estimate daytime

Author: Paolo Veronesi, software engineer at Paradox Engineering


In an increasingly complex and interconnected world, Machine Learning is proving to be highly beneficial for Smart Cities and any IoT application where large and high-dimensional volumes of data need to be processed, correlated, and actioned.

At Paradox Engineering we are piloting neural networks in specific use cases such as municipal solid waste collection or patient monitoring in hospital and clinics, both requiring medium to high computational powers. Neural networks are generally applied when it is possible to exploit large computing resources (like workstations or servers), but we are now exploring how to use these technologies in a diametrically opposed occurrence – thus with extremely limited resources (such as few hundreds KB of ram, and standard 32-bit processors with less than a hundred MHz).

Leveraging neural networks on embedded IoT devices is a new, emerging application. Our goal is to process data locally, on the embedded device, without sending information over the network: this allows the devices to work properly even in case of network failures, improves the scalability of the system and its overall security, as sensitive data are processed at local level.

The use case we are focusing is Smart Lighting. In a typical urban outdoor installation, streetlights are connected to a mesh network and managed by gateways, operating as border routers, network coordinators, and data concentrators. A central management software enables the remote management and control of the network, as well as of single or grouped devices.

Streetlights are expected to run smoothly even when the connectivity to the nearest gateway or the CMS is not available. To allow this, devices are configured to execute their routines based on current date and time, assuming the gateway provides this information. This is vital for streetlights to derive the ephemeris and calculate daily sunrise and sunset – according to this data, lamps can switch on/off and dim according to programmed schedules.

What happens if a node is isolated and cannot connect to a gateway, for example because of adverse environmental conditions or network topology? Or if the gateway is not installed at all? In such occurrences, the device should find a new way to calculate the current time. We decided to leverage data analysis and Machine Learning capabilities to let streetlights derive exact daytime from data collected from integrated environmental sensors.

Estimating time from parameters such brightness, temperature, or pressure, isn’t a simple task. Environmental conditions vary a lot from day to day, from season to season, with a level of complexity that it would not be possible to solve with a traditional programming approach.

Neural networks can efficiently manage time series and, if trained with an adequate quantity and variety of data, can provide an accurate answer to our question. In our experiment, we trained the system using data generated by a set of environmental sensors over the course of one year.

Results were really promising: isolated streetlights were able to process data locally and estimate the right daytime with an average accuracy of about 16 minutes – and perfectly execute their schedules. Without Machine Learning, this operational continuity would not have been possible.


Learn more about our Smart Lighting application: download our latest white paper (free registration required), and contact our Machine Learning experts!

smart lighting

8 things you wanted to know about Smart Lighting

Smart Lighting is one of the most mature smart urban applications - and its benefits have been widely discussed. We all know LED-based and remotely managed streetlights consume less power, contribute to climate change mitigation, and generally make cities safer, liveable and more attractive places.

However, it is sometimes difficult for city and utility managers to separate the hype from the reality and make a meaningful business case. In a newly published report - Smart Lighting: how switched on are you? (free download upon registration) - we answered 8 key questions that urban leaders might have so they can embark on their smart journey from a more informed position.

Let us offer a sneak peek...

How much money can cities really save with Smart Lighting?

Streetlighting accounts, on average, for 40 per cent of a city’s electricity bill. Savings due to smart technologies depend on a number of variables, but usually they are worth the case.

The municipality of Gijon in Spain deployed a public and interoperable IoT infrastructure in 2016. Streetlighting was the first urban application to be run on it, with an initial installation of 1,150 LED luminaires being monitored and actioned by our software management platform. The economic saving for city coffers was assessed at around €100,000 a year.

Another interesting example is the Tesserete‐Canobbio bike trail in Ticino, Switzerland, where local utility AEM upgraded existing luminaires to LED technologies, connected them to a wireless network and interfaced every lamp with a motion sensor, so lights automatically switch on and are dimmed to 100 per cent only when cyclists or pedestrians pass by. As a result, annual operating hours along the trail reduced from 4,300 to 600, and average costs decreased from 11.19 to 1.56 Swiss francs per light point.


Does Smart Lighting make the management, maintenance and repair of luminaires more complex and expensive?

