Search

Context: Historically, automotive manufacturers have adopted rigid requirements engineering processes, which allowed them to meet safety-critical requirements while integrating thousands of physical and software components into a highly complex and differentiated product. Nowadays, needs of improving development speed are pushing companies in this domain towards new ways of developing software. Objectives: We aim at obtaining a manager perspective on how the goal to increase development speed impacts how software intense automotive systems are developed and their requirements managed. Methods: We used a qualitative multiple-case study, based on 20 semi-structured interviews, at two automotive manufacturers. Our sampling strategy focuses on manager roles, complemented with technical specialists. Results: We found that both a requirements style dominated by safety concerns, and decomposition of requirements over many levels of abstraction impact development speed negatively. Furthermore, the use of requirements as part of legal contracts with suppliers hiders fast collaboration. Suggestions for potential improvements include domain-specific tooling, model-based requirements, test automation, and a combination of lightweight pre-development requirements engineering with precise specifications post-development. Conclusions: We offer an empirical account of expectations and needs for new requirements engineering approaches in the automotive domain, necessary to coordinate hundreds of collaborating organizations developing software-intensive and potentially safety-critical systems.

A silicon micromachined waveguide hybrid coupler for J-band (220-325 GHz) applications is presented. The coupling and phase shifting functions are realised using a 90° ridged waveguide phase shifter, a joint waveguide section and a port extension. The hybrid coupler is fabricated using a micromachined waveguide technology employing a double H-plane split that results in low losses. For the design band 230–300 GHz, initial VNA measurements indicate coupling coefficients of -3.2±0.4 dB, amplitude imbalance better than -0.5/+0.3 dB, phase imbalance better than -14°/+11° and an input reflection coefficient below -14 dB. To the best of our knowledge, this is the first time a low-loss micromachined hybrid coupler of this type is characterised within this frequency range. QC 20171211 SSF Swedish Foundation for Strategic Research, Synergy Grant Electronics SEl3-007

Heterodyne and direct detection radiometer receivers at 94 GHz have been realized in a 130 nm Si/SiGe BiCMOS process. They have been assembled into fully self-contained metal housings with a waveguide flange for the antenna connection. The receivers, which employ a SiGe LNA with below 4 dB noise figure, show overall responsivities of 620 µV/K and 380 µV/K respectively.

The currently important competences of university managers and faculty staff include not only teaching and assessing skills, but also ability to develop and to execute a strategy of consistent modernization of educational programmes according to the purposes of university. The aim of this paper is to propose a model of how the CDIO framework (CDIO — Conceive, Design, Implement, Operate) can be applied for designing and implementation of faculty development programme and for shaping a particular set of skills of managers and faculty staff for efficient and consistent modernization of engineering programmes. According to the CDIO context the programme provides complex preparation of the trainees to go through all stages of educational programme improvement: starting with setting programme objectives and learning outcomes to evaluating its quality. The authors describe the structure and implementation technology of the programme as well as the results of its approbation. The programme evaluation obtained from various stakeholder groups showed that the programme is relevant, has high practical value and high level of novelty for all categories of respondents. Besides, the survey results reveal the issues which are particularly challenging for Russian HEIs and identify the faculty skills which are mainly wanted to develop.

This paper presents an RF-MEMS SP4T switched LNA MMIC used in a 24 GHz beam-steering antenna module. The SP4T LNA has a measured gain of 10-16 dB at 16-30 GHz (NF=2.6-3.0 dB at 16-26 GHz) with up to 37 dB of isolation at 24 GHz when one of the four MEMS switches are closed while the other switches are open. To the authors' knowledge, this work represents a first-time realisation of a MEMS SP4T LNA MMIC used as an on-chip active beam-switching network in an adaptive antenna. The experimental results successfully demonstrate that the MEMS SP4T LNA can be used to control the antenna beam in 4 different directions between -55o and 45o in the same plane.

