Setting up a WIFI link between DL8RDS and DB0MHB

Aus DL8RDS Wiki
Wechseln zu: Navigation, Suche

This project is the supertarget of other projects:

And it is a sister project of:

1 Progress report

  • 2008-09-28 No good. Today it was a sunny autumn day and my friend Bernd DJ6PA was ready to help me over at DB0MHB. At my station, I placed my small tube-enclosed 13cm YAGI into a basket of clothes, directing it right towards DB0MHB, as I have clear sight to DB0MHB. Then I took the car and drove over to get the other side right. But what a disappointment. It was not possible to establish a link. Maybe it was up to the polarization, which I didn't consider at all right away (as a shortwave guy you don't have enough space anyway to care for polarization, but for SHF it seems really important). Maybe it was up to the fact that the tube-enclosed WIFI YAGI was really not worth the money. Bernd already tested it and it didn't really work well. Anyway, let's give it another try. BTW, we had a number of other problems too, mostly due to the reason that these damn WIFI engineers always use Reverse Polarity SMA instead of normal SMA. Another reason for the initial failure could have been the rather long coax cables on both sides, on my (DL8RDS) side about 5 meters, and on the DB0MHB side about 10 meters.
  • 2008-12-26 I have now purchased another YAGI over Ebay which was allegedly being used in another WIFI link scenario. This antenna should work better than the tube enclosed predecessor. I have also shortened the cables and I have constructed two Fonera routers to care for the link. Just a few meters of cables, even if H155 high quality ones, still eat up so much signal power that the WIFI transmitter won't deliver enough energy to the antenna and that the received signal strength won't make it into the receiver. So I decided to place the TRX very close to the antenna, only about 50 cm apart from each other. Today, I'm still waiting for a crimp compressor to fix the N plug, but that is a minor issue. I have also purchased a Linksys Power over Ethernet kit that provides the 5V which is required by the Fonera devices. The POE kit will help in case cabling should be difficult for any reason. Then everything should be in place for the next try.
  • Decided to build highly efficient DC converters into the link transceivers and to ensure that every device runs with 12V. The plan to use the builtin 5V was discarded because there are too many risks: Mix up the wiring in a bad moment and all the nice electronics goes smoky.
  • 2009-03-28 Success!! Today we managed to establish the link. The link distance is about 5 kilometers and the link quality is stable. The bandwidth we reached is a stable 36 MBit/s.
ath0     IEEE 802.11g  ESSID:"scramblescramble"  Nickname:""
         Mode:Managed  Frequency:2.412 GHz  Access Point: sc:ra:mb:le:sc:ra
         Bit Rate:36 Mb/s   Tx-Power:18 dBm   Sensitivity=1/1
         Retry:off   RTS thr:off   Fragment thr:off
         Encryption key:off
         Power Management:off
         Link Quality=17/70  Signal level=-79 dBm  Noise level=-96 dBm
         Rx invalid nwid:89896  Rx invalid crypt:0  Rx invalid frag:0
         Tx excessive retries:0  Invalid misc:0   Missed beacon:0

On my side, the antenna was even behind a window within the house. The window's glass is sure to have caused some further loss, but it had just a minor effect on the link quality. The 36 MBit/s is a great success and was very inspiring all over the evening. The next thing is to get the routing right. But since the proof of concept has been reached, I consider this project as finished.

  • 2009-08-28 We have been assigned HAMNET / ampr network IP addresses and hostnames. Now we can use the link productively. The only thing that's still missing is an official permission, but that's just a matter of officialeese...
dl8rds.ostbay.de.ampr.org <-> 44.130.59.169
db0bul.ostbay.de.ampr.org <-> 44.130.59.170
db0mhb.ostbay.de.ampr.org <-> 44.130.59.171

2 Transmission power

Here's a good article on power calculations: http://www.nwlab.net/tutorials/db/eirp-umrechnung-dbm-mw.html

And here's a calculator: http://home.in.tum.de/~prilmeie/wlan/db-umrechnung.xhtml

In order to estimate the used transmitting power, here's a calculation example:

X dBm = 10*log( Y mW / 1mW);
so 0 dBm = 1mW
10 dBm = 10mW
13 dBm = 20mW
18 dBm = 63mW
19 dBm = 80mW
20 dBm = 100mW
23 dBm = 200mW
30 dBm = 1000mW

An increase of 3 dBm doubles the transmitted power.

