XSelect a record to create:
When energy is not enough
How a miniaturized Energy Efficient Wireless device can turn a sensor into an IoT node
Published on 12/19/2016 | Technology
Although Semiconductors are showing optimistic market growth real opportunities to increase profitability, in a market dominated by the continuos shrinking of operative margins, are not at hand: the Semi is not more the fastest growing segment it was in the 20th century and there're few opportunities left for the industries to catch on with the new Digital Age and try to spin market: industrial, automotive and other market sectors are already saturated and growth is limited to commodities were pricing plays the major role.
One of those opportunities is perhaps the growth of Internet Of Things and the incoming (expected) revolutionary boom in connected things (Trillion Sensor Vision) that will demand billions of sensors to connect and readily deliver to Cloud based Infrastructures set of data to feed algorithms and provide innovative services to consumer.
One factor that is limiting the growth of IOT is the energy consumption: trillions of sensors requires that each sensors is deployed in the ambient and provided with extended autonomy to support unmanned operations over time, limiting cost of maintenance and avoid battery replacement. They must be capable to consume the least of the energy and operated by alternative power sources, cost less than ever and have a miniaturized form factor to fit in anywhere and everywhere. Energy in wireless sensors is measured as the amount of nJ per each single bit transmitted over the air. The most optimized technology available in the market consumes more than 60.000 nJ/bit: the Bluetooth Low Energy is the most popular device that apparently fit requirements of IOT sensor deployment. But a system operated through a BTLE normally requires regular battery replacement and at least 2 within a year: Semiconductors are trapped, there'snt yet a solution to this equation (?). Looks like new breakthroughs are required as well as advances in semiconductor.
The question is: how many years should a sensors be working unmanned to fit into the business cases of IOT and trigger the market? Or better: what is the breakthrough technology that will potentially trigger the demand of >1B sensor per year by 2020 the IOT is expected to accelerate?
“The Trillion Sensor vision” will come up in the very end: the recent EU and US regulations (US FMVSS138-2005 enforcing phase-in fitment of sensor in tires by 2007 and European EC661-2009 with phase-in by November 2014 for all new vehicles) in matter of mandatory fitment of sensor in tires are the first evidence of the revolution in the Internet of Things: collecting data from tire is not just a matter of warning drivers of tire inefficiencies, advise for better driving behavior or to reduce fuel consumption enhancing better rolling resistance, it is also the kick-off of the new age of digitalized technologies and innovations. One of the components that builds for the new digital world is the wireless sensor that stays at the bottom of the pyramid (first leg) and collect the necessary data to enable analytics in mobile monitoring, wearable, health, agriculture, automotive industrial and more generally in the Internet of Things. In order to get there industries and IT operators will need to outline the 3 layers that lay at the bottom of the IOT paradigm: first the Ultra Energy Efficient Wireless data collector, the least of energy needed by such sensors to capture and deliver a single bit of data and, the least but not the last, at what cost per single sensor.
At the other side of the equation there are Big Data Analytics and Deep Learning as the two high-focus of data science that are becoming increasingly important for organizations that aim at collecting massive amount of domain-specific information for product enhancement, technological advances, new intelligence and marketing approach, security and medical informatics. Deep Learning algorithms extract high-level, of complex abstractions as data representations through a hierarchical learning process. Complex abstractions are learnt at a given level based on relatively simpler abstractions formulated in the preceding level in the hierarchy. A key benefit of Deep Learning is the analysis and learning of massive amounts of unsupervised data, making it a valuable tool for Big Data Analytics. Deep Learning can be utilized for addressing some important issue in structured data collection including extracting complex patterns from massive volumes of data, semantic indexing, data tagging, fast information retrieval, and simplifying discriminative tasks. The general focus of Deep Learning is the representation of the input data and generalization of the learnt patterns for use on future un(fore)seen data patterns. The optimization of data representation has a large impact on the performances of data machine learners: a poor data representation is likely to reduce the performance of even an advanced, complex machine learner, while a good data representation can lead to high performance for a relatively simpler machine learner. There's a growing need for Deep Learning and Artificial Intelligence to get access to the physical data, to collect and fit them into the unknown variables of their algorithms and models applications of the virtualized layer: Ultra Low Energy Efficient Wireless Sensors is the domain of IOE that is regarded as the last piece of the puzzle to solve the equation. The globalized IT infrastructures are full of powerful cloud based platforms and ready to provide innovative services and applications of deep analytics but still the energy efficient, ubiquitous, unmanned and cost effective physical device, that collect the data they require to power up their engines, is not yet identified.
