Research Projects

Computational models for improved EMC design – eMC Hammer

Background and Motivation

The proposed research project is part of the research theme Electronics for Internet of Things at the EIS research environment. Several other running projects and project proposals are being part of this research theme. One of the already launched projects is EMC-NG that looks into new building practices for next generation of electronics with a special focus on EMC and 3D-printing. Another running project develops new education programs in electronics at advance level with an international exchange program. Both these projects are funded by the KK-foundation. Currently we from the KK-foundation also apply for an adjunct professor in electronics in order to strengthen the research group. The research theme Electronics for Internet of Things is part of Electronics Centre in Halmstad (ECH) where researchers together with involved industrial partners continue to develop a unique research direction for next generation building practice foreseen to become utterly important in the near future. The specific outcomes of research conducted at ECH include a better understanding of new innovative design and production methods to realize embedded systems suitable for fulfilling the vision of the pervasive computing revolution. A successfully developed network of interacting embedded systems must exhibit high reliability, including excellent EMI discrimination, at a low cost. Design and optimization from system integration point of view requires modular design systems comprising EMC prequalified modules for easy system integration.

The projects focus on electromagnetic modelling perfectly fit the ECH strategy promoting regional economic growth and the formation of an excellent research center in the area of electronics. This project continues the partnership with already involved partners, create new relations and lead to both increased scientific knowledge and business advantages. The coproduction conducted within this project have the goal to elevate the knowledge and the research conducted by the partners within this project to a high level of excellence using numerical modeling for solving electromagnetic problems. The expected outcome of the project is partners and companies able to utilize the power of numerical computation in their product development. We further expect the research group to develop an environment that can support and contribute to this development. The long term impact of introducing these methods change the way that these companies approaches electromagnetic design challenges and build a new knowledge base that will increase these companies competitiveness on the global market.

Research Problem and Approach

The starting point for this project is to verify the possibility to build models for simplified systems with single sources and coupling structures, where accurate measurements of the near-field and far-field of the sources are used as input into the model. A complication is that all the complex field vector components must be measured with sufficiently high spatial sampling rate and accuracy. A typical problem of EMC fault tracing is to successfully extrapolated from near field to far field, and for the coupling to other radiating structures to be established. Any feedback to electronics sources including non-linear elements makes this problem even more complex. The measurements described, both far-field and near-field, strongly relates to antenna design and antenna verification. Here we especially will look into antennas requiring fixed phase centers, i.e. the phase of the radiation pattern has no directional dependence. This is a desired property of antennas used for positioning using satellite systems (GNSS) or radar. For optimization of antenna shape with respect to the radiation (Maxwell) and geometrical effects on electron interaction in semiconductors (Poisson Boltzmann), it is desirable to develop numerical methods that allow easy and efficient modification of the models geometric design. Recently developed methods to cut elements in the finite element method allow such changes in an easy way and can be developed further for example for wave propagation problems. Model parameters have to be determined against experimental data using numerical inverse identification. The solution to a partial differential equation is then compared with the measured signal, and the parameters of the differential equation are varied to give the same signal. In this respect numeric surrogate models have to be developed and implemented. Numerical models for interface with eg transmission impedance may be necessary in cases with high impedance to avoid ill-conditioned problems. Such models have been developed for analogous situations in the continuum mechanics but have never been tested for wave propagation.

Standard finite element simulations depend on access to a computational mesh. The generation of a mesh, for complex models, is unfortunately still a complicated matter involving significant manual labor, making simulation of optimal design difficult since geometry changes often require re-meshing. In industrial applications usually only one mesh is used, and the reliability of the computational results is judged by experience or rules of thumb regarding the number of nodes per wavelength, a restricted approach that hinders efficient use of simulation technology. Another important trend is that a given geometry can be described by a multiplicity of different formats, from point based descriptions obtained by laser scanning, via triangulations used in geometry modeling software, to high fidelity parametric NURBS (Non-Uniform Rational B-Spline) representations that are standard in CAD software. These descriptions are then used to create computational meshes, by triangulating first the surface and then, given the surface mesh, meshing the volume. Even though this process is highly automated, the amount of manual work that has to be invested is still not to be neglected.

In this proposal we seek to apply a recently developed technology called CutFEM that enables high order representation of both the geometry and the electromagnetic fields in a given domain as well as on its surface. It is based on the ideas in two different approaches: (1) the X-FEM method of Hansbo and Hansbo [1, 2], extending to standard finite elements the partition of unity approach of Belytschko and co-workers [3]. In this method, the elements are cut by a real or artificial interface, and the approximation is allowed to be discontinuous across a discretization of the interface, inside the elements. Only zero order operators were considered on the discrete interface; the next step (2) to allow for the discretization of differential equations on the interface was taken by Reusken and co-workers in recent work [4, 5]. Their approach employs a three-dimensional background mesh cut by an interface, and a problem, in the form of a surface partial differential equation, is posed and solved on the discretized interface only.

