Category: Self-hosted

Posted on by

SOOTHSAYER™ 1.9 released

Our latest feature release improves offline setup and adds support for HF Skywave, interference analysis and indoor simulation.

Offline by default

With pre-built images, offline install is now enabled by default.

This was developed to both speed up the install, ensure all dependencies are present and correct and automatically assign the domain name / IP address. IP networking in particular was noted as common support issue so we’ve put effort into helping users get this right early on in the setup.

We use a lot of open source libraries in our stack and there is no guarantee they will be available or compatible at the point a user decides to build. This stung us after we launched 1.8 when a third party “npm install” command later failed causing us to release a patch (1.8.1).

A build-your-own option is available to users who wish to generate their own images using our compose script.

Offline install in 5 minutes

HF Skywave

Rather than reinvent the wheel, we integrated the popular VOACAP engine into our API as a new model option. You can use VOACAP via the web interface and API to make HF heatmaps or frequency predictions.

We incorporated feedback from VOACAP experts about the critical relationship between antenna height, take off angle, and the subsequent antenna pattern. The engine uses HF antenna patterns which are built on-demand by our API dynamically so you will get the corrected pattern for a given dipole height above the ground.

The HF model was calibrated against published papers and crowd sourced sounding data. You can read more about this in the HF blog.

ARM-64 support

Support for ARM-64 architecture has been officially added now. We field tested this portable edge server concept up a mountain and were very pleased with the power economy of the Nvidia Jetson.

Read more about the field test from earlier this year.

Phase Tracing interface

A new 3D interface with BYO models designed for urban simulations with multi-path, material properties, phase tracking and support for the glTF 3d model format.

This builds upon the 3D API which shipped with 1.8 last year.

Read more about Phase Tracing in the blog.

Enhanced interference analysis

Interference analysis had a significant upgrade in December to enable support for co-site interference and network analysis.

Now you can compare entire networks to see how they might interfere and get a visualisation of the impact of bandwidth and power on adjacent channels.

Read more in the interference blog.

Interference analysis using published spectrum sharing standards

Deprecated features

As a small software company we cannot add features indefinitely. We have therefore retired the following interfaces in line with CloudRF:

Google Earth desktop interface

This was getting harder to support with every new Javascript feature as it’s Webkit browser was frozen in time since they focused effort on the web based version. You can still export to KMZ and open in Google Earth.

TAK chatbot

We made this in 2021 to avoid writing an Android app as we have scars from Android development going back to 2012. The first version used the taky TAK server which was then rewritten for the OG TAK server. As a result we ended up learning a lot about TAK and our open source setup script became one of the most popular non-DoD TAK Server repos on Github.

We now have a signed ATAK plugin which does not need a TAK server so is much easier 🙂

Product Flyer

SOOTHSAYER-2025-flyer.hd_Download

Security

Anchore compatible SPDX-2.3 SBOM available upon request.

SHA1 Checksum: 

411f309d9e228b0238e6918d15bc4e5bd7803a93  SOOTHSAYER-1.9_OFFLINE.zip

SOOTHSAYER™ Changelog

1.9.0 04/2025

  • Update: API 3.23.1, UI 3.16.2, 3D UI 1.1.0, CPU Engine 1.16.0, GPU Engine 1.12.0, 3D Engine 0.2.1, AntennaWizard 2.0.5, Bouncyball 0.4.3, Analysis Utilities (QRM2 & SuperTool) 1.0.0, Documentation 2.9.0.
  • Update: pgAdmin container updated to 8.14.
  • Update: PHP updated to 8.4.
  • Feature: Support for HF Skywave.
  • Feature: Added binaries for new processing of interference and superlayer.
  • Feature: Added new 3D UI – available at /phase-tracing.
  • Feature: Added new install.sh, start.sh, stop.sh, and watch.sh scripts to simplify management.
  • Feature: Added support for importing pre-built containers for offline environments.
  • Feature: Improved automatic assistance with setting FQDN during install.
  • Feature: Improved conditional logic for GPU and non-GPU environments.
  • Feature: Reduced size of soothsayer-network from /24 to /28 to reduce opportunity for network subnet overlap.
  • Fix: Removed use of Docker hello-world for checking Docker permissions.
  • Fix: Improved timeout delays when serving cached tiles from MapProxy.
  • Fix: /dbadmin endpoint not working for non-standard HTTP/HTTPS ports.
  • Fix: Changing database password in the admin dashboard doesn’t change password within /dbadmin.
  • Fix: Updated default TLE_CATALOG_URL value.

API

3.23.1 (2025-04-03)

  • Fix: Points data endpoint failed when using legacy colour keys.

3.23.0 (2025-03-18)

  • Fix: BSA restricted receiver sensitivity unnecessarily.
  • Fix: BSA forced to FSPL and no diffraction for all requests.
  • Fix: Terrain data map throws a 400 at low zoom levels.

3.22.0 (2025-02-27)

  • Deprecation: Removed v1 API endpoints (key=value), deprecated since Dec 2022
  • Deprecation: TAK chatbot removed now we have an ATAK plugin
  • Feature: Added colour schema to URL export.
  • Feature: Added favourite sorting to antenna database.
  • Fix: Empty colour schema file being written to archive when requesting KMZ files.
  • Fix: Added handler to prevent creation of clutter profiles with system profile names.
  • Fix: Unable to delete clutter profiles which contain certain characters.
  • Fix: Improved validtions when creating a bad clutter profile.
  • Fix: Improved error message for GPU engine over ocean.

3.21.0 (2025-02-18)

  • Improvement: Increase maximum reflections limit to 25.
  • Change: Uploaded model textures are compressed on upload
  • Feature: Credit-based purchases removed from API.
  • Feature: Users on free plan can upload 3D models.
  • Feature: Added “Raw Request” to KMZ balloon.
  • Feature: Improved system clean-up of stale files.
  • Improvement: Increase maximum reflections limit to 25.
  • Improvement: Show “No LOS” with PPA PNG chart if LOS blocked with LOS model
  • Change: Uploaded model textures are compressed on upload
  • Fix: Improve validations on edges for BSA requests.
  • Fix: Custom colour schema support for negative step sizes.
  • Fix: Custom colour schema step size of 0 triggers a loop.

3.20.0 (2025-01-22)

  • Fix: Added restriction to /mesh and /merge endpoint where sites must be within 40 degrees of each other.
  • Fix: Input model filenames with spaces would fail 3D calculations.
  • Fix: 3D calculations sometimes incorrectly returning previous responses.
  • Fix: HF NVIS calculations not working.
  • Fix: Incorrectly translating hex code of ff to RGB value of 254 for red and green channels when creating a custom colour key.
  • Fix: Downloading greyscale calculations as SHP alters the source file which causes other analysis tools to fail.
  • Fix: Sending a request to /area without an output.out value results in a HTTP 500.

3.19.0 (2024-12-30)

  • Improvement: Better handling of streaming large files.
  • Feature: Added /3d/models and /3d/model/delete endpoints.
  • Feature: Antenna file uploading moved into Antenna Wizard.
  • Fix: MANET tool reporting validation errors with RCS, even when RADAR mode isn’t used.
  • Fix: Favouriting antenna patterns sometimes is not always reflected with the favourite button.
  • Fix: Uploading of some 3D models incorrectly failing validation.
  • Fix: Improved validation when uploading small and large models.
  • Fix: Some requests fail where a user agent header is not sent to the API.
  • Fix: Antenna database filters now reset the page number.
  • Fix: Some environment values not persisting after multisite or satellite calculations.
  • Fix: Validation error when using database noise on multisite requests.

3.18.0 (2024-12-12)

  • Enhancement: Allow custom per-bucket colour when creating custom colour schemas.
  • Enhancement: Upgraded /inteference to show JS ratio instead of just which site is strongest.
  • Improvement: Increased the speed of the /mesh endpoint, and added an alias /merge.
  • Improvement: Added controls to Antenna Wizard.
  • Fix: Colour schema editor doesn’t give warning if name is invalid.
  • Fix: Colour schema buckets will style the label according to the brightness of the background.
  • Fix: Some CPU calculations failing if input files contained spaces.

3.17.0 (2024-11-25)

  • Improvement: Added support for compressed glTF models
  • Improvement: Extended noise API to handle array of values.
  • Fix: Fixed points kmz url.
  • Fix: Landcover and buildings always returned as enabled from preferences.
  • Fix: Differing logic relating to noise limits.
  • Fix: SNR units reporting as BER when sensitivity out of range.

3.16.0 (2024-11-11)

  • Feature: Added HF Skywave APIs /hf/area and /hf/prediction

UI

3.16.2 (2025-04-03)

  • Fix: Maximum altitude of 120,000 to align with API.
  • Fix: Disable engine toggle when loading in a flight path.
  • Fix: Calibration mode has redundant button to enable multipoint links.
  • Fix: Download report failing on some browsers. Moved the report within a modal window.

3.16.1 (2025-03-31)

  • Deprecated: Google Earth layer removed.
  • Feature: Added example long format CSV file to help for automatic processing.
  • Fix: Antenna not set during interface load or during selection of a template.
  • Fix: Automatic processing fails if colour schema name is only numeric.
  • Fix: Clearing layers changes radius after HF Skywave model has been used.
  • Fix: BSA mode removes Tx marker to avoid confusion.
  • Fix: BSA mode disables adjustment of model and diffraction to avoid confusion.
  • Fix: Mesh download title was undefined.
  • Fix: Loading calculations from the archive didn’t set colour key correctly.

3.16.0 (2025-02-27)

  • Feature: Added “Select all” to archive.
  • Improvement: Give modal warning when deleting a network from the archive.
  • Improvement: Show “No LOS” with PPA chart if LOS is blocked and LOS model is selected.
  • Improvement: Show calculation_adjusted messages for interference tool.
  • Improvement: Clicking success/error icon with “Automatic processing” shows debug info.
  • Fix: No feedback when using archive functions without selecting a calculation.
  • Fix: ‘invalid’ metrics entries.
  • Fix: Environment values not set from template.
  • Fix: Gain from a template gets replaced after loading a template.
  • Fix: Retrieving a null antenna upon load.
  • Fix: Satellite mode does not reference API.
  • Fix: Colour key replaced with a dBm equivalent after using “Manage My Colours”.
  • Fix: Operator mode is giving option for custom antenna patterns.