The “Tire as a Sensor” concept: one domain of data analytics that can potentially advance the use of Intelligent Transportation Systems in the Complex Networks and Complex Systems is the Intelligent Tire Concept.
Since its mass production, tires have always been passive elements playing a crucial role in vehicle safety and stability. Although the science of tire compounding, design, and manufacturing has grown tremendously, its inclusion as part of the vehicle chassis control system has lagged all the other subsystems. In order to be able to provide the much needed tire road contact characteristics to the chassis control designer, it is required that the tire becomes part of the intelligent grid that provides the information to the controller. One approach on "Tire as a Sensor" concept, more widely and improperly known as the Intelligent Tyre, is to combine modeling, instrumentation, testing, signal processing and algorithm development interlaced with vehicle's and tire data: the research in such domain, although it can now largely benefit from new advances in Deep Learning through data fusion of the Virtual and Physical Layers, lack of essential information regarding the real state of the tire behavior and characteristics at road contact level. In this regards, after years of attempts to extract data using software extrapolation and indirect analytics, Tire Monitoring System now seems to foresee fitment or integration of miniaturized and flexible sensors in tire structure (or just embed them onto the internal surface or inner liner) to measure complex data processing, internal forces to characterize dynamics. Tire manufacturers, after a decades of research, are now making up their mind giving up the idea that data extrapolation can be processed by only using software and algorithms and begin to look at sensors in tires as source of Big Data to feed their needs of analytics and get access to innovative source of revenues. Tires are considered as a great sources of data readily available in vehicles but virtual dynamic estimation just in itself provide a limited information on vehicle's status and driving condition: tire patch contact area, slip angle, yaw rate, torques, wear and load transfer are among some of the parameters that could greatly benefit from the fusion of direct analysis of Physical data with the Virtualized layer.
Recently turned out that one sensor in each of the tire manufactured would mean several millions of sensorized tire delivered year by year collecting terabytes of data on a daily base; such data are than storaged and used by tire makers or ICT providers to pursue and foster innovative services to customers, innovative technologies and advances or just to fullfil needs of spinning new and innovative business models and get revenues up again.
On the other hand digital industry sees TMS or I-tires potentially as the first pillar of the so called Internet of Everythings (“trillion sensors vision”): being the rubber among the material with the highest volume of sale globally the “Tire as a Sensor” is capturing global attention on the 4 verticals of the vehicle’s Data Telematic such as passenger vehicle , truck and bus , Off the road and recently also Agriculture vehicles. In each one of such vertical the sensorized tire is bringing in lot of value, new business and technological advances: in fact measuring tire forces is by far the unquestionable challenge in data transmission and sensor technology ever since, in the ’80, first devices to measure temperature and pressure were installed on tire vehicles.
Measuring data in a tire is a challenge that requires not only a sensor to be designed extremely efficient in terms of energy consumption, not only such system to be robust and mechanically resistant to high shocks and temperature gradient, or highly miniaturized and designed for recyclability purpose, but it also need to be re-designed to replace Lithium compound battery cell with alternative energy sources to power up the system. Recyclability in tire today plays an important part of the manufacturing process: scrap tire rubber is used in the manufacture of new tires, playground surfaces, equestrian mats and rubberized asphalt among other products. Other cutting-edge manufacturers are combining scrap tires with materials such as scrap plastic to produce flower pots, roofing tiles and auto parts. A tire is a highly engineered and extensively designed product that is meant to be virtually indestructible under a variety of conditions. Because of this, tires are difficult to recycle: tire recyclers have invested millions of dollars in technologies and equipment to recycle tires, allowing scrap tires to play an important role in strengthening our economy and protecting our environment.
