Oddvar Søråsen

• Kaur, Sukhdeep; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2015). Characterization and Modeling of Nonlinearities in In-Plane Gap Closing Electrostatic Energy Harvester. Journal of microelectromechanical systems. ISSN 1057-7157. 24(6), p. 2071–2082. doi: 10.1109/JMEMS.2015.2470112. Full text in Research Archive Show summary

  This paper investigates in detail a micro scale in-plane gap closing electrostatic energy harvester with strong nonlinearities in squeeze-film damping, electromechanical coupling, and impacts on end-stops. The device shows softening response on increasing the bias voltage and saturation behavior on impact with end-stops at high enough acceleration amplitude. We demonstrate that a lumped model can adequately describe the measured nonlinear behavior for a range of operating conditions with nonlinear fluid damping force and impact force included in the model. While modeling capacitances, a finite-element method (FEM) is used to analyze fringing field effects on the capacitance variation for gap closing electrodes. The nominal capacitance is obtained from FEM analysis, for a range of under-cut values in the fabrication process treated as a free parameter in the model. The device modeled for linear and nonlinear squeeze-film damping force highlights the importance of nonlinear damping force to understand the device behavior over the range of operating conditions. With the compliant end-stops treated as spring-dampers and with proper choice of end-stop damping-coefficient, the model successfully captures the end-stop nonlinearities for a particular operating point and reproduces the dynamic pull-in phenomena at 8 V bias, and rms acceleration 0.6 g , as observed in the experiments. Thus, the model described in this paper reproduces the subtle nonlinear effects dominating the dynamics of an in-plane gap closing electrostatic energy harvester.

• Kuppireddi, Srinivasa Reddy; Sayanu, Pamidighantam & Søråsen, Oddvar (2013). Influence of SU8 Photopolymer for Silicon RFMEMS Packaging Applications. International Journal of Materials Science and Engineering. ISSN 2315-4527. 1(2), p. 79–81. doi: 10.12720/ijmse.1.2.79-81. Full text in Research Archive Show summary

  Epoxy based photopolymer–SU8 is evaluated as microwave packaging material for packaging of Radio Frequency Micro Electro Mechanical Systems (RFMEMS). Standard and Slotted Coplanar Wave guide (CPW) transmission lines are chosen as test vehicles. Transmission lines of lengths up to 10mm are formed on HRS substrates. The fabricated transmission lines are subjected to wafer level packaging. The measured S-parameter data shows a significant difference prior to and after packaging. Further, the effects due to the packaging could be model educing lumped components and the measured data are presented in this work.

• Le, Cuong Phu; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2013). Wideband excitation of an electrostatic vibration energy harvester with power-extracting end-stops. Smart materials and structures (Print). ISSN 0964-1726. 22(7). doi: 10.1088/0964-1726/22/7/075020. Full text in Research Archive Show summary

  An electrostatic energy harvester with two-stage transduction is investigated for enhancement of bandwidth and dynamic range. The harvester includes a primary proof mass with two main transducers and end-stops for the proof mass functioning as secondary transducers. In the small acceleration regime, the power is primarily obtained from the main transducers. In the high acceleration regime, the mass impacts the end-stops and actuates the secondary transducers, generating additional output power. The device is designed and fabricated using the SOIMUMPs process and has a total active area of 4 × 5 mm2. Under wideband acceleration at high levels, the experimental results show that the total output power increases to about twice the output power of the main transducers, while the 3 dB-bandwidth is enlarged by a factor of 6.7 compared to the linear-response bandwidth at low levels. In comparison with a reference device made with the same die dimensions, the two-stage device improves output power instead of saturating when the maximum mass displacements of both devices reach the same limit. Measurement of output power demonstrates that the device with the transducing end-stops give an efficiency of 23.6%, while this value is 14.1% for the reference device with the conventional end-stops, at an acceleration spectral density of Sa = 19.2 × 10−3 g2 Hz−1. The efficiency is improved about by 9.5% in the impact regime.

