Routing Analysis¶
The best route from one source to the best of several targets, over time
Routing Analysis reads a completed Connection Network and, at every time step, picks the best route from a source to any of the selected targets, using the criterion you choose. It then stores which target, which relays and which routes were used most, and draws the best route in 3D, in its own colour and thickness, over the Connection Network.
It does not invent links: only edges declared in the Connection Network, with their cached access windows, are used. A CN sensor used as a laser or dump antenna maps to its parent satellite: the route is drawn bus to bus, and a hop is open when any of that bus’s ISL terminals sees the neighbour. Sensors as terminals.
One best route per time step; consecutive steps on the same route form a segment.
Add from Mission Tree → Analysis → Routing Analysis.
When to use¶
| Use Routing Analysis when… | Prefer something else when… |
|---|---|
| You want to see which path a file or command would take, and when it changes | Declaring which links exist → Connection network |
| You have several ground stations (or satellites) and want to know which one is reached best, and how often | One pair's visibility or RF budget → Link |
| You want to compare criteria (shortest path, fewest hops, fastest file, CGR) on the same mesh | Full traffic with data volumes, queues and onboard storage → Latency / ISL routing |
Criteria¶
| Criterion | What it minimises | Links used |
|---|---|---|
| Shortest path (light delay) | Sum of the hop ranges, so the propagation delay. Ties go to fewer hops. | Links open at that instant |
| Fewest hops | Number of hops. Ties go to the shorter path. | Links open at that instant |
| Fastest file delivery (no onboard wait) | Arrival time of a file of the given size: transmission time (size / data rate) + light delay + node processing, on every hop. A hop counts only if one access window holds the whole transmission. | Links open while the file crosses them; nothing waits onboard |
| Earliest delivery, store-and-forward (CGR) | Earliest delivery time of the file with Contact Graph Routing: a node may keep the file and send it on a later contact. Same planner as Latency and Coverage. | Any future contact in the interval |
The first three criteria are real-time: they answer "if I sent it now, along links that are open now, which path is best?". CGR answers "if I sent it now, when does it arrive at the earliest?". CGR routes are planned on an idle network (no other traffic), as a planning tool for simulation.
Is CGR always used?
Latency and Coverage always move data with CGR (earliest delivery). In Routing Analysis CGR is one of the criteria, so you can compare it with the geometric criteria on the same network.
What one time step simulates¶
Every time step sends one hypothetical file from the source at that instant and keeps the best route for it. Steps are independent: nothing is queued, stored or consumed from one step to the next, so a 30 s step over one day is 2 880 separate "what if I sent it now" questions, not one file followed through the network. Consecutive steps that pick the same route merge into a segment; the reports count time per segment.
File size only matters for Fastest file delivery and CGR. With a small file both stay close to the shortest path. With a large one, links that close before the transmission ends drop out (fastest file) or the file waits for a longer contact (CGR).
In 3D the route shown at a given time is the segment covering that time. Nothing is drawn while there is no route.
Several targets¶
When several targets are checked, each time step routes to the best one under the criterion. A constellation target means any of its members. The results store how long each target, relay and route was used, so the most-used target answers "which station ends up doing the work?".
Time on each target — Walker Star 20/4/1, shortest path from SAT-01-01 to the four SC stations. Svalbard takes the largest share (polar passes); Hartebeesthoek the smallest.
Editor¶
Complete the Connection Network first, then set:
| Field (editor) | Meaning |
|---|---|
| Routing ID | Name in the tree and Dashboard |
| Active | Off → skipped on scene refresh |
| Network | The completed Connection Network |
| Source | Origin node (must be on that network). A sensor used as an ISL or dump terminal maps to its parent satellite. All Satellites also needs a member; that does not consume the other members as targets. |
| Targets | One or more destinations; the best one is used at each time. All Satellites means any other member; you can also check a specific satellite of the same constellation. Sensor terminals likewise map to the parent bus. |
| Apply connection-network directionality | Honour each edge as From→To only. Off uses both directions. |
| Criterion | How the best route is chosen (see above) |
| Time step [s] | Sampling step of the route (default 30) |
| File size / unit | Shown only for Fastest file delivery and CGR |
| Interval | Scenario window (usual) or a custom UTC range |
| Show route | Draw the best route in 3D after Run |
| Route color / width | Colour and thickness of the route over the Connection Network (default Yellow, 4 px) |
Apply Changes redraws the stored route. Run Analysis recomputes after the network or these fields change.
No route — missing edges, no access in the interval, endpoints not on the graph, directionality blocking the reverse hop, or (shortest path) no positions available for the nodes. Stale — the Connection Network changed; run Routing again.
Reports and plots¶
| Name | Content |
|---|---|
| Routing Summary | Availability, route changes, value and path length statistics, most-used target, relay and route |
| Routing Configuration | Criterion, step, file size, interval, network size |
| Routing Segments | Each best-route segment with start, end, target, hops and value |
| Routing Target / Relay / Route Usage | Time and share of each target, relay and route |
| Routing Route / Target Timeline | Gantt of the best route and of the target reached |
| Routing Metric / Hops / Path Length vs Time | Criterion value, hop count and path length at each step |
| Routing Target Share, Relay / Route Usage Bar, Hop Distribution, Metric CDF | Usage and distribution plots |
The plots below are the same Dashboard templates, from a 12 h shortest-path study (SAT-01-01 to any of the four stations).
Best-route path length vs time. Short stretches are a direct dump; spikes are multi-hop ISL chains when the bus is out of view.
Light-delay CDF of the best route. About 30 % of samples stay under 10 ms (own downlink); the tail is the longer mesh paths.