At tire recycling facilities the main piece of equipment is the tire shredder which uses powerful, interlocking knives to chop tires into smaller pieces. Shredding a tire at room temperature using such knives is called ambient shredding. Tires can also be shredded through a cryogenic process that uses liquid nitrogen to freeze them at a sub-zero temperature. Such temperatures cause the physical properties of the tires to change dramatically and become very brittle. The tire is placed in an enclosure in which powerful hammers smash the tire apart.
In 2012, scrap processors produced 1.1 billion pounds of crumb rubber that was used in the creation of new products ranging from sidewalks to horse tracks. Tire recycling is an economically sound, environmentally friendly activity that can contribute to the reduction of a product’s overall carbon footprint. In fact, the use of recycled rubber in molded products provides a substantial carbon footprint advantage over the use of virgin plastic resins, having between four and 20 times lower carbon footprint. The future for tire recycling is strong. Applications for scrap tire rubber — such as rubberized asphalt — have become recognized for their preferable properties and is gaining in prominence and widespread use. Many states already use rubberized asphalt when they design, reconstruct or repair their roadways and it is used for several simple and straightforward reasons: it can cost less, provide safety benefits and last longer than conventional asphalt.
Taking into consideration the entire process its crucial to understand a sensor embedded in a tire should not only be capable to resists to high acceleration shocks (>2000g), to resist to the widest temperature range (-40°C+125°C as required by Automotive standards) or to comply to needs of miniaturization but also to comply to the process of tire scrapping and recyclability. Scrap tires — generated when an old, worn tire is replaced with a new tire — are often dumped illegally in lakes, abandoned lots, along the side of roads and in sensitive habitats: by statistics each year approximately 100 million tires are processed by the recycling industry therefore this would mean to have millions of lithium coin cells attached to the tire inner liner that would be equally scrapped illegally with their carcass attached on it; Lithoum in tires could put at stake the environment more than ever before and for this (and many other reasons) the energy source plays an important role.
It is self explained that Lithium compound and battery coins are limiting the process of tire scrapping as it requires such material to be removed manually causing the entire process to increase in costs and therefore limiting the potentiality of the economic that lies in behind. Therefore it also goes in itself that to turn a sensorized tire into a component that best fits into a complex process of recycling requires development of a number of critical advances in technologies: the development of a suitable technology capable to deliver the concept of “Tire as a Sensor” is undubtedly one of the major challenge electronic industry has ever faced since Intelligent Tire concept has brought in. To find the hidden variables or unknown values in this process will lead to the solution of the equation that builds for the “Trillion Sensor Vision” and it will undoubtedly contribute to reduce request of energy to feed sensors hence enabling more advanced autonomous and ubiquitous solutions that could have impact in the entire cycle.
The quest to make human life safer and comfortable by studying the environment that surround ourselves has led to rapid advances in technology: electronic sensors have empowered us in this quest to study the environment. A sensor is a device that detects changes in a physical variable, such as barometric pressure, temperature and provides a detectable electrical/optical signal: a thermocouple converts a temperature difference at two points to an output voltage. The advancement in the sensing and processing technology has enabled new areas of study in medical research, automotive engineering, agriculture, consumer applications to name a few. For example, modern medical care routinely involves monitoring vital physiological signals, like the electrical activity along the scalp (EEG) to diagnose brain disorders. Just to give another example we know that automobiles of today are equipped with hundreds of sensors to enhance the safety and comfort of our journey. Cellphones today have numerous sensors embedded in vehicle’s chassis to improve user’s experience and improve the quality of driving. In the past decade, both industrial and academic interest in the development of fully autonomous sensor nodes has seen rapid growth. The number of sensors, in just the mobile market consisting of cell phones, tablets, cameras etc, increased from 10 million units in 2007 to about 3.5 billion units in 2012. They mainly consist of four sensors: microphones, gyroscope, accelerometer and compass. Moving forward, in a foreseable near future, sensors will be deployed in a diverse array of fields pointing to a world of trillion sensors by 2022.
