Embedded World 2020

•••2••• Innovationen New way to store quantum information Research results have the potential to revolutionise computing R esearchers at the National In- stitute of Standards and Tech- nology (NIST) have for the first time created and imaged a novel pair of quantum dots – tiny is- lands of confined electric charge that act like interacting artificial atoms. Such “coupled” quan- tum dots could serve as a robust quantum bit, or qubit, the fun- damental unit of information for a quantum computer. Moreover, the patterns of electric charge in the island can’t be fully explained by current models of quantum physics, offering an opportunity to investigate rich new physical phenomena in materials. Unlike a classical computer, which relies on binary bits that have just one of two fixed values – “1” or “0” – to store memory, a quantum computer would store and pro- cess information in qubits, which can simultaneously take on a mul- titude of values. Therefore, they could perform much larger, more complex operations than classi- cal bits and have the potential to revolutionise computing. Electrons orbit the center of a sin- gle quantum dot similar to the way they orbit atoms. The charged particles can only occupy specific permitted energy levels. At each energy level, an electron can occupy a range of possible po- sitions in the dot, tracing out an orbit whose shape is determined by the rules of quantum theory. A pair of coupled quantum dots can share an electron between them, forming a qubit. To fabricate the quantum dots, the NIST-led team, which includ- ed researchers from the Univer- sity of Maryland NanoCenter and the National Institute for Materials Science in Japan, used the ultrasharp tip of a scanning tunneling microscope (STM) as if it were a stylus of an Etch A Sketch. Hovering the tip above an ultracold sheet of graphene (a single layer of carbon atoms ar- ranged in a honeycomb pattern), the researchers briefly increased the voltage of the tip. The elec- tric field generated by the volt- age pulse penetrated through the graphene into an underlying layer of boron nitride, where it stripped electrons from atomic impurities in the layer and created a pileup of electric charge. The pi- leup corralled freely floating elec- trons in the graphene, confining them to a tiny energy well. But when the team applied a magnet- ic field of 4 to 8 tesla (about 400 to 800 times the strength of a small bar magnet), it dramatically altered the shape and distribution of the orbits that the electrons could occupy. Rather than a sin- gle well, the electrons now resid- ed within two sets of concentric, closely spaced rings within the original well separated by a small empty shell. The two sets of rings for the electrons now behaved as if they were weakly coupled quan- tum dots. This is the first time that research- ers have probed the interior of a coupled quantum dot system so deeply, imaging the distribution of electrons with atomic resolu- tion. To take high-resolution imag- es and spectra of the system, the team took advantage of a special relationship between the size of a quantum dot and the spacing of the energy levels occupied by the orbiting electrons: The smaller the dot, the greater the spacing. How information is stored has a huge effect on embedded systems. Photo: David Latorre Romero Too good to be real? Deep learning can fool listeners by imitating any guitar amplier Many popular guitar amplifiers and distortion effects are based on analogue circuitry. To achieve the desired distortion of the gui- tar signal, these circuits use non- linear components, such as vacu- um tubes, diodes, or transistors. As music production becomes in- creasingly digitised, the demand for faithful digital emulations of analogue audio effects is increas- ing. Professor Vesa Välimäki from Aal- to University explains that this is an exciting development in deep learning, “Deep neural networks for guitar distortion modelling has been tested before, but this is the first time, where blind-test listeners couldn’t tell the differ- ence between a recording and a fake distorted guitar sound! This is akin to when the computer first learned to play chess”. The main objective of the field of Virtual Analog (VA) modelling is to cre- ate digital emulations of these analogue systems which will allow bulky, expensive and fragile ana- logue equipment to be replaced by software plugins that can be used on a modern desktop or lap- top computer. A specific amplifier’s circuitry can be accurately simulated using cir- cuit modelling techniques, but the result is often a model that is too computationally demanding for real-time processing. Additionally, a newmodel has to be created for each amplifier being modelled, and the process is labour inten- sive. An alternative approach for VA modelling is “black-box” mod- elling. Black-box modelling is based on measuring the circuit’s response to some input signals and creating a model which rep- licates the observed input-output mapping. The study from which these results came, was based on the WaveNet convolutional neural network. The digital ampli- fier model is created using a deep neural network. Audio is recorded from a target guitar amplifier, and this audio is used to train the deep neural network to simulate that guitar amplifier. Alec Wright, a doctoral student, focusing on audio processing us- ing deep learning says, “The tests were conducted to validate the performance of models emulat- ing either the Blackstar HT5 Met- al or Mesa Boogie Express 5:50+ tube amplifiers. The models were created with a focus on real-time performance, and all of them can be run in real-time on a desktop computer”. All of this means that in the near future, all a guitarist will need to do is plug into their laptop that is running the deep neural plugin, and a thoroughly convincing vin- tage guitar amp sound will come from the speakers. It remains to be seen if guitar am- plifier purists will be willing to part with their beloved rigs, but this innovation paves the way for any audio enthusiast to digi- tally get the desired guitar sound, whether it be a Marshall, Fender, or anything else, in the studio. Replaceable? A Marshall amplifier. Photo: Made by Morro

RkJQdWJsaXNoZXIy NjM5MzU=