3.15.0 (2024-12-12)

  • Improvement: Intereference analysis lets you compare two networks
  • Improvement: Colour picker lets you define full custom values and colours
  • Fix: Interference analysis docs link updated
  • Fix: Loading some layers from archive throw antenna validation error when “Custom pattern” is selected.
  • Fix: Coverage analysis values have overly high precision.

3.14.0 (2024-11-11)

  • Feature: Added HF Skywave propagation model

3D UI

1.1.0 (2025-04-03)

  • Feature: Added GLB export menu.
  • Feature: Added a toggleable scale indicator.
  • Feature: Wireframe mode improved by removing interaction with light sources which create strange effects.
  • Fix: Wireframe mode is broken on some models.
  • Fix: Saving scene is not correctly loading from cached values.
  • Fix: Antenna pattern rendering not depth-checked.
  • Fix: Loading models from cache not working.

1.0.0 (2025-01-30)

  • Initial release.

CPU Engine

1.16.0 – 2025-03-19

  • Fix: Pathloss colour bins were offset.

1.15.0 – 2025-01-30

  • Fix: Opening a .ter file did not update its atime.

1.14.0 – 2024-12-12

  • Improvement: Antenna pattern interpolation.
  • Improvement: Write a .json file with metadata
  • Change: Write RGBA images so as to reduce post-processing

GPU Engine

1.12.0 (2025-03-19)

  • Feature: Added CUDA fast math flag to release builds.
  • Deprecation: Removed CUDA architecture 87.
  • Improvement: Calibration against CPU engine for different units.
  • Improvement: Values between the colour key and rxs are output as black.
  • Fix: SNR colour keys with negative numbers are now handled correctly.

1.11.0 (2025-01-30)

  • Improvement: Increase per-thread sleep to reduce power consumption.

1.10.2 (2024-12-12)

  • Fix: Removed antenna pattern artifact at azimuth 0 and 90.
  • Improvement: More accurate area/coverage calculation.

3D Engine

0.2.1 (2025-03-19)

  • Improvement: Refraction attenuation is now per meter
  • Fix: Fixed overflow resulting in random hot spots
  • Fix: Reflection attenuation was applied before the incident
  • Build: Added cuda fast math flag to release builds
  • Build: Removed cuda architecture 87
  • Build: Commented-out unused code to reduce binary size

0.2.0 (2025-01-17)

  • Performance: Only generate photons on collisions, not along the whole path
  • Improvement: Photons can be exported as a GLB for debug purposes
  • Improvement: Material refraction coefficient can be set
  • Fix: Photons were invalidated when the movement distance was to small, preventing refracting through more than one voxel
  • Fix: AllInRadius mode was summing all the photons that didn’t belong to a given transmitter ID, rather than all the ones that did
  • Fix: Refraction loss was not scaled by voxel size

0.1.1 (2024-11-12)

  • Improvement: Draco compressed models are now supported
  • Fix: Diffusion was not being passed on correctly
  • Improvement: Antenna pattern model size can be changed
  • Fix: Engine was not restarted on fatal cuda error
  • Fix: Idle CPU usage was too high

Posted on by

RF planning at the Edge

ATAK SOOTHSAYER plugin

Developments in containerised software and portable GPU hardware have enabled RF simulation at the network edge. Field testing has proven this powerful capability can serve multiple clients from a small battery for a sustained period of time.

Key findings

  • Offline operation for over 6 hours
  • Average 4W power consumption
  • Fan draws most power over a day
  • Nvidia Jetson power profile recorded
  • Phones are tricky to operate with gloves

Test setup

An Nvidia Jetson Orin nano 8GB processor was paired with a 10AH 12v battery in a portable case.

This was assembled and demonstrated earlier in January with this video.

The following software and services were installed to provide a representative stack:

Instrumentation was performed using the onboard tegrastats utility every second. This provided load data at a 1Hz refresh rate throughout the test. Two ATAK end user devices (Samsung S21 and S23) were associated with the WiFi AP and SOOTHSAYER server via the open source soothsayer plugin with distinct accounts for parallel processing.

Each phone was associated with the TAK server using TLS authentication. This served no purpose other than to show us a green light in the corner of the phone to prove IP connectivity which would be needed to use the (IP) plugin.

The last time we went field testing up a mountain we struggled with sound and wind so bought a lapel microphone so we could provide a detailed narrative whilst on a windy summit. This detail helped immensely with post analysis as we were able to identify details from the video which weren’t noted at the time.

Test results

A rugged 20km Scottish mountain route was selected over a period of 6 hours on a relatively mild day with still air temperatures in single digits celsius. Tests were performed periodically after long intervals to simulate real use with increasing difficulty: The first test being a single area calculation at 1MP resolution.

This was followed by a 4MP calculation which was later followed by a intensive stress test with two clients performing concurrent multi-site calculations. A multi-site calculation simulates several sites in a single API call and requires a GPU.

A final test was concluded at the car park where we were pleased to discover 2 of the 4 LEDs were lit on the battery. This video contains more detail of our hardware craftsmanship than we chose to share and is available upon request.

System startup

The server was prepared en-route and required it’s clock to be set manually since it had no internet access. This clock setting was necessary for both test results and SOOTHSAYER’s rate limiting which uses time stamps to ensure each user is waiting at least one second between API calls. Unlike the public system, SOOTHSAYER has no upper API limit.

The OS was ready after a minute and the software services ready after approximately another minute. The first, and largest (7W) power spike in the data was the first test from the plugin prior to departure at the car park. This was assessed to be larger than others due to Just In Time (JIT) compilation of the GPU kernel. The SOOTHSAYER GPU engine is noticeably slower for the first calculation for this reason, which is expected behaviour.

Test 1 – VHF handheld, 1 million pixels

This test was conducted an hour after startup whilst walking along. The template used was a VHF handheld radio with 10km radius and 20m resolution for a total of 1 million points. As you can see from the video it concluded quickly.

Calculating VHF mobile coverage on the move

Test 2 – VHF repeater, 4 million pixels

This test was conducted after 2.5 hours after climbing up to the snowline. A VHF repeater template with 20km radius and 20m resolution was used for a 4 million point calculation. This was noticeably slower than the previous test due to the increased computation required.

Calculating VHF repeater coverage up a mountain

Test 3 – Multi client stress test

This test was conducted on a cold summit after 3.5 hours of operation. This time, two clients were used to stress test the server with concurrent requests. The requests were more complex multi site API calls which models an entire network versus a single site. As well as GPU heatmaps, the CPU was employed to create links between the radios. Due to the steep topography and random test locations, not many viable links were displayed.

Server stress testing on a mountain top

Power consumption

The 1Hz data was collected with tegrastats and showed good power economy. The jetson fan kicks in around 43C and spins up then slows which is visible throughout the data as a saw tooth pattern.

The 3 degree temperate spike during the ascent at 2 hours occurred in a sheltered ravine. Our assessment is that the simple case was suitable for the winter conditions but would require a case with improved cooling for use elsewhere.

The notable temperate drop occurred on a mountain top during filming of test 3 when the box was exposed.

Test power and temperature data
Boot sequence showing idle load and fan load after 10 minutes

Summary

The test exceeded expectations on power economy and performance and showed that a (budget) portable battery could easily provide up to 12hrs of operation. At the end of our expedition there were 2 out of 4 LEDs still lit on our battery.

It validated the client-server design since the phone batteries were only slightly depleted throughout the test. Both phones finished with more than 75% battery capacity. Performing simulation on a phone is convenient but a poor choice for battery endurance especially and scalability due to the dependency on reference data such as LiDAR and antenna patterns which are provisioned on the server. Note that phone 2’s mapping was unprepared during test 3 but was able to render high resolution topography, sourced via the server.

Of note, the stability and power economy of both ATAK 5.2, TAK server 5.3 and our own plugin was good. We experienced no crashes or connection issues throughout the day which was welcome.

CPU calculations

In the first video, CPU tests were mentioned which we did plan to do. We tested CPU area calculations the day before and were pleased to find they can be almost as fast on a (6 core) Jetson for single sites which was a nice surprise. Compared with a GPU calculation, they use more power so are not as efficient. We did employ the CPU for link simulation on the mountain but elected to focus on the GPU compute capability and power economy.

Lapel Mic

We were stunned by the sound quality on the mountain as the wind was intense for test 3 and we were using an iPhone with a wireless lapel microphone. We recommend this microphone for field testing.

Look ahead

Applications for this network capability include Search and rescue, Mining, Agriculture, Government and Emergency services. Crucially, The edge concept challenges historic ways of working, whereby an expert with a powerful computer produces a plan which is later rendered obsolete by changing events.

With edge compute, planning becomes dynamic which is important since no plan survives contact with reality. Having the ability to redesign a communications plan locally reduces the planning cycle, saves bandwidth and improves spectrum awareness.

For autonomous systems, having integrated communications planning improves their chance of success and supports decision making beyond topographical route selection.

More information

SOOTHSAYER is a self hosted RF planning server with interfaces for different systems such as ATAK.

The software is available as containers for x86-64 and arm64 architectures and can be hosted in the cloud, on a laptop or an Nvidia Jetson. For more information see here or email sales@cloudrf.com

Posted on by

SOOTHSAYER 1.8 released

Our latest feature release adds new APIs and retires the virtual machine for a container-first solution compatible with more host systems which enables GPU sharing and greater control over both code and user data.

3D API

A new full 3D API and GPU engine designed for urban simulations with multipath, material properties, phase tracking and support for the glTF 3d model format.

The REST API can currently be interfaced with directly using scripts or via our Blender plugin. A new indoor/outdoor/underground interface to use this new API is under active development…

3D API documentation

3D API gallery

3D API Blender plugin

Satellite API

Improving upon our legacy Satellite route analysis capability, we can now model the entire coverage footprint using our GPU engine. This exceptionally hard problem required a rewrite of our GPU engine to achieve the “bounding box” capability needed to model a small area at high resolution from a transmitter 35,000km(!) away.

Satellite API documentation

General Purpose Model

Following field testing in the summer, we concluded empirical cellular path loss models are unfit for modelling UHF handheld radios. As a result we created a hybrid model from ITU-R P.525 which is frequency and height agnostic. The simple model is made accurate by the Deygout 94 diffraction model and tuned clutter which we support with new profiles.

