Are there any examples showing monte carlo simulation to find probability of line of sight for randomly distributed heights

monte carlo simulation

[David Goodmanson]

Hi SG,

how do you define successful line of sight in this situation?

[Sudeep Gyawali]

If, lets say there is a aerial vehicle flying above a building of height h and autonomous vehicle is distance d away from the base of aerial vehicle, and there is no obstruction between the uav and vehicle then it is successful line of sight for me.

[Sudeep Gyawali]

[Sudeep Gyawali]

What i am actually trying to do is finding the probability of occurance of height h1, h2.... When there is a line of sight between uav and vehicle.

[David Goodmanson]

[moved from answer (mistake) to comment]

so the target is at ground level, and there are a certain number of objects in between, at random disances with random heights?

reply: yes

[David Goodmanson]

if you put the car at the origin of coordinates and define the x and y distance to the uav as D and H respectively, the line of sight of course has slope H/D. Then let d_i be a set of distances from the car to buildings (the side of the building closest to the car), drawn from a suitable probability distribution. Similarly h_i is a set of corresponding heights, drawn from another suitable distribution. Then you can compare each ( h_i / d_i ) to H/D to see if line of sight is preserved. You might have to throw out some sets of d_i and try again if the difference of some of the d_i is smaller than the realistic spacing of buildings.

[Sudeep Gyawali]

so, according to what you have said: i can have the height_of_buildings as a uniform distributed numbers between minimum and maximum numbers(say 10m and 500m of the heights of building).

and the distances can also be a uniform distribution?

so, given this two array of variables are generated, then we compare each ( h(i)/d(i) ) with H/D meaning that we compare the : building-and-car-slope to drone-car-slope. if drone-car-slope is greater, then there is LOS otherwise NLOS.

[David Goodmanson]

yes, although the distributions can be uniform but could be other distributions as well.

Uniform distributions make some sense, but the distribution of building heights is not uniform. If you search 'distribution of building heights in cities' and similar phrases, interesting articles pop up.

[Sudeep Gyawali]

Thanks a lot, your response helped me a lot.

Also, how to plot the probability of line of sight with respect to height in same scenario. As now i am stuck in finding the probability?

Answer 1

[Anmol Dhiman]

Hi Sudeep,

I could not find any specific examples as you mentioned but thought this link would be helpful to you.

Hope it helps