Article 1

Air Particle Matter Pollution Assessment: Applying IoT and DLT Technologies to High Volume Air Samplers

Diego Suárez-Bagnasco

Aerosols are fine solid particles (particulate matter, PM) or liquid droplets in gas (usually air), whose origin can be natural or anthropogenic. Air PM pollution exposure is linked to diverse human health problems and also to many environmental effects. People living in urban environments are potentially more exposed to polluted air.

 


Article 2

Blockchain-enabled Edge Analytics for a Mobile Universal Biotesting Station

Sergii Kushch, Martin Hayes and Eoin O’Connell

The impact of COVID-19 on manufacturing, educational, and a varied assortment of locations has been immense. The use case described below proposes a Blockchain-enabled infrastructure for deployment on an edge gateway to secure a universal biotesting station that is capable of processing sensitive personal biodata at scale. The unit can be deployed for access to a health center, manufacturing facility, or a Higher Education laboratory facility.

 


Article 3Adoption of Real-Time Machine Learning for Cyber Risk Assessment in IoT Environments

Mohamed Rahouti and Moussa Ayyash

The Internet of Things (IoT) has emerged in the past as Internet extension and significantly changed our world. A tremendous amount of IoT applications has highly eased people’s daily lives and improved resource usage and allocation, e.g., power bank sharing, bike sharing, etc. However, the ingrained openness of underlying wireless systems renders these IoT entities vulnerable to a broad range of cyber risks, e.g., vulnerabilities of spectrum that can be a source of adversarial inference.

 


Article 4New Generation Alternative Sensors for IoT

Euclides Lourenço Chuma

The Internet of things (IoT) is a system of interrelated computing devices and mechanical and digital machines provided with unique identifiers (UIDs) and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction. Many types of sensors are used in IoT for various applications, such as measuring temperature, humidity, pressure, acceleration, proximity, etc. However, a new generation of alternative sensors is being developed, which can be connected to IoT so that they are used in the creation of autonomous vehicles, in health monitoring, and other research areas.

 

 

EVENTS & ANNOUNCEMENTS


Article 5

IEEE Internet of Things Initiative - Upcoming Events

IEEE 6th World Forum on Internet of Things - 2020
Join us virtually for upcoming events:
IoT Vertical and Topical Week: 14-18 September
Women in Engineering Panel: 23 September 

View more details 


Article 5

IEEE Internet of Things Magazine

Internet of Things Magazine logo The Internet of Things Magazine (IoTM) publishes high-quality articles on IoT technology and end-to-end IoT solutions. IoTM articles are written by and for practitioners and researchers interested in practice and applications, and selected to represent the depth and breadth of the state of the art. The technical focus of IoTM is the multi-disciplinary, systems nature of IoT solutions.

Become an author - Submit an article today!
Never miss a copy - Subscribe today! 


Article 5

IEEE Xplore®

Stay Connected to IEEE Xplore When Working Remotely
If your organization has an institutional subscription to IEEE Xplore® and you need to work remotely due to school and workplace closures, you can still access IEEE Xplore and continue your work and research while offsite. Try these tips for remote access or contact IEEE for help. IEEE is here to support you, making certain that your IEEE subscription continues to be accessible to all users so they can continue to work regardless of location. 

 

This Month's Contributors

Diego Suárez Bagnasco Received is BSc. Electronic Eng. in 2002 (Catholic University of Uruguay), PgD Marketing in 2006 (Catholic University of Uruguay), MSc. Biomedical Eng. in 2010 (Favaloro University, Argentina), Ph.D. Biophysical Sciences in 2017 (PEDECIBA, Univ. de la República, Uruguay).
Read More >>

Sergii Kushch holds a Ph.D. in Computer Science and a Master degree in Radio Engineering (Hons.) and is a Senior Member of the IEEE.
Read More >>

Eoin O’Connell is a Funded Investigator in the Irish national research centre CONFIRM.
Read More >>

Martin Hayes is Head of the Electronic & Computer Engineering Department (ECE) at the University of Limerick and is a funded investigator with CONFIRM, the SFI funded institute for Smart Manufacturing Research in Ireland.
Read More >>

Mohamed Rahouti received the M.S. degree and Ph.D. degree from the University of South Florida in the Mathematics Department and Electrical Engineering Department, Tampa, FL, USA, in 2016 and 2020, respectively.
Read More >>

Moussa Ayyash (M'98–SM'12) received his B.S., M.S., and Ph.D. degrees in Electrical and Computer Engineering.
Read More >>

Euclides Lourenço Chuma earned a degree in Mathematics (2003) from the University of Campinas (UNICAMP), a graduate degree in network and telecommunications Systems (2015) at INATEL, and MSc in electrical engineering (2017) at UNICAMP, and a PhD in electrical engineering (2019) at UNICAMP, SP-Brazil.
Read More >>

 

Contributions Welcomed
Click Here for Author's Guidelines >>

 

Would you like more information? Have any questions? Please contact:

Raffaele Giaffreda, Editor-in-Chief
rgiaffreda@fbk.eu

Massimo Vecchio, Managing Editor
massimo.vecchio@uniecampus.it

 

About the IoT eNewsletter

The IEEE Internet of Things (IoT) eNewsletter is a bi-monthly online publication that features practical and timely technical information and forward-looking commentary on IoT developments and deployments around the world. Designed to bring clarity to global IoT-related activities and developments and foster greater understanding and collaboration between diverse stakeholders, the IEEE IoT eNewsletter provides a broad view by bringing together diverse experts, thought leaders, and decision-makers to exchange information and discuss IoT-related issues.

Air Particle Matter Pollution Assessment: Applying IoT and DLT Technologies to High Volume Air Samplers

Diego Suárez-Bagnasco
September 15, 2020

 

Aerosols are fine solid particles (particulate matter, PM) or liquid droplets in gas (usually air), whose origin can be natural or anthropogenic. Air PM pollution exposure is linked to diverse human health problems and also to many environmental effects. People living in urban environments are potentially more exposed to polluted air.

The aerodynamic properties of particles determine how they are transported in air and how they can be removed from it, as well as how far they get into the air passages of the respiratory system [1] [2]. Particles with an aerodynamic diameter [3] equal or less than 10 µm (PM10) can be hazardous to human health, and those equal or less than 2.5 µm (PM2.5) can reach alveoli during respiration and then they may be absorbed into the bloodstream. WHO (World Health Organization) states that there is strong evidence to conclude that fine particles PM2.5 and smaller are the more hazardous in terms of mortality and cardiovascular and respiratory endpoints in panel studies [2].

To study air pollution, airborne PM in ambient air is collected using High Volume (HVAS) and Low Volume (LVAS) Air Samplers equipped with special filters. The main blocks of an air sampler are four: air inlet (usually an inertial classifier), special PM filter, air pump, and control unit.