You can read more about the testing and the model improvements here.

API change log

3.15.3 (2024-10-08)

  • Fix: Restored missing values to path and points endpoints.
  • Fix: Trilateration not working with GPU in some scenarios.

3.15.2 (2024-10-04)

  • Improvement: Admin pages improved for support of SOOTHSAYER 1.8.
  • Improvement: Tighter zoom restrictions on terrain map to reduce response size.
  • Improvement: Better default preferences eg. colours
  • Fix: /API/terrain does not load in some environments.
  • Fix: Resolved antenna rotation edge cases
  • Fix: Trilateration broken by numpy update. meh

3.15.1 (2024-09-30)

  • Improvement: 3D engine calculations are locked for a maximum of 30 seconds.
  • Fix: Stale locking of 3D engine calculations in the event of system crashes would prevent immediate calculation reattempt.

3.15.0 (2024-09-26)

  • Improvement: Added DEM Manager to the admin dashboard.
  • Improvement: Reorganized data folder.
  • Improvement: Refreshed the terrain and clutter map.
  • Fix: Removed Chart image and Network KML from network API response as they are not used.
  • Fix: Missing megavoxel from products SQL table.
  • Fix: GPU calculations with imperial units exceeded the radius.
  • Fix: Some antenna patterns being incorrectly interpreted as vertically aligned.
  • Fix: Deletion of uploaded files upon system errors.
  • Fix: GPU calculation failure when using user-defined clutter profile.
  • Fix: Processing engine is not stored as a value when logging requests.
  • Fix: Custom ant minimum value dropped from -50dB to -100dB to fix vertical patterns

3.14.2 (2024-09-10)

  • Improvement: Network API search radius increased to 0.5 degs for mountain WISPs .
  • Fix: Antenna elevation plane wrong for some patterns.

3.14.1 (2024-09-05)

  • Improvement: Include Altitude type when retrieving PPA JSON from archive.
  • Improvement: Clutter manager supports custom building density.
  • Improvement: Custom antenna patterns are smoother.
  • Fix: Fix mesh endpoint was ignoring blue and alpha.
  • Fix: Antenna downtilt inverted by recent change.

3.14.0 (2024-08-14)

  • Feature: Added output-bounded GPU calculations.
  • Feature: Added /satellite/area endpoint.
  • Fix: Fix antenna wizard plotting tilt direction and handle all valid ADF gain values.

3.13.0 (2024-08-01)

  • Improvement: Multipoints KMZ improved to show signal strength making use of bubbles along the route.
  • Improvement: 3D endpoint accepts ‘output.megavoxels’ as an alternative to ‘output.res’.
  • Improvement: SLEIPNIR 1.13 has black alpha background like GPU for easier GIS styling/analysis

3.12.0 (2024-07-12)

  • WARNING: ITM will be conservative with new clutter profiles.
  • Feature: General Purpose Model (#10)
  • Enhancement: Landcover trees now support up to 1.5dB/m for jungle
  • Enhancement: New ITU clutter profiles for CITY, SUBURB and FOREST
  • Enhancement: Refactored CORS header handling
  • Enhancement: 3D engine version and megavoxels added to admin page
  • Fix: Handle SNR 0 in templates
  • Fix: 3D API materials validation
  • Fix: KMZ PPA network link restored

3.11.0 (2024-06-19)

  • Feature: Added 3D engine processing endpoints: /3d, /3d/model/upload, /3d/viewer.
  • Fix: Error message if license ID has changed.
  • Legacy: Removed colour schema value of 8.

3.10.6 (2024-05-30)

  • Fix: ITM model cannot be used with altitudes above 3000m
  • Fix: Path KMZ metric heights were missing TxH
  • Gift: GPU enabled for Silver plans 🙂

3.10.5 (2024-05-17)

  • Fix: Multisite antenna tilt inversion.
  • Fix: Stop Google Earth UI opening clutter/antenna forms.

3.10.4 (2024-05-10)

  • Improvement: Recommend CPU engine for Ocean planning / no tiles.
  • Improvement: Optimised Ocean planning.
  • Fix: Handles when BSA Tx equals Rx.
  • Fix: Handle P.525 optimistic context (aka circles mode) with GPU.

3.10.3 (2024-04-26)

  • Improvement: Memory allocation improvements.
  • Fix: Super layer KMZ metadata was missing some values.
  • Fix: MANET network fails to save if it has unachievable links.
  • Fix: MANET KMZ would always be produced using dBm units.
  • Fix: MANET KMZ draws links which are below the sensitivity threshold.
  • Fix: MANET KMZ assigns black to links which sit on the colour bucket.
  • Fix: MANET KMZ links were not floating at altitude of nodes.
  • Fix: Default to 1m2 RCS value for multisite RADAR requests
  • Fix: multisite antenna template/custom logic aligned with area
  • Fix: multisite model.rcs only validated if using RADAR
  • Fix: Mesh PNG alpha bug.
  • Fix: Can save templates with AMSL height units.
  • Fix: Template name capped at 30 character length.
  • Fix: Antenna Pattern Import Tool fails if OEM already exists.
  • Fix: Downloading ADF pattern from ADF wizard broken.
  • Fix: Mesh PNG alpha bug
  • Fix: Can save templates with amsl heights
  • Fix: Template name capped at 30 chars

UI change log

3.13.4 (2024-10-08)

  • Fix: CPU engine unusable if interface loaded in GPU mode by default.

3.13.3 (2024-10-04)

  • Improvement: Better default preferences for model, clutter and colours.
  • Fix: GPU calculate unresponsive when reloading the UI after a GPU session.

3.13.2 (2024-09-30)

  • Fix: Disable MANET tool when using route analysis with mixed altitudes (flight path).
  • Fix: Loading imperial ASL path from archive doesn’t properly populate on the PPA chart.
  • Fix: Google Earth layer calculations returning “Bad API key” for some environments.
  • Fix: Remove broken window when clicking on a best server link.
  • Fix: “My Account” modal doesn’t load if the API returns an error.
  • Fix: Colour schema changing when enabling/disabling MANET tool even if the schema exists for the measured unit.

3.13.1 (2024-08-20)

  • Fix: MANET marker alt aligns with plumb line alt with imperial units
  • Fix: Engine selection no longer adjusted when enabling/disabling MANET tool
  • Fix: PPA appearing in air when reloaded from archive.
  • Feature: Added dropdown selection for bandwidth in operator mode rather than an input box. Input box is still valid in engineer mode.
  • Feature: Extended lower limit of bandwidth on input box down to 0.001MHz (1KHz) to match the allowed value in the API.

3.13.0 (2024-08-01)

  • Feature: Route analysis import supports CSV as per coverage / calibration etc.
  • Feature: Route analysis import supports altitude/height from CSV/KML/KMZ. Uses ‘points’ vs ‘area’.
  • Fix: Auto-processing CSV warns when mixing short and long format fields.
  • Fix: Fixed Auto CSV typo on example re Gain_dB which is now Gain_dBi.
  • Fix: Added missing “KML + PPA” option to quick download list.
  • Fix: Buildings value not persisting between interface reloads.
  • Fix: Noise floor value not persisting between interface reloads.
  • Fix: Auto CSV processing bandwidth supports 1KHz (Previously 100KHz)

3.12.1 (2024-07-16)

  • Feature: Engineer template mismatch with operator mode modal changed to go into console window.
  • Fix: Importing KML/KMZ polyline into “Coverage Analysis” fails.
  • Fix: Multipoint links button missing from engineer mode.
  • Fix: Custom antennas in templates ignored for operator mode

3.12.0 (2024-07-12)

  • Feature: Operator/Engineer mode select
  • Feature: General Purpose Model
  • Enhancement: Models list sorted into categories

3.11.5 (2024-06-08)

  • Improvement: No longer truncating long antenna filenames in pattern input box to make automated processing simpler.
  • Improvement: Automatic CSV processing accepts antenna ID as well as name.
  • Improvement: Include URL in error message when failing to load layer.
  • Fix: Marker and label disappear when image layer is hidden.
  • Fix: RSOD when importing MANET network.
  • Fix: MANET import when MANET mode enabled will disable MANET mode.

3.11.4 (2024-06-01)

  • Improvement: Updated Cesium to 1.106 (June 2023)
  • Fix: Applied 5 months of Cesium bug fixes 🙂

3.11.3 (2024-05-27)

  • Improvement: Bandwidth changes now adjust receiver sensitivity as well as noise
  • Fix: Imported MANET nodes were getting processed twice
  • Fix: Imported MANET nodes context menu didn’t open

3.11.2 (2024-05-21)

  • Fix: Custom antenna plots update after archive reload
  • Fix: Tx marker moves after archive reload to match site marker

3.11.1 (2024-05-14)

  • Fix: Imported MANET nodes from KML were not tracking 3D terrain
  • Fix: Can import a KML route and then a KML for MANET w/coverage analysis

3.11.0 (2024-05-07)

  • Improvement: Selection of “use case” on the “Import Data” dialog. Support for kml placemarks with routes
  • Fix: Uploading KML/KMZ to “Coverage Analysis” with a Placemark and LineString
  • Fix: Uploading KML/KMZ with more than one LineString would fail.
  • Fix: Uploading KML/KMZ with LineString coordinates closer than interpolation distance

3.10.2 (2024-04-29)

  • Improvement: BER uses SNR colours for consistency with API.
  • Improvement: Interference tool hides colour key.
  • Improvement: Restyled MANET links.
  • Fix: Best site analysis output sometimes does not get cleared when “Clear all layers” is clicked.
  • Fix: Marker labels are hidden when the view is tilted and hard to read.