This project’s prior research problems are:

  1. Methods for Maxwell’s equations with cut boundaries.
  2. Treatment of antennas, with application to shape optimization and inverse identification for electromagnetic wave propagation.
  3. Methods for Maxwell’s equations on surfaces, based on tangential calculus.

Goals

The main expected output (deliverable) from this project is a modelling toolbox, CutFEM4EM, which can be used for modelling and optimization of electromagnetic problems by end users. This toolbox should be possible to interface with best practice CAD systems. The output from an academic perspective is to attend at least three scientific conferences per year, produce several journal publications and attend industrial conferences in the area of electromagnetic modelling. The expected outcome is better modeling capabilities of the near field electromagnetic characteristics for treating electromagnetic interference in early design phases and dealing with EMI problems in a more efficient way. The expected long term impact is more efficient models and methods to be used for component model-based design and troubleshooting methods for future embedded electronics. The goals are to develop models for individual components easily integrated and parameterized to system models, and to enable EMC modeling of electronic systems at the design stage and to identify the cause of EMC problems. This type of tool not only allows for new construction methods for electronics but also gives an opportunity to significantly speed up the design cycle for embedded electronic systems and thereby reduce costs. The merging of mechanical and electromagnetic modelling is a very important part of future building practice of highly integrated electronic systems. It can be seen as an enabler for the envisioned pervasive computing revolution and especially for long tail IoT applications.

[1] A. Hansbo and P. Hansbo. An unfitted finite element method, based on Nitsche’s method, for elliptic interface problems. Comput. Methods Appl. Mech. Engrg., 191(47-48):5537–5552, 2002.

[2] A. Hansbo and P. Hansbo. A finite element method for the simulation of strong and weak discontinuities in solid mechanics. Comput. Methods Appl. Mech. Engrg., 193(33-35):3523–3540, 2004.

[3] N. Moes, J. Dolbow, and T. Belytschko. A finite element method for crack growth without remeshing. Internat. J. Numer.Methods Engrg., 46(1):131–150, 1999.

[4] M. A. Olshanskii and A. Reusken. A finite element method for surface PDEs: matrix properties. Numer. Math., 114(3):491–520,2010.

[5] M. A. Olshanskii, A. Reusken, and J. Grande. A finite element method for elliptic equations on surfaces. SIAM J. Numer. Anal., 47(5):3339–3358, 2009.

Cohabitant Radars – Realization study for reducing mutual interference impact of radars – a necessity for future short distance applications, where radars are brought in massive use.

Background and Motivation

The development in the last decades within electronics has been truly been amazing. The down scaling of the chip technology and mass production have enabled highly advanced functionality in a large range of products such as within mobile communication, GPS and digital cameras. The experience from these areas has opened up possibilities to implement sophisticated functionality in other areas as in this case in radar technology. As an example, car radar is an area which can benefit from more sophisticated algorithms and system solutions now possible to implement in hardware. These possibilities are in fact quite essential since a massive use of radar close by will bring the fundamental issues of radar mutual interference in focus. Unlike communication signals, radar signals are shaped by their requirement of avoiding leakage of the strong transmit signal into the sensitive radar receiver – normally this is achieved by pulsing. However this requirement on signal shape prevents effective measures to minimize mutual interference. This study will investigate the possibility (already demonstrated) of cancellation radar in which the necessary transmit receive isolation is obtained by leakage cancellation. Achieving this, waveforms can be freely selected and particular chosen to minimize mutual interference. Though car radars serve as an important application, the issue of interference is pressing for all types of radar and the project will also consider the larger issue of radar system cohabitation for radars in general. As for short range radar it will include radars operating with meter range, where interference issues are a crucial shortcoming for radar applications in sensors for guidance and control.

Research Problem and Approach

As said to approach the problem of radar interference the project proposes a new principle for radar operation. The new approach is based cancellation to obtain isolation between the received signal and the transmit signal. A very deep cancellation is necessary but is in fact enabled by the transmit signal being exactly known. By a suitable radar design including the appropriate calibration paths, the radar transfer characteristics can be precisely determined, enabling equally precise software defined cancellation filter. The possibility of this radar principle has been demonstrated experimentally by Saab. After the required signal processing a net cancellation of no less than 160 dB was achieved.

This study will develop on two or three fronts. There will be one activity studying waveforms exploiting the same frequency band but providing a minimum of mutual impact. There will be another activity considering the appropriate hardware implementation of the technology, in particular when it comes to small and miniature systems, suitable for mass production such as car radars. Thirdly there will be an activity of determining the performance requirement to be met in order to ascertain that the cancelation principle will work. It is anticipated that level of required isolation will reduce significantly for the shorter range systems, meaning that such designs may be well suited for realization in the form of e.g. ASIC circuits.

