D8_InfrastructureReturns_BackboneLinks - Radega1993/the-one-scenario-corpus GitHub Wiki

Scenario D8 — D8_InfrastructureReturns_BackboneLinks

1. Overview

Scenario ID: D8
Name: D8_InfrastructureReturns_BackboneLinks
Family: Disaster
Settings file: corpus_v1/05_disaster/D8_InfrastructureReturns_BackboneLinks.settings

Objective

Partitioned disaster layout where infrastructure/backbone links reappear mid-simulation via external events.

2. Scenario configuration (core features)

Values below come from analysis/data/features.csv (raw) and the mapping to the 23-core subset.

Feature	Value	Comment
world_area	40000000	Total simulation area (m^2)
aspect_ratio	0.4	min(width,height)/max(width,height)
N	80	Total nodes
nrofHostGroups	2	Number of host groups
speed_mean	0.8	Mean configured speed (m/s)
wait_mean	135	Mean pause/wait time (s)
mm_WDM	0	WorkingDayMovement enabled (1/0)
mm_RWP	0	RandomWaypoint enabled (1/0)
mm_MapRoute	0	MapRouteMovement enabled (1/0)
mm_Cluster	1	ClusterMovement enabled (1/0)
mm_Bus	0	BusMovement enabled (1/0)
mm_Linear	0	LinearMovement enabled (1/0)
transmitRange	10	Interface range (m)
bufferSize	50000000	Node buffer (bytes)
transmitSpeed	2000000	Interface speed (bytes/s)
msgTtl	10000	Message TTL
event_interval_mean	112.5	Mean Events1 interval
event_size_mean	65000	Mean Events1 size (bytes)
nrof_event_generators	2	Number of event generators
pattern_burst	0	Burst windows in traffic (1/0)
pattern_hub_target	0	Hub-target traffic pattern (1/0)
workDayLength	—	Not used in this scenario
ownCarProb	—	Not used in this scenario
clusterRange_mean	350	Mean cluster radius if ClusterMovement

3. Mobility model

World size: 10000, 4000
Base speed range: Not recorded
Base wait range: Not recorded

Clustered partitions with ExternalEventsQueue introducing late inter-partition links.

DTN implication

This mobility design creates a constrained-contact disaster regime where connectivity depends on temporal bridges, dense local clusters, or opportunistic relays rather than stable end-to-end paths.

4. Traffic pattern

Events.nrof = 2
Events1.interval = 45, 180
Events1.size = 10k, 120k
Group.msgTtl = Not recorded

Traffic is configured as emergency-oriented load with timing/size parameters aligned to this disaster narrative.

DTN implication

Under Epidemic routing, these parameters amplify trade-offs between urgency and congestion: short opportunities improve fast deliveries in contact windows but can sharply increase redundancy or message expiration when partitions persist.

5. Expected network behavior

Contact opportunities are heterogeneous and depend on movement structure (clusters/partitions/routes).
Delivery is limited when temporal bridges are weak or TTL is very short.
Overhead rises quickly when flooding meets dense local contacts.
Delay can be bimodal: near-instant inside local contact islands, very high across partitions.

6. Role in the corpus

This scenario represents a specific disaster communication regime inside the corpus, contributing diversity relative to Urban/Campus/Social baselines and complementing other Disaster scenarios with a distinct structural stressor.

7. Distinguishing characteristics

Disaster-focused configuration with explicit structural constraints.
Mobility/traffic coupling designed to stress store-carry-forward behavior.
Relevant for evaluating robustness under disrupted or intermittent connectivity.

8. Correlation with other scenarios (core 23)

Using the 23-core feature space (analysis/data/correlation_pearson_core23.csv):

Most similar (top 3):
- D2_PartitionedCity_MuleBridge — r ≈ 0.76
- S1_StrongCommunities_SeparateClusters — r ≈ 0.55
- T3_MixedBimodal_SmallAndLarge — r ≈ 0.36
Most different (top 3) (smallest |r|):
- C1_Campus_ClassChange — r ≈ 0.00
- T1_ManySmallMsgs_HighRate — r ≈ -0.01
- D6_ShortTtlCritical_5to10min — r ≈ 0.01

Full pairwise correlations are available in analysis/reports/correlation_core23_report.txt and analysis/data/correlation_pearson_core23.csv.

Interpretation

The nearest scenarios share the same main structural levers (movement model family, host-group structure, and traffic scale), while near-zero correlations typically correspond to scenarios governed by orthogonal drivers (e.g., extreme range/speed, map routing, or different TTL/load regimes).

9. Cluster assignment

In the Ward k=7 clustering on the 23-core feature space (cluster_assignments_core23.csv), this scenario belongs to:

Cluster 2.

10. Simulation outputs (optional)

If routing simulations have been run and metrics were extracted (analysis/data/output_metrics.csv):

Metric	Value
delivery_ratio	0.4607
latency_mean	2015.7523
overhead_ratio	79.3182
drop_ratio	22.976439790575917