No, that is exactly the opposite! By having all light points connected to same network, cities can fully monitor and operate luminaires remotely, leveraging a single software platform. Proactive to real-time detection of luminaire failures or out-of-the ordinary behaviours becomes possible, and technicians can be sent out only when and where needed, arriving on site already informed of and equipped to address the specific issue. As well as reduced costs and more streamlined operations, this also improves the quality of service and citizens satisfaction.


Can Smart Lighting generate revenues?

Smart lighting investments can pay for themselves and even turn into a revenue generation opportunity. In 2016, the city council of San Leandro in California, USA, started its smart journey by replacing around 5,000 streetlights with LED lamps and implementing a wireless IoT network with a centralised management and control system. Other applications and services were added over time to the network, such as integrated parking, public wireless internet and traffic video surveillance. Upon the initial spending of $5.2 million for energy- and water-saving equipment, it was calculated that the investment would generate $8 million savings over 15 years through reductions in energy and water use, and $1.5 million in positive cash flow over that time.

This happens when urban networks are designed as interoperable infrastructures, able to host multiple applications and launch public-private collaborations and more, they can create viable opportunities to monetise the data they generate.


Eager to read more? Download our paper and don't miss our free webinar on Thursday, July 1st 2021 at 9am (EST), 2pm (BST), 3pm (CET), 5pm (GST): we will discuss benefits and success factors of best-in-class Smart Lighting, providing an overview of some real-life urban experiences. Register today!

smart cameras

Intelligent surveillance to improve senior and patient care

Author: Riccardo Biella, software engineer at Paradox Engineering


Health systems face immense challenges in many countries around the world. Lots deal with an ageing population and need to control public spending for healthcare service, but in many cases patients outnumber available resources including basics such as hospital beds, doctors, and nurses. This shortage became far more relevant during Covid-19 emergency, with most countries struggling to provide adequate pandemic response and quality of care.

Smart technologies are enabling a whole new class of services to improve patient assistance, specifically the treatment of elderly and long-term patients, whether in hospital, in assisted living facilities, or at home. Connected healthcare systems and remote patient monitoring (RPM) technologies are being increasingly used to ensure constant surveillance, support professional caregivers in their daily duties, and relief family members.

Together with Minebea Intec and MinebeaMitsumi Sensing Division, at Paradox Engineering we are working on an innovative solution for the managed care industry: our goal is to finalize a secure and open technology platform for non-invasive patient monitoring in hospitals, clinics, nursing, care houses and home environments.

It is meant as an adaptive and knowledge building solution, able to generate patient-specific intelligence by collecting data from multiple sensors, correlating multiple parameters, and integrating a unique Artificial Intelligence (AI) engine for situation recognition, behavior analysis, long-term health evolution monitoring and personal relationships.

The solution provides local AI to prevent possible data leakages from the hospital environment and better protect patient privacy. We are currently piloting Machine Learning to improve some specific features, such as the detection of adverse events: by processing data coming from connected smart cameras and sensors, the system can recognize when something goes wrong and immediately alert caregivers in case of falls, lack of movements or other types of anomalies, ensuring constant monitoring of patients around the clock.

Another feature we are testing relates to room access control. We are using Machine Learning to elaborate images from connected smart cameras in order to recognize the number of people in a room and distinguish patients from hospital staff. This is a highly valuable feature to increase guest safety by detecting unauthorized access.

We started from an initial dataset made of hundreds of images of patients and doctors to train the system. PE Smart Cameras are now able to perform object detection tasks in real time, distinguishing people within the room and classifying them as patients or hospital staff, thanks to the use of a Deep Neural Network that can also recognize things such as scrubs, uniforms, nightgowns, medical instruments and postures.

Watch this short video to have a sneak peek:



Want to learn more? Contact our Machine Learning experts!


Smart Waste, Machine Learning enables bin filling level prediction

Author: Riccardo Biella, software engineer at Paradox Engineering


Annual waste production will increase by 70% to 3.4 billion tons by 2050, says the World Bank. At this growth rate, we will need about 3 worlds to absorb and dispose of our trash without compromising our health and the environment.

