Line of Site, LOS, Near Line of Site, nLOS, Non Line of Site, NLOS - Definitions etc

I am trying to get a handle on the terms LOS, nLOS and NLOS.
What they stand for on the face of it is easy enough.
For the record and in case I need correcting this is what these acronyms stand for.
URLS are to Wikipedia/Webopedia for want of a better starting point. Any comments welcome:

LOS = Line of Sight (Wikipedia) - Source and receiver can see each other
nLOS = Near Line of Sight (Webopedia) - Describes a “partially obstructed path between the location of the signal transmitter and the location of the signal receiver. Obstacles that can cause an obstruction in the line of sight include trees, buildings, mountains, hills and other natural or man-made structures or objects.”
NLOS = Non Line of Sight (Wikipedia) - This “occurs outside of the typical line-of-sight (LOS) between the transmitter and receiver, such as in ground reflections.”

Late entry - A more specific Cambium definition is given here:

So far so good.

Lets take PMP450. Its brochure makes a statement about LOS and nLOS capability:

(Apology I cannot find where I downloaded it from exactly. Got it off recently e.g. last 2 weeks)

Question: How do I know if my PMP-450 is functioning with nLOS? I know it is capable of such.

ANSWER1 = Well if I have a working link AND there is some stuff in the way. E.g. trees/foiliage. That I should be able to assess with a straightforward visual inspection. A nice example is given here:
Anyone using ePMP 425 for nLOS / NLOS PTP?
From that post is this photo apparently showing the line of site clearly and substantially obstructed:

ANSWER2 = This post What are the threshold values for nLOS and NLOS? seems to give a second possible answer. Basically the Excess Path Loss (in dB) loss tells us the extent to which the fresnel pattern is interrupted. by this definition
nLOS = 6-10dB loss
NLOS = 10dB or more
Excess Path Loss is calculated by the Link Planner. I am yet to investigate that tool.

So those are the 2 possible answers I have found. Both are reasonable enough. I would expect there to be come correlation between the 2. i.e. the Link Planner output would hopefully correspond with the real life scenario shown with the trees in the photo above. There is chatter I think about the Link planner in that post. I have not dug in too deeply.

Let me restate my question:
How do I know if my PMP-450 is functioning with nLOS?

Here are I guess some corollary questions that might better flag what I am wanting to know:

  • Is there anything on the controls/gui of the unit itself that will tell me?
  • Does the unit know if it’s fresnel is obscured and/or traffic is travelling by bouncing off stuff e.g. the ground/houses/etc ? (Or is it restricted to simply measuring a received signal strength?)

I hope this question makes sense!

I suspect there are good docs that I have not found yet. Please post links if so.
With thanks for any comments…

Here is an interesting post on the topic
900 MHz and nLOS, NLOS, and LOS

In my opinion, some examples of near LOS (nLOS) would be you can still see part of the transmission tower from the client radio… so maybe there’s a tree branch, or a post, telephone pole, etc. in the way.

Non LOS (NLOS) would be a complete blockage of LOS from the transmission tower to the client. This could be a single tree, a building, a hill, etc.

Radios don’t know or report if the RF path or signal is being attenuated by nLOS or NLOS factors. They only metric that they report that might be of value is the receive signal strength indicator (RSSI). One can make an educated guess as to the LOS conditions by looking at the radio’s RSSI’s and comparing it to a predictive reference point like cnHeat or LinkPlanner. That being said, there are other issues that can reduce or skew RSSI values… like the antennas being misaligned, snow/ice on the antenna’s radome, damaged/loose RF connectors or connectors with water in them, etc.

In any case, when in doubt, it’s always a good idea to model the link path using something like cnHeat or LinkPlanner, and then compare that to what you’re actually seeing in the field, and then make adjustments as needed.


I would extend Eric Ozrelic’s definitions to include in nLOS any fresnel zone intrusions that do not completely block the path like building or tree tops or passing through a notch in the trees or between buildings.
LOS would be with completely clear path without any incursions what so ever and no trees or buildings within the fresnel zone.

1 Like