Ontology modules (rationale)

From Semantic Sensor Network Incubator Group
Jump to: navigation, search

Defining Ontology modules

Pass 1

Following the agreed process for the development of the ontology, after the device ontology review and assessment of existing sensor ontologies, we now have to identify the bits that are missing or need revision.

The Key Sensor Concepts page describes the current structure of the sensor ontology and its central concepts. There are references like the MMI facet list and the SANY sensor taxonomy that give other perspectives on what a sensor ontology should cover -- see also SensorML and O&M.

This table is a condensed version of the extended minutes from our device ontology review. Members of the group can identify package(s) they're interested working on and enter their name in the "Who's doing it?" column. The "Use case(s)" column is meant to narrow the scope of the package.

The amount of work per package isn't equal; some are actually quite large while others are just have a look at something and add a couple of concepts to the ontology.

This is a work in progress - this table is being refined and updated, as is the design of what should be included.

Packages and responsibilities
Concepts for Scope of this Use case(s) Who's doing it? !
1 values and ranges has links into anything that requires a value and a unit ... Michael and Holger
2 UoM, time, location, platform should all probably be external, but what are the right things ... TBD
3 Mobility, In-situ-/remote-sensing and field of sensing describing where a sensor senses, how it can move and how it can change its field of sensing ... TBD
4 System structure sensors, components, systems etc, including granularity of sensors and processes. Note the difference in basic structure between for example the CSIRO, MMI and OOTethys ontologies ... TBD
5 Observation, Phenomenon, Measurement see discussion from recent weeks, how to align sensors and observations ... TBD
6 Operational/survival range and restrictions battery life, maintenance schedule, lifetime of device, on what and where can it be deployed, storage/processing capacity ... TBD
7 calibration ... ... TBD
8 operational availability availability, functionality status, etc ... TBD
9 Metadata manufacturing details, websites, identification ... TBD
10 Uncertainty/accuracy ... ... TBD
11 Events ... ... TBD
12 Sensor types See A3ME and OntoSensor for examples ... TBD
13 history of deployment, configuration, operating conditions, etc ... TBD

Pass 2

Table 1 - Groups and sub-groups of concepts

  • Table below based on pass 1 and on the categories from the Compton et al 2009 review paper allocated to the main groups described above (plus a 5th group for the rest and the imported ontologies)
Open (and closed) items (Ontology)
Groups Sub-groups Description
Sensor/Device - Sensor, systems, components and devices
System structure (components, deployment, configuration_ Sensors, components, systems etc, including granularity of sensors and processes.
Sensor types (sensor hierarchy) See A3ME and OntoSensor
History (history) of deployment, configuration, operating conditions, etc
Observation: O&M
Observation, Phenomenon, Measurement (data/observation, feature/quality, sampled medium) how to align sensors and observations
Capabilities What the sensor measures, how accurately it measures, how often it can measure and the field of view
Measurement Capabilities (response model) What the sensor measures, how accurately it measures
Mobility, In-situ-/remote-sensing and field of sensing (field of view/sensing) describing where a sensor senses, how it can move and how it can change its field of sensing (including discussion on compatibility with platform ontology?)
Operational availability availability, functionality status, etc
Operational/survival range and restrictions (operating conditions) battery life, maintenance schedule, lifetime of device, on what and where can it be deployed, storage/processing capacity
Uncertainty/accuracy (accuracy), Calibration
Process (action & process) Inputs and outputs, procedure/functions, describe how a measurement is made and how sensors (and other systems) are combined together
Action and process
Event
Imported ontologies (not sensor specific) where the sensor is, where it is measuring, its power supply what it is attached to, its physical dimensions and weight, its operating conditions and who made it
Values and ranges (dimension, weight, power supply, frequency) has links into anything that requires a value and a unit
UoM (units of measurement), time (time), location (location), platform (platform) should all probably be external, but what are the right things
Metadata (identity & manufacturing, contacting & software) manufacturing details, websites, identification