3.10.1 (2024-04-10)

  • Improvement: CSV processing supports GPU
  • Fix: CSV processing power_dBm label should be power_W

SOOTHSAYER change log

  • Feature: Add update.sh script.
  • Feature: Added Rocky OS support for scripts.
  • Feature: Improved OS detection for scripts.
  • Feature: Removes satellite container as all is now supported within the API and database.
  • Feature: Set DNS entry of containers which call out to the Internet to install packages. This can be overridden
    with the DNS_RESOLVER value in .env. Default of Google 8.8.8.8 is used.
  • Feature: Documented all of the .env values in the README.
  • Feature: 4326 map tiles are now cached rather than reprojecting 3857.
  • Update: API 3.15.4, UI 3.13.4, GPU 1.10.1.
  • Update: Changed 3D container to use docker.io/debian:12.5 rather than docker.io/nvidia/cuda:12.
    2.2-runtime-ubuntu22.04 to support CentOS Podman and reduce number of dependencies.
  • Update: pgAdmin container updated to 8.12.
  • Fix: Some containers missing network assignment which was create secondary Docker network.
  • Fix: Remove hello-world from list of stopped container after testing with docker run hello-world.
  • Fix: Increase timeout of frontend Nginx container to 10 minutes to match that of the API Nginx container.
  • Fix: Display administration dashboard login during setup.sh script.
  • Fix: Set port number for SERVICE_API environment value when using Podman during setup.sh script.
  • Fix: Commenting out values in .env with # can cause errors with some parts of the system.
  • Fix: Some endpoints fail to load with or without trailing slash.
  • Fix: Non-standard HTTP/HTTPS ports redirecting incorrectly on some endpoints.
  • Fix: OSM.xml updated to use the FQDN of the server.
  • Fix: Disabled execution of setup.sh after first time execution.
  • Fix: Nvidia containers respinning at fast rate if there are issues with nvidia-smi
  • Fix: TAK Server Chatbot directory permissions
  • Fix: ca.crt download location updated.

More information

For more information on SOOTHSAYER, our self hosted server, see here.

Posted on by

SOOTHSAYER server performance testing

Brown bears

We have lab tested three different size hardware profiles for running our SOOTHSAYERTM RF planning server to find out how they compare under load. These consumer profiles cater for different setups ranging from an enterprise with rack mounted servers, to a small office to a vehicle.

Enterprise server

This server is a standard Dell Poweredge R740 with an Intel Silver CPU and a 24GB Nvidia A5000 GPU running a Proxmox hypervisor. SOOTHSAYER 1.7 is installed as a virtual machine and LiDAR data is mapped via a network share.

Datasheet.

Mini desktop PC

This small form factor desktop PC is a HP z2 G9 mini with an Intel i9 CPU and a Nvidia A2000 GPU running Ubuntu 22 server.

SOOTHSAYER 1.7 is installed as a docker container and LiDAR data is local.

Datasheet.

Embedded PC

This portable server is an Nvidia Jetson AGX Orin with an ARMv8 64 bit CPU and a 2048 core GPU. The server has 3 variable power settings and was run in the modal 30W mode.

SOOTHSAYER 1.7 is installed as a docker container and LiDAR data is local.

Datasheet.

The tests

The tests used were designed to benchmark both the hardware and our software’s capability for high performance RF planning. We’ve picked challenges other tools would struggle to compete with, like Bullington diffraction and double digit megapixel resolutions.

High resolution area

The test parameters here were for a 5m resolution coverage heatmap out to 10km radius for a total image size of 16 mega pixels. This was repeated with and without Bullington diffraction and soft clutter data which is computationally expensive to show diffraction with soft clutter versus basic line-of-sight (LOS) speed respectively.

This would exercise our Area API.

Long range path profile

The test parameters here were for a 2m resolution LiDAR path out to a point 10km away. This would test 5000 points and would exercise our Path API.

Ten links at once

The test parameters here were for 10 random transmitters at 3km from a receiver using 2m resolution and Bullington diffraction. All links would be tested in a single API call to our Points API.

The results

All times are in seconds and were taken from the API response, excluding network latency and presentation in an interface.

TestServerMini PCEmbedded PC
Area w/Diffraction (CPU)261338
Area w/LOS (CPU)1710.924
Area w/Diffraction (GPU)6.713.1116
Area w/LOS (GPU)2.53.929
Path0.140.050.08
Links0.090.050.08
Table of results

The times didn’t fail to disappoint and threw up more than a few surprises. Unsurprisingly, cores matter when processing coverage and the fastest compute went to the largest GPU on the server.

When processing links, the CPU is critical and here the Intel i9 on the Mini PC excelled with a 50 millisecond compute time for multiple 2m LiDAR links. This faster-than-human reaction-time speed makes it suitable for dynamic planning with moving vehicles. The enterprise server disappointed with quick links due to the latency with the large data share where the LiDAR GeoTIFF tiles were stored. This latency was only noticeable with very quick calculations however.

The embedded PC performed admirably considering it was seriously under powered compared to the others at only 30W. It was able to model LiDAR links in 80ms and was only about 46% slower than an enterprise server at CPU calcs. Where it was noticeably slower was with processing the GPU area calculation. By increasing the power on the device to the 60W maximum the CUDA cores are doubled and from our testing we expect this would halve the GPU time.

Recommendations

For MANET link planning; an intel i9 CPU with an SSD is extremely fast

For high resolution area coverage; an enterprise grade GPU is unbeatable

For a small form factor host; the HP z2 G9 mini with an A2000 GPU is powerful

For value for money; the HP z2 G9 mini with an A2000 GPU is excellent

For low SWaP; the Nvidia AGX Orin 64GB delivers great economy

More information

For more information on self hosted RF planning, see our SOOTHSAYER page.

No load balancers with arrays of RTX gaming GPUs were used in this testing. We don’t need to do that!

Posted on by

SOOTHSAYER 1.7 released

SOOTHSAYER 1.7

Our latest major release of our private server, SOOTHSAYER, is ready. It includes six months of features, updates and bug fixes from CloudRF and features several customer sponsored capabilities including RADAR and Trilateration.

By popular demand, we now have a Docker enterprise solution so you can build your own containers or use our pre-built AWS template.

Docker support!

Thank you to all who gave feedback and feature sponsorship to help make this feature release. As you can see from the substantial new features and enhancements we continue to model the future of scalable APIs for multiple technologies and verticals such as Aviation and Counter UAS.

New in 1.7

RADAR model

The RADAR propagation model has a RADAR cross section parameter (m2) so you can model the effective range for detection of different sized objects with a RADAR, up to 90GHz, and 500km radius – horizon permitting!

It’s implemented in the API as model #8, both CPU and GPU engines and the user interface.

RADAR documentation: https://cloudrf.com/documentation/02_web_interface_intro.html#radar

Airport RADAR
Airport RADAR

Noise API

The noise API was developed from user feedback about the problem with varying local noise figures. Using a universal guessed value eg. -110dBm is not representative of the real world and especially the difference between a quiet rural and loud urban area for example. Now you can push in your own noise readings from radios or other sources either before or during planning. When modelling, live noise can be used by setting the noise value to ‘database’.

Noise CREATE API schema: https://cloudrf.com/documentation/developer/#/Manage/noiseCreate

Noise GET API schema: https://cloudrf.com/documentation/developer/#/Manage/noiseGet

Trilateration API

The Trilateration API was developed by popular request to accelerate and enable the process of geo-location of an unknown emitter. It will challenge conventional thinking about the accuracy of power based geo techniques by using accurate modelling and clutter data instead of circles. Our modelling has been field tested to below 8dB RMSE.

It requires receivers to be pre-modelled to enable rapid RSSI lookups using live receiver measurements. Using this two step process, results are delivered in milliseconds unless a receiver is moving in which case it’s coverage can be maintained using the fast GPU engine.

Trilateration API demo: https://cloud-rf.github.io/CloudRF-API-clients/slippy-maps/leaflet-trilateration.html

Height AMSL

Since our inception we’ve used height above the ground as most of our users are land based terrestrial planners. As we’ve gained more aviation, and RADAR, customers, barometric altitudes are now supported by request. The altitude type is specified in the request “output.units” key as before only now there are four possible inputs instead of two. Range is 1 to 120,000 m/f.

ValueDescription
mMeters above ground
m_amslMeters above sea level
fFeet above ground
f_amslFeet above sea level
API height measurement units

HF NVIS model

By request we’ve added a HF Near Vertical Incidence Skywave (NVIS) model. This models the first bounce from the ionosphere out to 500km and has an option for three layers (D, E, F) at differing refractive heights. This capability is supported in both our CPU and GPU engines and is particular valuable for teaching HF as it will give students an interactive HF tool to learn dipole patterns, the difference between day and night and critical frequency selection.

We have calibrated our NVIS model to align within 10dB of measurements taken from a 2012 research paper by Marcus Walden using a 5MHz NATO frequency in the UK. From this paper we selected one of the longer links at 210km where we used the median measurement value for August.

HF reliability animation
HF reliability animation

Bullington and Deygout diffraction models

Our single knife edge diffraction model has served us well for many years but cannot deliver the accuracy we aspire to once multiple obstacles are on the path. We have therefore invested substantial effort to add the much more complex Bullington and Deygout models to both our CPU and GPU engines. These greatly enhance simple propagation models as we proved during our LTE800 field test in the mountains earlier this year.

Deygout diffraction
Deygout diffraction


Automatic CSV processing in UI

From user feedback we created a solution to a problem whereby customers using managed IT systems were not able to install or execute our python API scripts but needed to batch process spreadsheets. We addressed this by adding a form within our web interface where CSV spreadsheets could be uploaded and automatically processed. It uses a much simpler format which combines with the current form settings like environment to execute API calls.

Documentation: https://cloudrf.com/documentation/05_web_interface_import_data.html#automatic-processing-process-a-spreadsheet

ITU-R P.1546 VHF/UHF model

This is a logically more advanced path loss model compared with legacy curves which is designed for terrestrial VHF and UHF planning. It’s conservative so we recommend the optimistic context with Bullington diffraction.

Multisite support for mixed AGL / AMSL units

After we implemented height above sea level for aircraft, we received feedback from customers using our multisite API that they would like to model transmitters above ground level and receivers above sea level. This is a common scenario for a ground-to-air network for example. We extended the multisite API to allow for mixed units so this can all be modelled in a single API call.

Testing

Our testing cycle is six months long, and starts with CloudRF where thousands of users, on every device imaginable, will test our API and interfaces to destruction. By opening it to the public via our free plan, we encourage many concurrent users, with diverse client software, to test our service and in doing so receive much more comprehensive testing than legacy products or GOTS software which only the contractor has tested.

Field testing is essential to validate the accuracy of our software and calibrate radio templates. After we implemented our new diffraction models, we took them to Scotland where we mapped out 22km of mountain LTE800 measurements. This valuable data improved the models and clutter profiles for UHF and validated our investment in improving accuracy.

