Saturday, 1 October 2011

Comparison Of Different Wireless Interfaces

Between most wireless networking technologies we use as of now, the one we directly operate the most is probably Wireless LAN or WiFi. 802.11 is a figure we all have seen. It is the set of instructions most modern Wi-Fi tech. is based upon, given by the team of engineers from IEEE in 1997. We’ve used Bluetooth, Infrared (cheap, portable mobile technology), and even I didn’t know I’ve made use of 802.11g as a fourteen year old child to run the net on laptop to game.

(My trusty old 802.11g data card)

What we see nowadays is mostly 802.11a, b, g and n. This post explores the differences between them. Now, theoretically, this is what happens:

· 802.11 , 2.4 Ghz frequency waves capable of only transmitting 2Mbps in short range.

· 802.11a, 5 Ghz regulated frequencies (expensive and undisturbed by other appliances),
upto 54 mbps in a short range. This became accepted in office networks.

· 802.11b 2.4 Ghz, 11 mbps unregulated frequencies. These are cheap to produce, more penetrative than 802.11a waves and thus have better interior range. Was mostly for domestic use.

· 802.11g 2.4 Ghz, 54 mbps, regulated, amalgamating best of both a and b, this has become the most common wireless due to affordability, decent indoor range. It is relatively safe in transmission.

Low frequency allows for greater range and makes it affordable.
“Backward compatibility” with 802.11 allowed for it to build on a the 802.11b user base, allowing new adapters to work on b signals.

· 802.11n 2.4/5Ghz tech aiming to use multiple transmitters and receptors (MIMO) for bandwidth speeds upto 300 mbps over maximum range.

(Note : this is not our own work, this is a preview/precap of a well-researched topic from listed sources)

Our experience:

Prateek and I paid multiple visits to the server room. We decided to measure the range, speed and frequency of the various connections there.

The sir there asked us to bring our own modem to test the same, and thus we got a Beetel 450T ADSL Wi-Fi modem. Apparently, a b and g Wi-Fi connection was available here. The router has its own limitations of range and speed (as we discovered on its Manual), it cannot test send over 24 MBPS thus disallowing us to check G and N networks. It could also only transmit b and g signals.

Another challenge that we faced was finding speed over range. And range over obstacles (walls, indoors).

Our current day laptops come with various Wireless adaptors whose settings can be seen and modified. Prateek’s Inspiron 15 R and my XPS 16 for example support b, g and n reception. This can be seen from
Control Panel -> Device Manager -> Network Adaptors -> Wifi Adaptors -> Advanced

In Ubuntu, we go to the Network Connections, Select Wireless. While adding a new connection with ad-hoc properties, we can specify the wireless standard we wish to use.

First , how to find frequency?

Coming from a gaming background, my first association with frequency was responsiveness, and thus PING. Ping or latency in getting a response from the server seemed to have a direct correlation with frequency of transmissions.

(Run -> cmd -> ping 192.168... {enter IP address}, check result)

Experiment: I speculated that smaller the ping, better the response, thus frequency must be 1/ping. Upon pinging the IP addresses of a few pc’s in the lab we found the delay to be about 0.08 ms.

Taking the inverse of 0.08ms, we only get a frequency of 1250 Kilohertz. This was nowhere close to being in Giga terms.

Testing the range of different Wi-Fi interfaces :

We used a 450TC1 router. If you have a basic mtnl/bsnl/airtel etc Wi-Fi enabled broadband connection, this is the default router supplied. We first type in your IP (internet protocol) address into the browser on a machine connected to your router via cable. This is the address your computer on the internet. This information is sent out through the cable to access/modify the settings of the router:

The user name and password settings are either the default settings of the device or those set by your service provider. I called up mtnl to inquire about mine.

Here is what the router settings menu looks like:

We went to Interface Setup à Wireless:

Important notes: Your Wi-Fi connection is secured by a simple encryption method called WEP or WPA; namely Wireless Equivalent Privacy and Wi-Fi Protected Access. They use relatively weak encryption algorithms over a security key (conventionally 40, 64 or 128 bit) to ensure outsiders cannot access your wireless. The latter uses TKIP encryption (Temporal Key Integrity Protocol) and is slightly better than WEP.

What we did for outdoor range:

I placed the router right outside the window (so that it remains unobstructed). Now I tried to set it to b, g and n exclusive modes respectively and physically carried my laptop around to check signal is detected up to what distance.