The PM retained in the filters can be studied using several techniques, depending on the objectives of the assessment (chemical, physical, and isotopic composition) [4] [5] [6]. Systematic periodic air sampling is needed to have a confident air quality assessment [7][8]. Usually, the mass concentration of particles (MC in μg/m3) and the standard mass concentration (SMC in μg/m3 std) are used to assess air quality. To accurately determine MC, it is important to measure instantaneous airflow and to calculate the total volume of air sampled in each sampling experiment. For SMC determination, real-time temperature, and pressure values at the sampling site must be measured to calculate normalized airflow related to EPA (Environmental Protection Agency) reference conditions [9][10].

Despite most of the modern air samplers have an electronic control unit, some models do not have any kind of digital flow recording or data transmission capabilities, nor environmental conditions measurement. In this situation, users usually must trust in the stability of airflow during the whole air sampling period (24 hours) when calculating MC.

In this article, we present an overview of the work done building a solution to enable access to the environment and flow data from existing HVAS with no digital data acquisition or transmission capabilities. The solution involves IoT, Cloud, and DLT (Distributed Ledger Technology) that are some of the technologies that are enabling and driving Digital Transformation. The main components are an electronic device called RDMA and a software application (called Enviro-AirSampling - ENVAS). This solution was implemented and is being used for air quality studies at the Laboratorios de Tecnogestion of the Ministry of Industry, Energy, and Mines (MIEM) of Uruguay. It brings valuable information to scientists for interpretation or air sampling results and a more accurate determination of MC and SMC. The RDMA was designed ab-initio to be an easy add-on to the existing Tisch Environmental HVAS owned by MIEM.

Developed IoT Device: RDMA

The RDMA measures and calculates volumetric flow rate, normalized volumetric flow rate, sampled air volume, normalized sampled air volume, external air temperature (T), dew point, relative humidity (H), ambient pressure (P), the average of T, H, P during each air sampling period and start and stop time of each sampling period. The measurements, internal (device) state information, and location coordinates are sent by the RDMA device to a custom implemented application called ENVAS running on the Cloud (Fig. 1). Communication between device and the Cloud is done using TCP/IP and TLS, with MQTT protocol used to transport data in JSON format.

The main components of the RDMA are an IP65 rated enclosure (containing the main electronics), an UV coated 3D printed external sensor shielding (housing the environmental sensors), an air pump on/off sensor, airflow signal input, and a power input cord. The microprocessor selected for the RDMA was the Espresiff ESP32-D2WD SoC (Xtensa dual-core 32-bit LX6) [11]. As the HVAS gets its energy from the power grid, this energy source is also available for the RDMA. WiFi was the selected RF communication system because the HVAS is used in urban areas with access to WiFi (and alternatively data transmission over 3G/4G). The RDMA (enclosure and the sensor shielding) is fixed with screws to the aluminum housing of the Tisch Environmental HVAS (Fig. 1). In this deployment, the target HVAS was the Tisch Environmental model TE-6070D-BLX-2.5 non-FRM high volume ambient PM2.5 air sampler (for gravimetric method) featuring mass flow controller (MFC) [12]. Tisch Environmental (Cleves, OH) is a company that designs and manufactures high-quality benchmark high volume air samplers (US EPA Federal Reference Method (FRM) and non-FRM).

Figure 1: Left: IoT device RDMA attached to Tisch HVAS. Center: Detail of enclosure and environmental sensor shielding. Right: Simplified diagram of the solution architecture showing RDMA devices, Enviro-AirSampling (ENVAS) application, other Cloud services, IOTA network, end-users, and administrator.

Figure 1: Left: IoT device RDMA attached to Tisch HVAS. Center: Detail of enclosure and environmental sensor shielding. Right: Simplified diagram of the solution architecture showing RDMA devices, Enviro-AirSampling (ENVAS) application, other Cloud services, IOTA network, end-users, and administrator.

 

Developed Application: ENVAS

The web-access cloud-based application (Fig. 2) enables the access of users to real-time environmental and flow data from each HVAS air sampler in the network, visualization of plots of the variables associated with each HVAS, location of HVAS in a map, and other environmental information (wind speed and direction, precipitations, and climate forecast). ENVAS processes the data received from each IoT device and sends automatic messages (e-mail) to users when needed.  Some of the e-mail reported events are: HVAS start/end sampling, volumetric flow out of range, and automatic final report of each sampling experiment. Also, the application records a summary of the key variables of each sampling experiment in JSON format in the IOTA Tangle [13] (Fig. 3). This functionality is useful for traceability, audit, and quality system purposes.

Figure 2: Screenshot of ENVAS main screen showing real-time data for one sampler (some private data is hidden in black).

Figure 2: Screenshot of ENVAS main screen showing real-time data for one sampler (some private data is hidden in black).

 

Figure 3: Screenshot of a transaction registered in the IOTA Tangle (accessed using Tangle Explorer) (some private data is hidden in black).

Figure 3: Screenshot of a transaction registered in the IOTA Tangle (accessed using Tangle Explorer) (some private data is hidden in black).

 

Final Considerations

This solution has been used satisfactorily for almost a year to date. Updates and improvements are being made as more field experience is gained and feedback is received from users during operation.

As a summary, the main benefits obtained with this solution are: a) enables more accurate measurement of total sampled volume and normalized total sampled volume, b) real-time visibility of the sampling process for each sampler in the network, c) brings complementary information useful for the interpretation and analysis of sampling results, d) automatic report generation for each sampling experiment, e) e-mail alarm/event notifications, f) DLT storage capability (useful for QS, traceability or audit purposes).

Future work may involve hardware improvement (for example, to enable the acquisition of new variables), integration of other new IoT devices, and the introduction of predictive algorithms trained using historical data series.

Acknowledgment

To Roberto Suarez-Antola for technical advice in particle transport in the atmosphere and mathematical modeling related issues during the design of the solution.