Pass 3 Groups and sub-groups

Packages and responsibilities
Groups Sub-Groups Object Description !
Base Data Utilities to capture data Data, Value (xsd:types), Interval
Base UoM Units and Quantities Dimension, Unit of Measurement, (Physical) Quantity
Base Time Time Time (absolute and relative)
Base Location Location and coordinate systems Math and Geospatial defs (absolute and relative)
Base DataQuality Uncertainty, data quality factors Characteristics of data, fitness to purpose (probability-based, other)
Base Geometry Basic geometrical concepts (static) Any type of 2D or 3D objects
Base SpatioTemporal Basic geometrical concepts (dynamic) Any type of 4D objects, including time series
Model ConstraintBlock Constraint, Condition (specification-form model) (Simple or Complex) Condition, Constraint having an influence on something
Model FunctionBlock Function with inputs and outputs (solution-form model)) A model
Model Parthood Aggregation, Compostion, Merology Different sorts of has-part relationships for the concepts defined above
Model System System structure (components,configuration_ Sensors, components, systems etc, including granularity of sensors and processes.
Model Process Process (action & process) Inputs and outputs, procedure/functions, describe how a measurement is made and how sensors (and other systems) are combined together. See SensorML process, SysML and DAQ in NEESGrid.
Upper Entity Entity (tuple of attributes), Physical objects (tupleof observable attributes), Abstract Object Anything: real, possible, or imaginary, which some modeller wants to talk about for some purpose. (DUL). Anything that is defined by its properties.
Upper Upper Upper classes (branches): main top classes and the relations between them Event, Agent (anything to import from DOLCE)
Upper ArtifactFunction Upper functions: Artifact functions High level principles (including actuators and sensors functions). Inlcuding Stimuli stuff for observationsand other relevant definitions from Garbacz, P. 2006. Towards a standard taxonomy of artifact functions.
Deploy PlatformSite Platform, Site Container for all siting and mobility specific characteristic
Deploy PlatformSite Platform, Site functions Functions for all types of platforms, sites
Deploy PlatformSite Platform, Site properties Properties for all types of platforms, sites (see FGDC Remote Sensing metadata)
Deploy PlatformSite Platform, site capabilities  
Deploy PlatformSiteRestriction Site or platform-related operational/survival range and restrictions (perating conditions) Ability of the platform to reach, maintain itself at the right point for sensing
Deploy PlatformSiteType Site and platform types (hierarchy) arranged in function of their siting and mobility-specific capabilities May have multiple sub-hierarchies in function of domais e.g. MMI platform ontology, Stream gages siting configuration (GSDQ), …
Deploy DeviceSupport Device relation to other objects, especially platform On what and where can it be deployed. How a device can be related to a site or to a platform (overlap with system structure, configuration)
Deploy Deployment Deployment (deployment, configuration) See definitions in NEESGRID
Deploy DeploymentRestriction Operational/survival range and restrictions (operating conditions) - operational availability Maintenance schedule, lifetime of device, platform,
Deploy DeploymentMetadata Who is doing the deployment, for what reasons (experimentation) See definitions in NEESGRID
Deploy DeploymentType Platform, Site, Deployment types (hierarchy) arranged in function of characteristics If needed (corresponds to scientific experiment templates). May overlap with other hierarchies (should be inferrred - not hard-coded)
Device Device Device Container for all device specific characteristic
Device Device Device function See Artifact functions and also primary and secondary equipment functions in NEESGrid
Device Device Device properties characterising the ability of the device to perform its function availability, functionality status,
Device Device Properties (dimension, weight) Properties for all types of devices
Device DeviceType Device types (hierarchy) arranged in function of the type of functions they perform May overlap with other hierarchies (should be inferrred - not hard-coded)
Device DeviceMetadata Metadata (identity & manufacturing,contacting & software) manufacturing details, websites, identification
Energy Energy Energy device Container for all energy specific characteristic
Energy Energy Energy function Whether the device can generate energy or consume energy or both (describe the energy flow)
Energy Energy Energy properties (power supply) Properties for devices requiring an energy source
Energy EnergyRestriction Operational/survival range and restrictions (operating conditions) for energy management battery life,
Energy EnergyDeviceType Energy device types (hierarchy) arranged in function of their energy characteristics May overlap with other hierarchies (should be inferrred - not hard-coded)
Communication Communication Communication device Container for all communication specific characteristic
Communication Communication Communication function Whether the device can send or receive data, messages, information or both
Communication Communication Communication properties (frequency, throughput) Properties for communication devices (overlap with system?)
Communication CommunicationRestriction Operational/survival range and restrictions (operating conditions) for communication Factors limiting the ability of the device to communicate
Communication CommunicationDeviceType Communication device types (hierarchy) arranged in function of their communication characteristics May overlap with other hierarchies (should be inferrred - not hard-coded)
Processing Processing Processing device Container for all data processing specific characteristic
Processing Processing Processing function Whether (and how) the device can process information
Processing Processing Data processing properties (e.g. computing power) Properties for processing devices (overlap with system?)
Processing ProcessingRestriction Operational/survival range and restrictions (operating conditions) for processing Factors limiting the ability of the device to process information: storage/processing capacity
Processing ProcessingDeviceType Processing device types (hierarchy) arranged in function of their data processing characteristics May overlap with other hierarchies (should be inferrred - not hard-coded)
Sensor Measuring Measurement (device), Sensing device, Sensor Container for all sensing/measurement specific characteristic
Sensor Measuring Physical principles (response model) on which the sensor is based What the sensor measures (generic defs) - check the overlap with Observation
Sensor SamplingCapability Spatio-temporal capabilities of the sensor (field of view, rate of sampling) How often it can measure, the field of view - check the overlap with Observation
Sensor MeasuringCapability Properties of measuring devices (accuracy, resolution, …) How the sensor measures (what characterises its performance)
Sensor MeasuringRestriction Measuring range and restrictions What impact the ability of a sensor to measure something
Sensor MeasurementDeviceType Sensor types (sensor hierarchy) arranged in function of the type of measurement they perform What the sensor measures (domain-dependent defs). See A3ME and OntoSensor. Overlap with other hierarchies (should be inferrred - not hard-coded)
Sensor Calibration Calibration TBD
Observation Observation Observation, Phenomenon, Measurement (data/observation,feature/qualitysampled medium) Container for observation characteristic (projection into O&M)
Observation Procedure Procedure Method used to do the observation
Observation SampledFeature Sampled feature As defined by O&M
Observation ObservationMedium Observation Medium (sampled medium) As defined by O&M
Observation ObservationSpecimen Observation Specimen As defined by O&M
Observation SamplingFeature Sampling feature As defined by O&M (see also NEESGRID)
Observation ObservationMetadata Metadata What is not specified elsewhere
Dataset Dataset Dataset Container for all dataset specific characteristic. See Information object in DOLCE, digital library defs, SERONTO
Dataset GeospatialDataset Observation data Container for O&M/SOS (results)
Dataset GeospatialDataset Geographic feature data Container for feature characteristic (projection into GML)
Dataset DatasetMetadata All the information provided in the header or registry of a datasest and not listed above Especially provenance, preservation/archive
Dataset DeploymentDataset Deployment (System + space/time variable configuration) = history (history) of deployment, configuration, operating conditions, etc