Upon activating b mode, (theoretical range is ~50m), I checked the signal received on my Laptop:

I walked outside and noted the longitude and latitude coordinates of the router and the range end point of the WiFi using a GPS device (Positioning Software on Nokia N97 mini). I used Google maps to evaluate the exact between the 2 points:

By Pythagoras Theorem, we get the range of the Wi-Fi as 39m which is close to the theoretical value, thus justifying our observation. It is also worth nothing that on the backside of the house (through 5-6 walls) the range was cut to just 20m approximately.

Then I tried this for g:

And again obtained the same range. This was theoretically backed; b and g have the same range.

Now, upon shifting to 802.11n, a slight hiccup was faced. 802.11n refuses to work on WEP and WPA2 security, possibly demanding better protection due to its large range and more susceptibility to misuse.

We could only activate 802.11n on RADIUS encryption,
RADIUS stands for “Remote Identification Dial In user service”.

It is supposed to be safer than WPA and WEP, but we did not know how to configure it. Nevertheless, we can detect signals (even if we cannot access the internet) in a range. I expected a range of about 65m (theoretical range is 70m), but the following signal died again just at the dairy itself.

It then struck me that I am actually standing at the horizon of the range of my tiny router, and not the limit of the signal range of 802.11n signals.

Now, most modern day routers come with support for b, g, and n networks, because they are backwards compatible. It was not possible to test the signals for a. It is also worth noting that while the transmitter is signal specific, a n-enabled wifi card will accept signals of all previous standards.

What we did for indoor range:

I used the software “Xirrus Wi-Fi Monitor 1.20”. It’s a freeware windows gadget available on the internet which collects information about Wi-Fi networks. It gives the interface of the Wi-Fi, its frequency, encryption and signal strength through Radar/Sonar like format.

To find the range I used the software’s locate feature. It Gives a sound beep with a variable frequency depending on the distance from the Wi-Fi router. So instead of searching the router with the locate feature, I used it for finding the range of different Wi-Fi interfaces.

(At 100% signal strength the beep frequency is the maximum, for 0% signal the beep dies.)

I placed the router 5m into the house, obstructed by about 4 walls and 3 doors, I walked outwards (away from the router) until the beep died out completely and signal strength appear 0% on the radar. During this process I also recorded the latitude and longitude coordinates of my router and the final point where the signal died out using my NOKIA C5-00 mobile’s GPS.

The video below illustrates the process for b and g type Wi-Fi interface:

Latitude, longitude coordinates of Wi-Fi router: 28°41’32’’ N, 77°9’54” E

Latitude, longitude coordinates of range end point for both b and g: 28°41’32’’ N, 77°9’55” E

Using Google maps, I found the distance between the two locations which came out to be 25m.

Comparing the observed data (~25m) to that stated on the internet (~30m) came to be very close thus verifying our observation.


Now for speed:

This was frustrating. It took us three visits to the server room to get this done. After failing to find software to do this, we decided to use our trusty However this posed its own problems, as Flash plugins over IIITD servers are forbidden. Once we returned because sir wasn’t there. The next time he allowed us to use his Cyberoam with special privileges, but on the very day the servers had very poor speed (0.05 Megabits per second!), thus not allowing us to conduct our test.

It was not possible to compare the speeds of these wireless standards as even the best commercially available internet connections in India do not exceed 7.2 mbps, much below the capacity of these wireless standards.

Testing different Wi-Fi interfaces for frequency:

However, after further research, we found that

Frequency: is the arbitrary band of frequency range to which a transmitted signal belongs to and is received. Thus testing if a connection is 2.4Ghz is futile, as unlike bandwidth speed, the frequency written on the box or adapter IS the actual frequency data signals are transmitted over. These are caught by a receptor designed to receive only this given frequency. At most variation occurs to a very slight extent in channels around 2.4 Ghz itself, such as Low, Centre, Upper Values where it may vary from 2.401 to 2.412 to 2.423 Ghz, (typical standardized Channel 1 by the IEEE itself).

Using Xirrus Wi-Fi Monitor 1.20 Gadget and switching router settings between b and g as done above, I recorded the frequency of the 2 interfaces:

For b interface:

For g interface:


We discussed the theoretical definition and creation of the Wi-Fi technology. We talked about how we could relate it to our personal experiences and then finally explored a few aspects vis range, frequency and speed of different Wi-Fi interfaces(a, b, g and n) and were satisfactorily able to compare our observations to theoretical results.

Contributed by:

Aditya Gupta and Prateek Mehra


Yahoo answers :

Wiki :

No comments:

Post a Comment