References

  1. Health and Environmental Effects of Particulate Matter (PM), United States Environmental Protection Agency (last accessed: September 2020) Available at https://bit.ly/2FmOKMF
  2. "Health Aspects of Air Pollution with Particulate Matter, Ozone and Nitrogen Dioxide" (2003), World Health Organization.
  3. Baron P. and Willeke K., “Aerosol measurement: Principles, techniques and applications”, 2nd Edition, Wiley-Interscience, 2005.
  4. Odino R., Gabrielli A., Piuma L, Suarez-Antola R., “Composición Elemental de las Partículas del Aire de una Zona de Montevideo – Uruguay”, 7th workshop of the International Center for Earth Sciences, 2011.
  5. Odino R., Reina E., Gabrielli A., Piuma L., “Elemental composition of airborne PM2.5 and PM10 during volcanic ash intrusion in Montevideo, Uruguay”, XRF Newsletter, N°23, September 2012.
  6. Seinfeld J. and N. Pandis “Atmospheric Chemistry and Physics: From Air Pollution to Climate Change”, 3rd Edition, Wiley, NY, 2016.   
  7. “Guidance for Network Design and Optimum Site Exposure for PM2.5 and PM10” (EPA-454/R-99-022)
  8. USEPA: 40 CFR Part 58, Appendix D to Part 58 - Network Design Criteria for Ambient Air Quality Monitoring
  9. USEPA: 40 CFR Appendix J to Part 50 - Reference Method for the Determination of Particulate Matter as PM10 in the Atmosphere
  10. USEPA: 40 CFR Appendix L to Part 50 - Reference Method for the Determination of Fine Particulate Matter as PM2.5 in the Atmosphere
  11. ESP32 Series Datasheet, V 3.2 (last accessed: September 2020) Available at https://bit.ly/3m98Niw
  12. TE-6000 Series Operations Manual Rev1 8-10-2010, Tisch Environmental Inc. (last accessed: September 2020) Available at: https://bit.ly/2Fk788M
  13. IOTA Homepage (last accessed: September 2020) Available at https://bit.ly/3k4ddoQ

 

Diego Suarez BagnascoDiego Suárez Bagnasco Received is BSc. Electronic Eng. in 2002 (Catholic University of Uruguay), PgD Marketing in 2006 (Catholic University of Uruguay), MSc. Biomedical Eng. in 2010 (Favaloro University, Argentina), Ph.D. Biophysical Sciences in 2017 (PEDECIBA, Univ. de la República, Uruguay). IEEE member. An active researcher in the framework of the National Investigators System (SNI) from the National Agency for Investigation and Innovation (ANII) of Uruguay. Currently, he is head at Electronics and Nuclear Instrumentation Laboratory of Laboratorios de Tecnogestión, Ministry of Industry, Energy and Mines (MIEM) of Uruguay, and founder and director at Omnia, Uruguay (contact: dsuarez@ieee.org)

 

 

Blockchain-enabled Edge Analytics for a Mobile Universal Biotesting Station

Sergii Kushch, Martin Hayes and Eoin O’Connell
September 15, 2020

 

The impact of COVID-19 on manufacturing, educational, and a varied assortment of locations has been immense. The use case described below proposes a Blockchain-enabled infrastructure for deployment on an edge gateway to secure a universal biotesting station that is capable of processing sensitive personal biodata at scale. The unit can be deployed for access to a health center, manufacturing facility, or a Higher Education laboratory facility.

IoT technologies must be trustworthy so they can enable digital transformation. The goal of a connected system is to provide a better understanding of the physical world, leading to more accurate decision making such as the automation of physical tasks based on historical information and knowledge, or improved outcomes. Distributed ledger technology (DLT) provides a trusted, immutable ledger on which organizations can transmit and store valuable information based on internal operations or interactions with the organization’s environment, including the Industrial Internet of Things (IIoT) devices.

The use case that is presented here is a platform-agnostic approach to the ad-hoc connection of new IoT devices that allows for the secure construction of a holistic, patient-centered digital record. There is a growing requirement that institutions take steps to assure that access is only granted to citizens that pose no threat to the health of others. The Internet of Things (IoT) is a relatively simple concept that seems to be designed for such a public health crisis. IoT uses sensors to collect data in the physical world then transports this data to web-based compute and storage platforms [1]  analyses the biodata using trusted algorithms and is capable of adaptation using machine learning and artificial intelligence techniques [2]. The goal of such a connected access system is to provide a better understanding of the individual who is making the request and requires more accurate decision making that is informed by the use of automation, broad-spectrum testing, synthesis of historical information and knowledge that can yield improved outcomes for all citizens. A requirement is that any proposed IoT solution be secure and accurate. IoT security concerns can be found on all levels of the IoT stack. Trust is always one of the main challenges with any new technology and any change.

At the same time, edge computing is providing the capability to transform the way data is being handled, processed, and delivered from millions of devices around the world – in real-time. The explosive growth of internet-connected devices – the IoT ‘ecosystem’ – along with new applications that require real-time computing power is driving the mass deployment of edge-computing systems. Faster networking technologies, such as 5G wireless, enable edge computing systems that accelerate the creation or support of real-time applications, such as video processing and analytics, self-driving cars, artificial intelligence, and robotics. Edge enabled use cases now exist that are capable of correlating data from multiple input protocols and mediums, undertaking local processing such as data compression, and/or real-time decision making using Artificial Intelligence algorithms [3], providing connectivity to TCP/IP via IP Networks, Wi-Fi or radio access networks (RAN) and interfacing to proprietary networking and industrial connectivity protocols.

The use case here considered is a universal biotesting station that is proposed to enable certifiably safe entry to a secure/important resource. Such a resource can range from a Manufacturing facility where certifiably safe operation is required to a University teaching laboratory where public health guidelines mandate that tracing protocols are in place for safe operation. This use case exhibits the following features: Continuous operation, Minimal Operator/Technician intervention, High testing rates/subject throughput. Implementation of institutional access protocols. Local dynamic determination of rule-based testing procedures depending on real-time diagnostic data, high throughput, minimal outage, latency issues, or WAN connectivity to the supporting Cloud infrastructure. Personal data is stored in a protected, secure fashion where access is strictly controlled using Distributed ledger technology (DLT).

DLT is proposed as a mechanism for the development of a trusted, legally compliant, immutable ledger on wherein organizations can transmit and store valuable information based on internal operations or interactions within an organization’s environment, including the IIoT devices.

Healthcare, Secure Access, Public Health, and DLT

The design of access control mechanisms for healthcare systems is challenging. These mechanisms deal with sensitive data and must guarantee confidentiality within a statutory framework. Specific access control policies must be in place to access a subject’s personal health records (PHRs). Moreover, the PHR’s integrity must be guaranteed and cannot be modified without a clear, immutable ledger entry. evidence. Cohort safety, be that of a patient, student, or classmate must be guaranteed and thus Doctors, Health service professionals, or Higher Education Authorities must be able to access information quickly and without interruptions in case of health care emergency or for contact tracing purposes as part of a public health intervention. Flexible and frictionless access control will always introduce attendant moral hazard problems. Indeed, as reported in [4], 58% of attacks involved insider or ‘Person in the middle’ attacks - insiders are often the biggest threat to an organization. The motives range from simple curiosity about a friend or family member, or students wishing to game the laboratory access system, right through to a malevolent intent to damage a patient by revealing some sensitive data or financial gain (e.g., receiving an insurance payment by using a stolen diagnosis). Thus, access control mechanisms must strike the right balance between throughput, permissions, and restrictions. Given the aforementioned issue with insiders, all the access of a healthcare organization must be tracked through a secure non-repudiation logging system that will identify cases of privilege abuse in such a fashion that it will protect company interests while simultaneously deterring employees from improper behavior.

