Peer-review record

IoTSyn Smart City and Public Spaces Edition: a specification-driven, physics-informed generator of synthetic urban IoT data with a multi-hazard decision layer

Amir Duhair  ·  ORCID 0000-0002-6831-0342

Faculty of Science and Technology, University of Jijel, Jijel, Algeria

Manuscript v4.4.0 Submitted for peer review

Abstract

Research on smart cities and urban public spaces is constrained by the scarcity, cost, privacy sensitivity, and irreproducibility of real sensor data. We present IoTSyn Smart City and Public Spaces Edition, a public web service that generates coupled synthetic Internet-of-Things datasets for urban public spaces. It combines declared models of meteorology, urban heat, air quality, environmental noise, pedestrian presence, and traffic. The specification fixes stochastic algorithms, random-consumption order, named Mersenne Twister substreams, and a canonical serialization profile. In the tested reference environment, identical configurations reproduce byte-identical artifacts. The generator separates physical truth, sensor observation, and decision output so that reported derived quantities can be recomputed. A multi-hazard decision layer produces risk states, communication messages, and episode-level behaviour-change strategy objects while distinguishing data-derived from template-derived fields.

The generator is accessible at iotsyn.com. Citable research-artifact records provide scenarios, contracts, and reported outputs; the proprietary implementation source remains undistributed.

Reported findings

Stated as they appear in the manuscript, including the negative result.

Reproducibility

In the tested reference environment, identical configurations reproduce byte-identical artifacts. Known-answer vectors cover the random number generator, sampling, solar geometry, pollutant dynamics, and risk aggregation.

Directional scenarios

Eleven directional scenarios satisfy their declared internal signatures and produce zero domain-contract violations.

Sensor-error sensitivity

Under the specified synthetic observation-error models, a 30 % i.i.d. log-scale error yields a 4.38 % mean decision-flip rate, whereas 10 % stationary calibration drift yields 9.78 %. Alarm decisions attenuate i.i.d. noise but degrade under calibration drift.

Held-out external comparison

A held-out comparison with City of Melbourne observations met none of the predeclared diurnal-profile criteria for pedestrian-flow proxy, air temperature, or PM2.5 across 30 seeds. It therefore provides no evidence of cross-site transfer.

What this evidence does and does not establish

Establishes

  • — Implementation verification: the generator executes its stated model.
  • — Byte-level reproducibility in the tested reference environment.
  • — Declared internal signatures hold across eleven directional scenarios.

Does not establish

  • — Site realism, or transfer to any specific real deployment.
  • — Calibration against a named city, sensor network, or population.
  • — The held-out Melbourne comparison met none of the predeclared diurnal-profile criteria across 30 seeds, and is reported as a negative result.

Evidence packages

Permanent, versioned deposits. Each carries its own DOI and is independently citable.

Revision R12 2026-08-03 · CC BY 4.0

IoTSyn v4.4.0 Smart City and Public Spaces evidence package, revision R12: stress sweeps, empirical baseline, verification audits and external-comparison data

10.5281/zenodo.21580464 →
Frozen v4.4.0 CC BY 4.0

IoTSyn v4.4.0 public research artifacts: scenarios, verified outputs, contracts and external-comparison data

10.5281/zenodo.21480873 →

10.5281/zenodo.21480872 is the all-versions concept record only. Cite a specific version DOI, never the concept DOI.

Reproducing a reported run

Every published record declares the four values needed to regenerate it. Byte-identity is claimed only for the tested reference environment declared in the specification.

domain

One of six generator domains

master_seed

Decimal seed, SHA-256 named-substream derivation

parameters

Declared scenario configuration

engine_version

Per-dataset; legacy domains are not v4.4.0

Cite this work

The manuscript is under review and has no assigned DOI yet. To cite the underlying evidence today, use the R12 record above. This block updates automatically once the preprint deposit is public.

APA
Duhair, A. (2026). IoTSyn Smart City and Public Spaces Edition: a specification-driven, physics-informed generator of synthetic urban IoT data with a multi-hazard decision layer. Manuscript submitted for publication.
IEEE
A. Duhair, "IoTSyn Smart City and Public Spaces Edition: a specification-driven, physics-informed generator of synthetic urban IoT data with a multi-hazard decision layer," 2026. Manuscript submitted for publication.
BibTeX
@article{duhair_iotsyn_smartcity_2026,
  author  = {Duhair, Amir},
  title   = {IoTSyn Smart City and Public Spaces Edition: a specification-driven, physics-informed generator of synthetic urban IoT data with a multi-hazard decision layer},
  year    = {2026},
  note    = {Manuscript submitted for publication},
  urldate = {2026-08-04}
}

Keywords

synthetic data smart cities urban public spaces Internet of Things reproducibility thermal comfort air quality agent-free simulation decision support

Correspondence: duhair.amir@univ-jijel.dz