History – The Inception of Smart Dust
The advancements in Very Large Scale Integration have led us to the inception of neural motes or as we call it – smart dust. The foundation of smart dust dates back to 1998 when Dr. Kris Pister of the University of California, Berkeley developed an autonomous wireless sensor of five cubic millimeters, about the size of a grain of rice with a sensor, power source, analog circuitry and a programmable microprocessor.
The notion of smart dust eventually is to create a sensor so small, it is dust-like. Inventions have generated a neural mote that is about 20 micrometers to a millimeter. (Pister, Pister, and Boser, n.d.,). Smart dust is made up of minuscule motes or MEMS called microelectromechanical systems. MEMS are tiny devices that contain cameras, sensors and communication mechanisms to process and transfer data to a base node. This allows them to collect information on various aspects like pressure, humidity, sound, light, vibration and more with their sensors and communicate them to a cloud or other MEMS for numerous applications. These applications being, surveilling enemy borders for military purposes, monitoring activities in inaccessible regions, transport of animals or fragile goods – keeping a tab of their health, environment conditions and track them, monitoring crops to determine fertility and the list goes on. Smart dust is imperceptible, analogous to a micro sized drone, it is at the millimeter level currently but the goal is to produce a speck-sized sensor integrated with IoT and Swarm Intelligence – the back bone being advanced VLSI for designing low-power, flexible, radio frequency CMOS sensor architecture and networking.
Understanding Smart Dust
A node that contains embedded sensors, robots, or other devices that can sense vibration, magnetism, chemicals, light, temperature, and other stimuli is termed as a single smart dust. Smart dust can communicate and/or report data to a base computer, the cloud or other nodes. The rise of smart dust is a consequential event of miniaturization as well as a way to access information from unreachable regions. The initial motivation behind smart dust was to use it for the military applications expressed above, and ultimately led to the evolution of Internet of Things and Swarm Intelligence. Swarm Intelligence or Collective Intelligence is if AI was applied to a swarm, it is the coordinated group behavior without an external controller or leader. This is already seen in nature with bee colonies, bird flocks, fish schooling or ant colonies. In a fish school, every individual fish swims elegantly in the exact same direction and speed based on its perceived speed, distance and direction of neighboring fishes and objects. There is no single fish supervising the group. This depicts an advanced cognitive performance that results in self-organized information exchange in groups. In smart dust there will be billions of motes present in order to create a collective “dust” of motes that move in a single direction, exactly like that of dust being blown. Integrating smart intelligence to the motes allows for this to happen so MEMS can coordinate with one another. Internet of Things is a technology where devices exchange data with other devices via the internet and sensors. IoT is preferable for smart dust to detect stimuli like humidity or torque.
For several decades, the multibillion-dollar MEMS sector has been expanding, with significant markets in medical sensors, process control sensors, automotive pressure sensors and accelerometers. Many of these sensor processes now exhibit exponentially diminishing size/power/cost curves because of recent technological advancements. Micromotors are also created using different MEMS sensor technologies; millions of these micromotors are employed in commercial projection display systems like the Texas Instruments Digital Micromirror Device. (Pister, Warneke, and Liebowitz 2001, 8)
- Corner Cube Retroreflector
The entering beam reverses direction after being reflected three times, once by each surface, back again towards the source exactly where it came from. This is used for passive optical transmission.
A sensor is a device that detects input of any kind from the physical world and responds to it. Light, heat, motion, moisture, pressure, and a variety of other environmental phenomena can all be inputs. The output is often a signal that is translated into a display that can be seen or heard by humans at the sensor site or that is electronically sent over a network to be read or put via more processing. Sensors play a crucial role in the internet of things (IoT). They are responsible for creating an environment for collecting and processing data to be monitored, managed and controlled.
- Photodiode – Optical Receiver
A p-n junction or PIN structure is a photodiode. An electron-hole pair is produced in a diode when a photon with sufficient energy impacts the device. Also known as the inner photoelectric effect, this process. The photo diode allows optical data reception
- Steering Mirror
Beam steering is a method of changing the direction of radiation from the main lobe. This is deployed in acoustics to direct the sound from loudspeakers to a particular location. Beam steering is deployed by a steering mirror in smart dust for active transmission.