The primary enabler for the popularity of sensors has been the miniaturization in the sensing technology possible due to the advancement in the field of electronics. Future emerging applications of sensors will demand further miniaturization along with a near complete energy autonomy and low cost in manufacturing: thematics such as innovations and new advances in processing technology to enable miniaturization, cost reduction, architecture and circuit innovations to enable low energy consumption of the sensor interface readout circuitry are at the base of the paradigm of the IOT and EOT age.
The Tire as a Sensor concept - Big Data in vehicles: Incoming technological shift at OEM level, lead by autonomous and connected car evolution, brings traditional system to monitor pressure in tires (TPMS) into the age of “tire as a sensor” concept (TMS): there’s a new requirement to interconnect the tire to the vast array of vehicle’s sensors and provide the vehicle more data relative to dynamic forces and tire parameters. Measuring these parameters directly from tires offers the potential for significant improvements in safety, vehicle performances and provides new services such as active tire logistic, tire predictive maintenance, fleet management and new business opportunities based on vehicle data collection. A new generation of intelligent sensors to be attached onto the tire inner liner are currently investigated to collect data directly internally to the carcass.
Tires, as part of a connected vehicle, are an important piece of transformation of driving from standpoints of safety, fuel efficiency and mobility. Some more immediate reasons that push the evolution of tire industry have to do with political imperatives of CO2 emissions mitigation and related economic significance of high fuel costs; tire manufacturers are under pressure from OEMs to contribute meaningfully to these efforts, and tire sensor might become a critical component of their compliance effort to meet fuel economy standard (together with lower weight, increased aerodynamics, smaller engines, new powertrains like hybrids and electric vehicles and others).
The energy factor: Traditional Tire Monitoring Systems (TMS) are based on Frequency or Amplitude modulation schemes where digital information are transmitted through the changes of a constant carrier signal. ASK and FSK schemes have typically an average current consumption of 10 to 15 mA: because of their relatively high power consumption the small lithium battery cell (usually 20 mm in diameter), that are used to power up the system, allow limited system's lifecycle. The new functionalities that are necessary to monitor tire’s condition and internal forces are not yet feasible especially in the automotive temperature range (-40°+125°C) using such system's therefore new advances in hardware, firmware, data telemetry, data transmission or product's scaling are necessarily required. For such purpose tyre manufacturers and integrators foresee the use of innovative technologies of “Ultra Energy Efficient Wireless Sensor” (uEEWS) devices and adoption of advances in Deep Learning, sub-systems of Artificial Intelligence and Machine Learning or alternative approach to Fusion Sensors to provide suitable solutions and comply to the energy factor.
Thanks to advances in domains such as sensor fusion (physical and virtual layer), data analytics and new techniques, such as Machine Learning (ML) and Artificial Intelligence, algorithms to characterize forces applied to tire patch contact area could be provided to further enhance vehicle’s intelligence spinning interaction with existing vehicle’s sensory architecture (that heavily rely on tire and vehicle kinematic formulation and changes in quantities of applied forces) and at chassis control systems level so that to reduce overall system’s hardware complexity and overcome the energy issue.