  • 10 dB more means power x 10
  • 20 dB more means power x 100
  • 30 dB more means power x 1000

Here's a picture of the 13cm WIFI Yagi antenna on my side. In the background you can see the "Münchshofener Berg" (Münchshofen Mountain), 530m above NN. The counterpart of the link, DB0MHB, is located about 30 meters below the peak.

13cmWIFIYagi.jpg

Here's my power estimate:

18 dBm TRX power
+ 15 dBi estimated ANT gain
- 3 dBm cable and other setup loss
-----
30 dBm

which equals up to 1 Watt e.i.r.p. in the beam. So clearly illegal for normal people since the maximum legal e.i.r.p. in Germany is 20 dBm. And yes, it is legal for ham radio operators.

3 Legaleese

I quote in German from http://www.tino.cc/amateurfunk/afu-wlan.html

Das IEEE 802.11b-Übertragungsprotokoll stellt Kompatibilität zwischen Geräten unterschiedlicher Hersteller im gesamten drahtlosen Netzwerk über mehrere Funkzellen hinweg sicher. Funknetze nach 802.11b arbeiten auf Frequenzen von 2,4 bis 2,4835 GHz, im so genannten 2,4-GHz- oder auch ISM-Band (ISM steht für: Industrial, Scientific, Medical). In diesem lizenzfreien Bereich funken ebenfalls zahlreiche andere Geräte wie zum Beispiel Mikrowellen oder Bluetooth-Produkte.

Der WLan-Frequenzbereich ist als ISM-Band freigegeben, d.h. kein Nutzer hat Anspruch auf Schutz vor Störungen durch andere, legal betriebene Anlagen!

Bei einer Bauanleitung für WLAN-Antennen habe ich vor kurzem folgendes gelesen:

"Der wohl bedeutenste Nachteil von selbstgebauten WLAN-Antennen ist rechtlicher Natur. Die Sendeleistung von 100 Milliwatt darf bei WLAN-Geräten nicht überschritten werden. Bei Antennen mit einem niedrigen Gewinn ist dies nur selten der Fall, aber Antennen die sowohl sende- als auch empfangsseitig einen hohen Antennengewinn haben, überschreiten diesen Grenzwert oft um ein Vielfaches. Da Laien meist nicht in der Lage sind den Antennengewinn selbst zu berechnen ist der Bau und die Inbetriebnahme von Antennen ausschließlich Funkamateuren und anderem Fachpersonal vorbehalten. Nur diese Personen dürfen eine selbstgebaute Antenne in Betrieb nehmen, man muss also bei einer selbstgebauten WLAN-Antenne alles von einem Spezialisten überprüfen und anschließen lassen. Ein weiterer Nachteil von selbstgebauten Antennen ist die oft schlechte Qualität der Teile die zusammen mit ungenauer Verarbeitung zu starken Qualitätsschwankungen führt. Oft sind selbstgebaute Antennen völlig unbrauchbar weil Teile schlecht zugeschnitten wurden oder beim Zusammmensetzen ungenau gearbeitet wurde. Der Einsatz von Eigenbauten lohnt in der Praxis meist nur, wenn man selbst Fachmann ist oder wenn man Fachpersonal kennt welches einem die Überprüfung und die Installation zu einem guten Preis durchführt und beim Bau der WLAN-Antenne zur Seite steht." Theoretisch darf also die maximale Sendeleistung (e.i.r.p.*) nicht mehr als 100mW betragen.

Soviel zur Theorie. Beschäftigt man sich aber einmal mit den Frequenzplänen, so findet man hier den Frequenznutzungsplan der Bundesnetzagentur. Dort steht unter Eintrag 279004 WLAN nocheinmal die maximal zulässige äquivalente Strahlungsleistung 100mW e.i.r.p. bestätigt. Liest man nun weiter zum Eintrag 279005, findet man hier für den Funkdienst AMATEURFUNK den Frequenzteilbereich 2400 - 2450 MHz zugewiesen. Das heisst, für lizenzierte Funkamateure gilt für die Kanäle 1 - 8 die Einschränkung der Sendeleistung nicht:

Funkamateure dürfen also dort mit einer Leistung von bis zu 75W senden,

WLan-Karten liefern nur 30mW (15dBm). Um hier Störungen zu vermeiden sollte man einen der Kanäle 9-13 wählen.