- Power Source
Smart Dust gets its power from thick-film batteries or solar energy via a solar cell consisting of a CMOS chip. A supercapacitor might serve as the mote’s power source. A significant technological problem is developing submillimeter-sized energy storage devices for ever smaller microelectronic components. However, scientists are still able to make them smaller and smaller, taking in consideration the example of a nano supercapacitor, which is the size of a speck of dust but has the voltage of a AAA battery. In the 2001 Energy and Performance Considerations for Smart Dust paper by K.S.J Pister the most simple energy storage was decided to be the lithium energy cell with an energy density of 300 W-Hr /kg. (Doherty et al., 2001, 13) It also asserted solar energy to be the most available source of energy because photoelectric energy can come even from indoor lighting and not just the sun. Sunlight producing 1mW/ conversion energy is best around 30%.
The working of the smart dust majorly relies on the communication aspect. Active and Passive transmission of information deals with the exchange of data between mote and base source. Passive communication is the transfer of data amongst motes. Micro controllers control Smart Dust Motes. These micro controllers include small sensors for capturing different kinds of data and these sensors are operated by timers. The microcontroller identifies the tasks that the mote must perform. In order to preserve electricity, it regulates the power going to various system parts. The sensor sends a signal to the microcontroller. Depending on the type of sensor, different data is processed – light, temperature, vibration, pressure, and acceleration. Following processing, it saves the outcome to memory. For information about any device willing to interact with it, the device employs an optical receiver.. The timers decide whether the mote should be powered or not when sending or receiving packets of information. When a timer ends, a portion of the device powers on to perform a task before turning off. When one of these timers expires, the sensor is then powered on, a sample is taken, and a digital word is generated. If the information is valuable, it may be saved straight in the SRAM or the microcontroller may be activated to process it in a more sophisticated manner. The timer restarts counting when this task is finished and everything is powered off once more. The microcontroller will create a packet holding sensor data or a message in response to a message or another timer expiring and deliver it using either the laser diode or the corner cube retro-reflector, depending on which it possesses. The message may instruct the mote specifically to do something, or it may simply be a message being sent from one mote to another as it travels to a certain destination. . The mote is capable of receiving a variety of packet types, including those containing fresh computer code that has been stored. The receiver is managed by another timer. After a certain amount of time, it is powered down yet again if it doesn’t see one. This ensures that the power is conserved with minimum wastage of energy.
An Outlook on CMOS and SoC Technology
An embedded method incorporates an analogue to digital converter (ADC) to transform an analogue signal into a digital format in order to gather data about the outside environment (data acquisition system.) Typically, an analogue voltage serves as the input signal, while a binary number serves as the output. Performance and adaptability of the ADC are essential since it serves as the interface between the sensed environment and the sensor network in its entirety. In the 2003 paper by Kris Pister and Boser, [MD Scott, BE Boser, KSJ Pister IEEE Journal of Solid-State Circuits 38 (7)] The switch network and SAR allow control of the ADC. In a computer network, a switch is a device that links other devices together. To allow communication between several networked devices, many data cables are inserted into a switch. Simple nMOS devices were used to build the switch network, together with pMOS devices for the supply reference and CMOS switches for the input signal. SAR is an abbreviation for Successive Approximation Register and is a type of ADC. It is obvious that the CMOS shrinking trend should continue. CMOS was created utilizing planar (2D) technology in the early 1970s. However, in order to adhere to Gordon Moore’s concept, downsizing and new technologies are crucial. Now, CMOS integrated circuits are designed using non-planar technology. By shipping at capacity with its leading edge (5nm) and comparable (7nm) designs that outperform Intel, TSMC has overcome Intel. Polysilicon has replaced the material originally utilized, which was aluminum. According to IBM and Intel, other metal gates have returned with the development of high-dielectric materials in the CMOS process for the 45 nm node and lower sizes. (Cook, B.W., S. Lanzisera, and K.S.J. Pister. “SoC Issues for RF Smart Dust.” Proceedings
of IEEE 94.6 (2006): 1177–1196. Web.) Midway through the 1990s, work on highly integrated sensor mote components began, leading to multichip systems that could be combined to form motes. For early Smart Dust motes, passive optical communication was investigated to reduce energy consumption. In addition to an 8-bit ADC, an optical receiver, a corner cube reflector passive optical transmitter, a light sensor, an accelerometer, a multi-voltage solar cell power supply, and limited processing, the smallest optical mote to date has a 4 mm diameter. The Spec mote was a later model of sensor mote that had an integrated CPU, SRAM, RF transmitter, and 8-bit ADC on a single CMOS chip.
