Sure, lower power is a good thing, and IoT clearly needs much lower BW. But players in this field seem to be missing the big picture: people want to control and benefit from their Things. What's missing is not so much layer 1, but layer 7 or so.
The current Things tend to just try to toss our data onto some vendor site, where we rely on the site to make the data worthwhile. We don't really get to even see our own raw data, though sites often make some version of it available. Why shouldn't the model be that our Things report to *our* storage, which we conditionally permit to also flow to vendor sites. Or maybe third-party sites - after all, why would a Thing vendor necessarily have any particular skill at websites, or making data from one Thing meaningful to another vendor's Things...
Wi-fi looks to progress more. Even if we look around the gadgets we use in home wi-fi hasa presence everywhere. Home automation is an area that has been in news from quite some time but has not made a revolutionary impact. May be the interface used and what are commonly supported in consumers home devices were quite different. But yes quite agree on the overloading the bandwidth part.
Home automation can be made a reality in countries like america and many parts of europe but in countries like india where even the digital set top box for viewing tv channels are yet at their basic levels home automation will have to go a very long way.
Given that its only one engineer, its hard to tell whether its just a "me too" trying to find another band to put WiFi on, or whether there's a real need.
802.11/WiFi originated from the desire to give Ethernet mobility, hence WLAN. Using it on 900MHz only seems to go contrary to this given the fact that WLAN requires high throughput. Over the years, WiFi steadily went up in bandwidth from 22MHz to 40MHz, 80MHz and 160MHz. Going to smaller bandwidths will get customers less throughput and add to confusion. I realize that 802.11/WiFi has good marketing potential (whereas Zigbee isn't recognized by most consumers), but this will be confusing to consumers.
My company Argenox Technologies does 900MHz designs, and I can tell you that one of the main reasons we pick Sub-GHz designs is to avid WiFi congestion. Here comes 802.11/WiFi to grab more bandwidth and disturb other devices, just like it happened in 2.4GHz.
WiFi chipsets and modules such as TI's CC3000 are steadily gaining steam and although power consumption is an issue, they are easy to use. Unless I needed extreme low power or Mesh networking, I'd pick them over Zigbee to integrate IoT.
Not sure why this should make a difference to anyone but WiFi vendors, however. It's true that 802.11 can be applied to any number of roles other than that of "home PC network." Why not? Assuming you can make the electronics very small, cheap, and low power, it becomes a very flexible wireless LAN that could in principle accommodate snesors, refrigerators, toasters, and the light switches in your house.
It's more important to understand the Internet Protocol that has to ride on top of this, seems to me, than to focus on the link layer protocol.
Blog Doing Math in FPGAs Tom Burke 23 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 ...