A biotesting use case requires the ability to dock multiple devices using different communication standards and protocols (i.e., Bluetooth, WiFi, LoRA, and 4G/5G). [5, 6]. However, there are still many open questions that need to be addressed so that any proposed solution is scalable, reliable, and secure for integrated use together (Figure 1).

Figure 1: A distributed Edge deployment.

Figure 1: A distributed Edge deployment.

 

The design prototype supports multiple physical layer communication protocols received from a variety of medical-grade test-station devices (e.g. basic temperature, heart rate, blood pressure, saliva testing, blood tests, visual cognition, etc.) that are securely labeled for edge-based analysis and network layer training before transmission to a central database. Edge analytics facilitates testing regimes that can evolve dynamically according to clinician requirements but that are also optimized to make the best use of Health Service Resources. The digital record database will be secured using a distributed ledger that is deployed on all testing devices via the edge gateway (Figure 2). Case records will evolve so that deviations can be monitored over time, class lists are checked dynamically so that access protocols are updated and a secure registry is kept in line with the public health tracing regimen.

Figure 2: Prototype Automated Health Testing Station.

Figure 2: Prototype Automated Health Testing Station.

 

Final Remarks

This project will detail the specifics of system design and implementation, will detail preliminary use case feedback in a busy higher education setting, and will also detail performance and validation information from the new state of the art SFI funded Confirm Smart Manufacturing 5G testbed that is being rolled out at the University of Limerick during Autumn 2020. The authors will provide test results including data throughput to a central cloud-based registry, access decisions based on a significant deviation of subject data from stored biodata, new testing as per clinical need, time-series information being made available to management in realtime with minimal additional impact on lab activity.

References

  1. J. Chin, V. Callaghan and I. Lam, "Understanding and personalising smart city services using machine learning, The Internet-of-Things and Big Data," 2017 IEEE 26th International Symposium on Industrial Electronics (ISIE), Edinburgh, 2017, pp. 2050-2055, doi: 10.1109/ISIE.2017.8001570.
  2. E. O’Connell, D. Moore, and T. Newe, "Challenges Associated with Implementing 5G in Manufacturing," in Telecom, 2020, vol. 1, no. 1, pp. 48-67: Multidisciplinary Digital Publishing Institute.
  3. Vermesan, O. et al., 2017. “Internet of robotic things : converging sensing/actuating, hypoconnectivity, artificial intelligence and IoT Platforms.” In: J. B. Ovidiu Vermesan, ed. Cognitive Hyperconnected Digital Transformation : Internet of Things Intelligence Evolution. Gistrup: River Publishers, pp. 97-155
  4. Protected health information data breach report. Verizon (last accessed: September 2020), Available at https://vz.to/3k9eMCd
  5. S. Kushch and F. Prieto-Castrillo, "Blockchain for Dynamic Nodes in a Smart City," 2019 IEEE 5th World Forum on Internet of Things (WF-IoT), Limerick, Ireland, 2019, pp. 29-34.
  6. S. Kushch, S. Ranise and G. Sciarretta, "Blockchain Tree for eHealth," 2019 IEEE Global Conference on Internet of Things (GCIoT), Dubai, United Arab Emirates, 2019, pp. 1-5.

 

Sergii KushchSergii Kushch holds a Ph.D. in Computer Science and a Master degree in Radio Engineering (Hons.) and is a Senior Member of the IEEE. He worked on the development and optimization of microchips structure, human factor assessment in cybersecurity, and on the development of Blockchain technology for use in IoT, WSN, and mobile devices. Currently, Dr. Kushch works as a Senior Research Fellow in the Irish National CONFIRM research centre and the Department of Electronic & Computer Engineering at the University of Limerick, focused on the implementation of new wireless technologies in industrial environments. Additionally, he works with modifications of Blockchain technology for securing and protecting personal data and developing an implementation of Blockchain for WSN and IoT.

 

Eoin O ConnellEoin O’Connell is a Funded Investigator in the Irish national research centre CONFIRM. He is an academic in the Department of Electronic & Computer Engineering at the University of Limerick. Dr. O’Connell holds a Ph.D. in the wireless integration of fibre optics sensors, he has a Master in Business Administration (MBA) and a Bachelor’s degree in Telecommunications (Hons.) from the University of Limerick. Dr. O’Connell is also a graduate from Limerick Institute of Technology, where he received an NCEA Diploma and an NCEA Certificate in electronic engineering. His research interests are in the areas of Cellular communications, IoT, System Security vulnerabilities, edge processing techniques, Data Management, sensor development, and Wireless Sensor Networks with a particular focus on the corresponding issues of interoperability and scalability.

 

Martin HayesMartin Hayes is Head of the Electronic & Computer Engineering Department (ECE) at the University of Limerick and is a funded investigator with CONFIRM, the SFI funded institute for Smart Manufacturing Research in Ireland. Previously, Martin acted as a Research Director for the Intelligent Power Management and Control branch of the Circuits and Systems Research Centre within the Department. He is the UL representative on the Irish Signals and Systems subcommittee of the Royal Irish Academy. Martin holds a BE and ME from the University of Limerick and a Ph.D. from Dublin City University. His research interests lie generally in the area of Artificial Intelligence and Systems Theory. Martin’s current funded research work focusses on the intelligent use of system resources within smart manufacturing or educational environments.

 

 

Adoption of Real-Time Machine Learning for Cyber Risk Assessment in IoT Environments

Mohamed Rahouti and Moussa Ayyash
September 15, 2020

 

The Internet of Things (IoT) has emerged in the past as Internet extension and significantly changed our world. A tremendous amount of IoT applications has highly eased people’s daily lives and improved resource usage and allocation, e.g., power bank sharing, bike sharing, etc. However, the ingrained openness of underlying wireless systems renders these IoT entities vulnerable to a broad range of cyber risks, e.g., vulnerabilities of spectrum that can be a source of adversarial inference.

Moreover, as our reliance on wireless/connected devices and radios has also been dramatically increasing, public safety, business operations, socializing and navigation as well as critical national communication infrastructures have become more vulnerable to cyber threats. Hence, policymakers and industries have recently begun to perceive that the expansion of connected devices and their cyber susceptibility results in a large malicious and inferential risk. As a result, there is a need for identifying and comprehending these risks in order to elaborate efficient security solutions. Lastly, the economic impact of IoT devices and associated security vulnerabilities are further growing by artificial intelligence (AI) integration into human-computer interaction (HCI), e.g., banking, insurance, etc. Consequently, cyber risks are growing in term of both frequency and acuteness.

Why Real-Time Machine Learning (RTML)?