This thesis presents solutions and studies related to the design of reconfigurable and wideband receiver circuits for system-on-chip (SoC) radiometer front-ends within the millimetre-wave (mm-wave) range. Whereas many of today’s mm-wave front-ends are bulky and costly due to having discrete RF components, single-chip receiver modules could potentially result in a wider use for emerging applications such as wireless communication, short range radar and passive imaging security sensors if realised with adequate performances and at a lower cost. Three main topics are considered in this thesis, monolithic integration of low-loss RF-MEMS (Dicke) switch networks and switched LNAs in MMIC/RFIC foundry processes, designs of SiGe wideband (IF) amplifier and broadband power detectors up to W-band (75-110 GHz).Low-loss and high isolation GaAs and SiGe RF-MEMS switch networks were designed and characterised for the 30-110 GHz range. A GaAs MEMS Dicke switch network has a measured minimum loss of 1 dB and maximum isolation of 19 dB at 70-96 GHz, respectively, making it a potential candidate in Dicke switched radiometer receivers. Furthermore, single-chip 30 GHz and W-band MEMS Dicke switched LNA designs have been realised for the first time in SiGe BiCMOS and GaAs mHEMT processes, respectively.For a targeted 94 GHz passive imaging application two different receiver topologies have been investigated based on direct-detection and direct-conversion (heterodyne) architectures. An optimised detector design fabricated in a 0.13 μm SiGe process achieves a more wideband input matching than earlier silicon W-band detectors and is competitive with reported III-V W-band detectors in terms of a higher responsivity and similar NEP.A SiGe 2-37 GHz high-gain differential (IF) amplifier design achieves a more wideband matching and an order of magnitude higher linearity than a recent single-ended SiGe LNA. The SiGe IF amplifier was integrated on-chip with a power detector in a 5-35 GHz IF section. Their broadband properties compared with other IF amplifier/detector RFICs, make them suitable for W-band down-conversion receivers with a larger pre-detection bandwidth and improved sensitivity. The experimental results successfully demonstrate the feasibility of the SiGe 5-35 GHz IF section for high performance SoC W-band radiometers using a more wideband heterodyne receiver architecture.

This paper presents the results of some W-band power detector and wideband (IF) amplifier circuit designs made in 0.25 μm and 0.13 μm SiGe BiCMOS processes. Two 0.25 μm SiGe wideband power detector and amplifier RFICs present an NEP =1-2 pW/Hz1/2 at 85-101 GHz and s21=10-19 dB at 2-32 GHz, respectively. To the authors' knowledge, the proposed SiGe detector design reports the widest Sn bandwidth (s11≤ -10 dB 84-104 GHz) among SiGe based W-band detectors. Two 0.13 μm SiGe detector/amplifier circuits show a simulated NEP=0.2-0.9 pW/Hz1/2 at 75-100 GHz and S21= 19-21 dB at 5-30 GHz. The presented SiGe power detectors and (IF) amplifiers are targeting broadband applications such as passive imaging.

This paper presents the circuit designs and results of a wideband power detector and an on-chip slot antenna fabricated in a 0.25 μm SiGe BiCMOS process. The W-band SiGe detector has close to 20 GHz of operational bandwidth (s11≤−10 dB at 75–92 GHz) and a responsivity of 3–5kV/W (NEP = 10–16 pW/Hz1/2) at 83–94 GHz. The SiGe on-chip slot antenna design covers a wide bandwidth (70–110 GHz) with 2 dBi of simulated gain at 94 GHz. The realised wideband SiGe BiCMOS power detector and on-chip antenna are targeting highly integrated single-chip RF front-ends for broadband sensing applications up to 110GHz.

This paper presents mm-wave capacitive RF-MEMS based Single-Pole-Double-Throw (SPDT) switches fabricated in a SiGe BiCMOS process technology. Three different SiGe RF-MEMS based SPDTs (targeting the 40–80 GHz range) present 3–4 dB of in-band losses and up to 20–25 dB of isolation, respectively. The measured in-band attenuation of the characterized SiGe MEMS SPDTs corresponds to a loss of 2–3 dB when close to 1 dB of combined losses within the RF pads has been removed. The experimental s-parameter data was obtained from RF-tests of more than 300 characterized SPDT switch networks indicating a high fabrication yield and process repeatability of the fabricated SiGe MEMS circuits. The validated SiGe MEMS switch circuits can enable single-chip reconfigurable ICs for wireless communication and sensing applications up to 100 GHz.

We present an experimental study of some GaAs MMIC based RF-MEMS switch networks used here as limiter test circuits and that were tested also within a standard low-frequency package. The small and large signal results for such discrete as well as packaged GaAs MEMS limiter MMICs shows possibilities for obtaining a very wideband RF power limiter functionality within small circuit dimensions. Self switch actuation tests made on a packaged GaAs MEMS limiter circuit showed a power limiting effect of 11 dB for an RF input power of 16 dBm at 4 GHz.

