iAutomate response...

[Michael McSharry]

I played with the CheaperRFID and was not able to achieve my objective of proximity detection. I believe the best you will ever be able to achieve is presence detection. With multiple receivers you have a better chance at reliable presence detection that is also somewhat responsive to indicating when the tag has moved out of range. I developed an adaptive algorithm with xapmcsRFID and provided a couple control knobs for user tweaking. I had collected reliabilty data early on, but now do not do anything with it.

One of the hardest issues you have to deal with is a mixed environment of short range and long range transmitters. The long range transmitter, when will swamp the receiver so it turns its gain down. The short range tags will not have enough umph to be heard by the receiver now that its sensitivity has been turned down.

For proximity detection you obviously need multiple receivers, but even at that the probabiltiy of a missing transmission is sufficiently high that timeout adjustments need to be made relatively large so near real time proximity is not possible. My objective was to track a moving tag from receiver to receiver, but the implementation of CheaperRFID did not allow it.

I think the people that use it to detect when a car is at home or away or similiar uses will tend to have the most success.

From what I had read the iAutomate solution was also marginal. Much like the use of Voice Recognition for automation. A lot of appeal for the concept, but the reality is that it only works in specific controlled scenarios.

 

[AnthonyZ]

Got my "goodies" yesterday from CheaperRFID. So this morning, I opened Hyperterminal and had all my addresses "learned" and devices labeled in a matter of minutes. So easy to work with I want to scream. I wish all the gear I used was this easy. My wife is fascinated by it and thus, time to get creative with app's. I'm wanting to announce when it's time to replace batteries, so I am trying to wrap my mind around how I will programmatically differentiate between "not true" (as in, the car is absent) and "battery needs replaced". Any ideas are appreciated.

 

[bradsjm]

I'm wanting to announce when it's time to replace batteries, so I am trying to wrap my mind around how I will programmatically differentiate between "not true" (as in, the car is absent) and "battery needs replaced". Any ideas are appreciated.

Only thing I can tell you is that when the battery gets low you tend to get bad IDs received so if you watch for "unknown" ids being received then it may be an indication one of the devices has a low battery. The CQC driver I wrote just ignores the bad IDs and increments an error counter.

 

[hucker]

I'm using it for car present detection as other posters have noted and for garbage can notifications on garbage/recycling days.

Using my own code + xpl.

Don't like the battery issues. It is a breeze to work with.

 

[ano]

I think we need to start exploring more RFID solutions. Nothing against the CheaperRFID, it certainly is a good option for the cost-constrained home tinker person, but outside of that, I'm not so sure. I have two HomeSeer with two iAutomate receivers, and about 8 iAutomate tags, but can tell you that even if Peter supported this system, its also less than ideal. Its not that coverage is short, its that its not consistent. Tags come and go into and out of range even when nothing in the house is even touched AND when the tags are well withing range. Even just a few feet from the receiver. Just buggy software or a bad system? I don't know, but its not 99.995% reliable. If two receivers can't cover my 3000 sq. ft. house, I hate to think how many you need in a 200,000 sq. ft. warehouse for what this system is designed for. (Receivers are about $500 each)

Then there is the battery issues. Tags have a reasonable price, but the batteries are non-replaceable and they are claimed to last 3-5 years. In reality its closer to 2, maybe 3 if your lucky. So now instead of replacing a $2 battery your replacing a $40 tag every 2 - 3 years, which means unsticking the old one, and re-memorizing the new one. I'm even willing to overlook the tag problem if the whole thing actually worked, but it doesn't.

I was thinking of contacting these guys:
http://rfidusa.com/

They carry many RFID solutions including the Wavetrend receivers and tags that make up the iAutomate system. Maybe they can recommend something that can be molded into what we are looking for?

 

[Rupp]

I think we need to start exploring more RFID solutions. Nothing against the CheaperRFID, it certainly is a good option for the cost-constrained home tinker person, but outside of that, I'm not so sure. I have two HomeSeer with two iAutomate receivers, and about 8 iAutomate tags, but can tell you that even if Peter supported this system, its also less than ideal. Its not that coverage is short, its that its not consistent. Tags come and go into and out of range even when nothing in the house is even touched AND when the tags are well withing range. Even just a few feet from the receiver. Just buggy software or a bad system? I don't know, but its not 99.995% reliable. If two receivers can't cover my 3000 sq. ft. house, I hate to think how many you need in a 200,000 sq. ft. warehouse for what this system is designed for. (Receivers are about $500 each)

Then there is the battery issues. Tags have a reasonable price, but the batteries are non-replaceable and they are claimed to last 3-5 years. In reality its closer to 2, maybe 3 if your lucky. So now instead of replacing a $2 battery your replacing a $40 tag every 2 - 3 years, which means unsticking the old one, and re-memorizing the new one. I'm even willing to overlook the tag problem if the whole thing actually worked, but it doesn't.

I was thinking of contacting these guys:
http://rfidusa.com/

They carry many RFID solutions including the Wavetrend receivers and tags that make up the iAutomate system. Maybe they can recommend something that can be molded into what we are looking for?

Go for it and let us know what you find out.

 

[jay]

Looks like a real company!