Our API is regression tested daily and our models have a custom test harness to validate the many permutations of path loss models, environment contexts, diffraction models and parameters. As the number of models and inputs has grown we are relying on automation to ensure outputs are consistent and within parameters for the model(s).

Our user interface on CloudRF is instrumented with third party error handling software which automatically triages bugs for us. Through this we are able to identify issues early before customers are aware. This works especially well with our crowd sourcing strategy since we see a greater variety of clients than legacy or GOTS competitors who do not have the confidence to do genuine crowd sourced testing.

For hardware and hypervisor compatibility we have invested in a wide variety of systems and GPUs ranging from low end consumer GTX cards to enterprise grade devices like the A5000 and A100. We test SOOTHSAYER virtual machines on Proxmox 8 and ESXi 8 with different CPU architectures, network profiles and resource profiles.

Custom clutter

More information

Get in touch for a demo and pricing today at support@cloudrf.com

Posted on by

SOOTHSAYER 1.6 released

The latest release of our private server, SOOTHSAYERTM is now available with bigger limits, better reference data, VSphere 8 enterprise GPU support, enhanced security controls and improved performance.

SOOTHSAYER-1.6-flyer-Download

GPU clutter attenuation

The GPU engine can now model clutter attenuation in the same way as the CPU engine, only much faster. This functionality works for our global land cover for forests, global buildings data, and even custom clutter for BYO obstacles. Clutter profiles for regions and seasons are configured through the interface as normal.

Global buildings

We have replaced OpenStreetMap buildings with Microsoft buildings, generated from machine learning trained on satellite imagery. This provides much better data for planning in rural areas and countries which traditionally have suffered from poor data coverage. In the comparison below in Khartoum, Sudan, MS buildings produce a much more conservative result owing to their comprehensive coverage with OSM’s crowd-sourced data lacks.

A SOOTHSAYER server now has global buildings which can be side loaded by country into the data folder or fetched on-demand via api.cloudrf.com when connected to the internet.

Antennas and utilities

Previously SOOTHSAYER’s antenna database was sparse but now it ships with over 500 patterns from major OEMs as well as improved tools to validate, convert different text formats, import ADF patterns and search. Pattern images have been retired for dynamic plots direct-from-source so what you see is what you get and the custom plots are indistinguishable from OEM patterns in both the UI and outputs.

The new look database lets you find patterns quicker than ever with responsive filters. Type the first few letters of what you’re looking for to reduce search results.

The rating feature lets your team score patterns so “the good ones” are always top of the list.

Enhanced MANET limits

Limits for the Multisite API and it’s flasgship MANET tool have been enhanced. You can now do 500km radius / 310 Mile MANET sites and networks that are 2000km wide.

If you try to create a network across the Atlantic Ocean this won’t work….yet.

VSphere 8 support

SOOTHSAYER has been successfully tested on enterprise servers (HPE) with VSphere 8 and a Nvidia T4 GPU. VSphere 8 uses UEFI firmware (instead of BIOS) by default which is becoming the norm as enterprise moves towards trusted platform architectures.

It has also been tested with Proxmox, Virtualbox and Azure environments. We recommend Proxmox for SOOTHSAYER for the widest GPU support (RTX etc) and VSphere for enterprise GPUs.

You can install the VM on a GPU laptop with Linux but this is only recommended for people who like a challenge. We have a tutorial video and instructions available for this.

ATAK plugin

It comes with a free TPC signed plugin for ATAK 4.8: https://github.com/Cloud-RF/SOOTHSAYER-ATAK-plugin

Simple data storage with a second drive

You don’t need to mount an SMB share for your data anymore! (You can if you want to though…)

Just attach a second disk to the VM and reboot. You decide the size. Your terrain data and user files will be saved to this disk. The database remains encrypted inside the first disk.

If you don’t then you can still use it but will get warnings telling you about disk limits it in the CLI:

Change log

1.6.0

API v3.8.10

3.8.10 (2023-08-16)

  • Improvement: New default dBm and dB colour keys
  • Improvement: Enhanced admin dashboard for server
  • Improvement: Auto-rotate function for misaligned patterns refined
  • Fix: A private server will fetch MS buildings from CloudRF before OSM
  • Fix: Path and route ATAK data packages not working in ATAK.
  • Fix: Downloading .ant returns NAN and INF.
  • Fix: GPU elevation pattern downtilt aligns with CPU

3.8.9 (2023-08-05)

  • Improvement: More verbose errors for engine/data failure
  • Improvement: Handling and logging corrupt DTM files

3.8.8 (2023-07-31)

  • Improvement: MANET KMZ now returns points and heatmap
  • Fix: Refactored antenna pattern parsing
  • Fix: Fixed bug in antenna auto-alignment
  • Fix: Sorting MANET files to use latest when requesting KMZ

3.8.7 (2023-07-25)

  • Improvement: MANET networks in the archive show average value for each of the columns.
  • Improvement: Added support for MS 2m buildings via buildings option
  • Improvement: Using buildings will now only go to OSM as a last resort
  • Improvement: System clutter profiles adjusted to make vegetation more dense
  • Fix: Remote tile download not working with DTM tiles.

3.8.6 (2023-07-21)

  • Improvement: Custom antenna optimisations
  • Feature: Redesign of antennas database to make use of ADF files.
  • Fix: Handling legacy .ANT patterns not referenced to 0dB PEP

3.8.5 (2023-07-05)

  • Improvement: GPU can now do 500km radius with a minimum 180m resolution
  • Improvement: Multisite limit increased to 2000km for wide networks
  • Improvement: Logarithmic custom antenna pattern radiation

3.8.4 (2023-06-23)

  • Improvement: GPU calcs can now use custom clutter
  • Fix: Storing antenna gain before feeder loss adjustment
  • Fix: area and coverage values returned in KMZ balloon not being calculated.
  • Fix: When using PPA tool network name is appended with _PPA to separate from other calculations in the archive.
  • Fix: Interference tool not working with GPU calculations.

3.8.3 (2023-04-24)

  • Feature: Added session endpoint to get current load state of API in the response of a PNG image.
  • Feature: Area calculation site and network parameters now accept Latin letters which are not in the ISO basic Latin alphabet.
  • Feature: Added remaining_calculations_since_last_purchase to my-metrics API.
  • Feature: Added edges argument to a BSA request to allow an area to be defined for a BSA area which is cut out during processing.
  • Feature: Requesting tiff file type on an area request returns both 3857 and 4326 projections.
  • Improvement: Requesting tiff file type via the archive is now compressed to reduce size and transfer time.
  • Improvement: Additional error reporting and logging.
  • Fix: Antenna search form now shows all patterns for a given OEM when name is selected even if search string not present in description/filename
  • Fix: site is not a required field.
  • Fix: Google Earth throws an error upon first loading a coverage layer in relation to the PPA tool. This was causing errors as the PPA tool is called immediately after the loading of the coverage layer of which both the transmitter and receiver locations were the same.
  • Fix: Some temporary files not being removed during cleanup process.
  • Fix: User clutter files being copied to wrong location when using GPU processing engine.
  • Fix: Google Earth throws an error upon first loading a coverage layer in relation to the PPA tool. This was causing errors as the PPA tool is called immediately after the loading of the coverage layer of which both the transmitter and receiver locations were the same.
  • Fix: User clutter files being copied to wrong location when using GPU processing engine.
  • Fix: Submitting receiver.lat or receiver.lon to area endpoint causes calculation failures.
  • Fix: Timings no longer output for error messages on production.
  • Fix: credits returned from preferences not factoring in requested calculations.
  • Fix: Loading KMZ into Google Earth would immediately fire off a PPA calculation.
  • Fix: Calculations made in Google Earth would always have TEST network name.

3.8.2 (2023-03-28)

  • Improvement: Polarity for custom patterns
  • Fix: Rx gain NULL after using interference tool
  • Fix: All KMZs contained the GE ppa tool
  • Fix: Removing gain from ADF patterns so all are referenced to 0dB PEP
  • Fix: Using Signal to Noise units with a receiver sensitivity of 0 dB caused an issue
  • Fix: Google Earth PPA tool could fail when both markers were stacked up. Now fails gracefully.
  • Fix: Antenna model and OEM search filters was using OR vs AND

3.8.1 (2023-03-02)

  • Improvement: GPU engine has improved Huygens diffraction
  • Improvement: Colour steps can now be any value between -30 and +30
  • Improvement: New buildings are now fetched synchonously so will block for request #1 up to a max of 30s
  • Improvement: Extended bandwidth lower limit from 10kHz (0.01MHz) to 1kHz (0.001MHz).
  • Improvement: Extended noise floor lower limit from -140 to -150 dBm.
  • Improvement: Requesting a network which doesn’t exist in a mesh calculation will indicate up to 3 network names which have been used recently.
  • Improvement: Chatbot authentication process simplified. The user ID will always be the ID of the Chatbot.
  • Improvement: Extended bandwidth lower limit from 10kHz (0.01MHz) to 1kHz (0.001MHz).
  • Improvement: Extended noise floor lower limit from -140 to -150 dBm.
  • Commercial: Ibiza is no longer the anonymous-full-spectrum-party-island due to abuse. A plan is required to test.
  • Fix: Tightened restrictions on users on free plan.
  • Fix: Interference tool failing to produce result.
  • Fix: Bug in SLEIPNIR JSON report for very low power levels eg. 0dBm with feeder loss. Fixed as 1.6.2
  • Fix: Points megapixel calculation ignores radius in favour of point distance. Hard limit of 64MP
  • Fix: Confusing error messages when running a mesh calculation.
  • Fix: Chatbot outputs any returned error messages when a multisite API request fails.
  • Fix: Tightened restrictions on users on free plan.
  • Fix: Interference tool failing to produce result.
  • Fix: Show the callsign of the device which is authenticated with Chatbot as this is more useful than a device ID.
  • Fix: Requesting 3857 projections for GPU calculations (viewshed, BSA, multsite) returning HTTP 404.
  • Fix: Selecting a bad antenna ID would throw a HTTP 500 with no output error message.
  • Fix: PPA URL outputs will return a secure encrypted URL so as to now expose information via the public URL.
  • Fix: Multisite endpoint reporting validation errors for some imperial units.
  • Fix: Multisite endpoint not honouring imperial units.