Exploration and analysis of data volumes of IoT entities, including, but not limited to ubiquitous cameras and sensors, is bountiful. The overflow of such data enables a grand scope for exploring the associated cyber risks, especially with the assistance of machine learning mechanisms. Unlike traditional machine learning, where the Cloud is an option for integrating its modular application, real-time machine learning (RTML), aka online or incremental learning, can be of a great potential herein as the cloud-centric IoT environments are expanding following data movement and overhead (e.g., energy concerns). Therefore, it can be advantageous to leverage reliable IoT entities to conduct various inference operations, instead of constantly transferring massive amounts of raw data to the cloud. Additionally, the non-dynamic nature of machine learning model training can be inefficient in processing dynamic IoT data in-situ systems, and thus weak accuracy of predictions may occur. Furthermore, another insightful benefit in deploying IoT-enabled cyber threats detectors using RTML is the low power platform and fast prototyping.

Model Visioning

As a branch of machine learning, RTML is capable of enhancing the classification and prediction accuracy in malware and cyber risks detection based upon learners, whose outputs can be integrated for alert decision. Here, ensemble learning and joint decision process, if integrated, present another opportunity to enhance the decision accuracy of RTML classifiers and assist with devising learning techniques while guaranteeing better generalization performance.

The key idea behind designing an efficient IoT-enabled threat detector using RTML is based upon identifying appropriate features to describe raw data. As depicted in Figure 1, a feature reduction module can be exploited after the feature extraction stage to cut down the number of low-level features. Here, an examination of the correlation attribute can be used on the training dataset to identify essential architectural parameters and different characteristics of applications. Building upon this, a feature scoring technique can be used to score the categorized features by their relevance to the detector target variable.

Serving the Predictions

So far, only incremental learning models are being considered, but we also want to serve real-time predictions. A recommendation here is to use a linear learner. To make a prediction, only the current context and the parameters we have learned are needed (i.e., no need to know the parameters, architecture of the network including the choice of activation functions). In a typical Redis use case scenario, the linear learner can learn on examples through the Redis FIFO queue, it can then periodically save the model parameters back into Redis. The key point here is that saving (and indeed retrieving) the parameters should be an atomic operation (e.g., using the SET/GET commands in the case of Redis example). For more complex interactions, a locking mechanism may be used, whereas the key requirement is that all model parameter changes are updated in real-time to avoid wrong or low accuracy predictions.

Figure 1: Real-time machine learning-based cyber risk detection framework.

Figure 1: Real-time machine learning-based cyber risk detection framework.

 

It is sensible to separate the logic of training and inference. Assuming the model parameters are updated atomically, serving up predictions takes place as follows; (1) preprocess the context to make it suitable for inference, (2) retrieve model parameters atomically, (3) calculate the linear combination of model weights and context features, and (4) return result. Another key reason for using an in-memory store becomes apparent as we look at the logic here: for every prediction request, we need to retrieve new model parameters, since they may change rapidly. A traditional relational database solution is likely to be slow for such online learning purposes.

Future Directions and Trends: Flexibility Comes at a Price

As IoT applications have become more diverse today in evolving smart cities, RTML can be exploited in IoT environments to improve their reliability and security in particular. Existing machine learning and deep learning solutions are computationally and energy expensive. A grand challenge here is how to merge or integrate existing solutions of traditional machine learning into an incremental online learning/RTML to boost the overall accuracy of cyber risk detection. A combination of these mechanisms can guarantee a balance between the cost of energy and computation and accuracy of detection results which will be a vital assurance for next-generation IoT systems. Consequently, the integration of RTML in such systems will require overhauling the entire stack of communication within the application and physical layers. Lastly, traditional security threats (e.g., malware) detection and prevention mechanisms, such as signature and/or semantics-based detectors, are software solutions and result in a notable computational cost and overhead. Therefore, threat detection mechanisms could be improved by deploying simplified classifiers regardless of the detection approach is being deployed.


 

Mohamed RahoutiMohamed Rahouti received the M.S. degree and Ph.D. degree from the University of South Florida in the Mathematics Department and Electrical Engineering Department, Tampa, FL, USA, in 2016 and 2020, respectively. He is currently an Assistant Professor, Department of Computer and Information Sciences, Fordham University, Bronx, NY, USA. He holds numerous academic achievements. His current research focuses on computer networking, software-defined networking (SDN), and network security with applications to smart cities.

 

Moussa AyyashMoussa Ayyash (M'98–SM'12) received his B.S., M.S., and Ph.D. degrees in Electrical and Computer Engineering. He is currently a Professor at the Department of Mathematics and Computer Science, Chicago State University, Chicago. He is the Director of the Center of Information and Security Education and Research (CINSER). His current research interests span digital and data communication areas, wireless networking, visible light communications, network security, Internet of Things, and interference mitigation. Dr. Ayyash is a member of the IEEE Computer and Communications Societies and a member of the Association for Computing Machinery. He is a recipient of the 2018 Best Survey Paper Award from the IEEE Communications Society.

 

 

New Generation Alternative Sensors for IoT

Euclides Lourenço Chuma
September 15, 2020

 

The Internet of things (IoT) is a system of interrelated computing devices and mechanical and digital machines provided with unique identifiers (UIDs) and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction. Many types of sensors are used in IoT for various applications, such as measuring temperature, humidity, pressure, acceleration, proximity, etc. However, a new generation of alternative sensors is being developed, which can be connected to IoT so that they are used in the creation of autonomous vehicles, in health monitoring, and other research areas.

This new generation of sensors together with artificial intelligence gathers more data and accurate information to be used in IoT systems that will make them more powerful, bringing benefits to mankind.

Microwave and mmWave Radar Sensors

Microwave and mmWave radar sensors can be used in the applications for detecting, positioning, and tracking of humans, animals, or generally moving objects without compromising their privacies. Microwave and mmWave radar facilitate the rapid detection of the position of nearby objects with high sensitivity and high accuracy even when there are environments interferences.

Microwave and mmWave radars have some unique features and they work within a particular range, i.e., operate in the range of few centimeters to a few hundred meters without a direct line of sight (for example, through drywall or plywood). Further, they are also highly adaptable to environmental conditions such as darkness, sunshine, smoke, fog, or haze [1].

Technology companies like Infineon and Vayyar are manufacturing mmWave radar sensors that use frequencies above 60 GHz and are sensitive enough to sense even a heartbeat [2-3]. Therefore, it is possible to use this type of sensor to gather a lot of data and information.

Therefore, these types of sensors can replace traditional cameras in many situations with so many advantages. The main advantage is that people’s privacy is protected during its operation and its ability to operate efficiently in environments where there are light and sound interferences as well as adverse atmospheres.

Figure 1: Some applications of mmWave radar sensors.

Figure 1: Some applications of mmWave radar sensors.

 

Light Detection and Ranging

Light detection and ranging (LiDAR) is a device that combines pulse of light (laser) and sensors of light (photodetector) to make digital 3-D representations of the target or surrounding environment. The LiDAR compute distances (ranging) by illuminating the targets with laser light and measuring the reflection with a sensor. Targets made up of different materials at different distances will generate different laser return times and wavelengths and these data can then be used to make 3-D representations of the target [4-6].