Cities are increasingly looking for innovative solutions to improve municipal solid waste collection and treatment to preserve both public health and urban livability. At Paradox Engineering, we have extended our flagship platform PE Smart Urban Network to include a ready-to-use Smart Waste application for City and utility managers.

The news is, we are integrating Artificial Intelligence and Machine Learning algorithms to enhance this solution and make it even more performing. While traditional algorithms convert input to output using some given rules, we chose to have Machine Learning to build an accurate predictive model based on sample data.

First step is to equip bins with Paradox Engineering-MinebeaMitsumi 6LoWPAN sensors and allow municipal waste operators to collect and monitor data about the level of filling, date and time of latest waste collection, and bin location information.

Thanks to Machine Learning techniques, we are evolving the system from a raw data collection platform to an actionable prediction solution: the bin level prediction system can receive data from smart bins’ filling sensors and process it, providing an estimate of the date when the bin will reach its capacity limit.

Did you know that bins are generally emptied when they are only 40% full? With our bin filling level prediction system, operators can dispatch trucks only when and where the bins are close to full, thus optimising solid waste collection, reducing the number of truck rolls and the mileage associated with them, which in turn decreases pollution and congestion.

The bin filling level prediction system leverages a deep neural network that is trained on a dataset containing historical data of smart bins. Features considered during software training are based on the current level of filling correlated with time data, as well as geographic location data.

More complex logic will soon be integrated to teach the bin identify relevant points of interest nearby, such as supermarkets, stadiums, stations, or hospitals – and predict waste input upon variables such as people density, calendar, festivities, holiday seasons, and more.

Continual Learning techniques allow the model to continuously evolve combining new data and accumulated knowledge. This means that, by analysing filling patterns, operators can also take data-driven decisions about the quantity, capacity, and location of containers. Bins can be positioned to fit different urban scenarios and adapt within a few days to changing conditions, ie. different residents’ habits or behaviours, like the increase of packaging-related trash due to the rise of e-commerce and home delivery in pandemic time.

Thanks to PE Smart Urban Network and Machine Learning algorithms, cities can truly improve waste collection and management, enhancing livability while protecting public health and the environment.


Learn more about our Smart Waste application: download our white paper (free registration required), and contact our Machine Learning experts

machine learning

Machine Learning: a great opportunity for IoT platforms

Authors: Riccardo Biella and Paolo Veronesi, software engineers at Paradox Engineering


World is increasingly complex and interconnected, generating more data traffic than ever before. A recent IDC report predicts connected IoT devices could reach almost 75 billion globally by 2025 and generate about 79.4 zettabytes of data.

This huge amount of information contains an incredible value, but it must be stored, processed, analyzed, and correlated to unlock its full potential. Is human effort enough for this incredible task? Probably not. That’s why Artificial Intelligence (AI) is increasingly being leveraged to elaborate such a large and growing amount of data, with more accurate and faster results than traditional computer programming.

Some years ago, Andrew Moore, Former-Dean of the School of Computer Science at Carnegie Mellon University, defined AI as “the science and engineering of making computers behave in ways that, until recently, we thought required human intelligence”. AI technology is inspired by human behaviors and cognitive processes, but it also leverages knowledge from other biotic processes such DNA mutations or the chemical decay of substances.

Although often mentioned as a synonym, Machine Learning is something different: it is a branch of AI focused on developing computer algorithms and building applications that learn from data and automatically improve their accuracy over time through experience.

Examples of Machine Learning applications are all around us. Every evening, while driving home after work, digital assistants search the web and play music in response to our voice commands, our watches monitor our heartbeat, our cars keep us on track and suggest stopping for a coffee if we are too tired.

Human brains have the extraordinary skill to learn from experience and, thanks to Machine Learning, algorithms will more and more sharpen this ability and provide better results through information. How far can Machine Learning go? Difficult question, but one thing is sure: data play a key role since their quality and quantity influence the accuracy of results, so they are a truly critical element in any Machine Learning progress.

Traditional algorithms convert input to output using some given rules, while Machine Learning algorithms build a model based on sample data to make predictions without being explicitly programmed to do so.