Energy sources and new material for powering micro-sensor nodes - the Energy Harvester (EH): Micro and nanotechnologies have already made possible the fabrication of small, low cost and good distributed intelligence. However energy autonomy keeps being one of the most desired enabling functionalities in the context of off-grid applications an wireless sensor networks. In many of such applications wired power is not feasible, and because of the modest energy involved batteries are normally used to power up devices. However, battery replacement will eventually become impractical (economically, environmentally, and logistically) not only for sensor networks in remote places or harsh environments, but also for more standard applications if the number of nodes explodes exponentially. Harvesting energy, tapping into environmentally available sources such as heat and vibrations in a foreseeable future are the best solution in unmanned scenarios of predictive maintenance and data collection. Furthermore, coupling those harvester devices to secondary batteries to buffer enough energy, to account for the power demand peaks required by the communications of wireless nodes, could be a quite enabling solution to provide extended energy autonomy. Silicon is an abundant material amenable to micro and nano-structuration and silicon processing is an economy of scale technology apt to mass production. Although downsizing is not generically favorable to energy applications, this problem can be off-set by the capacity of silicon processing to integrate high density features in small volumes. This best in knowledge is required to figure out the correct dimensioning of such devices and select best technological choice in design and fabrication route that fulfill the power needs to overcome the constraints posed by those applications. As a consequence, while the focus is placed on the devices themselves, system level integration issues are to be thoroughly considered as well as the general application-wise approaches to all the elements required by an eventual autonomous working sensor node in such a way to consider which of them may have an impact for the final device architecture.
Vibration Energy Harvesters are resonant systems and work only efficient if their resonance frequency is matched with the excitation frequency. Silicon based microsystem energy harvester have normally high resonance frequencies and are therefore difficult to match with excitation frequencies found in applications. During tire rotation the centripetal force, at the contact patch area, drops to zero resulting in a shock excitation. The height and length of the shock depends on the speed and tire dimensions. The energy harvester will be excited by the shock and starts to oscillate at its resonance frequency (ring-down mode). At 40km/h the shock amplitude could be several tens of grams. The shock acceleration has a quadratic dependence on car speed.
Figure 1: Acceleration evolution during tire rotation. Shocks occur at the contact patch in the tire where the centrifugal force drops to zero (see figure on the right).
Speaking of standard Tire Pressure Monitoring Systems (TPMS) main parameters to sense are pressure and temperature of the air inside carcass. The measurement frequency depends, still taking into account impact on system’s energy requirements and budget, on how fast is necessary to detect a possible puncture or tire deflation due to diffusion (losses of air over time due to natural leakage of molecules): 5 to 10 seconds measurement is typically the rate. Pressure measurement is a procedure that takes around 50μs and the mean energy necessary to sustain such action is between 1-2mA@3V (3mW - 6mW for the 50μs timeframe). Temperature measurement is, on the other hand, a procedure that takes up to 50μs and the mean energy necessary to sustain such action is around 1mA@3V. Wireless transmission of the data requires some energy, of course. However, using adequate settings of duty cycle and a Position Pulse Modulation (PPM modulation scheme) technology, the transmission of a complete message could take less than 0.9mA@3V average (20-25mA@3V pulses and nearly 0mA consumption of the radio block and microprocessor between pulses), and an average power of <3µW for the whole cycle making feasible the use of Energy Harvester to power up the system. For this calculation a block of measurement + transmission of 6ms transmitted every 10s has been considered: of course the longer the cycle rate the less the energy required to run the system.
In a real use case scenario a system transmits around every 1 minute (unless a puncture with fast leak is detected) and the total energy budget will depend on the average vehicle operating temperature: however high temperatures leads to an inevitable increased energy consumption of microprocessor as well as, for certain extents, of the RF block and consequent inefficiencies of energy conversion in a harvesting scenario. In the picture of Figure 2 an overview is shown of the schematic building blocks of an autonomous TMS module.
Figure 2: Schematic TMS system with the different building blocks.
RF communication: as those used to run TPMS devices, the Low Power Short Range Device working on the ISM (Industrial, Scientific and Medical) frequencies (such as the 434, 315, 868 or 915 MHz or 2.4 GHz) refers to one class of devices that best fits requirements of IOT: for the purpose of collecting data in a suitable and efficient way they are engineered taking into account the lowest duty cycle to minimize power consumption. One of the controversy that stay at the top of the paradigm of energy, and is also linked to the best technological choice to fit requirements for system’s durability over time, is whether it should have bi-directional or mono-directional features but, generally speaking, it goes in itself that bidirectional functionality requires larger amount of energy. In order to keep within the required average life time it appears that, using a small battery of 20mm in diameter and a capacity of approximately 200mAh, the power consumption of the system should be in the order of less than 10µW and for such reason it is clear the system architecture will necessarily need to adopt Ultra Energy Efficient modulation and probably mono-direction features.