UI v3.8.8

3.8.8 (2023-08-15)

  • Update: Satellite database updated to 01/08/2023.
  • Fix: Calculate ERP after switching antenna pattern
  • Fix: BSA colour key no longer disabled when BSA fails

3.8.7 (2023-07-31)

  • Feature: Added a download KMZ button to the MANET tool.
  • Fix: Skips polygons and polylines with no coordinates when uploading KML as clutter.
  • Fix: Bit error rate and modulation drop downs missing in Google Earth.
  • Fix: GPU engine enabled no longer hides the play button to allow calculations to be made without changing settings.
  • Fix: Gracefully fail with a message when unable to load path profile results from the archive.
  • Fix: Gracefully fail with a message when an image could be loaded properly for some scenarios.
  • Fix: Multi-point results hover in the sky when turning off 3D terrain.

3.8.6 (2023-07-13)

  • Fix: Buildings on the Path Profile Chart were all being bunched up on the left when land cover was turned OFF.
  • Feature: “My patterns” antennas makes use of recent updates to API endpoints and draws pattern plots rather than outputting JPEG images.
  • Fix: Antenna menu labels tidied and widths standardised. Horizontal is now H, Vertical is now V.

3.8.5 (2023-06-23)

  • Fix: Handling blocked sentry script with link to adblocker FAQ
  • Fix: Metric/Imperial logic in Multisite tool
  • Fix: MANET links not showing if no preference set
  • Fix: ERP not recalculated when reloading from archive
  • Fix: Deleted archive images now throw a error

3.8.4 (2023-05-23)

  • Enhancement: Loading page with async loader for large JS assets
  • Enhancement: Added “Hide Node” functionality to the MANET context menu.
  • Enhancement: Sets the default antenna gain when selecting an antenna.

3.8.3 (2023-04-24)

  • Fix: Removed “Plane Earth Loss” from models list.
  • Fix: Using “Line of Sight” model automatically sets diffraction to “Off”.
  • Fix: Improved error handling and reporting.
  • Fix: Issue when changing the colour schema while best site analysis is processing a drawn polygon.
  • Feature: API load status light added in top navbar upon page load.
  • Fix: Marker pin on MANET nodes at top of plumbline rather than at ground level.

3.8.2 (2023-03-28)

  • Enhancement: Welcome screen while loading ui.
  • Enhancement: Added reference data import dialog for CSV,KML,GeoJSON.
  • Enhancement: Increased 1000 row limit for CSV to 60k (Make sure your device has lots of memory if you use this)
  • Enhancement: BSA KML polygons via reference data form
  • Enhancement: MANET KML locations via reference data form
  • Enhancement: KML reference points via reference data form
  • Fix: Issue when entering invalid DMS coordinates.

3.8.1 (2023-03-03)

  • Improvement: Upgraded CesiumJS to the latest version.
  • Fix: Mouse hover tooltip value lookup to use the correct schema for multiple layers with different schemas.
  • Fix: Error messages from API for MANET tool not being output.
  • Fix: MANET failed links are styled which is confusing, particular when zoomed out and dashed line is not visible.
  • Fix: Best server tool able to misfire on GPU area calculation upon click.
  • Fix: Allow for imperial units with MANET heatmap which adds locking to the dis value to prevent unit mismatch.

More information

To request more details see here, or to get a quote or a demo VM, get in touch from your official email.

Posted on by

System updates – 3.6 (Nov 22)

Here’s a roundup of what’s changed since June. As always it’s been a busy period of development which we’ve matched with a focus on maintenance and bugfixes. Our backlog of feature and change requests stands at 28 for the UI and 23 for the API with all priority issues resolved.

The deck is now clear for a dynamic network planning feature which will be integrated in December.

UI changes, June to November

400km radius
Custom WMS
JSON templates
Custom WMTS
Satellite data

With 19 new features/improvements and 69 bugfixes in this period, the user interface has matured considerably due in large part to user feedback from the community. We’ve said no to a few feature requests also which is necessary to keep the interface technology agnostic and easy to use. That doesn’t mean we can’t support complexity or special waveforms directly in the API.

  • Radius increased to 400km for aircraft (3.0)
  • GPU engine integrated with best server and CSV analysis tools (3.0)
  • New 300m resolution for airborne broadcasting (3.0)
  • Received Signal Received Power (RSRP) option added to measured units (3.1)
  • Account information dialog (3.2)
  • GPU Best Site Analysis (3.2)
  • Custom WMS and WMTS map sources (3.4)
  • Satellite database updated (3.4.2)
  • Multi azimuth antenna patterns (3.5)
  • Route, multipoint and path path tools updated for all possible measurement units (3.5.1)
  • JSON templates system (3.6)

The long surviving Google Earth interface lives on and has been given a fresh wind thanks to code compatibility libraries which mean we no longer have to worry about breaking it with new Javascript. Long live Google Earth!

UI Changelog

## Version 3.6.1 (2022-11-28)

- Fix: Layer from archive would not load when clicking on the layer name or selecting "Add tower(s) only to map" button.
- Fix: Processing modal window getting stuck when using the Interference or Super Layer tools.
- Fix: Display the power of items loaded from the archive in the selcted power unit.
- Fix: Dynamically move navigator controls down to make space for imagery layer list.
- Fix: Prevent crash when disabling path profile analysis before calculation completes.
- Fix: Refreshes displayed clutter after deleting all clutter using the clutter manager.
- Fix: Selecting a template in the Google Earth interface crashes.
- Fix: MANET nodes sometimes clipped by the terrain resulting in partially hidden markers.
- Fix: Best site analysis tool sometimes goes into "zombie" state under bad network conditions.

## Version 3.6 (2022-11-10)

- Feature: Templates functionality makes use of the new API endpoints and JSON files.
- Feature: Added button to download the JSON template for both custom and system templates.
- Feature: Added modal to add/remove favourite system templates.
- Feature: Dropdown select of templates indicates if the template is a system or custom template by prepending with either "System" or "Custom" respectively.
- Fix: Styling issues on custom template save/delete modal.
- Fix: Reorganised "My Account" modal as was becoming cluttered. Logout button moved to top and managers split into a single row. API usage moved to the bottom.
- Fix: Some API errors which are returned as an array of values would not be parsed as an array, instead the error displayed would be a generic "Request failed." message which wasn't indicative of the error.
- Fix: Don't fire a GPU request when loading a GPU template.
- Fix: Adjusting the `rxs` field manually would not update the receiver sensitivity slider.

## Version 3.5.1 (2022-10-31)

- Feature: Route, Multipoint and Path tools able to display SNR, dBuV and RSRP units as well as dBm.

## Version 3.5.0 (2022-10-18)

- Feature: Multi-azimuth antenna support by sending a comma-separated list of values, for example, `0,120,240`.

## Version 3.4.2 (2022-09-21)

- Update: Satellite database updated to 2022-09-22.
- Feature: Fly to PPA when loading from the archive.
- Fix: Custom pattern vertical beamwidth not visible in the Google Earth interface.
- Fix: Profile drop down not populated and opening the clutter manager in the Google Earth interface.
- Fix: Sensitivity is loaded when selecting a template.
- Fix: Map attribution interfered with satellite timeline controls.
- Fix: Loading PPA from archive and then hovering over chart would cause a crash.

## Version 3.4.1 (2022-09-14)

- Fix: Template saves coax type.
- Fix: Problem toggling the azimuth mode in the Google Earth interface.
- Fix: Problem loading a template in the Google Earth interface.
- Fix: Switching between templates with engine set to GPU and CPU runs area calculation with incorrect engine.
- Fix: Saving a custom map URL cuts off part of URL after an "&".
- Fix: Suspends error logging when library blocked by client.
- Fix: Handle missing preferences.

## Version 3.4.0 (2022-08-31)

- Feature: Custom WMS and WMTS URLs for imagery layers.
- Feature: Attribution text lists API providers.
- Fix: Pruned dead and unsuitable map sources, such as "Blue Marble".
- Fix: Problems with editing coordinates.
- Fix: Antenna patterns not loading properly in some browsers.

## Version 3.3.0 (2022-08-23)

- Feature: Sets free space model with diffraction off and shows warning for best site analysis.
- Feature: Select last network when opening archive to match the last row.
- Feature: Make clutter go to ground.
- Fix: Improved area calculation error handling.
- Fix: Selecting a network in archive drop down when some items have no network would cause a crash.
- Fix: Legacy colour key images not displayed.
- Fix: Super layer KMZ quick-download button fixed for "Merge network"

## Version 3.2.0 (2022-08-15)

- Feature: Account Information dialog.
- Feature: GPU Best Site Analysis.
- Fix: Updates to colour schemas to match recent changes to colour schema API.
- Fix: Buildings layer cannot be removed.
- Fix: Receiver sensitivity icons missing.
- Fix: Creating a clutter profile does not refresh the list.
- Fix: RSRP breaks MANET.
- Fix: Opacity slider doesn't work.
- Fix: MANET reload by name click and best server for imperial units mi/f.
- Fix: Crash clicking layer checkbox while layer is loading.
- Fix: Default RSRP schema not always selected.
- Fix: Interference tool is trying to map colours to colour key.
- Fix: Crash saving clutter polylines with less than 2 coordinates or polygons with less than 3 coordinates imported from kml.
- Fix: Crash downloading MANET metrics report when node evauation failed.
- Fix: Power unit in MANET node details dialog displayed as undefined.
- Fix: Node percentanges on MANET metrics report sometimes NAN.
- Fix: Points CSV validation no longer works.
- Fix: Entity already exist in collection when disabling MANET.
- Fix: Crash adjusting bsa sliders when result loading failed.
- Fix: BSA images not recoloured on colour schema changed.
- Fix: Custom antenna polar plot not displayed in Google Earth.