Als lizenzierter Funkamateur kann man WLan natürlich als Amateurfunkstelle deklarieren und ist somit nicht mehr an die Leistungsbeschränkung und die Meldepflicht (!) für Richtfunkstrecken bei der Bundesnetzagentur gebunden. Der Netzwerkname sollte in diesem Fall das Rufzeichen beinhalten und (um die Bandgrenzen nicht zu überschreiten) sollte ein Kanal zwischen 1 und 8 gewählt werden.


  • e.i.r.p.: e.i.r.p. (Equivalent Isotropic Radiation Power – äquivalente isotrope Strahlungsleistung) ist ein theoretischer Wert, der die Leistungsaufnahme eines imaginären, isotropen (kugelförmigen) Senders angibt, die nötig wäre, um in alle Richtungen die gleiche Abstrahlungsleistung zu erreichen, wie bei einem gerichteten Sender. Die Angabe der Abstrahlung in Watt EIRP ist also unabhängig von unterschiedlichen Antennencharakteristiken und deren entfernungsabhängigen Feldstärkeverlauf, wie z.B. bei Rundstrahlern, Sektorantennen, Richtantennen, etc. und gibt bei WLAN eher den Gewinn in Bezug auf eine theoretische Normale von 0 dB wieder.


Here's an overview of channels within the 2,4 GHz range:

Channel 01 : 2.412 GHz
Channel 02 : 2.417 GHz
Channel 03 : 2.422 GHz
Channel 04 : 2.427 GHz
Channel 05 : 2.432 GHz
Channel 06 : 2.437 GHz
Channel 07 : 2.442 GHz
Channel 08 : 2.447 GHz
Channel 09 : 2.452 GHz
Channel 10 : 2.457 GHz
Channel 11 : 2.462 GHz

4 Recommendation for non licensed readers

Before you read my summary, have a look at this excellent German article: http://www.brennpunkt-srl.de/reichw.html

There are WIFI boosters around on Ebay. There are also kindof YAGI antennas around with 5 meters of cable attached. Forget all that stuff. Why?

  • Boosters are illegal in most countries. And they're unnecessary. Don't just think of the transmitted signal, but also think of the received signal. The latter is generally by dimensions weaker. The best transmission signal is worthless if it's blocked by some obstacle in the way or if the antenna is so unselective that the signal disappears within the noise level. There is a wisdom that says: A good antenna is the best booster. It's like all those dumb as a rock CB radio operators running illegal kilowatt boosters, wondering why they cannot be heard across Europe while HAM operators operate worldwide with just a few watts. It's all about being able to hear well! It's not about being able to shout!
  • Keep your coax cabling extremely short. Every meter (!!!!!!) will eat up about 3 dBm, even with good cables. So if you have 15 dBm and 5 meters of cable, NO signal will come out at the end. It's hard to reach some extra gain, but it's easy to prevent unnecessary loss. Since decibel is a logarithmic and not an absolute measure, it's hard to speak in millivolts, but the received signal at the antenna is surely just a few millivolts, compared to the transmitted signal. Figure out that the received signal will not run for 5 meters to disappear below the noise level. It will be lost already after a short distance.
  • In order to keep legal, DO use a good antenna. Make sure your antenna has a big gain. Then calculate precisely, and reduce your transmission power to a degree that does not exceed the 20 dBm. The law does not say anything about highly selective receiver systems.
  • Since we're talking about receivers. There are active and passive systems. By active systems I am talking about RX signal boosters that sit right on the antenna and pick up weak signals. They are legal, but it depends on the exact model if their VOX is fast enough. In any case, they're a good idea to push your sensitivity to extremes, but they only make sense if you already own a good antenna. It's stupid to combine a bad antenna and a RX booster. That's a waste of money. If you're interested in antennas, have a look here: http://www.hb9bs.ch/images/stories/pdf/Antennentechnik/
  • Talking about antennas: 2,4 GHz is handy. The antennas you construct don't eat up all your garden like on short wave. At most they eat up your working table. But they have a disadvantage: It's very critical to pick the right length of the radiator element. Fractions of millimeters do count! And if you cut off half a millimeter too much, the resonance frequency of the dipole will no longer be optimal for your desired frequency. The optimal tool to measure resonance frequencies is a dip meter, but a SWR meter will also do if you can use a range of transmitting frequencies and draw a resonance diagram (which will show a 'dip'). Bad enough, there are hardly any SWR meters around for 2,4 GHz. And there's something else: You need a stable reference transmitter to do your measurements. The packet stream of WIFI is not a good signal source, it's too unsteady to reach optimal results.
  • Need help? Ask your local amateur radio club. There are people who can help you. Radio is a fantastic hobby.