Goals

Overall goal is to assess the feasibility and advantage of cancellation radar, for various fields of application. The advantage would depend on the complexity, reliability and cost of implementation, as well as what possible improvements regarding nearby radar operation is obtained. All these aspects will be covered in a final report. On particular goal is to identify new applications for radars (like guidance and control in e.g. robotics) not previously considered, since the possibilities of small and short range radars critically depends on the continuous and drastic improvements in miniature electronics.

Prepare Halmstad – Establishment of a prepare program for industrial PhD studies within the area of electronics 

Bacckground and Motivation

There are a need to increase the understanding of PhD studies and enable more people in SME companies to qualify themselves for PhD studies. A milestone is to get more companies to understand the value that a PhD student can bring to a company in the form of knowledge, competence, and as a door opener to new networks and knowledge environments.

Approach

The project is performed in coproduction with external actors, divided in three activity blocks: A) planning: mandatory courses and elements and possible complementing courses. B) Development:  Connection to present research is strengthen, templates (course plans, contracts etc.) and marketing materials are prepared.   C) Establishment: Company contacts are intensified.  Interviews with possible students and assessment/admission are planned. A company employed carries out a pilot prepare program to extract experiences. Marketing is conducted throughout the entire project.

Effects and results

Anticipated effects are a better understanding of PhD studies in SME companies within the region and an understanding of the advantage a company can have of industrial PhD student. Anticipated long term results are that the knowledge level in SME companies will increase and that their network connectivity to the global technology arena increase, which will increase the competitiveness for regional SME technology based companies on the global market.

Elektronikdesign för framtidens ”Internet of Things” – Utveckling av avslutande del av civilingenjörsprogram inom elektronik i samverkan med näringslivet

Bakgrund

Civilingenjörer utbildade på svenska lärosäten har haft utomordentligt stor betydelse för svenskt näringsliv och därigenom för Sveriges ställning som tekniskt ledande nation. Många civilingenjörer arbetar nära teknikutveckling, men i och med grundutbildningens bredd återfinns de lika ofta i ledande positioner på företag. Det är ingen överdrift att påstå att det idag råder stor brist på civilingenjörer inom det mycket viktiga området elektronik. Denna brist är allvarlig inte minst med tanke på att det sker en explosionsartad utveckling i världen inom detta område med helt nya typer av elektronikkomponenter och system baserade på nya material och innovativ banbrytande teknik som exempelvis RFID eller nanoteknik, såväl som helt nya typer av byggsätt med additiv 3D-printing. En viktig inriktning inom utbildning, forskning och företagande i Halmstad berör inbyggd intelligent elektronik som kommunicerar och samverkar, i många fall genom innovativ radioteknik. Verksamheten kring denna inriktning har med tiden blivit allt mer konsoliderad, och kan idag anses ingå i det stora område som beskrivs som Internet of Things (IoT), ett framtidsscenario där 100-tals miljoner unikt identifierbara objekt är sammankopplade i ett komplext internet-liknande system.

Denna utveckling av framtidens elektronik kopplar starkt till befintlig forskning inom KK-profilerna CERES och CAISRvid Högskolan i Halmstad. Centrala forskningsfrågor finns bl a inom följande forskningsområden:

  • Trådlös industriell kommunikation och fordonskommunikation
  • Korthållskommunikation
  • Kognitiv radioelektronik
  • Strömsnål RF elektronik/sensorer
  • Nya tillverkningsmetoder för elektronik som t ex 3D-printing
  • EMC kompatibilitet
  • Beräkningsteknik (multifysik) för komplex integration av modern elektronik

En annan stark forskargruppering vid Högskolan i Halmstad, som vid den senaste externa forskningsutvärderingen initierad av KK-stiftelsen (ARC-13) utsågs till excellent, beforskar nya typer av material och design av komponenter och sensorer baserade på nanostrukturer för framtidens inbyggda elektronik. Centrala forskningsfrågeställningar här finns bl a inom följande forskningsområden:

  • Nya banbrytande halvledarmaterial för elektronik, fotonik och spinntronik
  • Ny LED teknik för t ex solid state lighting
  • Optoelektronik för on-chip och chip-to-chip kommunikation
  • Optiska sensorer för avbildning och övervakning
  • Solceller

Forskningen inom nanoelektronik bedrivs i stark samverkan både med Lunds Universitet (nmC@LU), forskningsinstitut (Acreo AB) och företag (IRnova, SolVoltaics och GLO).