Pass 4 SSN Ontology

Some changes to the ontology split into "modules" have been made during the final phase of development to match the documentation page structure and the scope of the examples supplied by the XG participants.

See the page introducing the Ontology structure.

The table below lists for each sections and modules:

  • the initial discussions about it (when available)
  • the documentation page for the module
  • the examples page for the module
Documentation roadmap
Section Module Discussion Final documentation Examples
Upper Upper Alignement to DUL Upper Ontology Foundational Layer
Skeleton Skeleton SSN Skeleton Skeleton
Model System Model#System SSN-UniDeployment
Model Process Model#Process SSN Wind Sensor and Agriculture Meteorology Sensor Network
Sensor Measuring Key Sensor Concepts Sensor#Measuring SSN Smart product
Sensor MeasuringCapability Modelling of Accuracy, etc Sensor#MeasuringCapability SSN Smart product
Observation Observation F2F minutes Observation#Observation SSN Smart product
Deploy Deployment Deployments, Systems and Devices Deploy#Deployment SSN Uni deployment
Deploy PlatformSite Deployments, Systems and Devices Deploy#PlatformSite SSN Uni deployment
Deploy OperatingRestriction Operating Model Deploy#OperatingRestriction SSN Uni deployment
Base Data Base#Data SSN Uni deployment
Base Time Base#Time
Base ConstraintBlock Base_ConstraintBlock (Condition) SSN Wind Sensor
Device Device Device
Energy EnergyRestriction Operating Model Energy#EnergyRestriction

Modules and Mapping to use cases

A set of use cases have been identified as motivational use cases for the work that has been done in the context of this group. They can be roughly classified into four main groups, according to the following two axes:

1) User classification: from users who directly manage sensors and sensor networks; through to users who use and manipulate data (but may not have any direct interest in sensors).

2) Technology: from Semantic Web through Sensor Web to Sensor Network.

The following table and figures provide different viewpoints on this classification, providing names for these four groups of use cases (data discovery, device discovery and selection, provenance and diagnosis, and device operation tasking and programming). The reason to identify several use case categories is that each one requires different levels of detail for the modelling of the sensors, the features they observe, their capabilities, their environment and their conditions of use. These use cases will be described next.

Uses cases categories

The figures below illustrates which modules are more important for each category of use cases defined above.

Use Case 1: Data discovery and linking

Use cases and ontology modules

Comment: there are many ways to find the datasets containing the right observations. Some discovery strategies, via the features or the observed properties are less dependent on sensor-related knowledge.

Use Case 2: Sensor device selection and discovery

Use cases and ontology modules

Comment: in this use case, measuring capabilities are more important to identify the sensing assets.

Use Case 3: Provenance and diagnosis

Use cases and ontology modules

Comment: operational restrictions become more important, especially the ones which impact sensor performances.

Use Case 4: Tasking, programming

Use cases and ontology modules

Comment: to operate sensors and sensor networks, it is important to better manage the environmental conditions and system features more generally and to know what are the communications or the energy requirements for each sensors.