## Version 3.1.0 (2022-07-18)

- Feature: Added Received Signal Received Power (RSRP) as an option to the "Measured Units" dropdown.
- Fix: Deleting multiple selected networks will now give the name of the networks rather than the ID.
- Fix: Selecting a PSK/QAM modulation will only show supported PSK/QAM bit-error-rate values, and likewise when selecting a LoRa modulation.
- Fix: Deleting a network from the archive now gives a better modal message rather than a browser `alert()` window.
- Fix: Occasional crash when computing azimuth and tilt for the mouse tootltip for satellite mode when the position has not been computed yet.
- Fix: Can't find variable `Map` in Google Earth interface.
- Fix: Crash in Google Earth when logging error information.
- Fix: Prevents crash in Google Earth interface when changing clutter profiles.
- Fix: Formatting issue in Google Earth.
- Fix: Helper dialog describing channel noise adjustment when changing bandwidth
- Fix: Tx Height AGL and RF Power textboxes extending over unit drop downs.
- Fix: Crash in Google Earth interface when you change properties that would update the map location when in web UI.

## Version 3.0.1 (2022-07-06)

- Feature: Added support for 180m and 300m resolutions.
- Fix: GPU is populated in dropdown of processing engines when GPU isn't available.

## Version 3.0 (2022-07-04)

- Feature: Radius limit now 400km.
- Feature: GPU processing no longer handled through WebSocket, instead goes through the same API as CPU processing for simplicity. Removed button to close GPU processing engine, instead you now just switch back over to CPU mode from the dropdown in the "Output" menu.
- Feature: Integration of GPU engine with other tools including best server and CSV import.
- Fix: Colour schema list gets populated with some empty values when using GPU mode.
- Fix: Prevents failure on area calculation due to invalid input from logging an error.
- Fix: Interference dialog not always populating networks.
- Fix: Duplicate entry for gpu in the engine drop down.
- Fix: Some antenna patterns were not being saved in templates.

API changes, v3.0 June to 3.6 November

With 30 improvements and features and 28 bugfixes, the API has largely remained the same from a developer’s perspective but it now has more consistency between the CPU and GPU engines, more choice with defining antenna azimuths and more useful error messages to help developer’s make better choices and save time.

Under the hood, the ATAK interface has had a major refactor to support customer’s official TAK Server’s with our infamous chatbot. A new chatbot management interface has been especially created which can validate certificate chains using OpenSSL to warn of issue before you get to them eg. “This certificate is invalid” or “This key is incorrect” or “Cannot ping your TAK server”.

Offline JSON file templates
Multi-Azimuth requests
Best Site Analysis for ATAK

GPU integration

The new GPU engine has received several updates throughout this period to add models, gains, antenna patterns, multi-azimuth patterns and output styling inline with the CPU engine, SLEIPNIR.

In September, the powerful Best Site Analysis capability was integrated with ATAK. When the SOOTHSAYER bot is connected to the TAK server, any polygon sent to it will generate a BSA layer.

Due to demand, the maximum radius for the GPU engine was increased to 150km to support broadcasting. During GPU testing we were generating 100 mega pixel (100,000,000 points) heatmaps in a few seconds which is too large for a client. We are working on a scalable solution to visualise images of this resolution as an image pyramid.

SLEIPNIR mods

The SLEIPNIR CPU engine is now at version 1.6.1 and has conditional smoothing of coarse DTM, knife edge diffraction from clutter (note we already do this for surface models) and slightly (3dB) more optimistic diffraction following feedback from a mountain rescue team.

Received Signal Received Power (RSRP) is not native to the engine instead of the API and the reliability variable can now be used for non-ITM models to provide a 10dB tuning margin where 50% is 0dB and 99% is 9.9dB path loss.

The maximum radius has been increased to 400km for airborne platforms. To support the larger radius, we’re also offering a new low resolution of 300m.

API management

The V1 “URL” API will be retired at the end of this year. Users who have integrated with this API should migrate to the new JSON API as soon as possible. The new JSON API has been powering the web interface for almost two years.

Reference: https://cloudrf.com/documentation/developer/

API Changelog

## Version 3.6.0 (2022-11-25)

- Deprecation: Added deprecation message to `v1` API endpoints. The `area` and `path` will be removed on the 31st December 2022. Please migrate any scripts to the new JSON API and see https://cloudrf.com/documentation/developer/
- Fix: Chatbot welcome-storm with other chatbots.
- Fix: Obsolete paths on terrain coverage map and HTML PPA response.
- Fix: Improvements to automation testing.
- Fix: Sending malformed JSON to a `v2` API endpoint now returns a useful error message.
- Fix: Deprecated `v1` API endpoints were incorrectly counting API credits.
- Fix: Power value of 1mW for GPU fails.

## Version 3.5.0 (2022-11-10)

- Feature: Moved from database-driven to JSON-driven templates with new API endpoints to manage custom templates.
- Feature: System templates can be retrieved and saved as a favourite using new endpoints.
- Fix: Race condition on GPU processing files as its so damned fast.
- Fix: Public shares for GPU layers needed reprojecting to EPSG 3857.

## Version 3.4.0 (2022-10-18)

- Feature: Gain support for GPU engine.
- Feature: Antenna support with azimuth and tilt for GPU engine.
- Feature: Multi-azimuth support for antennas by sending a comma-separated list of values in the `antenna.azi` value, for example, `0,120,240`.
- Fix: Path tool on TAK successfully returning only about 50% of the time.
- Fix: Area requests to `v1` API returning colour key data in improperly formed string values.
- Fix: `clutter` command on TAK chatbot working again.
- Fix: Resolved issue with custom colour schemas and low resolution screens.
- Fix: Sanity check ADF files.
- Mod: Default clutter templates adjusted. Urban height reduced to no more than 2m.
- Mod: GPU engine max radius capped at 10km (CPU max = 400km).

## Version 3.3.0 (2022-09-22)

- Feature: Support for TAK server 4.7 through Chatbot and ATAK.
- Feature: Extended response information when executing `id` to Chatbot through ATAK.
- Feature: Support for GPU BSA on Chatbot with the multipoint tool through ATAK.
- Feature: Added new chatbot manager to spin up chatbots on demand which interact between TAK servers and API.
- Feature: Extended preferences functionality to allow for custom imagery.
- Feature: Delete all clutter button added to clutter form.
- Fix: Processing engine saved in templates.
- Fix: Commercial restrictions messages now return HTTP 403 (Forbidden) rather than HTTP 400 (Bad Request).

## Version 3.2.0 (2022-08-15)

- Feature: System colour schemas no longer able to be deleted by a user so that users will always have a list of schemas to choose from.
- Feature: Height ceiling increased to 120km over 60km.
- Feature: When using Greyscale GIS colour schema no longer returns validation error for mixing outputs and schema.
- Feature: Improvements to the PPA KMZ export with fresnel support and new icons for transmitter, receiver and obstructions.
- Fix: Improvements to the functionality of the colour schema generator.
- Fix: Creating a BER colour schema fails with a HTTP 500.
- Fix: Update to `preferences` stripping out underscores from colour schemas.
- Fix: Google Earth layer no longer allows to manage colour schemas. Added notice with workaround instructions.

## Version 3.1.0 (2022-07-18)

- Feature: Improvements to `preferences` to extend to additional values and give information about a users plan.
- Feature: Improvements to precendence of preferences, where in some cases a previous request values will be taken over a users preferences.
- Feature: Colour palette bucket limit increased to 75.
- Feature: Return request to `preferences` as a correct JSON response.
- Feature: Users with a private server will now be indicated when they make a request to `preferences`.
- Feature: Colour pallette creator extended to support Bit-Error-Rate measured units.
- Feature: Bit-Error-Rate modulation extended from 10e-6 to 10e-8.
- Feature: Allowed support for Received Signal Received Power (RSRP) measured units with an `out` value of `6`.
- Feature: Added default RSRP colour key.
- Feature: SLEIPNIR 1.6.1 calculates and returns (LTE) RSRP if selected
- Feature: SLEIPNIR 1.6.1 uses reliability for non-ITM models with a 10dB range. 50% = +0dB, 99% = +9.9dB
- Fix: Successful deletion of networks in the archive are returned with a success `message` rather than `error`.
- Fix: Correct names of plans when making a request to `preferences`.
- Fix: BER colour palette in JSON response was missing unit.
- Fix: Improvements to the default `PATHLOSS.dB` colour key.
- Fix: Hata model high-plateau issue resolved in SLEIPNIR 1.6.1 

## Version 3.0.1 (2022-07-06)

- Feature: Resolution supported up to 300m.

## Version 3.0 (2022-07-04)

- Feature: Improvements to GPU processing with the use of remote processing nodes.
- Feature: GPU produced images are now styled in the API based on the specified colour schema.
- Feature: Extended `data` API to retrieve a tile based only on tile name.
- Fix: Logic issue with users mixing credits and plan was incorrectly refusing service to users with expired plans, but spare credits.
- Fix: Custom clutter profiles are created with the list of 101 to 199 clutter codes.
- Fix: Obtaining height data from OpenStreetMaps would sometimes fail.
- Fix: KMZ download failing on equator

That’s all folks.

Posted on by

Bandwidth and noise

As communications systems have grown in capacity, they have expanded in physical bandwidth in an increasingly congested RF spectrum.

Effective digital communications planning requires more than knowledge of antennas and propagation fundamentals. It now needs an intimate understanding of bandwidth and noise to co-exist and communicate efficiently.

Unfortunately, this key aspect of modern RF systems is often taught badly, or in some cases not at all, leading to often unwelcome surprises for equipment operators in the field. As a technology and market agnostic platform we’ve observed poor bandwidth knowledge in many markets, but notably MANET and 5G, both of which are accelerating the deployment of wideband systems, often with little to no planning beyond a topographic map study.

Both radio markets are evolving from narrower channel technologies, which in the case of MANET and the VHF Combat-Net-Radio (CNR) it replaces was measured in KHz so they need to update their theory training content and associated software to convey these potentially complex topics to novice students in a digestible manner.

Increasing bandwidth increases noise, which reduces coverage

Teaching noise

As bandwidth increases, so does channel noise. This simple concept might seem easy to remember for a student but without visual aides, and since the demise of analog systems; audible aides, it is hard to demonstrate in practice.

A good teacher may show visual aides like noise charts, FFTs, spectograms and a bad one may show some Johnson-Nyquist formulas buried within an all-day powerpoint which is not helpful except for getting paid.

FFT showing a narrow signal and wideband interference

A student can tick the right box on their exam(s) but spend their career wasting bandwidth and struggling to establish communications because they believe that big is better – it isn’t, or worse still, that bandwidth has no effect on the coverage since that’s a function of transmitted power and/or height. Having an intimate understanding of the interplay between bandwidth, receiver sensitivity and thermal noise will make spectrum users more efficient, effective and considerate.

