Science and IT news

Researchers devise efficient power converter for internet of things by reducing resting power consumption by 50%

The “internet of things” is the idea that vehicles, appliances, civil structures, manufacturing equipment, and even livestock will soon have sensors that report information directly to networked servers, aiding with maintenance and the coordination of tasks. Those sensors will have to operate at very low powers, in order to extend battery life for months or make do with energy harvested from the environment. But that means that they’ll need to draw a wide range of electrical currents. A sensor might, for instance, wake up every so often, take a measurement, and perform a small calculation to see whether that measurement crosses some threshold. Those operations require relatively little current, but occasionally, the sensor might need to transmit an alert to a distant radio receiver. That requires much larger currents. Generally, power converters, which take an input voltage and convert it to a steady output voltage, are efficient only within a narrow range of currents. But at the International Solid-State Circuits Conference last week, researchers from MIT’s Microsystems Technologies Laboratories (MTL) presented a new power converter that maintains its efficiency at currents ranging from 500 picoamps to 1 milliamp, a span that encompasses a 200,000-fold increase in current levels.“Typically, converters have a quiescent power, which is the power that they consume even when they’re not providing any current to the load,” says Arun Paidimarri, who was a postdoc at MTL when the work was done and is now at IBM Research. “So, for example, if the quiescent power is a microamp, then even if the load pulls only a nanoamp, it’s still going to consume a microamp of current. My converter is something that can maintain efficiency over a wide range of currents.”Paidimarri, who also earned doctoral and master’s degrees from MIT, is first author on the conference paper. He’s joined by his thesis advisor, Anantha Chandrakasan, the Vannevar Bush Professor of Electrical Engineering and Computer Science at MIT. Packet perspective The researchers’ converter is a step-down converter, meaning that its output voltage is lower than its input voltage. In particular, it takes input voltages ranging from 1.2 to 3.3 volts and reduces them to between 0.7 and 0.9 volts.“In the low-power regime, the way these power converters work, it’s not based on a continuous flow of energy,” Paidimarri says. “It’s based on these packets of energy. You have these switches, and an inductor, and a capacitor in the power converter, and you basically turn on and off these switches.” The control circuitry for the switches includes a circuit that measures the output voltage of the converter. If the output voltage is below some threshold — in this case, 0.9 volts — the controllers throw a switch and release a packet of energy. Then they perform another measurement and, if necessary, release another packet. If no device is drawing current from the converter, or if the current is going only to a simple, local circuit, the controllers might release between 1 and a couple hundred packets per second. But if the converter is feeding power to a radio, it might need to release a million packets a second. To accommodate that range of outputs, a typical converter — even a low-power one — will simply perform 1 million voltage measurements a second; on that basis, it will release anywhere from 1 to 1 million packets. Each measurement consumes energy, but for most existing applications, the power drain is negligible. For the internet of things, however, it’s intolerable. Clocking down Paidimarri and Chandrakasan’s converter thus features a variable clock, which can run the switch controllers at a wide range of rates. That, however, requires more complex control circuits. The circuit that monitors the converter’s output voltage, for instance, contains an element called a voltage divider, which siphons off a little current from the output for measurement. In a typical converter, the voltage divider is just another element in the circuit path; it is, in effect, always on. But siphoning current lowers the converter’s efficiency, so in the MIT researchers’ chip, the divider is surrounded by a block of additional circuit elements, which grant access to the divider only for the fraction of a second that a measurement requires. The result is a 50 percent reduction in quiescent power over even the best previously reported experimental low-power, step-down converter and a tenfold expansion of the current-handling range. “This opens up exciting new opportunities to operate these circuits from new types of energy-harvesting sources, such as body-powered electronics,” Chandrakasan says. “This work pushes the boundaries of the state of the art in low-power DC-DC converters, how low you can go in terms of the quiescent current, and the efficiencies that you can achieve at these low current levels,” says Yogesh Ramadass, the director of power management research at Texas Instruments’ Kilby Labs. “You don’t want your converter to burn up more than what is being delivered, so it’s essential for the converter to have a very low quiescent power state.”

Micromote now one cubic millimeter computer with a megabyte of flash memory using a few nanowatts of power

In 2015, the Michigan Micro Mote constituted the first complete, operational computer system measuring as small as two millimeters across. “To be "complete," a computer system must have an input of data, the ability to process that data - meaning process and store it, make decisions about what to do next – and ultimately, the ability to output the data.” Prof. Blaauw explained. “The sensors are the input and the radios are the output. The other key to being a complete computer is the ability to supply its own power.” In 2017, the systems are one cubic millimeter. There are several types and are a line of the world’s smallest computers. They have one megabyte of flash memory. Their broader goal is to make smarter, smaller sensors for medical devices and the internet of things—sensors that can do more with less energy. Many of the microphones, cameras, and other sensors that make up eyes and ears of smart devices are always on alert, and frequently beam personal data into the cloud because they can’t analyze it themselves. Some have predicted that by 2035, there will be 1 trillion such devices. “If you’ve got a trillion devices producing readings constantly, we’re going to drown in data,” says Blaauw. By developing tiny, energy efficient computing sensors that can do analysis on board, Blaauw and Sylvester hope to make these devices more secure, while also saving energy. Micro mote designs now use only a few nanowatts of power to perform tasks such as distinguish the sound of a passing car and measuring temperature and light levels. They showed off a compact radio that can send data from the small computers to receivers 20 meters away—a considerable boost compared to the 50 centimeter range they reported last year at ISSCC Another micro mote they presented at the ISSCC incorporates a deep-learning processor that can operate a neural network while using just 288 microwatts. Neural networks are artificial intelligence algorithms that perform well at tasks such as face and voice recognition. They typically demand both large memory banks and intense processing power, and so they’re usually run on banks of servers often powered by advanced GPUs. The Michigan Micro Mote contains solar cells that power the battery with ambient light, including indoor rooms with no natural sunlight, allowing the computers to run perpetually. This line of “smart dust” devices includes computers equipped with imagers (with motion detection), temperature sensors, and pressure sensors. They are the culmination of work initiated by Blaauw and Sylvester on very low-power processing for millimeter-scale systems. An Astonishing Lack of Power A key breakthrough in the size/power matchup came with the Phoenix processor in 2008. The Phoenix processor is miniscule at 915 x 915µm2, and boasts ultra-low operating voltage and a unique standby mode that results in an average power consumption of only 500pW. (Consider that 1pW is the average power consumption of a single human cell.) {source}<iframe width="853" height="480" src="" frameborder="0" allowfullscreen></iframe>{/source}