• Kuppireddi, Srinivasa Reddy & Søråsen, Oddvar (2013). Electrostatic actuation gap reduction method and analysis for square plate resonator on SOI substrates, 2013 Symposium on Design, Test, Integration and Packaging of MEMS/MOEMS (DTIP). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467344777. p. 6–9. doi: 10.1109/smicnd.2013.6688687. Full text in Research Archive Show summary

  This paper describes a simple process method for obtaining submicron capacitive electrode gaps for lateral vibrating square plate resonators and the resulting advantages. A novel electrostatic actuation method is proposed to achieve submicron electrode resonator gaps below the fabrication limitation given by conventional optical lithography. The structures are built of single crystal silicon using silicon-on-insulator wafers without complex process steps. The process sequence and simulation results for submicron electrostatic gaps are reported.

• Kuppireddi, Srinivasa Reddy & Søråsen, Oddvar (2013). High aspect ratio lateral electrode nano gap rectangular plate micro-resonator novel process. In Navickas, Romualdas (Eds.), Proceeding of the NORCHIP 2013. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781479916474. p. 1–4. doi: 10.1109/NORCHIP.2013.6701998. Full text in Research Archive
• Kuppireddi, Srinivasa Reddy & Søråsen, Oddvar (2013). Lateral MEMS square plate resonator post-process nano gap creation method and study. In Dascalu, Dan (Eds.), Proceedings of the IEEE International Semiconductor Conference (CAS). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467356701. p. 319–322. doi: 10.1109/SMICND.2013.6688687. Full text in Research Archive
• Le, Cuong Phu; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2012). Microscale electrostatic energy harvester using internal impacts. Journal of Intelligent Materials Systems and Structures. ISSN 1045-389X. 23(13), p. 1409–1421. doi: 10.1177/1045389X12436739. Full text in Research Archive Show summary

  This article presents a new concept for electrostatic energy harvesting devices that increase output power under displacement limited inertial mass motion at sufficiently large acceleration amplitudes. The concept is illustrated by two demonstrated electrostatic energy harvesting prototypes in the same die dimension: a reference device with end-stops and an impact device with movable end-stops functioning as slave transducers. Both devices are analyzed and characterized in small and large excitation regimes. We found that significant additional energy from the internal impact force can be harvested by the slave transducer. The impact device gives much higher, up to a factor of 3.7, total output power than the reference device at the same high-acceleration amplitude. The bandwidth of the response to frequency sweeps is beneficially enlarged by up to a factor of 20 by the nonlinear mechanisms of the impact device.

• Le, Phu Cuong; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2012). Comparison of transducing end-stops with differentstiffness in mems electrostatic energy harvesters. In Allen, Mark G. (Eds.), The 12th international workshop on micro and nanotechnolgy for power generation and energy conversion applications : powerMEMS 2012. Transducers Research Foundations. ISSN 9780974361192. Full text in Research Archive Show summary

  This paper presents experiments on electrostatic energy harvesters with different designs of end-stop transducers, giving output power improvement and bandwidth enhancement under displacement-constrained operation. The end-stops function as additional electrostatic transducers collecting additional power from internal impacts. We investigate the influence of end-stop stiffness on output power by comparing two devices with different end-stop stiffness. The comparison shows that the compliant end-stops result in higher output power than the stiff end-stops do and increase the acceleration-amplitude range where transducing end-stops are beneficial. Compared to their linear response, the two variants enlarge the 3dB-bandwidth with factors up to 15.9 and 13.7 respectively.