 

[ano]

Looks like a real company!

I wrote to them and we'll see what they say. I know that that they have the protocol for the Wavetrend receivers so with that its possible we could add interfaces to various home automation systems to the Wavetrend receivers (same one iAutomate uses), and therefore bypass any further reliance on iAutomate. That company also sells tags for Wavetrend which work with what iAutomate has, so if you need to purchase more tags, you might want to purchase them here instead of iAutomate.

The Wavetrend stuff might be the most cost-effective, but maybe it just was the iAutomate "layer" that has been causing the problems.

 

[ano]

He was their reply. I guess not unlike I would have expected. I don't think the HA community requires that much hand-holding, but I can understand their concern. If we can get the interface spec for Wavetrend, for example, we could design our own interface, but we would need a point person, so a company like this would only need to deal with one knowledgeable person, not many unknowledgeable ones. In a way that is the role Peter had, but that didn't work out well. :lol:

-------------------------
Thank you for your inquiry.
I have seen the iAutomate system advertised a couple of years back. Dynasys does offer the Wavetrend tags and readers typically for industrial applications. The main shortcoming of Wavetrend products, as you have mentioned, is the lack of easy to use software. Most of our applications are developed from the low level protocol specification which is not user friendly.

I do not feel that the home automation market is where I want Dynasys to be. It will likely generate many customers buying small amounts of hardware, wanting cheap pricing and over loading us with requests for technical support. I do not believe it would be profitable.

I appreciate your interest but I do not have any product line that would be simple enough and cheap enough to satisfy a hobbyist RFID market. I have enough trouble already dealing with engineering students and their science fair projects.

I am available for further discussion if you wish to call.
Thanks
Bob

Bob Scher, CEO
[email protected]

www.RFIDusa.com
www.dyna-sys.com

Dynasys Technologies Inc.
800 Belleair Road
Clearwater, FL 33756 U.S.A.

 

[Ranger Digital]

While that is an understandable response from bob I think he could have been a bit more tactful as a CEO. Dont get me wrong, I respect those that can call a spade a spade, LOL. But found his reply a little less professional than a that of a CEO.

Regardless, point taken.


eKey was at CEDIA. Wonder if their response would be the same? Given they were at a custom installer convention, maybe they have a proper solution for us. Might be worth taking a gander .

http://www.ekeyusa.com/
http://www.ekeyusa.com/pdf/ekey%20TOCA%20S...Description.pdf

 

[electron]

It sounds like if someone is willing to deal with support/software development, and only deal with the company when buying hardware, it might work. Hardware could be bought in group buys in order to increase the size of an order, making it more interesting to them.

 

[TCassio]

I currently have the cheaperRFID stuff and use it to determine who is home and who is not. I used to use it to open my garage doors when I arrived home. This proved to be to un relaible as I have a mix of 40m transmitters and 8m transmitters but they don't work well together. I looked at the more expensive iAutomated solutions and realized that it was not that much more reliable than the cheaperRFID stuff for the cost.

Here is an idea:
Someone with basic knowledge of electronic circuits could design a circuit that would give an open for two seconds and then a close for two seconds, and repate over and over. This circuit could then be adapted to be used with the DS10A door/windows sensors. These devices have pretty good range and are very reliable. I've been using them on a couple of doors for years. DS10A's along with the WGL W800 reciever could prove to be a reliable, cheap solution.

I use Homeseer and I don't think it would be to difficult to have Oman modify the ACRF plugin for this seniaro.

Any thoughts.

 

[electron]

I have had a motion sensor/palmpad in my car for several years now, and range wise, it will work. I used to use it for a 'honey I am home' routine, but always wondered how well it would work with the DS10A's since the signal is more secure than the X10 RF signal. Another option I am considering is modifying one of those GE wireless door sensors, so the signal is really secure, and if the range is anything like the keyfobs, then it should be a really good solution. I just need to find some time

 

[Squintz]

I would reverse engineer the IAutomated/HS plug-in to get the protocol but I don't own HS2.x and the last time I used HS1.x it caused my PC all kinds of problems.

I'm thinking that reverse engineering the protocol should be possible since it simply communicates over the RS232 port. An RS232 port sniffer program which I have should be able to read what's being said. With enough trial and error you could figure it out.

Or all of us who have spent thousands on these IAutomated readers could buy a dev kit from wavetrend.

I was hoping the Miracle would have shiped already. I have 3 readers and about 12 tags from the early beta days and I can't use them with anything other than HS.

 

[ano]

I would reverse engineer the IAutomated/HS plug-in to get the protocol but I don't own HS2.x and the last time I used HS1.x it caused my PC all kinds of problems.

I'm thinking that reverse engineering the protocol should be possible since it simply communicates over the RS232 port. An RS232 port sniffer program which I have should be able to read what's being said. With enough trial and error you could figure it out.

Or all of us who have spent thousands on these IAutomated readers could buy a dev kit from wavetrend.

I was hoping the Miracle would have shiped already. I have 3 readers and about 12 tags from the early beta days and I can't use them with anything other than HS.

That is certainly a possibility, but maybe not the best option. Wavetrend does offer a the spec of the protocol for free to "qualified developers" but I'm not so sure how you become one of those.