Machine Learning offers the best performance when processing and analyzing large and high-dimensional volumes of data, making it particularly useful and interesting in the Internet of Things (IoT) world due to the large number of devices collecting and sharing non-homogeneous data. After going through a statistical analysis process, those data are a treasure trove for the development of new smart applications.

In an IoT network we can acknowledge three different layers, each one characterized by the availability of different resources and data where Machine Learning is applicable:  IoT devices, edge devices and the cloud.

With the advent of new technologies, it is possible to integrate the usage of neural networks and Machine Learning techniques at all IoT levels to perform complex data analysis locally. IoT devices, such as sensors typically constrained, can locally process data without sending them over the network, reducing latency and avoiding security problems. Moving to the cloud, it is possible to carry out more complex analysis due to the greater computing power and leveraging data coming from different sources, thus opening the IoT world to new and highly innovative use cases.

At Paradox Engineering, we have integrated the most recent Machine Learning techniques into our IoT platform components, enabling valuable use of Machine Learning at different levels, from cloud to edge and IoT devices, making the most of their peculiarities and functionality.

Our IoT solutions have been extended with a Machine Learning development framework to allow these technologies to be used in multiple application cases in a simple and transparent way. Developers can easily integrate Machine Learning as enabler for new information-driven functionalities or to enhance existing ones.


Want to learn more? Contact our Machine Learning experts!

open source mesh network

Open data models and open source mesh networks for smarter cities

Author: Nicola Crespi, Chief Innovation Officer at Paradox Engineering


Interoperability is paramount for truly smart cities. We have been widely discussing how important it is for a city to rely on openness and get rid of proprietary technologies and data formats [read our white paper to learn more].

When cities and utilities design an interoperable urban infrastructure, they typically look at existing IoT international connectivity standards such as LoRaWan, NB-IoT or WI-SUN: good starting point, but they should be aware these standards only deal with the network layer and the messaging protocol of the IoT solution, so they won’t provide any specify data model or application payload for urban use cases.

The result is, in most occurrences the solution provider goes back to a proprietary data model or application layers, and the customer is locked to its technology and need to chase expensive API integrations.

That’s the key reason for Paradox Engineering and MinebeaMitsumi joining the uCIFI Alliance. We have always been advocating open standards and are committed to create multi-supplier, open solutions for Cities and other smart environments.

Today we are leading uCIFI’s Technical Working Group and we contributed to the public release of the first unified data model for cities and utilities – a significant step forward to ensure full interoperability between IoT devices connected to the same network. A common data model allows connected devices from one vendor to be replaced by equivalent equipment from a third party without requesting any software integration.

The uCIFI Data Model leverages the Open Mobile Alliance’s Light Weight M2M (LwM2M) registry to describe smart devices (reference objects) and their attributes (resources) to unlock the true potential of Open Cities with reduced costs and a superior guarantee of investments’ sustainability. Our solution has successfully passed the interoperability tests for LWM2M implementation during the 2021 OMW Virtual TestFest.

Moreover, the uCIFI Data Model makes it easy for suppliers adopting uCIFI on their products and devices to become TALQ compliant and get related certifications.

What’s next? We are working with uCIFI on the open source mesh network. The mission here is to define and provide an open-source mesh reference implementation that standardizes the smart city application layers on top of any 6LoWPAN/Wi-SUN certified mesh network.

Smart city applications may have specific user cases and requirements in terms of data bandwidth, device reachability and commissioning, device-to-device communication in outdoor spaces (ie. a presence detection sensor that increases the light level on a remote luminaire), but these application services are now implemented only by proprietary solutions on the market.

The goal of uCIFI is to standardize this smart city application layer on top of 6LoWPAN/Wi-SUN networks, defining the protocol and process to be used for the device and network commissioning, end-to-end security and device-to-device communication, thus guaranteeing a full end-to-end interoperability.


Want to learn more? Contact our expert Nicola Crespi!

unified data model

uCIFI Alliance releases unified data model for Smart Cities and utilities

Active members of the uCIFI Alliance together with our parent company MinebeaMitsumi, we are proud to share the public release of the first unified data model to provide interoperability and interchangeability between connected devices to unlock Smart Cities, reduce cost and guarantee investments’ sustainability.