• Michaelsen, Jørgen Andreas; Ramstad, Jan Erik; Wisland, Dag T & Søråsen, Oddvar (2011). Low-power Sensor Interfacing and MEMS for Wireless Sensor Networks. In Tarannum, Suraiya (Eds.), Wireless Sensor Networks. IntechOpen. ISSN 9789533073255. Full text in Research Archive
• Ramstad, Jan Erik & Søråsen, Oddvar (2011). Modeling and Design of Higher Order, Multi-Mode, Multi-Port MEMS Resonators in 90 nm CMOS. Procedia Engineering. 25, p. 1521–1524. doi: 10.1016/j.proeng.2011.12.376. Full text in Research Archive Show summary

  Abstract This work presents various RF mixer-filter MEMS resonators implemented directly in CMOS. Three different composite resonator designs have been made in a 90 nm CMOS process. These resonators move laterally in two different modes made possible by the flexible internal routing and stimulation of the multi-port creating these second mode mixer-filter capabilities. By utilizing post-processed self-assembly electrodes, narrow gaps between the electrodes and the resonator is achieved for increased electrostatic coupling. Based on three different resonator designs, 4th order filters are implemented by using a coupling beam connecting pairs of resonators. The higher order mechanical filters are driven differentially.

• Aunet, Snorre & Søråsen, Oddvar (2011). Editorial. Microprocessors and Microsystems: Embedded Hardware Design (MICPRO). ISSN 0141-9331. 35(7), p. 593–594. doi: 10.1016/j.micpro.2011.10.001. Full text in Research Archive
• Ramstad, Jan Erik & Søråsen, Oddvar (2010). Higher order FFSFR coupled micromechanical mixer-filters integrated in CMOS, Proceedings of the 28th Norchip Conference, Tampere, Finland, 15. - 16. November, 2010. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781424489725. Full text in Research Archive
• Kaur, Sukhdeep; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2010). Numerical analysis of nonlinearities due to rigid end-stops in energy harvesters, PowerMEMS 2010: Proceedings of The 10th International Workshop on Micro and Nanotechnology for Power Generation and Energy Conversion Applications. Micropower generation group. ISSN 9789073802889. p. 367–370. Full text in Research Archive Show summary

  This paper gives a numerical analysis of nonlinearities due to rigid, elastic mechanical end-stops in a vibrating energy harvester. Both an eigenvalue problem approach and time domain simulations are used. We consider two rigid end-stops limiting the motion of a proof mass. The dynamics are linear between the impacts which cause strong nonlinear behavior. Even though the output power varies little with change in acceleration amplitude, we find that there can be drastic changes in the proof mass motion with change of amplitude. In a wide range of acceleration amplitudes the phase space trajectories are very complicated and have periods much longer than the period of the vibration, some appearing chaotic. Only in a small range of amplitudes do we find simple periodic motion.

• Ramstad, Jan Erik; Kjelgård, Kristian Gjertsen; Nordbø, Bård Eirik & Søråsen, Oddvar (2009). RF MEMS front-end resonator, filters, varactors and a switch using CMOS-MEMS process. In Bourouina, Tarik; Courtois, Bernard; Ghodssi, Reza; Karam, Jean Michel; Soma, Aurelio & Yang, Hsiharng (Ed.), Design, Test, Integration and Packaging of MEMS/MOEMS 2009. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9782355000096. p. 170–175. Full text in Research Archive Show summary

  The paper describes the design of multiple RF MEMS front-end components that could replace off-chip units in future transceivers. The components can effectively be combined and cofabricated with CMOS circuitry. A CMOS-MEMS approach is used, which means that the microelectromechanical parts are made by multi-layer structures in a standard CMOS process. The ASIMPS procedure offered by CMP uses the CMOS processing at STMicroelectronics, and a simple post-CMOS release-etching step is performed at Carnegie Mellon University, USA. The paper presents a resonator, resonating filters, electrothermally operated varactors and an experimental switch. The components are modeled and simulated by using CoverntorWare FEM software. A test chip is currently in production.

