GNV’S TRAINING MITHRA
Make an Enquiry

  • Smart Sensors in IoT

    • 21 May 2022
    • 1045 Views

    Smart Sensors in IoT

    Table of Content

    Sensing the physical phenomena is a key aspect in any Internet of Things application and sensors help exactly to do the same.

    Sensors:

    Both sensors and actuators are collectively called ‘Transducers’. Transducers are the devices that convert the energy of one kind into the energy of another kind. Some common transducers are as follows:

    Quality Being Measured The input device (Sensors) Output device (Actuators)

    Light

    Light Dependent Resistor (LDR)

    Photo Diode

    Phototransistor

    Solar cell

    Lights and Lamps

    LEDs and Displays

    Fibre Optics

    Temperature

    Thermocouple

    Thermistor

    Thermostat

    Resistive Temperature Detectors (RTD)

    Heater

    Fan

    Force or Pressure

    Strain Gauge

    Pressure switch

    Load cells

    Lifts and Jacks

    Electromagnetic vibrations

    Position

    Potentiometer

    Encoders

    Reflective/slotted opto-switch

    LVDT

    Motor

    Solenoid

    Panel Meters

    Speed

    Tacho – generator

    Reflective/slotted optocoupler

    Doppler effect sensors

    AC and DC Motors

    Stepper Motor

    Brake

    Sound

    Carbon microphone

    Piezoelectric crystal

    Bell

    Buzzer

    Loudspeaker

    A sensor is a transducer that converts a physical phenomenon into an electrical signal. Sensors provide the interface to the physical world through electrical devices, such as microcontrollers/microprocessors, computers, etc. Sensors may have multiple transducers. For example, for a sensor that has to convert something in the form of chemical energy into electrical energy, if it is difficult to convert directly from chemical to electrical energy, it uses more than one transducer. Transducer 1 will convert chemical energy to heat and transducer 2 will convert heat to electrical energy.

    smart sensors

    There are many types of Sensors and they are classified based on output type, method of detection, means of detection, conversion phenomenon, active or passive, etc., and with the advent of the Internet of Things, these sensors have undergone evolution and as a result of that today we have ‘Smart Sensors‘, where we are integrating electronics into a sensor to make it an intelligent sensor.

    What is a Smart Sensor?

    A sensor producing an electrical output when combined with interfacing electronic circuits is known as “Smart Sensor", it is a combination of both sensor and actuator.

    The smart sensor will have intelligent features and some electronics that can perform:

    • Data conversion
    • Bi-directional Communication
    • Take a decision
    • Perform Logical operations

    Sensor + interfacing circuit = smart sensor

    Smart Sensor Functional Diagram:

    A Smart sensor senses measurand - physical quantity, property or Condition to be measured and Signal condition and storage unit has Analog to Digital converters which converts the signal into a digitally readable form and stores in its memory and further processes it like aggregating, error checking, etc., before sending to microprocessor or microcontroller.

    Click here to Learn Data Science Training in Chennai

    smart sensors

    Smart Sensor Functions:

    Smart sensor carries four functionalities

    • Measurement

      Measurement is a common functionality of any sensor where the sensor is expected to detect a physical signal and convert the signal into an electrical signal. Signal will also do some signal processing if it is an integrated sensor. Smart sensors have additional features like Correction compensation. Let us consider an example of correction compensation – Offset (An offset means that the sensor output is higher or lower than the ideal output.)and gain (Gain is a ratio between input and output) are two parameters of sensors mostly adjusted during fabrication and calibration cycle (Sensor calibration is a method of improving sensor performance by removing structural errors in the sensor outputs) are done in the factory itself, based on that adjustment in offset and gain is done. These adjustments usually change in time for different reasons, requiring the device to be recalibrated. Smart sensors may have the correction functions in its memory and processors may apply for compensation from time to time that means they would be able to do Self-calibration.

      Diagnosis - This is the inherent ability for a smart sensor to determine if it operates properly. With the IoT system complexity steadily increasing, malfunctioning components must be identified as quickly as possible.