Behovet av kompetens inom elektronikområdet bland företagen i Halmstadregionen och intresset för forsknings- och utbildningssamverkan med Högskolan i Halmstad inom avancerad elektronikutveckling ökar. Företagen HMS Industrial Networks AB, NIBE AB och Trefoten Development AB arbetar mycket aktivt med Högskolan inom utvecklingen av ett nytt Elektronikcentrum i Halmstad (ECH). Dessa företag är också medsökande företag inom ett av KK-stiftelsen beviljat forskningsprojekt inom 3D-printing av elektronik (KK projektet: Electromagnetic Compatibility for Next Generation of Embedded Devices)

Metod

Mot denna bakgrund kan man konstatera att medan forskningen växer sig allt starkare inom inbyggd elektronik och behovet av kompetens inom området bland företagen i Halmstadregionen ökar, saknas välutvecklad grundutbildning på avancerad nivå inom samma område. Vi avser därför utveckla ett helt nytt civilingenjörsprogram inom elektronikdesign. Studenterna skall inom detta program få möjlighet att utveckla gedigna kunskaper kring design och utveckling av banbrytande ny elektronik, sensorer och material för att tillgodose näringslivets allt högre krav på spetskompetens. Utbildningens koppling till företagen ger även studenterna förutsättningar för att utveckla kunskaper inom innovation och entreprenörskap.

Vi ser stora vinster med att denna utveckling sker i nära samarbete med företagen inom ECH vilka också har deklarerat ett mycket starkt intresse för att såväl bidra till uppbyggnaden som att delta i genomförandet av nya ingenjörsprogram inom elektronik på avancerad nivå.

Ett stort antal företag och andra aktörer deltar tillsammans med företagen inom ECH ovan och Högskolan i Halmstad i nätverket Framtidens elektronik i Halmstad, ett större nätverk för generell utveckling av elektronikområdet i regionen. Nätverket är en stark stödgrupp för förnyelseprocessen kring utbildning inom elektronik vid Högskolan i Halmstad. Inom ramen för denna samverkan har även dessa företag visat stort intresse för att bidra till planering, utveckling och genomförande av avancerade utbildningsprogram inom elektronikområdet.

Mål

Denna ansökan avser att i samverkan med dessa företag utveckla den senare delen (de två sista åren) av ett civilingenjörsprogram som svarar mot Högskolans forskning och företagens kompetensbehov inom elektronikdesign. Detta omfattar utveckling av:

  • Kurser på avancerad nivå inom elektronik med inriktning mot IoT
  • Former för fortsatt utbildningssamverkan med företagen inom utbildningen
  • Infrastruktur för undervisning inom utbildningen (utrustning, anpassning av labmiljöer och projektrum)
  • Former för studentrekrytering och marknadsföring av utbildningen

Electromagnetic Compatibility for Next Generation embedded systems (EMC NG)

Background and Motivation

This research project is a strategic effort to increase the knowledge of the next generation of EMC requirements and possibilities in the context of two important and ongoing technology trends: the pervasive computing revolution, also known as Internet of Things (IoT), and new production methods, known as additive manufacturing. There is a need for understanding the new EMC environment and the requirements of electronics and embedded systems, which will follow in the footprints of the pervasive computing revolution. The pervasiveness of embedded systems, smart objects and IoT will be fulfilled by using the next generation of building and production practice. One of the most interesting emerging technologies in this context is additive manufacturing (3D printing) which has the potential to realize extreme integration of electronics and embedded systems.

Research Problem and Approach

The proposed research project is built around the EMC challenge of next generation embedded systems where high end research meets industrial best practice and innovation of new production methods and use of new materials. The focus is to conduct a comprehensive assessment, from an EMC perspective, of using additive manufacturing for integrating embedded electronics. The electromagnetic properties of the materials used in additive manufacturing are to be investigated, examples of important material parameters are: granularity, conductivity, permittivity, and permeability. Different additive production processes may provide varying structure granularity, spatial resolution, and a wide range of materials. However, there is a lack of standardized specifications for mechanical as well as electromagnetic properties enabling a good prediction on how manufactured parts will perform. To increase the opportunities of innovation in the area of additive manufacturing the following research questions have been formulated:

  • How can we provide the necessary knowledge and control of processes, structures, and materials used in additive manufacturing to produce commercially useable devices?
  • What electromagnetic properties do additive manufacturing provide, and does it provide new features usable for further integration of embedded electronics while sustaining EMC performance? What possibilities are there for extreme integration of electronic components, like antennas, for future wireless systems by applying additive manufacturing?

Goals

Besides the goal of achieving answers to the research questions, secondary goals of the project are: develop or improve laboratory measurement methods and models that more accurately reflect new requirements related to the pervasive computing vision; and develop best practice and recommendations for implementing these techniques into new EMC test standards. Accelerate the adoption of additive manufacturing and 3D printing technologies in the electronics sector and to increase competitiveness.