• Dalbro, Martin; Eikeland, Erik; Veld, Aart Joakim in't; Gjessing, Stein; Lande, Tor Sverre & Riis, Håvard Kolle [Show all 7 contributors for this article] (2008). Wireless Sensor Networks for Off-shore Oil and Gas Installations. In Braem, Bart; Benveniste, Mathilde; Dini, Cosmin; Fortino, Giancarlo; Karnapke, Reinhardt; Mauri, Jaime Lloret & Monsi, M. Sohrab H. (Ed.), Second International IEEE Conference on Sensor Technologies and Applications, 2008. SENSORCOMM '08. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9780769533308. p. 258–263. Full text in Research Archive Show summary

  Underwater development and production of oil and gas needs networked sensors and actuators to monitor the production process, to either prevent or detect oil and gas leakage or to enhance the production flow and yield of the wells. Sensors used today are large and expensive, both to manufacture and to deploy sub-sea. We suggest sensor networks created by inexpensive and small wireless sensor nodes in a redundant multi-hop topology. Preliminary designs and results are shown for two such nodes using ultrasound and infrared light as wireless commutation.

• Søråsen, Oddvar & Ramstad, Jan Erik (2008). From MEMS devices to smart integrated systems. Microsystem Technologies : Micro- and Nanosystems Information Storage and Processing Systems. ISSN 0946-7076. 14(7), p. 895–901. doi: 10.1007/s00542-007-0523-x. Full text in Research Archive Show summary

  The smart integrated systems of tomorrow would demand a combination of micromechanical components and traditional electronics. On-chip solutions will be the ultimate goal. One way of making such systems is to implement the mechanical parts in an ordinary CMOS process. This procedure has been used to design an oscillator consisting of a resonating cantilever beam and a CMOS Pierce feedback amplifier. The resonating frequency is changed if the beam is bent by external forces. The paper describes central features of this procedure and highlights the design considerations for the CMOS-MEMS oscillator. The circuit is used as an example of a “VLSI designer” way of making future integrated micromechanical and microelectronic systems on-chip. The possibility for expansion to larger systems is reviewed.

• Søråsen, Oddvar & Ramstad, Jan Erik (2007). From MEMS Devices to Smart Integrated Systems. In Chakrabarty, Krishnendu; Courtois, Bernhard; Jung, Erik; Kempe, Volker & Laur, Rainer (Ed.), Design, Test, Integration and Packaging of MEMS/MOEMS 2007. EDA Publishing / DTIP 2007. ISSN 9782355000003. p. 134–140. Full text in Research Archive Show summary

  The smart integrated systems of tomorrow would demand a combination of micromechanical components and traditional electronics. On-chip solutions will be the ultimate goal. One way of making such systems is to implement the mechanical parts in an ordinary CMOS process. This procedure has been used to design an oscillator consisting of a resonating cantilever beam and a CMOS Pierce feedback amplifier. The resonating frequency is changed if the beam is bent by external forces. The paper describes central features of this procedure and highlights the design considerations for the CMOS-MEMS oscillator. The circuit is used as an example of a “VLSI designer” way of making future integrated micromechanical and microelectronic systems on-chip. The possibility for expansion to larger systems is reviewed.

• Arhaug, Ole-Petter & Søråsen, Oddvar (2006). Designing H-shaped Micromechanical Filters. Journal of Physics: Conference Series (JPCS). ISSN 1742-6588. 34. Full text in Research Archive Show summary

  This paper investigates the design constraints and possibilities given when designing a micromechanical band pass filter for intermediate frequencies (e.g. 10 MHz). The class of filters are based on coupled clamped-clamped beams constituting an H-shaped structure. A primary beam can electrostatically be activated in one of its different harmonic modes, setting the filter center frequency. The motion is transferred to an accompanying beam of equal dimensions by a mechanical coupling beam. The placement or coupling points of the quarter-wavelength coupling beam which connects the vertically resonating beams is critical with respect to the bandwidth of the filters. Of special concern has been to investigate realistic dimensions allowing the filters to be processed by an actual foundry process and to find out how the choice of materials and actual dimensions would affect the performance.

