<p>The proliferation of efficient edge computing has enabled a paradigm shift of how we monitor and interpret urban air quality. Coupled with the dense spatiotemporal resolution realized from large-scale wireless sensor networks, we can achieve highly accurate realtime local inference of airborne pollutants. In this paper, we introduce a novel Deep Neural Network architecture targeted at latent time-series regression tasks from continuous, exogenous sensor measurements, based on the Transformer encoder scheme and designed for deployment on low-cost power-efficient edge processors.


We have prepared a synthetic dataset to detect and add new devices in DynO-IoT ontology. This dataset consists of 1250 samples and has 35 features, such as feature-of-interest, device, sensor, sensor output, deployment, accuracy, unit, observation, actuator, actuation, actuating range, tag, reader, writer, etc.


The LEDNet dataset consists of image data of a field area that are captured from a mobile phone camera.

Images in the dataset contain the information of an area where a PCB board is placed, containing 6 LEDs. Each state of the LEDs on the PCB board represents a binary number, with the ON state corresponding to binary 1 and the OFF state corresponding to binary 0. All the LEDs placed in sequence represent a binary sequence or encoding of an analog value.


The Here East Digital Twin was a six month trial of a real-time 3D data visualisation platform, designed for the purpose of supporting operational management in the built environment. 


1. View the project video for context: https://vimeo.com/311089492

2. Review details of the Envirosensor implementation: https://github.com/virtualarchitectures/ICRI_Envirosensor

3. One option for analysing the data is using Spark and Python in a Jupyter Notebook as outlines in the following example GitHub repository: https://github.com/virtualarchitectures/ICRI_Envirosensor_Data_Analysis


Academic spaces are an environment that promotes student performance not only because of the quality of its equipment, but also because of its ambient comfort conditions, which can be controlled by means of actuators that receive data from sensors. Something similar can be said about other environments, such as home, business, or industry environment. However, sensor devices can cause faults or inaccurate readings in a timely manner, affecting control mechanisms. The mutual relationship between ambient variables can be a source of knowledge to predict a variable in case a sensor fails.


The dataset contains information about roses cultivation in greenhouses. It is aimed at identifying corrective actions to improve the roses state. Data acquisition was done with an autonomous robot incorporating sensors such as: soil humidity, light, temperature and humidity, and CO2.


1. Title:  Roses greenhouse cultivation database repository (RosesGreenhDB)


        Updated 17/07/2019  by Wilmer A. Champutiz


2. Sources:

     (a) Creators: Edison A. Fuentes-Hernández and Paul D. Rosero-Montalvo 

     (b) Date: July 2019


3. Relevant Information:

 The present dataset contains information about roses cultivation in greenhouses.

 It is aimed at identying corrective actions to improve the roses state.

 Correspondingly, the target variables (labels) are as follows:


1  Soil without water

2  Environment correct

3  Too much hot

4  very cold


Data acquisition was done with an autonomous robot incorporating sensors such as: soil humidity, light, temperature and humidity, and CO2. Resulting dataset is imblanced.


4. Number of Instances:  300 (125 soil without water

                               28 correct environment 

                               90 too much hot

                               57 very cold)


5. Number of Attributes: 5 numeric predictive attributes and the class


6. Attribute Information:

   1. Soil humidity in analog-digital conversion

   2. Light in lux

   3. Temperature in °C 

   4. CO2 in  analog-digital conversion

   5. Humidity relative 


   6. Class: 

           1  soil without water

           2  environment correct

           3  too much hot

           4  very cold


7. Missing Attribute Values: None


This dataset contains the resuts of an experiment in which an electronic nose implemented with six MOX sensors acquired samples of explosives in raw and combined states.

As for the collection of samples, a random experimentation was carried out in order to avoid that data generates any memory effect that could influence the results. Raw TNT and gunpowder data were taken in amounts of 0.1g to 2g. Soap and toothpaste were also used to be mixed with the explosives. In the end, we took samples of the explosive substances in raw and combined states.


The pressure sensors are represented by black circles, which are located in the three zones of each foot. For the left foot: S1 and S2 cover the forefoot area. S3, S4, and S5 the midfoot area. S6 and S7 the rearfoot or heel area. Similarly, for the right foot: S8 and S9 represent the forefoot area. S10, S11, S12 the midfoot area. S13 and S14 the heel area. The values of each sensor are read by the analog inputs of an Arduino mega 2560.


Each label correspond to:

Label         Position                                             Pressure on sensors

1                normal footstep                                Left foot: S1, S2, S3, S4, S6, S7

                                                                           Right foot: S8, S9, S11, S12, S13, S14


 2                flat footstep                                     Left foot: S1, S2, S3, S4, S5, S6, S7

                                                                           Right foot: S8, S9, S10, S11, S12, S13, S14


 3                cavus footstep                                 Left foot: S1, S2, S3, S6, S7

                                                                           Right foot: S8, S9, S12, S13, S14


Pressure position sensors and actuators are represented by red large and blue circles on the seating, respectively. Black circles display ultrasound sensors in the backrest, yellow arrows show output peripheral, and red arrows indicate peripheral input into the analog-digital converter.S1= Preasure sensor, value 0-1023 decimalS2= Preasure sensor, value 0-1023 decimalS3= Preasure sensor, value 0-1023 decimalS4= Ultrasonic sensor, value 0-15 cm


Each label correspond to:

Label         Position                                             Possible health problems

1                Right position                                    No harm


2                Higher pressure or right side             Respiratory issues, muscle imbalance stress on liver, stomach and right kidney


3                 Higher pressure or left side              Respiratory issues, muscle imbalance stress on liver, stomach and left kidney


4                 Higher forward pressure                   Knee issues, back pain and stress on abodmen