    • Configuration

      The configuration in terms of technological, functional, operational aspects include Active Installation Support - sensor would be able to alert on alignment errors or installation errors

      • Configuration of operational modes- Based on the environmental conditions the sensor can be configured into different modes such as low power mode
      • Configuration of fault identification, response

    • Verification

      The verification function may offer services, such as continuous supervision of the intelligent sensor’s behaviour, using a set of supervisory equipment/circuits implemented in the sensor. Supervisory results can be stored in a database (eg: FIFO, Flash) & systematically updated and available for maintenance purposes. It offers the necessary services for diagnosis, allowing the user to locate faults whenever they are detected.

    • Communication

      Another important feature of the smart sensor is communication and this is a bidirectional – sensor to the main microprocessor or microcontroller and vice versa. Smart sensors will have mechanisms for synchronization communication between sensors and microcontrollers and they will support different data rates. Industrial Smart sensors come with a plug and play mechanism, like a CAN bus mechanism where they can be connected to the network and removed as easily as possible and also mechanisms to identify their own identity in the network.

    Smart Sensors Networks/ Wireless Sensor Networks

    Many high-end applications are designed by integrating IoT connectivity into Smart sensor technology. We are calling such connected Smart sensors as Smart Wireless Sensor networks or WSNs. These WSNs will sense, measure and collect information from their environment or from the object (that has to be measured) and they transmit to the outside world based on a user-defined decision process.

    Some examples of Wireless Sensor nodes or Smart sensor nodes:

    nodes in smart sensor IoT

    In the future, we may expect smart sensors which Modify their behaviour to optimize the collection of data from the external world along with advanced learning capabilities like machine learning that Combines -

    • A prior knowledge (available before experience)
    • Adaptive learning (based on experience)

    These sensors understand the environment they are put into and they can manage a wide range of conditions using their functional features like auto-calibration - these calibrate themselves without any further connections, compensated measurements that is they will compensate errors in the measurements and also, they evaluate their own health.

    Smart Sensors have their own components integrated onto the same PCB (Printed Board Circuit) by using these Smart sensors one can reduce production testing costs since they have firmware-based signal processing capabilities, processed data validation, power management, and multi-sensing capabilities. This results in improved reliability and performance.

    Watch Free Videos on Youtube

    Smart Sensors Networks Applications

    • Industrial

      In industries machines and equipment are monitored and controlled for pressure, temperature, humidity level, and also for vibrations. A Smart Sensor can monitor all these parameters at one go and also connects to the network without any other hardware assistance. This helps to maintain machinery and also ensure safety for employees handling the machinery.

    • Finger Recognition

      A fingerprint sensor scans and captures a digital image of the fingerprint pattern. The image captured is called livescan. Using that live scan a biometric template will be created and stored for matching.

    • Pattern Recognition

      When the sensor detects the contours of an object, it compares with them and also with models in a reference image.

    • Telecommunication

      A smart card similar to SIM card, called a Wireless Identity Module (WIM), Using this card e-commerce transaction can be done with 100 percent security using encryption and digital signature.

    • Smart Dust

      Smart dust is a hypothetical wireless network of tiny microelectromechanical (MEMS) sensors, robots, or devices, which can detect (for example) light, temperature, or vibration. The devices will eventually be the size of a grain of sand, or even a dust particle, with each mote having self-contained sensing, computation, communication, and power.

    • Biomedical Applications

      Many smart sensors for biomedical applications have also been developed by using chip technology .e.g. biochips Cyto-sensor micro-physio-meter: biological applications of silicon technology.

    • MEMS and Process Control

      MEMS(Micro-Electro-Mechanical Systems) are very small physical systems. MEMS sensors are a combination of electrical and mechanical components. MEMS uses a modified integrated circuit (computer chip) fabrication techniques and materials to create these very small mechanical devices.

    • Defence Applications

      Smart cameras can detect objects, perform crowd pattern analysis, secure zone intrusion detection and so on by using advanced software analytics and report alarms using IP network facilities in them. Smart Sensors are also used in monitoring EMI fatigue loading, thermal cycling vibration and shock levels, corrosive environments.

Recent Posts