• Vistnes, Kjetil E. & Søråsen, Oddvar (2005). Reconfigurable Address Generators for Stream-Based Computation Implemented on FPGAs. Proceedings, International Parallel and Distributed Processing Symposium (IPDPS). ISSN 1530-2075. Full text in Research Archive

View all works in NVA

• Kuppireddi, Srinivasa Reddy; Pamidighantam, Sayanu; Janardhana, V.; Søråsen, Oddvar; Roy, JS & Kulkarni, RG (2012). Evaluation of SU8 photo polymer for microwave packaging applications. In Lande, Tor Sverre (Eds.), Proceedings of the 30th Norchip Conference, NORCHIP12. IEEE Press. ISSN 9781467322225. p. 1–4. Full text in Research Archive Show summary

  Epoxy based photo polymer – SU8 is evaluated as microwave packaging material for packaging of Radio Frequency Micro Electro Mechanical Systmes (RFMEMS). Standard and Slotted Coplanar Waveguide transmission lines are chosen as test vehicles. Transmission lines of lengths up to 10mm are formed on ST-Cut Quartz substrates. The fabricated transmission lines are subjected to wafer level packaging. The measured S-parameter data shows a significant difference prior to and after packaging. Further, the effects due to the packaging are modeled using lumped components and the measured data are validated with simulations in Advanced Design System. The measured and simulated data agree favorably with each other, for both before and after packaging

• Aunet, Snorre & Søråsen, Oddvar (2011). Special Issue on Norchip 09. Microprocessors and Microsystems: Embedded Hardware Design (MICPRO). ISSN 0141-9331. 35(7), p. 593–594. doi: 10.1016/j.micpro.2011.10.001. Full text in Research Archive
• Le, Cuong Phu; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2011). An electrostatic energy harvester with power-extracting end stops driven by wideband vibrations. In Cho, Young-Ho (Eds.), Proceedings of the 11th Workshop on Micro and Nanotechnology for Power Generation and Energy Conversion Applications, PowerMEMS 2011, 15-18 November 2011, Seoul, Republic of Korea. Cell Bench Research Center, KAIST. ISSN 9788996759102. p. 122–125. Full text in Research Archive Show summary

  This paper presents an electrostatic energy harvester with two-stage transduction, giving increased bandwidth and dynamic range. The harvester includes a master proof mass with the main transducer of the device and a slave transducer functioning as an end-stop for the proof mass. Both transducers are within the same device which is designed using the 25µm thickness SOIMUMPs process with a total active area of 4×5 mm2. In a small acceleration regime, the harvester behaves as an ordinary transducer with output power primarily obtained from the master. In a high acceleration regime, the master motion impacts the slave transducer and is then confined to a maximum displacement amplitude. The slave transducer is actuated by the internal impact force, generating additional output power when the master motion reaches the limit. Under wideband acceleration at high levels, the experimental results show that the total output power is increased to about twice the output power of the main transducer, while the 3dB-bandwidth is enlarged by a factor of 5.

• Kaur, Sukhdeep; Halvorsen, Einar; Søråsen, Oddvar & Yeatman, Eric M. (2010). Numerical Analysis of Nonlinearities due to Rigid End-stops in Energy Harvesters. Full text in Research Archive
• Dalbro, Martin; Veld, Aart Joakim in't; Gjessing, Stein; Riis, Håvard Kolle; Lande, Tor Sverre & Søråsen, Oddvar (2008). Building a test-bed for wireless sensor networking for under-water oil and gas installations. Full text in Research Archive
• Søråsen, Oddvar (2008). Romteknologi ved Institutt for informatikk. Full text in Research Archive
• Vistnes, Kjetil E. & Søråsen, Oddvar (2005). Reconfigurable Address Generators for Stream-Based Computation Implemented on FPGAs. Full text in Research Archive Show summary

  This paper describes the design and implementation of an address generator for stream-based computation. The unit can generate addresses by a 1, 2 or 3-dimensional mapping from a linear data string in memory. A processing unit will get the required data in a continuous stream without empty time slots, even when switching between addressing algorithms. Each algorithm is specified by a set of parameters loaded into FIFOs in background. The unit is specified by VHDL, simulated, synthesized and implemented on an FPGA of type Xilinx Virtex-II Pro. A speed of 144 MHz is obtained for generating 36 bit addresses. Ideas for expanding the flexibility of the unit is discussed.