LiDAR is commonly used in many applications to generate high-resolution 3-D maps, such as in surveying, geography, geology, seismology, forestry, atmospheric physics, laser guidance, and airborne laser swath mapping.

Multispectral and Hyperspectral Imaging

It is known that the human eye can view or see the color of visible light mostly in three wavelength bands (red, green, and blue). But it is possible to obtain and view images in the bands that are beyond the visible region using new technologies. Multispectral and hyperspectral imaging divide the spectrum into many more wavelength bands., The recorded spectra in hyperspectral imaging have fine wavelength resolution and cover a wide range of wavelengths (Bands), while multispectral imaging measures spaced spectral bands.

Hyperspectral sensors detect objects using a wide band of the electromagnetic spectrum and as a result, objects leave unique “fingerprints” in the electromagnetic spectrum, which are known as spectral signatures. These “fingerprints” enable identifying the materials a scanned object or target is made of. Therefore, it is possible to know and obtain the chemical composition of the material at distance with a hyperspectral camera [7-8].

AI, Sensors, and IoT

The new generation of sensors as demonstrated earlier generate a huge amount of data requiring intelligent processing systems. Two systems are used for processing these huge amounts of information using artificial intelligence: cloud and edge [9-11].

Companies like Xilinx and Nvidia are developing solutions for both systems. When processing is carried out in the cloud, special cards are installed on the servers to speed up the deep learning processing method. When processing is performed on the edge, close to the sensor, processors specialized in deep learning with low energy consumption is used. In both cases, IoT solutions can accelerate the artificial intelligence process of the system thereby improving data quality.

Figure 2: AI Cloud vs AI Edge.Figure 2: AI Cloud vs AI Edge.

 

Final Remarks

Sensor fusion is the combination of sensory data received or data derived from disparate sources such that the resulting information has less uncertainty than that would be possible when these sources were used individually. It is important to know that several categories or levels of sensor fusion are commonly used. The sensor fusion level can also be defined based on the type of information used to feed the fusion algorithm [11-12].

Apart from the traditional sensor fusion model, currently, there are many modern methods based on artificial intelligence that can simultaneously process sensor data in many channels (such as the hyperspectral image with hundreds of bands) and merge relevant information to produce classification results.

In summary, the most important thing is to know that many innovative technologies are emerging that can be applied to IoT systems so that a more robust and effective system can be built to serve mankind.

References

  1. Z. Zhao, et al., “Point Cloud Features-Based Kernel SVM for Human-Vehicle Classification in Millimeter Wave Radar”, IEEE Access, 2020, v.8
  2. S. Dong, et al., “Doppler Cardiogram: A Remote Detection of Human Heart Activities”, IEEE Transactions on Microwave Theory and Techniques, 2020, v.68, i.3
  3. T. Lauteslager, et al., “Coherent UWB Radar-on-Chip for In-Body Measurement of Cardiovascular Dynamics”, IEEE Transactions on Biomedical Circuits and Systems, 2019, v.13, i.5
  4. J. Shaw, “Lidar Instruments and Applications”, IEEE Conference on Lasers and Electro-Optics (CLEO), 2017
  5. A. M. Wallace, A. Halimi, G. S. Buller, “Full Waveform LiDAR for Adverse Weather Conditions”, IEEE Transactions on Vehicular Technology, 2020, v.69, i.7, pp.7064-7077
  6. T. Sang, S. Tsai, T. Yu, “Mitigating Effects of Uniform Fog on SPAD LiDAR”, IEEE Sensors Letters, 2020
  7. W. Sun, Q. Du, “Hyperspectral Band Selection: A Review”, IEEE Geoscience and Remote Sensing Magazine, 2019, v.7, i.2
  8. M. Parente, J. Kerekes, R. Heylen, “A Special Issue on Hyperspectral Imaging”, IEEE Geoscience and Remote Sensing Magazine, 2019, v.7, i.2
  9. A. M. Ghosh, K. Grolinger “Deep Learning: Edge-Cloud Data Analytics for IoT”, IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), 2019
  10. D. Liu, et al., “HierTrain: Fast Hierarchical Edge AI Learning With Hybrid Parallelism in Mobile-Edge-Cloud Computing”, IEEE Open Journal of the Communications Society, 2020, v.1, pp.634-645
  11. M. Chiang, T. Zhang, “Fog and IoT: An Overview of Research Opportunities”, IEEE Internet of Things Journal, 2016, v.3, i.6
  12. M. A. Al-Jarrah, et al., “Decision Fusion for IoT-Based Wireless Sensor Networks”, IEEE Internet of Things Journal, 2020, v.7, i.2
  13. P. Ferrer-Cid, et al., “Multisensor Data Fusion Calibration in IoT Air Pollution Platforms”, IEEE Internet of Things Journal, 2020, v.7, i.4

 

Euclides Lourenco ChumaEuclides Lourenço Chuma earned a degree in Mathematics (2003) from the University of Campinas (UNICAMP), a graduate degree in network and telecommunications Systems (2015) at INATEL, and MSc in electrical engineering (2017) at UNICAMP, and a PhD in electrical engineering (2019) at UNICAMP, SP-Brazil. His research interests are microwave, millimeter-wave, photonics, bioengineering, sensors, wireless power transfer, and telecommunications.

 

 

Comments

2020-11-03 @ 6:30 PM by Amirthalingam, Balamurugan

Hi,

Would like to discuss more on AI sensor .Please let me know the medium to communicate on the same
 
Thanks,
Bala.

SmartTags-enabled Fast-Moving Consumer Goods: Creation and Management

Suparna De and Nenad Gligoric
November 11, 2020

 

New types of sensors that are printable and can collect, sense, and read environmental parameters of relevance to the product and its use are becoming reality. For fast-moving consumer goods (FMCG) this represents the basis for the creation of a new generation of supply chains which, in combination with GS1 Digital Link global specifications standard, makes it possible to identify each product item, track it and monitor it on an item level.

These capabilities offer opportunities for informed management of assets and innovative consumer engagement by transforming consumer goods into digital assets. The TagItSmart [1] project funded by the Horizon 2020 EU research program, has researched and piloted the use of QR-codes printed with functional inks and printed NFC (Near Field Communication) tags with sensing capabilities to create ‘SmartTags’. The project made solid ground on the research topic and development of the SmartTag technology, which is, because of its simplicity and standardized approach, adopted and used by many industry players.