The uCIFI Alliance open-source model targets key Smart City applications such as streetlighting, water metering and distribution monitoring, waste management, parking, traffic monitoring, air-quality monitoring, smart buildings, as well as safety and security. It also supports common and more complex services, including multicast grouping, sensor-to-actuator dynamic control and a calendar-based control program to standardize edge computing for IoT sensors.

With this unified data model, a series of 30 connected sensors and devices used for smart-city projects are described on the Open Mobile Alliance’s Light Weight M2M (LwM2M) registry. These reference objects and associated attributes are modelized using the LwM2M format and can be used on any constrained IoT network. They also can be implemented freely by any vendor in any product to meet cities’ tender requirements for openness and interoperability. Vendors may join the alliance to benefit from the certification program, access test tools and other software resources and to contribute to more interoperable object models.

The unified data model is a great news for Cities and utilities: they can avoid vendor lock-in and have more flexible, wider and future proof sourcing, taking immediate advantage of this open-source, multi-transport, multi-supplier and cost-efficient tool to break silos when designing and implementing their IoT projects.

As we advocate open standards and are continuously committed to create multi-supplier, interoperable IoT solutions, we welcome uCIFI Alliance unified data model and look forward to supporting Cities and utilities in their smart journeys. 


A quarter of streetlights to turn smart by 2030

Investments in LED and smart streetlighting are not likely to be halted, even in pandemic time. According to Northeast Group’s forecasts, global market is expected to reach a value of USD 28.1 billion over the next decade, including spending for LED and smart lighting infrastructure, related sensors, and management software. Software alone will account for over $200 million in annual recurring revenue by 2029.

Today there are about 326 million streetlights all over the world, and this should grow to over 361 million by 2030. We know a quarter of all streetlights globally have already been converted to LEDs and over 10 million have been connected to smart networks. If forecasts are right and investment trends are confirmed, we will have about 73% LED-based lamps and up to 25% smart streetlights in the world by 2030.

Advantages of Smart Lighting are very well known. Installing energy-efficient LED lamps immediately makes Cities save up to 70% in power consumption and related costs. Up to 25% more power can be saved if enabling full remote management and control of single or grouped luminaires by connecting them to a sentient IoT network, introducing key features such as scheduled on/off switching, and adaptive dimming.

There is more. Nowadays streetlights are a canvas for urban innovation, improving not only sustainability and efficiency, but creating opportunities for mobility, public safety, tourism, and overall quality of life.

Being a sort of nervous system for any City, streetlight networks connect almost any district and street with access to power. That’s why they can easily become a rich sensor platform, collecting vital data for a myriad of urban applications. Example of services that can be deployed together with Smart Lighting include traffic light controls, traffic and district video surveillance, air quality and environmental monitoring, pervasive WiFi and broadband connectivity.

Looking at streetlights as the backbone of smarter communities urges City managers to have a broad, far-sighted strategy, and possibly make use of new financing models. As reported by Cities Today, infrastructure investment funds, energy services companies (ESCOs) and urban management consultants are playing a growing role in financing and carrying out smart streetlighting projects. This might become an interesting option for municipalities facing budget shortages due by Covid-related economic issues.

As Cities seek flexible funding options and new public-private partnership models, they increasingly need to demonstrate Smart Lighting and other urban applications will generate a measurable return, and turn into an opportunity for the benefit of all, including investors.

How to make sure smart investments create both efficiency and new revenue streams? How to make sure they will support and stimulate future growth? In our experience, interoperability and openness are the best answers.

When a City builds on a standard-based infrastructure such as PE Smart Urban Network, it takes advantage a single wireless network and a single central management suite to remotely control any Wireless IoT and Wireless IoT Highspeed application, with full interoperability in device, data, and application management. There are no constraints on future expansion and integration of additional services and third-party systems, and lots of opportunities to innovate and enable forward-looking data monetization models.

Eager to read more? Download our paper ‘A Smart City is an interoperable City’ and review our ondemand webinar.

medical care

How technology is changing medical care

Since the start of the Millennium, technology has been increasingly part of our everyday life. While less than 7% of the world was online in 2000, today over half the global population has access to the Internet. Similar trends apply to cellphone use, that passed from 740 million subscriptions worldwide in the early 2000s to 8 billion in 2020, meaning there are now more cellphones in the world than people.

