Now that we have covered the theory behind modeling a link, let's take a real world example. Some time ago, the company for whom I work wanted to setup a wireless link to Ashland from Columbia, a distance of 16.7 kilometers (10.3 miles). We wanted to replace a 30 mbit/s metro-Ethernet line that was costing us far too much money with one we controlled, and only had to pay for the once. We also knew that we needed to increase capacity substantially, and should aim for at least 60 mbit/s of one-way performance.
Based upon forum advice, we decided to pair up two Rocket M5 radios, with 30 dB dishes. We were also co-locating on a busy site, so we ensured that we included RF Armor to reduce self-interference at both ends.
We started by preparing a link budget:
|Transmitter Power||21 (reduced, we want MCS15!)|
|Transmitter Antenna Gain||30|
|Receiver Antenna Gain||30|
|Free Space Path Loss||-132.17|
|TOTAL (RSSI)||-51.17 dB|
|Receive Sensitivity||-75 dB|
|Fade Margin||23 dB|
This is an excellent link-budget, showing a very viable link! Since we are using the same antenna at both ends, there is no need to run the reciprocal link – they should perform the same. With perfect alignment, we should see an RSSI of -51 dB on each end. That's far (23 dB) above the required sensitivity for the MCS encoding we wish to achieve – so that looks great.
Next, we ran a Fresnel Zone calculator to ensure that we can achieve appropriate clearance. The first Fresnel-zone at this distance (5.8 ghz frequency) is 14.6 meters (48 feet). 60% of the zone is therefore 8.76 meters (28.7 feet). Therefore, to achieve a link without obstructions, we need to be at least 28 feet above any potential obstacles along the path. We checked this with Radio Mobile (see the next chapter!), and confirmed our feeling that this was not a problem. We were connecting a very tall building with a cellphone tower, and were far above the trees and houses in-between the two points.
Armed with this knowledge, we purchased the equipment and installed it. Installation went mostly smoothly, in a hot Missouri afternoon, and soon we had a functioning link between the two sites. We achieved an RSSI of -54 dB – slightly worse than the prediction, but once we hit the -55 dB region we stopped trying to improve our aim: the link was satisfactory. It has been running ever since, and we rely on it every day.
A few months ago, our large fade-margin came to our rescue. A cellphone company who shall remain nameless installed larger antennas on our site (we are co-located with several other companies), and partly obscured one of our antennas. We raised a stink and this was corrected, but in the mean-time our signal dropped to the -62 dB region. This was still well within the signal required to achieve the speeds we need, and justified the use of large dishes (rather than, say, a NanoBridge M5 pair) – we had plenty of margin for error, and our link remained functional while we talked with the cellphone company to correct their abuse of their lease (the problem was largely corrected, but it took... a while. Large cellular carriers have little impetus to play nice, sadly).« Chapter 4: Antennas Patterns and Downtilt Up To Contents
If you like this book, please consider making a donation to the author, Herbert Wolverson. If there is sufficient interest in this e-book, we'll post more chapters.