Environment-reactive SmartTags with Functional Inks

With NFC and RFID (Radio Frequency Identification) solutions proving expensive for consumer-packaged goods, TagItSmart has pioneered the use of dynamic QR-codes that are printed on consumer goods using environmentally reactive inks that change appearance according to selected conditions, e.g. temperature, humidity, light intensity. The growing use of smartphones equipped with cameras facilitates an off-the-shelf solution for scanning the dynamic QR-code information as well as lifecycle tracking of the item through seamless observation measurements of the generated smart tag data. This visual change on the product packaging, as well as individual product identification using the GS1 Digital Link Standard [2] allows not only tracking of every individual item from the factory to recycling, but also enables additional information about the item to be communicated to consumers. This includes visual clues to consumers that a product is at its optimum consumption temperature, monitoring of condition-sensitive goods during transport (e.g. vaccines, meat, and dairy goods), alerts to retailers that an item is close to its best-before date, and also if an item has been properly recycled. In a nutshell, this opens up exciting possibilities for consumer engagement, supply chain tracking, and product recycling.

Service Platform for SmartTags Management

The ecosystem around such SmartTags-enabled FMCG is driven by the TagItSmart service platform that includes semantic models for capturing the characteristics of the SmartTags and reusable semantically enabled workflows for their creation and management. The platform includes SmartTags (either QR-codes with functional inks or NFC tags) which augment products or physical entities with sensing capabilities and allow their virtualization as virtual entities for monitoring the product lifecycle.

The approach uses semantic repositories for modeling the physical world objects as virtualized entities as well as users of the platform, together with their contextual information and location to enable internal or external services to be composed. The resulting workflows are the result of a composition process which starts on a service template front-end, exploiting the well-known Node-RED wiring tool, involves a service manager which oversees resolving service parameters semantically, based on a context (e.g. users and locations) and outputs a fully-fledged workflow execution engine based on AWS Lambda instances in charge of enforcing the service composition and hereafter handling the requests in an efficient and scalable fashion. The solution has been implemented in Node.js and empowers users to deploy new services, virtualizing the Internet of Things real-world objects such as industry FMCG in their consumer-oriented ecosystems from production to recycling. This Composition-as-a-Service (CaaS) approach is smart, efficient, and fully customizable with any service. It allows the definition of on-the-fly new service functions and to put them in line in the execution chain or handle service unavailability or errors in the composition process in a user-friendly and verbose fashion. The end-to-end workflow in the service platform is illustrated below with an example of the Magnum ice cream use case developed in TagItSmart, which implemented a smart, environmentally reactive packaging solution as well as an end-user experience application.

SmartTag-enabled Ice Cream

The TagItSmart technology can be showcased in numerous scenarios, with an ice cream scenario being noteworthy, due to the possibility to showcase most of the SmartTag’s technology benefits. An irreversible functional ink is used to create the SmartTag, capturing temperatures larger than −10 °C with a timestamp set to 30 minutes. The sensing capability of the SmartTag, supported by the ink, is specifically designed for the ice cream scenario to satisfy specific requirements, as shown in Fig. 1:

Figure 1: Ice cream SmartTag environment parameter definition.

Figure 1: Ice cream SmartTag environment parameter definition.

 

The SmartTags are encoded with a QR code to identify and encode the temperature-sensitive property and an image to show the user that the ice cream is good to be consumed. Next, a mobile phone with a web browser is selected as the scanner for the SmartTag. The identifiers for the designed SmartTags are created and the resulting tags are printed.

The next step is to model the objects that will be labeled with the SmartTags, i.e. the ice cream virtual entity, which is achieved through a VE-front end in the platform (Fig. 2). The front end is driven by the VE semantic model (detailed in [3]).

Figure 2: Application front end for describing the Ice cream VE.

Figure 2: Application front end for describing the Ice cream VE.

 

The user experience with the SmartTags is also an important part of the application development process. It will typically involve how the user is going to scan the SmartTag and how the information generated is going to be used and presented. This is developed as a Web application and integrates the mobile phone scanner and application workflow through the correspondent SDKs and libraries, as shown in Fig. 3.

Figure 3: The Magnum user experience application.

Figure 3: The Magnum user experience application.

 

To ensure the smooth creation of the tags, the TagItSmart project created a semantic model for describing the environment-reactive properties of the QR codes and NFC electronic tags. The model incorporates the materials’ composition of the physical products, the characteristics of the functional ink, such as what environmental conditions it reacts to (e.g. temperature, humidity, time-lapse, etc.), the accompanying relevant state changes (e.g. color and visibility changes), as well as observation measurements together with their spatial description. The semantic model is detailed in [4] and the instantiation of the SmartTag for tracking the ice cream lifecycle is shown in Fig. 4.

Figure 4: Semantic instantiation of SmartTag for ice cream lifecycle tracking.

Figure 4: Semantic instantiation of SmartTag for ice cream lifecycle tracking.

 

The semantic model guides the selection of the required functional ink for the QR code to be printed, based on what environment parameter needs to be monitored, together with the identification of the thresholds and tolerance levels.

References

  1. H2020 TagItSmart Project, Grant Agreement 688061. https://bit.ly/3n7saIy.
  2. Digital Link GS1 Standard. [Online]. Available: https://bit.ly/2U4ssDd
  3. "D4.1. Resource and Service Modelling Specification," in "TagItSmart Public Deliverable," TagItSmart! Smart Tags driven service platform for enabling ecosystems of connected objects, 2017. [Online]. Available: https://bit.ly/36gwRsz
  4. N. Gligoric et al., "SmartTags: IoT Product Passport for Circular Economy Based on Printed Sensors and Unique Item-Level Identifiers," Sensors, vol. 19, no. 3, 2019.

 

Suparna DeSuparna De is a Senior Lecturer in Computer Science and Networks at the University of Winchester, UK. She obtained her Ph.D. in Electronic Engineering from the University of Surrey. She led the work on developing semantic models for capturing the characteristics of the Smart Tags and to provide decision-support mechanisms for connecting their lifecycle data to semantically enabled workflows in the H2020 TagItSmart project. Her research interests include large-scale data analytics in Web of Things scenarios, machine learning, social computing, and Semantic Web technologies.

 

Nenad GligoricNenad Gligoric is one of the pioneers of the IoT scene in Serbia, working as a software engineer in Ericsson and as a project manager in DunavNET on more than 10 EU, FP7 and H2020 projects. He was one of the technical managers of the H2020 TagItSmart project.

 

 

 

 

Smart Building PLC Testbed Leveraging IoT Network Technologies

Vaishnavi Rajini Mohan, Krunal Patel, Manoj Kumar, Ramkrishna Pasumarthy, and Paventhan Arumugam
November 11, 2020

 

 

IEEE 1901 standard for narrowband Power Line Communications (PLC) over indoor and outdoor electrical wiring supports data rates of up to 500 kb/s. Also, recent amendments like IEEE 1901a-2019 provides enhancements for the Internet of Things (IoT) applications.

There are many worldwide successful PLC deployments for smart grid applications such as smart metering, substation automation, street lighting, BMS, EV charging infrastructure, etc.

