⌘ PHASE/FLOWRESEARCH
BEYOND STATIC TRAFFIC / INTERACTIVE RESEARCH

A network is
never still.

Change the conditions.
Follow the packets.
Discover what the evidence says.

▶ Watch the English narrated film ↗
MININET / RYU / OVS
OPENFLOW 1.3
01 — NETWORK LAB
ILLUSTRATIVE NETWORK
INSPECT THE NETWORKClick a node to explore its role.
CONTINUOUS COMMUNICATIONCONCEPTUAL PACKET TIMING · NO LIVE CONNECTION
02 / CHANGE THE NETWORK PHASE

Keyboard: focus the slider and use ← → to change phase. Each phase updates the illustration and reported H28 values below.

WHAT YOU’RE SEEING

The everyday baseline.

Continuous UDP and ICMP communication represents normal network activity across the testbed.

H28 / REPORTED PHASE MEANSExplore measurements ↗
6.0508ms / MEAN RTT
9.2999Mbps / TCP GOODPUT

Measured H28 phase means; not live readings. Topology and workload controls change the illustration only.

This interactive model explains the supplied paper. It does not execute attacks, detect live intrusions, or establish independently verified malicious ground truth.

BUILT ON AN OPEN NETWORK STACKMininet+Open vSwitch+Ryu+OpenFlow 1.3
01 / THE FRAMEWORK

Real dynamics.
Traceable observations.

A change in network behavior is not automatically an attack. This framework separates operational conditions from workload activity, so every observation keeps its context.

01
⌘

Configure

Build line or star topologies with Mininet, programmable switches, and a Ryu controller.

TOPOLOGY + SEED
02
≋

Generate

Run benign and bounded synthetic workloads through stable, constrained, and recovery phases.

6 WORKLOAD SCENARIOS
03
⌁

Observe

Align flow counters, RTT samples, UDP measurements, and timestamped experiment events.

MULTI-SOURCE TELEMETRY
04
▤

Validate

Preserve run identity, mark transition windows, and label activity using execution evidence.

EVIDENCE-GROUNDED LABELS
02 / EXPLORE THE DYNAMICS

One network.
Three different realities.

Switch phases to explore the network’s measured response. Values below are phase means from the paper’s H28 evaluation; the animation is an illustration.

NETWORK TOPOLOGY
Reference conditions. Traffic flows under stable network settings.
H28 / MEASURED PHASE MEANS
Mean round-trip time
6.0508ms
TCP bulk goodput
9.2999Mbps
ICMP loss 0.00%Conditions / phase 12

H28 aggregate results are not separated by topology in the paper. Topology controls change the illustration only. Zero observed ICMP loss does not establish perfect loss fidelity.

TRAFFIC CINEMA / FROM CONCEPT TO CONTEXT

See the behavior.
Follow the evidence.

A detailed English narrated walkthrough of the framework, six workloads, dataset construction, and reported H23/H28 results. Includes synchronized on-screen captions, chapter navigation, and a full transcript.

PHASE/FLOW ORIGINAL · RESEARCH EXPLAINERENGLISH VOICE / 14 CHAPTERS

Press Play to hear the English narration. Captions are included in the picture.

Read the full English transcript

English narration uses a synthesized voice. This is an authored visual explanation of the paper, not recorded experimental footage. Packet timing, ports, and geometry are illustrative. No real attack execution or live network connection.

03 / CONTROLLED WORKLOADS

Explore the workload.
Not just the label.

Select a scenario to see its packet behavior, then switch network phases to explore the context. Animations explain concepts; reported results remain in the measured findings.

BENIGN WORKLOADB00
PACKET BEHAVIOR / ILLUSTRATIVE
Benign communication Synthetic workload

The everyday baseline.

Continuous UDP and ICMP communication represents normal network activity. Its observations provide context across changing network conditions.

EVIDENCE CONTEXT

Activity labels follow recorded workload intervals. Network constraints are tracked separately from benign activity.

04 / REPORTED FINDINGS

Measured, not
assumed.

Separate experimental campaigns examine dataset construction, prospective classification, and network fidelity.

H14 / DATASET CONSTRUCTION216

observation intervals · 20 columns

24 runs · 72 observations in each split
H23 / PROSPECTIVE EVALUATION144

new intervals · 16 additional runs

Five frozen models · no retraining
H28 / NETWORK FIDELITY36

phase-specific network conditions

12 runs · stable → constrained → recovery
H23 / MODEL COMPARISON

How the frozen models performed.

0%50%100%

Synthetic probing classification, not independently verified malicious attack detection. Source: Table IV.

H14 / RUN-DISJOINT SPLITS

Keep experiments
independent.

Training 728 runs
Validation 728 runs
Test 728 runs

Complete run IDs stay in one partition to reduce leakage from correlated observations.

Each split: 60 class-0 + 12 synthetic probing observations.
↗

Strong evidence starts with clear boundaries.

Gradient Boosting reached 89.58% balanced accuracy; Random Forest had the highest recall at 87.50%. Statistical significance between models was not established. Independent malicious ground truth and generalization to unseen attack families remain unverified.

05 / THE ACTUAL DATASET

Every interval.
Its original context.

Explore the supplied H14 export directly. Filter by partition, operational phase, topology, and scenario. Each row preserves its experimental run ID.

216 observations · 20 columns · 24 distinct runs. The H14 export contains four scenario values: idle, udp_burst, tcp_probe, and lowrate_tcp. These describe this dataset’s coverage; the paper’s broader framework defines six workloads. Label 1 means synthetic probing; label 0 means not synthetic probing.
Loading the verified H14 export…
H14 observation intervals · selected fields from the original 20-column export
Run IDPartitionScenarioPhaseIntervalProbe labelRTT (ms)Delivery (Mbps)
Inspect all 20 column names and source checksum

Run IDs do not overlap between train, validation, and test partitions. The checksum identifies the original full CSV. The displayed delivery measurement is an H14 field; it is distinct from the H28 TCP bulk goodput evaluation.

07 / REPRODUCIBLE BY DESIGN

Take the data.
Trace the result.

Download the supplied exports and source tables. Original ZIP packages are preserved unchanged, including their reports, audit context, and scientific limitations.

Independent malicious ground-truth adjudication remains pending. These packages support controlled synthetic-workload analysis and network-fidelity findings within the reported experimental scope.

08 / THE RESEARCH PAPER

Go beyond
the overview.

Beyond Static Traffic: A Dynamic and Phase-Aware SDN Framework for Evidence-Grounded Intrusion Detection Dataset Generation

Read the full paper ↗Download PDF ↓
THE AUTHORS

Dr. S. Bharath Bhushan · Ganga Bharath · Pathepur Dikshith
Badveli Guru Pala Nivesh Reddy · Chitteti Durga Sai