On the www.lan23.ru are updated page for building wifi swr meter. mcWit was able to find replacement of the BAT62-03W Schottky diode used by pe2er.nl in his design. New diode is hsms-2822
There are several changes in the elements in due diode change. I can’t understand everything written on the site. But everything is there and all is build and tested.
Now just need to wait for more documentation.
Here is a picture of the finished device :
18 or 19dBi - no mater for this one 
. Several dipoles printed on PCB board. Have more dipoles You get more dBis. Will it be 10 12 or 24 - You decide how much do You need.
This is defiantly homemade design. I have one small antenna whit only 4 dipoles - and it works great, so this antenna should work fine. Also will be easy to be rescaled for other bands - 2.4Ghz or 5Ghz only change dimension of the dipoles.
19dbi panel wifi antenna link to picture
Here is the photo from the other side of the pcb board and other part of feeding element
Picture is from this address : diy.wifi-live.pl/node/4 where also have blueprints for less dbi antenna .
On www.zero13wireless.net also has several files posted by sushisan .
Link to files 18dbipcbdipoleantenna.zip
At las time a lot of hits came from a Poland there are some good wifi forums but I cant find online translator 
Here one interesting antenna from diy.wifi-live.pl
This is some strange Biquad and Yagi hybrid.
Dimensions :
/odległości:
deflektor- 36
/19
quad- 34
/17
direktor1 33
/17
direktor2 32
/17
direktor3 31
/17
direktor4 30
/17
direktor5 29
There are no results or testing for this antenna.
Are You remember this one : http://www.digdice.com/?p=138
This looks very crazy, oddly promising 18 dbi gain.
BUT now there are more info about this, as promised by admin of the GUW forum.
Here is all info in spanish :
http://www.guw.cl/sitio/index.php?option=com_content&task=view&id=110&Itemid=36
The author of this one is Ross Anderson and the original idea for this antenna is from 1984 year the good thing is that all the information about this one is here.
There is interesting version of this antenna build by the author. On his site he describes the how to build this panel antenna from zing fence. The original 2400mhz version is here : http://home.comcast.net/~ross_anderson/2400MHzAntenna.htm
The idea and the info : http://home.comcast.net/~ross_anderson/CurtainQuad.htm
Here some pictures from the em simulation from GUW site :
So as I see this, I can accept that this antenna works. On the site is said that this antenna outperform many commercial ones.
The diagram of this antenna and look from in front :
More pictures of this antenna build and tested for 1296Mhz and 2400mhz :
Interesting about this is the using of speaker wire for matching transformer :
“… Solder a 1+ inch piece of speaker wire to the N connector. One wire goes to the center pin; the other wire to the outside part of the connector. I drilled a small hole in the connector flange for the wire to be inserted into and soldered in place. Leave about 3/4 inch of insulation on the speaker wire (as measured where the two wires are together). This is the transformer, transforming the antenna impedance to about 50 ohms.”
There is also PDF tutorial on the GUW site : guw panel 18dbi.pdf
So this is promising design which I should test soon too. Will be interesting to find out will this design works if it is printed on PCB?
