do about it?

Flash Poll
Radio
LATEST ARCHIVED BROADCAST
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Top Comments of the Week