[1st-mile-nm] Information request on wireless reliability and frequency conflicts

Fri Dec 19 18:22:36 PST 2008

Thanks Bob, much appreciatted

This started on a project we where working on in Phoenix doing due 
diligence on a data management system with multiple wireless inputs.

It was plagued with errors shut downs and faults and so I did a complete 
inventory of the buildings wireless systems and also did wireless 
frequency detection mapping on several layers to see what was what.

Multiple signal detection at 802.11b ( yes its up to 2.4ghz - 2.6 was a 
typo ) and 900 htz was detected outside of the buildings frame but with 
potential interference to the remotes together with 5.8 .

The second weekend we where there suddenly the system started 
functioning correctly and the cross signals disappeared and I went to 
the high rise next door and a financial company was in the process of 
shutting down there entire business and turning of all the MISIT 
networks which were using Canopy and extensive wi - fi with boosters all 
over the place . In the same clients building on one floor all of the 
5.8ghz cordless phones kept dropping signals and malfunctioning and 
after the Canopy wi fi experience we went looking for a similar issue 
and found that the floor used 802.11n 5.8ghz airport extreme all over 
the place, turned of the airport bases and the phones work fine turn 
them on and the phone system cordless sucks. We also found that multiple 
systems are set up with little MISIT support and multiple medical 
systems are set up by the companies who supply the equipment and the 
SCADA in turn is set up by a maintenance contract team and they do not 
talk or coordinate with each other period. An error arises and fingers 
are pointed

After this we contacted several other hospitals who are having similar 
problems and asked if anyone had done a signal conflict analysis and 
test and whether this was part of any systems integration due diligence 
and of course its not and they had not. We also then went to several 
other clients in the area and drew the same conclusions as they had been 
having loads of interference.  This is especially pertinent on signal 
transmitters and boosters.

It would seem like everything else that 900 was used until it got 
crowded then up we went to 2.4 now thats getting crowded we go to 5.8

Go ask if anyone has done real world testing and analysis with best 
practices and the answer seems to be no hence our question

Needless to say this is scaring the crap outa the insurance carriers

( : ( : pete

Peter Baston

*IDEAS*

/www.ideapete.com/ <http://www.ideapete.com/>

Bob Knight wrote:
> Another note - polarization may be your friend in crowded spectrum.
>
> Bob
>
>
> Bob Knight wrote:
>> You need to look at the WISPA archives relative to Motorola Canopy. 
>> While it is the largest deployed base in that community, it also is 
>> considered to be the largest problem interference-wise, at least in 
>> the 2.4 (not 2.6) Ghz band. In our area, though, they're using 900 Mhz.
>>
>> In 2.4, there are 11 channels (in the US, at least). Ideally (and 
>> nowadays, this is far from ideal given the spectrum pollution), one 
>> wants things 5 channels apart, especially when hidden node problems 
>> are anticipated. The problem is that hidden node is going to be a 
>> problem pretty much no matter what measures you take.
>>
>> WiMax can be used in various bands from 2-66 Ghz. 2.5 and 3.5 are 
>> licensed spectrum and it can be used in 5.8. It seems to be suited to 
>> 700  Mhz as well.
>>
>> Our experience on 5.8 is very positive. We use it for our backhauls. 
>> One nice thing about this spectrum allocation is that the channels 
>> are non-overlapping. Our backhaul sink for the DS-3 has 5 different 
>> links coming in with no interference.
>>
>> Admittedly, we are rural. But I guarantee that if the Santa Fe 
>> spectrum in 5.8 (or Eldo for that matter) were crowded, we'd see it. 
>> And we don't.
>>
>> So where are you seeing problems?
>>
>> Bob
>>
>> peter wrote:
>>  
>>> We are involved in several projects investigating wireless conflict 
>>> signals within corporations and hospitals and are finding some 
>>> strange results which could impact Wi-Fi and Wi - Max usage and 
>>> indeed the whole paradigm first mile wireless use.
>>>
>>> Basically this is what we are seeing
>>>
>>> Open wireless frequencies that have been typical used are 900 - 2.6 
>>> and now 5.8ghtz.
>>>
>>> 802.11 of various types operates in both the 2.6 and 5.8 sectors ( 
>>> 5.8 is 802.11n and 802.16 wi max )  the 900 htz range is packed and 
>>> so is the 2.8 ghtz
>>>
>>> Inside of a hospital for instance multiple monitoring equipment 
>>> types and portable phone systems operate on the same frequencies and 
>>> so does a huge amount of SCADA operations.
>>>
>>> Although channel switching is supposed to prevent signal conflict 
>>> wi-fi  signal boosters and all wimax whether static or mobile is 
>>> continually channel hunting and knocking everything else of of the 
>>> channel making many types of systems non usable
>>>
>>> It would seem that a major problem is manufacturers testing their 
>>> equipment in a silo arena and this would also be relevant to the 
>>> white space transmission requests before the FCC at present
>>>
>>> This looks like a huge problem and if anyone is experiencing the 
>>> same issues please contact us
>>>
>>> Apparently Motorola with their Canopy system and Intel ( wi-fi and 
>>> max ) did a large amount of research  but when the conflict issues 
>>> arose chose to ignore the results or assumed that channel frequency 
>>> switch would fix the problem but its not
>>>
>>> ( : ( : pete
>>> -- 
>>>
>>> Peter Baston
>>>
>>> *IDEAS*
>>>
>>> /www.ideapete.com/ <http://www.ideapete.com/>
>>>
>>> 3210 La Paz Lane
>>>
>>> Santa Fe, NM 87507
>>>
>>> /Albuquerque// Office: 505-890-9649/
>>>
>>> /Santa Fe// Office: 505-629-4227/
>>>
>>> /Cell: 505-690-3627/
>>>
>>> /Fax: 866-642-8918/
>>>
>>> /_Mailto:pete at ideapete.com <mailto:pete at ideapete.com>_/
>>>
>>>  
>>>
>>>  
>>>
>>> ------------------------------------------------------------------------ 
>>>
>>>
>>> _______________________________________________
>>> 1st-mile-nm mailing list
>>> 1st-mile-nm at mailman.dcn.org
>>> http://www2.dcn.org/mailman/listinfo/1st-mile-nm
>>>       
>> _______________________________________________
>> 1st-mile-nm mailing list
>> 1st-mile-nm at mailman.dcn.org
>> http://www2.dcn.org/mailman/listinfo/1st-mile-nm
>>   
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www2.dcn.org/pipermail/1st-mile-nm/attachments/20081219/6f1e54b3/attachment.html>