Bandwidth MHzThermal noise (dBm)
0.1-124
1-114
2-111
4-108
8-105
16-102
32-99
64-96
Bandwidth thermal noise table based on a temperature of 21C

Which waveform is best?

Comparing digital radios is complicated due to the myriad of features, waveforms and software.

Given a particular waveform it will have characteristics such as a minimum Signal-to-Noise Ratio (SNR) value which it requires to achieve a symbol rate necessary to deliver a fast data link for example. This dB value must sit proud of the noise floor so if the noise floor is high at -90dBm, coverage will be reduced and conversely, by taking it to somewhere quiet eg. -110dBm, the coverage will improve by 20 decibels – a huge difference.

To compare waveforms precisely, the same noise floor should be simulated, initially, with fixed values to eliminate random error in field testing. The sensitivity values will be somewhere between 3 and 20dB depending on what the waveform and target Bit-Error-Rate (BER) is.

Bit Error Rate (BER) describes the ratio of errors in a data stream. An ok value is 10-3 or 1 bad bit in a 1000 or 0.1%. This increases with noise until a signal is unrecoverable. For more on BER see an older blog here.

For ground radios designed for noisy environments, a BER of 10-2 (1 error in 100 or 1%) is used here for extracting our planning thresholds from a chart of SNR curves. For an airborne system without obstacles this could be higher, for example 10-5.

Signal to Noise Ratio for different modulation schemas against error rates.

A narrow waveform eg. QPSK gives better coverage and works better in noisy conditions. This is the fallback telemetry mode used in many data radios.

A wide waveform eg. QAM64 is capable of better throughput and delivering high bandwidth streams such as HD video.

The best radio is one which can use different waveforms to satisfy both coverage and capacity.

Modelling bandwidth: A tutorial

Modelling RF Bandwidth and noise

Quick reference guide

A quick reference guide for using bandwidth and noise is available here. For other guides see here.

221025_SNR_quick-guideDownload

Conclusion

Bandwidth and noise is essential knowledge for anyone deploying wideband systems or comparing waveforms.

RF theory training can be enhanced (and needs to be) with visual tooling to let students quickly observe the impact of different inputs in a controlled environment with templates to minimise user error.

For information on how SOOTHSAYER can help with signals training see here.

Posted on by

Dynamic network planning with hardware APIs

Dynamic MANET network planning

Location aware radios

Modern digital radio systems often have Application Programming Interfaces (APIs) for remote management. They also commonly have Global Navigation Satellite System (GNSS) modules for location awareness which enables “network maps” of where the nodes are. These vendor maps are great at showing where nodes are now (provided they are in coverage!) but what they lack is the ability to plan where nodes could be moved to.

When you combine these key features with our open standards modelling API you get a powerful new capability which allows you to observe the network now, and by moving or adding a simulated node, see the network in the future.

Case study – Trellisware

TW-950 Shadow

TrelliswareTM are a market leader in Mobile Ad-hoc Networks (MANET) whose versatile radios work where others fail due to their spectrum efficient TSM waveform. It’s currently in service with Government, Commercial and First responder markets.

The radios have a comprehensive API and integrated software services which allows for remote operation on a IP based network.

We were loaned some TW-950 Shadow radios to explore API integration. In a few days we were able to create an API client (in Python) to interface between the Trellisware API and our RF modelling API on SOOTHSAYER 1.3 which has a MANET planning tool.

Our Python client would interface with the radio network via the Ethernet connector on the side of a TW-950 radio from where the API would expose node metadata.

We would fetch this metadata (as JSON) and package it into a JSON document compatible with our Points API, which powers our MANET and route analysis tools. In the interface’s MANET tool, a new ‘play’ button pulls in the network document and models it through our fast API. Each link is tested using the real parameters, with minimal user interaction.

The flowchart is depicted below.

Ethernet adapter for connecting the hardware to SOOTHSAYER

Once the data is displayed in the SOOTHSAYER interface, an operator can choose to move a node by dragging it or add an entirely new node, using settings of their choice, into the mix. This new node will be modelled alongside the rest of the nodes (many-to-many) to visualise what the impact of the new node will be.

Trellisware (Blue) and CloudRF (Yellow) APIs and components

A faster OODA loop

The Observe, Orient, Decide, Act (OODA) planning loop is an established concept which determines success in fast moving communications environments such as a fire or Police incident.

By combining real-time situational awareness with live modelling we can exercise scenarios and reach sound decisions much faster than by either guessing or using trial and error as is often the case in tactical communications in complex environments. The benefit is faster more efficient use of limited resources.

A Trellisware network map enhanced with two planned nodes to the right.

Demo: Do we launch the drone?

Deploying a drone equipped with a radio is a guaranteed way to fix MANET network issues, especially in cluttered urban environments. It’s resource intensive and risky as the drone is valuable, the radio, and network it enables, is arguably more valuable plus the battery life is very limited so this asset must be used sparingly, which is where planning comes in.

Placing a repeater drone “overhead” is not needed, unless you’re using it for observation also. If the drone has a vertical dipole, overhead would actually be the worst place for it due to the donut shaped antenna pattern as some early OEMs learnt the hard way.

A better place for a communications relay drone is at altitude but on the edge of the target network where it will be at less risk (and will present less risk to net members below it in the event of failure) but it will still be able to serve as an effective repeater.

As recent MANET demos have proven, this repeater could be effective from several miles away.

Demo of dynamic MANET planning in SOOTHSAYER with Trellisware radios

Look ahead

Now that we have a simple standards based, repeatable, method for adding live radios we’ll roll this into future SOOTHSAYER releases, the next of which is 1.4, scheduled for Q3 2022.

Customers can make their own plugins in any language to get their hardware data packaged as a JSON document for either SOOTHSAYER’s MANET tool or direct to the points API which powers it. You can find example scripts in various languages on our Github site and we are available to assist with bespoke integrations.

References

API examples: https://github.com/Cloud-RF/CloudRF-API-clients/tree/master/APIv2

API specification: https://cloudrf.com/documentation/developer/swagger-ui/

Contact

If you are a hardware manufacturer or integrator looking to add this type of capability into either your own map or ours, get in touch as we know what we’re doing and it will save you years of R&D.

Email support@cloudrf.com for more information on how you can achieve dynamic planning quicker.

Posted on by

Connecting smart cows to moove data

Smart cow

Smart farming

Smart farming is using Internet of Things (IoT) technologies in agriculture to enable efficient use of resources.

For this blog we’re focused on cattle farming on large, fence-less farms in New Zealand. The farms in question are vast and remote so connectivity options are limited. This is why an off-grid sub-GHz LPWAN network is ideal due to its long range and the requirement to only send small, infrequent, packets of data.

For the solution to be cost-effective compared with Satellite, as little infrastructure as possible is needed which in this case is a LPWAN gateway on a pole, some collars for the herd and an app to manage the system via a web service.

Siting the LPWAN gateway(s) properly is critical to achieving not only coverage across the farm(s) but to reduce the number of gateways, which reduces complexity and cost.

Sub GHz LPWAN on the farm

An 868MHz LPWAN signal can go for many miles under the right conditions. We know this well from powering the Helium LPWAN network’s planning tool, Helium Vision, where people can communicate data 50 miles with a fraction of a watt of RF power and an omni directional antenna.

Despite it’s useful diffraction properties which enables it to work non-line-of-sight (NLOS), it’s still sensitive to obstructions so clutter on the farm such as buildings and trees needs modelling accurately. CloudRF has 10m Landcover for New Zealand from the European Space Agency and 10m DSM from the LINZ Geospatial agency.

These data sets are adequate for most outdoor scenarios but are not fine enough to model a farm complex of buildings, such as tall grain silos, metal sheds and seasonal obstacles. For high resolution you could source your own surface model, as our customer Halter did…

Farm buildings and silos

Use case: Halter

Halter are a novel agri-tech startup focused on cattle management with a unique solar powered collar.

They needed accessible RF planning software to help their engineers site LPWAN gateways. Having used and liked Cloud-RF, they needed higher resolution surface models of the farms, and no pesky API restrictions!

They also planned to build their own tools on top of our powerful physics based API which is smart as it allows their R&D team to focus on their primary product, and not waste time reinventing the wheel.

Their options were either buy expensive commercial data or self generate data using a drone and photogrammetry software such as Pix4D. Given the prohibitive cost of high resolution commercial LiDAR, it would only take a few jobs to make a return on the purchase of a decent drone!

https://halterhq.com/

Halter purchased a private Keyhole Radio server from us which included the API they needed. The server runs as virtual machine and crucially, lets them import their own terrain data.

They were quickly able to import high resolution, organic data into their server as GeoTIFF files. This allowed them to work with data which was very current, even hours old, so would be an accurate model of tree heights and man made obstructions.

The terrain format accepted by Keyhole Radio and SOOTHSAYER is GeoTIFF, Int16 resolution and WGS84 (EPSG:4326) projection.

LPWAN coverage on a farm in New Zealand

1m resolution

It wasn’t all plain sailing though, they found that there was a limit to the physical tile sizes our server could use caused by memory. The solution was to reprocess the large tile into smaller tiles to make it digestible.

A 5000 x 5000 GeoTIFF at Int16 resolution will require 50MB of disk space. If this is 5m LiDAR, the physical width is 25km x 25km. Our engine can super-sample, so if you used this tile, but requested 1m resolution, it would create a raster in memory measuring 25,000 x 25,000 pixels which would need 1.25GB of memory.

For 1m resolution however, tiles measuring 1000 x 1000px would only require 2MB of disk and memory. You may need to load in a few, lets say 16, to do your model but that’s still only 32MB.

You could also resolve this by increasing the memory available to the server but it’s recommended to prepare data into smaller parcels. We support 1m resolution in our API but don’t hold a lot of 1m data sets due to their substantial cost and size. If you already have 1m data, a Keyhole Radio or SOOTHSAYER server is the answer.

1m resolution

Summary

Cloud-RF’s powerful API is ideal for efficient smart farming.

Our private servers will let you take it to the next level with terrain data you can source yourself, no API restrictions and as a bonus, they work without an internet connection!

Finally, all our jokes are offal.