This article describes the progress made during the first project year of the FP7 EU-project MEMS-4-MMIC that focuses on the merging of innovative RF-MEMS switch designs with established MMIC-technology. The ultimate goal is to include RF-MEMS switches into a standard commercial foundry process. Such switches will then become standard building blocks for MMIC circuits such as transistors, resistors and capacitors. Besides the development and qualification of the MEMS-MMIC designs and processes, MEMS-MMIC components are designed for specific applications (phased arrays, switch beam networks, etc.) that will benefit from this type of building blocks. First prototypes of latter components have been realized and measured in the first year of the project, and are presented in this paper. (11 refs.)

In this paper, we present GaAs MMIC based reconfigurable RF MEMS impedance matching networks for highly integrated (potentially single-chip) frequency-agile LNAs and adaptive multi-band front-ends. Such GaAs MMIC based RF MEMS LNA matching networks have been realized with a frequency tuning range of 40% (10–16 GHz and 15–23 GHz, respectively) and 1–3 dB of in-band losses. Simulated tunable LNA results based on measured data of GaAs MMIC MEMS matching circuits (and simulated data of MEMS matching networks made on quartz) show the potential of achieving a high gain and low in-band noise figure over such wide tuning ranges.

In this paper, we present different types of reconfigurable RF MEMS based matching networks intended for frequency-agile (multi-band) LNAs. Measured results of 2-bits matching networks show a centre frequency tuning range of 2-3 GHz (10-13%) around 20 GHz and 1.5-2.0 dB of minimum losses. Simulated tunable LNA results based on measured data of the RF MEMS matching networks show the possibilities of achieving similar high gain, good matching and low NF over the whole tuning range. The results demonstrate the potential of using RF MEMS switches for the realization of tunable LNAs at microwave and millimetre-wave frequencies.

In this paper we present the main scope and initial results obtained within a pan-European research activity aiming at a successful monolithic integration of innovative RF MEMS switches into a standard GaAs MMIC process technology. Measured results of a fabricated ohmic contact type of RF MEMS switch developed by the OMMIC foundry show promising results e.g. in terms of low RF losses (0.2-0.4 dB is obtained from DC up to 30 GHz). The experimentally verified GaAs based RF MEMS switch has further been used to design low-loss MMIC building blocks such as switching, matching and phase shifting circuits for various RF applications up to 40 GHz. (12 refs.)

This report summarizes some of the recent results at FOI with respect to multi-band and reconfigurable front-ends for flexible and multi-functional RF systems. Firstly, we report on a frequency agile X-band smart skin digital beamforming antenna based on using an 8-10 GHz tunable active filter and an image rejection mixer in a receiver GaAs MMIC. Secondly, we have also investigated the possibility of using RF MEMS based reconfigurable matching networks for realizing tunable bandpass LNAs with even wider tuning ranges (e.g. 6-10 GHz could be possible according to our simulations). Finally, we study a system concept for a Ka-band multi-functional electronically steerable antenna (ESA) on a small UAV based on using sub-arrays with low-loss MEMS phase shifters. The results show that adequate RF performance (in terms of 2 dB of average losses at 35 GHz) can be possible to achieve with a Ka-band 4-bits MEMS phase shifter design made on quartz.

We have studied the possibility to realize tunable RF front-end filters and LNAs for wide-band phased arrays using a 0.25 mum UTSireg (ultra thin silicon) CMOS low cost technology. Measured results of an X-band passive filter-bank and a C-band active tunable filter show that adequate performance can be obtained (in terms of relatively low losses or high gain and out-of-band rejection over wide tuning ranges) while using small circuit areas. The UTSireg CMOS based RF components we present in this paper could contribute to the miniaturization and integration of RF front-ends in phased arrays.

We study possibilities of implementing two key components for the RF front-ends of adaptive multi-functional phased arrays using a SiGe BiCMOS low cost technology. Measured data of an L-band to X-band LNA together with simulated results of a 5.2-5.5 GHz tunable bandpass LNA (active filter) show that adequate performance (in terms of high gain, low noise figure and power dissipation) potentially can be achieved while using small circuit areas. We further propose the use of low loss RF MEMS based TTDs as frequency tuning elements in tunable bandpass LNAs to be able to achieve wider center frequency tuning ranges (up to 20-50%) and also lower values of noise figure and power dissipation, respectively

A 7.9-9.7 GHz tunable active monolithic microwave integrated circuit (MMIC) filter intended future on-chip X-band radar receiver front-ends is presented together with measured and simulated results. Typical measured filter data over the agile frequency band show a maximum gain of 11-16 dB, a noise figure of 6 dB, an input-referred third order intercept point of 0 dBm and 20-23 dB of out-of-band rejection at 2 GHz below the filter center frequency.

