A Spin Wave-Based Approximate 4:2 Compressor: Seeking the most energy-efficient digital computing paradigm

Mahmoud, Abdulqader; Vanderveken, Frederic; Ciubotaru, Florin; Adelmann, Christoph; Hamdioui, Said; Cotofana, Sorin

doi:10.1109/mnano.2021.3126095

Cited by 4 publications

(4 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Motivated by the SW interaction potential, researchers have investigated different SW logic gates and circuits [7][8][9][10][11][12][13][14][15][16][17][18] . In 10 , a Mach-Zehnder interferometer has been utilized to design the first experimental SW NOT gate, whereas the Mach-Zehnder interferometer has been used to build different single output gates such as Majority, (N)AND, (N)OR, and X(N)OR gates in 10 .…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, the unique parallelism feature of the SW technology has been utilized to demonstrate multi-frequency logic gates 8,12 . On the circuit side, different circuits have been illustrated at the conceptual level 14 , simulation level 7,[15][16][17] , and physical millimeter scale level 18 . In [19][20][21][22] , benchmarking of beyond CMOS technologies, e.g., SW, Graphene, SpinFET, has been performed by modelling these devices as a capacitor and parasitics, which is not quite appropriate for the SW case.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Would Magnonic Circuits Outperform CMOS Counterparts?

Mahmoud¹,

Cucu-Laurenciu²,

Vanderveken³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Would Magnonic Circuits Outperform CMOS Counterparts?

Mahmoud¹,

Cucu-Laurenciu²,

Vanderveken³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Would Magnonic Circuits Outperform CMOS Counterparts?

Mahmoud¹,

Cucu-Laurenciu²,

Vanderveken³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In the early stages of a novel technology development, it is difficult to provide a comprehensive assessment of its potential capabilities and impact. Nevertheless, some preliminary estimates can be drawn and are certainly of great interest and in this paper we follow this line of reasoning within the framework of the Spin Wave (SW) based computing paradigm. In particular, we are interested in assessing the technological development horizon that needs to be reached in order to unleash the full SW paradigm potential such that SW circuits can outperform CMOS counterparts in terms of energy consumption. In view of the zero power SWs propagation through ferromagnetic waveguides, the overall SW circuit power consumption is determined by the one associated to SWs generation and sensing by means of transducers. While current antenna based transducers are clearly power hungry recent developments indicate that magneto-electric (ME) cells have a great potential for ultra-low power SW generation and sensing. Given that MEs have been only proposed at the conceptual level and no actual experimental demonstration has been reported we cannot evaluate the impact of their utilization on the SW circuit energy consumption. However, we can perform a reverse engineering alike analysis to determine ME delay and power consumption upper bounds that can place SW circuits in the leading position. To this end, we utilize a 32-bit Brent-Kung Adder (BKA) as discussion vehicle and compute the maximum ME delay and power consumption that could potentially enable a SW implementation able to outperform its 7nm CMOS counterpart. We evaluate different BKA SW implementations that rely on conversion- or normalization-based gate cascading and consider continuous or pulsed SW generation scenarios. Our evaluations indicate that 31nW is the maximum transducer power consumption for which a 32-bit Brent-Kung SW implementation can outperform its 7nm CMOS counterpart in terms of energy consumption.

show abstract

Self‐Imaging Based Programmable Spin‐Wave Lookup Tables

Gołębiewski

Gruszecki

Krawczyk

2022

Adv Elect Materials

View full text Add to dashboard Cite

oscillations propagating without charge transfer in magnetic materials. [2][3][4][5] This property, combined with a wider range of available frequencies than in electronics and the ability to encode information both in amplitude and phase, makes SWs an important candidate for an information carrier in a new generation of computing devices, where Joule-Lenz heat and other obstacles may be significantly reduced. [6][7][8][9] Magnonic circuits (systems utilizing SWs) [10,11] may consist of waveguides through which SWs propagate, [12][13][14][15] and interference areas at crossings, for example, for creating majority gates. [16][17][18][19][20] The waveguides may also couple with other waveguides [21,22] to implement a logical operation. In this manner, it has been possible to demonstrate a 32-bit magnonic full adder [21] and SWbased approximate 4:2 compressor. [23] Another strategy is to use wide ferromagnetic film areas for SW operation and narrow waveguides as SW inputs. This approach was used to redirect [24][25][26] and process SWs. [27][28][29][30][31][32] Operation of these systems is based on the interference of incoming SWs. Therefore, a local modification of the medium (the magnonic equivalent of the refractive index) in which SWs propagate is crucial to design and optimize its functionality. It was recently shown that it could be achieved by an introduction of defects in so-called inverse design approach, [32] placement of programmable magnetic elements on top of that region, [30] or utilization of noncolinear magnetization textures. [33][34][35] This interference based strategy appears promising also for the realization of physical neural networks operating on SWs. [30,33] Thereby, interference effects open a promising avenue for the development of SWbased beyond-CMOS solutions.A plane wave passing through a system of periodically spaced obstacles (diffraction gratings or holes) interferes, creating a characteristic diffraction pattern in the near field, reproducing the grating image at specific distances from the input apertures. This phenomenon is known as the Talbot or the selfimaging effect, and was observed for light already in the 19th century. [36] The resulting interference pattern is called a Talbot carpet, and we have recently theoretically demonstrated that this effect can also occur for SWs. [37] The properties of Talbot carpets created by SWs strongly depend on material parameters, geometry, type, thickness of a magnetic material, and on dynamic parameters such as wavelength, orientation, and the value of an external magnetic field.Here, we exploit the self-imaging phenomenon occurring in a thin ferromagnetic multimode waveguide with SWs introduced by periodically spaced single-mode input waveguides. SWs entering the multimode waveguide have a controllable phase. In particular, we present a new class of reprogrammable magnonic blocks implementing array indexing operations.Inclusion of spin waves into the computing paradigm, where complementary metal-oxide-semiconductor devices are still at th...

show abstract

A Spin Wave-Based Approximate 4:2 Compressor: Seeking the most energy-efficient digital computing paradigm

Cited by 4 publications

References 46 publications

Would Magnonic Circuits Outperform CMOS Counterparts?

Would Magnonic Circuits Outperform CMOS Counterparts?

Would Magnonic Circuits Outperform CMOS Counterparts?

Self‐Imaging Based Programmable Spin‐Wave Lookup Tables

Contact Info

Product

Resources

About