Researchers Reverse Hall Coefficient – Medieval Mail Armor Inspired Development of Metamaterial with Novel Properties

Scientists of Karlsruhe Institute of Technology (KIT), however, were inspired by medieval mail armor when producing a new metamaterial with novel properties. They succeeded in reversing the Hall coefficient of a material. The Hall effect is the occurrence of a transverse electric voltage across an electric conductor passed by current flow, if this conductor is located in a magnetic field. This effect is a basic phenomenon of physics and allows to measure the strength of magnetic fields. It is the basis of magnetic speed sensors in cars or compasses in smartphones. Apart from measuring magnetic fields, the Hall effect can also be used to characterize metals and semiconductors and in particular to determine charge carrier density of the material. The sign of the measured Hall voltage allows conclusions to be drawn as to whether charge carriers in the semiconductor element carry positive or negative charge. Mathematicians already predicted theoretically that it is possible to reverse the Hall coefficient of a material (such as gold or silicon), i.e. to reverse its sign. This was expected to be achieved by a three-dimensional ring structure resembling medieval mail armor. How-ever, this was considered difficult, as the ring mesh of millionths of a meter in size would have to be composed of three different components. Christian Kern, Muamer Kadic, and Martin Wegener of KIT’s Institute of Applied Physics now found that a single basic material is sufficient, provided that the ring structure chosen follows a certain geometric arrangement. First, they produced polymer scaffolds with a highest-resolution 3D printer. Then, they coated these scaffolds with semiconducting zinc oxide. The result of the experiment: The scientists can produce meta-materials with a positive coefficient, even though their components have negative coefficients. This sounds a bit like the philosopher’s stone, the formula, by means of which medieval alchemists tried to convert one substance into another. But here, no conversion takes place. “The charge carriers in the metamaterial remain negatively charged electrons,” Christian Kern explains. “Hall measurements only make them appear positively charged, as the structure forces them to take detours.” Kern admits that this discovery so far is of no practical use. There are sufficient solids with both negative and positive Hall coefficients. But Kern wants to continue research. The next step will be the production of anisotropic structures with a Hall voltage in the direction of the magnetic field. Normally, Hall voltage is directed vertically to current and magnetic fields. Such unconventional materials might be applied in novel sensors for the direct measurement of magnetic field eddies.

New brain implant design is meant to restore vision to the blind

Experiments that let a paralyzed person swig coffee using a robotic arm, or that let blind people “see” spots of light, have proven the huge potential of computers that interface with the brain. But the implanted electrodes used in such trials eventually become useless, as scar tissue forms that degrades their electrical connection to brain cells Next month, tests will begin in monkeys of a new implant for piping data into the brain that is designed to avoid that problem. The project is intended to lead to devices that can restore vision to blind people long-term. Researchers at Harvard Medical School will use a new kind of implant that will go beneath the skull but can rest on the surface of an animal’s brain, instead of penetrating inside the organ. An array of microscopic coils inside the hair-like device can generate powerful, highly targeted magnetic fields to induce electrical activity at particular locations in the brain tissue underneath. The implant will also be tested when placed inside brain tissue. The device will be used to stimulate the visual cortex of the monkeys to try and re-create the activity normally triggered by signals from the eyes—creating the sensation of sight without the eyes’ input. Ultimately, the goal is to use the implant to convert signals from a camera into brain activity. Unlike conventional electrodes, the coils' effectiveness shouldn't degrade over time. Magnetic fields aren't impeded by tissue forming around an implant as electric currents are. The three-year project is supported by a multi-million dollar grant under the BRAIN initiative, created by President Obama to improve scientists’ understanding of how the brain works. Todd Coleman, an associate professor at the University of California, San Diego, says that the new approach is promising, although it will be some time before it becomes clear how exactly it could be used in humans. If the technology proves useful, its use cases don’t have to be limited to the brain, he says. “There could be very nice applications in other parts of the body,” says Coleman. He suggests the tiny coils could be used to modulate activity in the system of more than 100 million neurons associated with the human digestive system, for example, to help people with conditions in which the gut doesn’t move food along as it should. Casse says he is interested in exploring use of the technology on the vagus nerve in the chest to control symptoms of PTSD.