In this paper, we present a 7.9-9.7 GHz on-chip radar receiver front-end intended for a digital beamforming X-band smart skin phased array antenna. This agile single-chip receiver front-end could potentially enable a significant size and cost reduction of the microwave receiver modules in such an adaptive frequency hopping radar system. Measured results show a close to adequate performance.

We present a neural network based approach for identifying salient features for classification in neural networks. Our approach involves neural network training with an augmented cross-entropy error function. The augmented error function forces the neural network to keep low derivatives of the transfer functions of neurons when learning a classification task. Such an approach reduces the output sensitivity to input changes. Feature selection is based on the reaction of the cross-validation data set classification error due to the removal of the individual features. We compared the approach with two other neural network based feature selection methods. The algorithm developed outperforms the methods by achieved a higher classification accuracy on three real world problems tested.

In this paper, we present theoretical estimations of the spurious-free dynamic range for first and second order MMIC filters based on two different recursive active filter topologies. For a given relative filter bandwidth together with given values of the noise figure and the third order output intercept point of the low noise amplifier used in the recursive active MMIC filters, we estimate the highest possible filter spurious-free dynamic range when the filter noise figure is approaching that of the low noise amplifier.

One application area of automatic computer analysis of colored images is quality control of multicolored pictures in newspaper printing. The multicolored pictures in newspapers are made by printing cyan, magenta, yellow, and black dots on each other in screens with different angles. During the printing process, the operator needs to control the amount of ink of the different colors to achieve the desired result. One important factor which influences the result is the percentage of the area covered by ink of the different colors in every part of the printed picture. This can easily be determined if one is able to recognize the color of every pixel in computerized image of the print. The authors look at how neural networks of different type and different unsupervised learning techniques were combined to produce a hierarchical architecture with classification accuracy high enough to use in print quality control. >

This paper presents a colour image segmentation method which attains a high segmentation accuracy even when regions of the image that have to be separated are very similar in colour. The proposed method classifies pixels into colour classes. Competitive learning with "conscience" is used to learn reference patterns for the different colour classes. A nearest neighbour classification rule followed by a block of fuzzy post-processing attains a high classification accuracy even for very similar colour classes. A correct classification rate of 97.8% has been achieved when classifying two very similar black colours, namely, the black printed with a black ink and the black printed with a mixture of cyan, magenta and yellow inks.

A fully integrated X-band radar receiver front end based on only one down converting stage has been designed. The two chip MMIC (monolithic microwave integrated circuit) front end includes an active tunable bandpass filter in combination with an image rejection mixer. A novel digital time shifter is utilised to accomplish the tunability of the bandpass filter. The tuning range of the filter is between 8.7 and 9.8 GHz. With the intermediate frequency (IF) equal to 360 MHz, the total image rejection of the front end exceeds 45 dB. These results indicates that a compact one chip X-band radar receiver front end can be achieved.

A novel frequency tunable X-band MMIC filter design based on a balanced cascaded second order recursive active filter topology is presented together with simulation results. The simulated results indicate that this filter design can achieve a performance adequate for future advanced on-chip radar receivers.

The possibilities and the limitations of tunable narrow band recursive active MMIC filters with respect to future adaptive on-chip radar receiv- ers are investigated. We compare the measured and he simulated results of two recursive active MMIC ilters with the corresponding theoretical results obtained using our analytical filter models. A good agreement between these results is generally found. Using our analytical filter models we also estimate the noise and intermodulation properties of the low noise amplifier used in the filters that are necessary in order to achieve a filter performance adequate for future adaptive on-chip radar receivers.

Presents an approach to determining colours of specks in an image taken from a pulp sample. The task is solved through colour classification by an artificial neural network. The network is trained using possibilistic target values. The problem of post-processing of a pixelwise-classified image is addressed from the point of view of the Dempster-Shafer theory of evidence. Each neighbour of a pixel being analysed is considered as an item of evidence supporting particular hypotheses regarding the class label of that pixel. The experiments performed have shown that the colour classification results correspond well with the human perception of colours of the specks.

In this paper, we present a novel analytical model for recursive active microwave integrated filters. For a given value of the third order output intercept point of the amplifier used in the filters, we show how the values of the third order output intercept points of two different filter topologies are related to the Q-values of the filters. Also, for a given Q-value, our results show which of the two filter topologies should be used to obtain the lowest possible filter noise figure in combination with the highest possible filter third order output intercept point.