Depending on the type of application the suitable modulation for Energy Efficient Wireless Sensors in the IOT are the standard ASK (Amplitude Shift Keying), the FSK (Frequency Shift Keying) and the Pulse Position Modulation (PPM): since the major challenge is the reduction of the energy needed to transmit each single bit of information (meaning how many nJ are necessary to transfer each single bit of information over the air) it appears that the Pulse modulation is the one that fits at the best the requirements: in Pulse Position Modulation schemes the energy is concentrated in micro-pulses and only for time lapse of few microseconds, which means the overall resulting energy budget is a fraction of the energy used to transmit bit of data using standard constant carrier FSK (Frequency Shift Keying) or ASK (Amplitude Shift Keying) schemes used in wireless sensors.
Here are described the characteristics of these low energy requirements:
· Pulse Modulation (4-PPM).
· 0mA energy consumption of radio block between pulses.
· 64 bits of transmitted data each message.
Figure 3 – Basic sketch of energy consumption during a message generated by a micro.sp TX
Figure 3 shows the energy consumption during the transmission of a message generated by a PPM transmitter. The X-axis represents the time in ms; the Y-axis represents the energy consumption in mA. The red line shows the average energy consumption during a complete message: since PPM runs on micro-pulses the system is optimized to absorb higher current only during pulse generation whilst in between of them the energy is close to a negligible value: between each one the pulses the whole system enters in a kind of sleep mode, allowing a great optimization of the energy budget. Even thought the maximum current required for the pulses transmission is 20mA or higher, the average current budget is less than 900µA which corresponds to less than 2.7mW @ 3V DC necessary to send a 64-bit message. Taking into consideration the TPMS use case to monitor pressure and temperature in tires the average power required by the TX RF transmitted unit is 2.7mW, with a peak power level as high as 75mW: the total energy budget over the duty cycle therefore is only 3.5µW at 3V supply voltage that, in comparison with the energy required for the ASK or FSK communication is a negligible value.
Fusion Sensor the Virtual and Physical layer as the game changer in IOT: A Virtual Sensor is a mathematical model that fuses together multiple sensors readings to estimate new measures that can’t be measured directly with a Physical Sensor. On the other hand a Physical Sensor, that is a device that can gather the collection of data to create and calibrate a Virtualized layer, is the necessary hardware that input the initial parameters for the algorithms to run properly and adjust mathematical models and equations: in other words the Virtual and Physical layer are symbiotic each other and their mix is an important condition to leverage the science of Deep Learning and of the Artificial Intelligence.
Just to give an example: SSE (Side Slip Estimator) is a virtual sensor that could measure the sideslip of a vehicle with, in some case, a precision of 0.5 degree. Since there’s not physical device that can be installed on a production vehicle, to measure in a reliable way such side slip angle the design shall take into consideration interlacement with other sensor readings such as tire sensors, gyroscopes, axis sensors or other physical data, that are normally available at vehicle's chassis level through CAN bus, and combine them with the hidden layers to fit in the algorithms.
The reliability of the Virtual Sensor is strictly dependent on the algorithm modeling and generally relies on the process used for the generation of the Virtual Layer but it also relies on the data feeding that the Physical layer is capable to deliver: the more the physical data quantity is fed into the mathematical model the higher the precision of algorithm will come out. Virtual sensors are available on standard productions and are mainly used as a feedback for control systems and machines: they are provided with a formal mathematical stability check and with a diagnostic control layer to ensure that the Virtual Sensor Algorithm will lay inside the design limits in any working condition. System Requirements and Computing Resources in Virtualized sensors arrays are dependent both on the Virtual Sensor algorithm and the specific design: in automotive the Virtual Sensors are adapted for embedded processors and OS. Most of the applications run on standard Automotive ECUs using a fraction of the available processing power (from 1% to 10%). Technological choice in System Computing Resources depend on several parameters and is a matter of research in Model-Based Design Reliability Analysis that consists in creating a mathematical model of the system by using physical equations to determine accuracy of mathematical model, measurements, fault detection analysis. The Model-Based Design approach is a sub-system that when tuned by a Data-Driven approach through interlacement between the physical and virtual layers could potentially optimize system's performances. In other words looks like that the "fused device", that could rise out from the symbiosis of the Physical and Virtual layers, is the actionable process that could veste the role of game-changer: in a foreseeable future this Ultra Energy Efficient Wireless Sensors might be seen as the last piece of the puzzle capable kick in the "trillion sensors vision":
“micro.sp®”: a breakthrough technology that enable the trillion sensor vision in the Internet of Things.