In India, there are small-scale PLC deployments such as the AMI pilot project by Tripura State Utility[1]. However, the potential of PLC in the modern context of IoT and other application domains such as smart grid deployments is not fully exploited in India. IEEE Standards Association (IEEE SA) under the Industry Connections (IC) Activity has initiated a program towards setting up multiple Power Line Communications (PLC) testbeds in India engaging various stakeholders from Industry, Academia, and Utilities. In this article, we report how companies can leverage PLC for a smart building use case that is being executed as part of the IEEE PLC Testbed in India initiative.

Smart Building Systems

Building systems are a  significant source of energy consumption. Traditional ways of monitoring and control rely on centralized schemes, which have become obsolete. The world is moving towards distributed systems, assisted by significant advances in IoT enabled sensing technologies.  The need is to combine the vital information from the IoT smart building sensing platforms, with a novel mathematical framework exploiting optimization theory and machine learning, to control a distributed network of HVAC nodes optimally. To handle the computational burden arising from this big data, we need to exploit cloud computing tools for fast and energy-efficient processing. Power Line Communication (PLC) is a technology that enables data communication over power cables. This means that two devices connected by electrical power cables can simultaneously be powered up and, once switched on, communicate via the connecting wires. In general, wired communication is more reliable than wireless with PLC supporting two-way communications. Networks based on PLC will provide electricity network operators with intelligent monitoring and control capabilities without any additional network cabling. Operators will monitor electricity consumption throughout the grid in real-time, implement variable tariff schedules, and set limits on electricity consumption to manage peak loads better.

Building systems, especially commercial ones, for long have been argued to be one of the best suited to provide ancillary services to manage demand-supply mismanagement in the electric grid. To enable any building, systems must be equipped with smart monitoring systems coupled with efficient control and optimization techniques. In our Smart Building PLC pilot deployment jointly being executed by ERNET India, IIT Madras, SLS Pvt. Ltd., and STMicroelectronics, we have developed a low power multisensory platform, comprising air quality, visibility, sound, temperature, occupancy sensors integrated with PLC technology (which is well suited to work in an indoor building environment) forming the backbone of communication modules to create a “nervous system” for building system environments.

Smart Building PLC Testbed

Figure 1: Smart Building PLC Testbed Deployment Architecture.

Figure 1: Smart Building PLC Testbed Deployment Architecture.

 

As shown in Figure 1, our smart building deployment comprises Power Line Communication sensor nodes to measure temperature, indoor air quality, luminosity, PIR sensor for occupancy detection, smart plug to measure power consumption data. The PLC end node transmits sensor data to a cloud server using the Nebulae IoT platform[2], an interoperable framework to integrate various IoT applications. The components in the Nebulae platform include NebuLink Middleware, NebuLink Nodes, and NebuLink Gateways. The NebuLink Node is a PLC or Wireless Edge device equipped with IPv6 connectivity for sensors and devices. The NebuLink Gateway provides seamless cloud connectivity for the IoT end-nodes implementing heterogeneous IoT technologies like 6LoWPAN, ZigBee, PLC, LoRa, and WiFi.  The NebuLink middleware component provides SDK interfaces to IoT end-nodes and gateways for ease of application development in helping with connecting, operating, and automating IoT device management using a single platform. The middleware platform supports secure operations, embedded control, sensing, and easy device management.  Various application layer protocols for cloud connectivity such as MQTT, RESTful APIs, and WebSocket are supported. The PLC nodes used in our deployment are based on ST8500 SoC[3] that supports a programmable PLC model compliant with FCC, CENELEC, and ARIB regulations. It supports PLC protocols such as PRIME and G3-PLC with a signal bandwidth up to 500 kHz. ST8500 has an ARM CORTEX M4 core having analog and digital front ends supporting the integration of communication interfaces other than PLC as well. This makes it ideal for applications where PLC and RF may also be required simultaneously as in building controls and monitoring.

Analyzing the Smart Building Data

Building systems comprise a major portion of the total energy demand in India. Conventional Diesel based generators were traditionally employed to meet the additional demand for energy, at enormous environmental costs. With real-time information, our multi-sensory platform enables us to optimize operations by freeing up capacity by improving buildings' energy efficiency.  A cloud-based platform enables fusion and visualization of data generated via different sensors. The real-time data generated is used to predict or detect occupancy, fault detection and in some cases also predict and prevent faults from occurring by incorporating appropriate data-driven techniques and finally design optimization algorithms to reduce maintenance costs and overall energy consumption, without compromising user comfort. Finally, to allow remote decision making and control, a dashboard and an Android-based application enable the user to monitor various parameters and quantities related to the overall building environment. 

A smart building automation system will help increase energy efficiency and create a healthy environment for its users. This setup increases productivity as it continually monitors building use and adapts the techniques to ensure that the users get the facilities they need. The data generated provides critical insights that help planning and efficient usage of the resources. With a long-term analysis of the data generated, problems can be solved even before they occur, thus saving much time. Our further work would involve exploring opportunities for the deployment of PLC based smart building solutions in some of India's government buildings.

[1] https://bit.ly/2JOGeYZ

[2] https://bit.ly/3pdkxSK

[3] https://bit.ly/32oyl2X


 

Vaishnavi Rajini MohanVaishnavi Rajini Mohan is an intern at IITM / ERNET working on the project Smart  Building PLC Testbed leveraging  IoT network technologies. She has completed B.Tech in Computer Science from Amrita Vishwa Vidyapeetham,  and is looking to pursue a career in data science.

 

Krunal Patel Krunal Patel is an Embedded Product Architect and Manager with over 14+ years of developing embedded products from concept to functional prototypes. His experience lies in identifying optimum technologies for architecture and design of solutions encompassing a complete life cycle in the areas of Renewable Energy, Consumer Electronics, Traffic system,  Medical,  Smart grid, Industry 4.0 based factory automation, building, and home automation.

 

Manoj KumarManoj Kumar is Director of System Research and Applications at STMicroelectronics, India providing a worldwide mission to build and support reference designs and system solutions in the areas of Connectivity (BLE, Sub GHz, Power Line Communication), Metering, Motor Control, Power & Lighting, and IoT. He is also leading ST Partner Program in India with a mission to provide global visibility to Indian partner companies.

 

Ramkrishna PasumarthyRamkrishna Pasumarthy is an associate professor at the Department of Electrical Engineering, and the Robert Bosch Center of Data Science and AI IIT Madras India. His research interests are in the area of Network Science, Data Driven Control, and the Internet of Things. He is also a co-founder of iMov motiontech, a startup that produces wearable devices for healthcare applications.

 

Paventhan ArumugamPaventhan Arumugam is Director of R&D at ERNET India (under Ministry of Electronics & IT) who is also chair of the IEEE-SA industry connections activity on PLC Testbed in India.  He has interests in building practical IoT systems leveraging interoperable standards and protocols.