A smart dust mote is an integrated package of sensors, optical receiver, beam steering mirror and other control circuitry. An integrated circuit that incorporates the majority or all of the parts of a computer or other electronic system is known as a system on a chip or system-on-chip. The typical, motherboard-based PC design, which divides components depending on purpose and links them via a central interface circuit board, is in contrast to SoCs. SoCs components are designed in high-level programming languages such as C++, MATLAB or SystemC and converted to RTL designs via high-level synthesis tools such as C to HDL. The evolution of System on Chip (SOC) design is driven by advancements in RF CMOS technology, high density and low power techniques in VLSI, design reuse, and reconfigurable approaches. SOC is highly useful in portable devices like smartphones, tablets, and laptops since it enables us to consolidate all necessary components into a single chip. These very advances have led us to the inauguration of Smart Dust. (Pannuto, n.d., #) The system-on-chip (SoC) technology improvements that have recently been made are profiled in the issue of Microelectronics TOE, which also offers strategic insights on IP, growth factors, difficulties, and application range. The 10 nm manufactured SoC from Mediatek, the SoCs for autonomous vehicles from NVIDIA, the semiconductor IP platform for performance improvement from Ultra SoC, and the fifth generation of EyeQ SoCs from Mobileye for intelligent vehicle systems are among the innovations covered.
There are numerous continuing discussions about the development of IoT devices and their impact on IT operations and security as IoT devices play a bigger role in business. No IoT device, however, is as outlandishly incredible as “smart dust,” which is evidence that IoT shrinkage appears to be approaching in the field of nanotechnology. Current advancements have established smart dust in the MEMs technology and are making way towards nanotechnology – NEMs. It is amazing to consider all the ways IoT first entered the enterprise IT landscape. We already have a large number of Internet-connected sensors, control appliances, and home automation devices, and their numbers are constantly growing. Even more puzzling, though, is where scientists are taking this technological idea. The advent of nanotechnology is undeniable, the Ministry of Electronics and Information Technology has taken up some major initiatives to promote Nanoelectronics research and innovation in India. At prestigious institutions around the nation, significant Nanoelectronics Centers of world standards have been developed. These Centers’ cutting-edge nanofabrication equipment has grown in popularity both in India and overseas. A flagship MeitY Indian Nanoelectronics Users Program-Idea to Innovation (INUP-i2i) programme is also being implemented at the Centre of Excellence in Nanoelectronics (CEN) at the IISc, IIT Bombay, IIT Delhi, IIT Madras, IIT Kharagpur, and IIT Guwahati. This has given the R&D community across the nation a fantastic opportunity to access cutting-edge nanofabrication facilities for conducting research. Several research papers and numerous breakthroughs for which patent applications have been made as a result of the research activities to date. A few start-up Indian companies in India have been granted licenses to use some of the technology produced at the CENs. Monitoring of the solar system’s planets’ and moons’ weather and seismic activity. Researchers are also investigating a brand-new design of a space telescope with a laser-controlled particle-swarm aperture that is released from a canister. The second stage of the “orbiting rainbows” project, which aims to merge space optics and smartdust, or autonomous robotic system technology, is funded by NASA’s Innovative Advanced Concepts Program. The lack of sufficient power on the small footprint and the difficulty integrating power systems into these highly scaled devices have been the key problems that researchers have been battling. IoT systems must rely on power conversion from outside sources like thermal, vibrational, light, or radio waves because battery storage density has not kept up with Moore’s law scaling trends.
Thousands of sand-grain-sized sensors that can detect ambient light and temperature make up smart dust and they each have wireless communications equipment attached to them. If you place a lot of them close to one another, they will automatically network. These sensors, which could be mass-produced for pennies each, could be installed all over homes and office buildings. A test was conducted on the newest smart dust particle, which has a volume of only 16 cu mm. It collects data from a photo-detector, transmits it using the CCR, and is powered by solar cells. Therefore, smart dust is coming. Micro-fabrication technique is used to integrate mechanical components, sensors, actuators, and electronics on a common silicon substrate to create Micro-Electro-Mechanical Systems (MEMS). While the micromechanical components are made using compatible “micromachining” processes that selectively etch away segments of the silicon wafer or add new structural layers to form the mechanical and electromechanical devices, the electronics are made using integrated circuit (IC) process sequences (e.g., CMOS, Bipolar processes). A complete system on a chip is realized through MEMS. VLSI technology will continue to fuel electronics innovation because of the growing demand for miniaturization, portability, performance, reliability, and versatility. Smart Dust is the proof of concept for the upcoming advancements in the VLSI sector.