STE Industries: Ste industries is an advanced fabless R&D center that, using core technologies in the space of data telemetry, is able to create innovative advances across industries, delivering unmatched unique solutions in the emerging sectors of energy efficient wireless sensors applied to the Internet of every things and big data collection
Micro.sp®: is a patented Innovation in short range radio frequency and refers to an innovative pulse schemes that enables highly miniaturized wireless sensors with very low power consumption equivalent to 2-3 orders of magnitude better than conventional system. The “micro.sp®” technology unlocks next generation battery powered IOT Wireless Sensors: with a total energy requirement of less than 220nJ/bit, when compared to other standards truly allows system’s lifetime up to 10 years by using the smallest CR1225 12mm 48mAh capacity lithium cell and leading research in the space of energy harvesting as a power source. “micro.sp®” technology, is a fully industrialized and commercially available device in the markets such as automotive, smart city, smart agriculture and vehicle monitoring and demonstrated cost savings largely below competitors’ target.
History: STE Industries is a Spin-off of STE srl, an Italian company founded in 1965 as an early Italian pioneer in radio frequency research, and focusing on Ultra Low Power Wireless Sensors for data telemetry and the Internet of Things. STE Industries is currently working with a number of major research institutes and organizations to develop vibration and thermal based energy harvesting to power up wearable sensors and Ultra Low Power wireless nodes. Recently STE Industries has begun, through its licensee Bridgestone (market leader in tire manufacturing) the rollout of battery powered Ultra low energy devices for trucking fleets to measure pressure in tires (TPMS) and, through the Inpulse® joint venture (see Inpulse agritech for reference) the roll out of innovative systems for Smart Agriculture.
This choice is because vehicles operators are more sensitive to TPMS benefits, namely from one to five percent improvements in fuel efficiency (due to correctly inflated tires) and the reduced downtime (early detection of slow leaks and puncture) for saving fuel and improved economy scale in the transportation and supply chain. Additionally in Smart Agriculture, because of the impact it will have in the incoming digitalized world, industries are looking at reaching up to new concepts of “agri-tech” and 4.0 supply chain to meet demand in agriculture yeld, new policies in food demand preservation and related health requirements to meet growth of worldwide population by 2050.
Technology: The Ultra Energy Efficient Wireless Sensor- which the company calls “micro.sp®”- consumes less than 500 nA with a total system’s budget, depending on the application specific requirements, of less than 2μW that is well below conventional wireless sensors. The underlying “micro.sp®” technology uses very short radio frequency pulses for wireless communication. The length of the pulses is on the micro-scale, typically between 2-5 μS, and make use of a dedicated pulse position modulation (PPM) scheme to efficiently transmit in the short range sensor’s data such as pressure, temperature, battery life and many other values that can be monitored through standard transducers. Additionally, overall size is also reduced: in example whereas conventional systems to monitor pressure and temperature in tires (TPMS) consists of a dedicated plastic box beneath the tire stem containing electronics components, STE Industries TPMS incorporates the completecircuitry into the tire stem itself.
The “micro.sp®” technology meets requirements in many of the market verticals such as industrial, building automation, lighting, smart city a smart tracking of assets thus enabling new business models opening new opportunities of collecting data reflecting what are the modern needs to improve industrial processes and optimize data processing in the IOT and big data mining domain.
This article was originally posted on LinkedIn.