-----Original Message----- From: Mustafa Hasan [mailto:[EMAIL PROTECTED] Sent: Tuesday, July 15, 2008 6:08 AM To: Adams Jon; [EMAIL PROTECTED] Cc: [email protected] Subject: Re: [6lowpan] Regarding coexistence with 802.11n
>2. The current research on this issue is interestingly biased, some are saying its not a big >problem (e.g ZigBee Alliance) and others (e.g. Z-Wave Alliance) are saying its a big issue. <ADAMS> Beware of wild claims %^). At least in the US, the 902MHz band has enjoyed a huge resurgence of uses as well as a strengthening of licensed uses. As many do not realize, the 902-928MHz band has some licensed users including location services and amateur radio, in addition to a huge and growing amount of license-free users. For the license-free users, it's the location and monitoring services (LMS) though, that is most concerning from a usability point of view. Back in the early 1990's, before the price of GPS receivers dropped much below US$1k, several companies in the US were able to get licenses to occupy 6MHz wide channels in the 902MHz band with fairly high power mountaintop or towertop radios. The object was to provide trilateration services to mobile receivers in delivery trucks, theft recovery services for automobiles, etc. Well, the price of GPS dropped more quickly than LMS, and the companies went silent, the licenses pretty much worthless. In the last few years, there has been a LOT of lobbying of the FCC to repurpose these licenses to allow the licensees to do pretty much whatever they want with their frequencies, including wireless broadband, etc. This is a great concern as they are not subject to the power limitations of Part 15 and in fact at least one licensee wants to exact some significant restrictions on 900MHz Part 15 devices. Wireless broadband internet service providers have discovered the excellent propagation and penetration of 902MHz. It is now common to find, especially in suburban or rural areas, wireless ISPs delivering broadband in the 902MHz band, in 5, 10 and even 20MHz bandwidths. The legal limit there is +36dBm EIRP as well. Lastly, one significant and continually growing use of the 902MHz band is wireless metering for electric, gas and water meters. My house has a 900MHz wireless gas meter that blurps out a packet every minute or so. In dense housing areas the aggregation can be impressive. US railroads have deployed 900MHz readers for RFID tags that are on about 80% of the railcar fleet (about 8M railcars) and those readers transmit essentially a CW carrier at a +44dBm level for several minutes while the train passes the reader. RFID readers in stores and commercial establishments are increasingly using 900MHz for illuminating tags using similar power levels. Amateur radio is making greater use of the 902MHz band (they're a primary licensed user). While it's FM narrow-band comms for the most part (but there is digital QPSK and GMSK as well), the EIRP is typically in the +43 to +50dBm range for mobile, portable and tower/mountaintop. The 900MHz band is no panacea. The users there are often digitally modulated, with power levels from Part 15 levels (+36dBm EIRP) and up. A truly obsolete technology (narrow-band FSK, traditionally 9600 baud, with some usage now at 40k baud, single RF channel) like Z-Wave cannot compete with any of the above uses. Not only is Z-Wave exceedingly slow and inaccurate in assessing the availability of the channel, its packets are 40-80ms long, and it isn't detected by the digitally modulated systems out there, so they punch holes in the packet. FSK requires a significant SNR (10-15dB) to demodulate successfully, so any interference can threaten its link performance. The sole Z-Wave 908.42MHz frequency happens to be in the middle of the mobile transmit side for ham radio, so the ham down the street can saturate a Z-Wave receiver without even knowing. It's no wonder that the new Z-Wave 401 chip is purported to support the 2.4GHz band. Now, all of this doesn't mean that somehow the 2.4GHz band is much better - like Christian and Ben remind us, there's plenty of uses in the 2.4GHz band that aren't too worried about who else is out there. However, both in-lab extreme testing, real-world testing, and anecdotal evidence show that 15.4, 802.11, microwave ovens, Bluetooth and cordless phones aren't necessarily incompatible. Witness the success of 15.4 systems and demonstrations in huge conference arenas like CES in the US, where the density and average power level of 802.11g systems (nodes on nearly every channel, even though that doesn't make much sense due to overlap) conspire only to generally increase latency of the 15.4 system by a few milliseconds at worst. Our 15.4 systems here at work behave well even with our very high usage and density of 802.11g systems. Use of 15.4 systems is most successful for low-duty cycle (<1%), moderate latency (10ms) uses. As Ben points out, there is significant concern by many that 802.11n has the potential to not play very fair with the channel, and make life difficult for the rest of us. This is a good reason for those of those with voting rights in 802.11 to stand up and fight against this threat to the 2.4GHz band. </ADAMS> Sincerely, Jon Jon Adams Business Development, Wireless Connectivity Operation Freescale Semiconductor 2100 E Elliot Rd MD EL 542 Tempe, AZ 85284 +1 480.628.6686 mobile +1 480.413.3439 office [EMAIL PROTECTED] _______________________________________________ 6lowpan mailing list [email protected] https://www.ietf.org/mailman/listinfo/6lowpan
