Heatmap and Radio Signal strength value

[manunc]

Hello, I need advise to decide which is the best option in order to:

• create a geolocation heatmap displaying a radio signal strengh in dBm (from 0 to -140 dBm)

I try to use MAPS functionnalities with grid aggregation but there is only 2 options to display data:

• count: does not display the signal strengh data in this case
• sum: (of the signal strengh) not Ok because it can have more point in a region and not many in another although the signal can be good in the second region

it is said in the documentation:
Only count and sum metric aggregations are available with the grid aggregation source and heat map layers. Mean, median, min, and max are turned off because the heat map will blend nearby values. Blending two average values would make the cluster more prominent, even though it just might literally mean that these nearby areas are average.

So how to reach the goal ?

Thanks by advance for your advise, I would lto know if Kibana can help me on this.

[Nathan_Reese]

This sounds like a really interesting use case. Thanks for posting your question.

What metric would make the most sense for your use case? What are you trying to symbolize with your heatmap?

The Maps application currently limits heatmap metrics to count and sum but the list can easily expand the list if needed.

[manunc]

Thanks for your answer.
In fact this is simple.

The metric I need to symbolize is a field value (ie: rssi: -20 dBm for instance)
Each coordinates point will have a RSSI value associated (each document)
I need to display a heatmap with a range of colors/rssi_value from blue (low value: -140 dBm) to red (hot value: -20 dBm)

This is usual in radio monitoring activity to have a map like this.

Is it clear?

[Nathan_Reese]

How many documents are you trying to visualize? The Documents source displays documents from Elasticsearch but is limited to the first 10000 matches. If you are trying to display large number of documents, then its recommended to use the Grid aggregation source to avoid showing incomplete results. Grid aggregation source groups documents into buckets and then returns metrics derived from the set of documents in each bucket.

Below are the color bands used by heatmap.
0.1, royalblue
0.3, cyan
0.5, lime
0.7, yellow
1, red

There are several options for your use case.

1. You could use multiple layers combined with layer filtering to show your documents styled to the heatmap color bands. Below is an example showing web logs sample data set with five layers, one for each color band. Your data would look better because the web logs sample data is random while your data set will be clustered since its sensor readings. You can try this with Kibana 7.2 and any version higher.

2. We recently merged a new feature that will be included in 7.4 that allows for defining custom color ramp scales. You could use this feature along with Grid aggregation symbolizing the results as grid rectangles to get the desired effect. Below is an example using the logs sample data set were each grid is symbolizing the average bytes for all documents in the grid cell.

Would either of those solutions work for your use case?

[manunc]

Thanks, the 1st solution works!