As pointed out by World Economic Forum, technology accelerated innovation in a variety of domains, and had an astonishing impact on healthcare, with unprecedented progress in disease and treatment research (the rapid vaccine development we have seen for Covid-19 would have been inconceivable a few years ago), in surgery and medical care delivery.

Telehealth has been on the rise in recent years, and new systems for remote patient monitoring and virtual assistance are being used. Some of the most popular solutions require patients to have wearables to monitor key health parameters and share data in near real-time with control systems or medical professionals. That’s an effective way to manage chronic conditions such as diabetes with continuous glucose monitors, or prevent heart failures by tracking blood pressure, pulse rate, and perfusion index.

Other remote patient monitoring platforms leverage contactless technologies to provide non-invasive surveillance and balance accurate data collection with user comfort. This is the goal of the technology Paradox Engineering is developing for and with Minebea Intec and MinebeaMitsumi Sensing Division: we are working on an IoT-based platform for patients and seniors to enjoy high quality of health when treated in hospitals, care houses, or at home, receiving continued assistance and taking advantage of fine-grained health data analytics without the need of invasive probes, body sensors, or wearable medical equipment.

Our care environment will feature a complete hardware and software solution, and provide clinics, nursing houses, and healthcare professionals with an end-to-end technology platform leveraging data collection, delivery and presentation. Our care environment will be intelligent, as we are developing an adaptive learning system, powered by artificial intelligence and machine learning, to detect and analyze irregular medical events, understand individual medical history, and support information-driven medical decisions.

The past two decades have seen technology advancing medical research. Now we are contributing to unleash the power of digital information for a more reliable and intelligent medical care delivery.

Smarter waste management to improve quality of urban life

Living in a clean, healthy and safe city is fundamental for most people around the world. Essential services such as waste management are taken for granted but when something goes wrong, waste quickly turns into a highly visible issue and emotive subject.

Pre-pandemic data stated that in Europe each person produces nearly half a ton of municipal waste per year, which means that every week more than 20 kg of municipal waste is generated per household. Preliminary reports from the US show that Covid-19 saw commercial waste decrease about 16 percent, while residential waste increased up to 25 percent in most cities.

This posed serious challenges for the proper treatment of waste: preventing distortions in waste management, including efficient collection and recycling, is crucial for public health and safety, as well as for the environment. Waste operators are working hard to ensure quality of service and at the same time keep costs under control.

“Most cities used to have a reactive approach to dealing with waste management, thus investing only in light of emergency situations, or when pushed by regulatory compliance issues. Increasingly asked to contribute to recycling targets and the circular economy, cities are now becoming proactive and looking at waste management not as an expenditure item, but as an opportunity to improve the quality of urban life,” says Gianni Minetti, CEO at Paradox Engineering.


What are typical pain points of municipal waste management?

Waste collection typically accounts for 10 to 25 percent of a municipality’s budget, and it represents high costs and poor performance almost everywhere. Independent studies reveal that 15 percent of bins are generally over-full, increasing health and environmental risks, and making town centres and districts ugly.

Cities tend to solve this by increasing the frequency of collection. Consequently, bins are emptied when they are only 40 percent full, so waste collection costs rocket. Further, it is difficult to monitor bins in a timely way, including replacing containers that happen to be vandalised or moved without authorisation.


How can IoT-based solutions help? 

Smart waste solutions leveraging Internet of Things (IoT) technologies allow trash bins to be remotely connected and monitored, with data showing the fill level and the date and time of the latest waste collection, and generating alerts in case of fire, vandalism or unauthorised bin movements.

Data is key for making any smart decision. By analysing bin-generated data, and correlating it through an intelligent routing software, waste operators can predict when containers will need emptying and dispatch trucks when really needed, or when the city prefers. This improves the quality of waste collection, generates efficiency and savings, and adds relevant benefits in terms of health, safety and liveability,” explains Minetti.

Read more on Cities Today and download our latest insight report to learn more about PE Smart Urban Network and our new Smart Waste application.

Any question? Contact us!