• Søråsen, Oddvar; Furuberg, Liv & Johansen, Klaus Magnus (2004). 3D Modelling of MEMS Components for Coupled Elastic - Electrostatic Characterization. Full text in Research Archive Show summary

  This paper presents a method for modelling MEMS components whereby the designer can perform precise electrical, mechanical, and coupled electrical and mechanical simulations. Both surface and bulk micromachined components can be modelled at a geometrical level by defining simplified pseudo processes for a commercial CAD tool such as CovertorWare. The procedure for this approach is described with emphasis on the bulk case. A motivating example of a simple comb-finger structure in an accelerometer shows that this method is required to obtain adequate precision. A long thin bulk micromachined mirror array, which is not easily described analytically, was modelled by using a pseudo process and the coupled field problem of its elastic-electrostatic behaviour has been investigated.

• Li, Shaomeng; Tørresen, Jim & Søråsen, Oddvar (2004). Improving a Network Security System by Reconfigurable Hardware. Full text in Research Archive Show summary

  Improving network security systems by using reconfigurable hardware is an important research field since the speed of the Internet is increasing. In this work, we have implemented stateful TCP inspection in a Field Programmable Gate Array (FPGA) to help alleviating a bottleneck in network intrusion detection systems (NIDSs). Today�s software based NIDSs (e.g. Snort) show inefficiency and even fail to perform for the faster Internet. Implementing stateful TCP inspection in FPGA aims at achieving an efficient, fast NIDS in general, � Snort specifically. By dividing a TCP connection into two flows, one to the Server and another to the Client, monitoring of the TCP flow can be sped up. A TCP flow deals not only with a single packet but also with multiple packets over the network. Reassembly of those packets is one of the main tasks that the TCP flow monitoring should accomplish. By parallelizing the tasks of reassembling TCP packets on the Server and the Client on an FPGA, the performance of stateful TCP inspection can be greatly improved. The performance obtained by this work is a throughput of 2.75 Gbps.

• Nilsen, Geir; Tørresen, Jim & Søråsen, Oddvar (2004). A Variable Word-Width Content Addressable Memory for Fast String Matching. Full text in Research Archive Show summary

  This work deals with off-loading some time critical parts in the process of performing intrusion detection from software to reconfigurable hardware (FPGA). Signatures of known attacks must typically be compared to high speed network traffic, and string matching becomes a bottleneck. Content Addressable Memories (CAMs) are known to be fast string matchers, but offer little flexibility. For that purpose a Variable Word-Width CAM for fast string matching has been designed and implemented in an FPGA. A typical feature for this CAM is that the length of each word is independent from the others, in contrast to common CAMs where all words have the same length. The design has been functionally tested on a development board for a CAM of size 1822 bytes (128 words). This design processes 8 bits per clock cycle and has a reported maximum clock speed of 100 MHz. This gives a thoughput of 800 Mbit/s.

• Johansen, Klaus Magnus; Søråsen, Oddvar & Furuberg, Liv (2004). Computed Electro-Mechanical Behaviour of Long, Narrow Silicon Mirrors. Full text in Research Archive Show summary

  A mirror array with seven long, narrow silicon mirrors has been manufactured in the MultiMEMS Multi-Project-Wafer process.The mirror array was designed for a bar-code reader and is more resistant to mechanical fracture and requires lower actuation voltages than a single, larger mirror. A commercially available, low-cost thin film metal print on a ceramic substrate is used for packaging and actuation. The mirrors are tilted by electrostatic forces from thin-film electrodes placed along the edges of the mirror, a vertical distance of 10 micrometers from the mirror. Coventor simulations of the coupled elastic-electrostatic behaviour of the mirrors are presented. A model process of the SensoNor manufacturing process is created and the lithographic mask layers are defined. This input creates a 3D model of the mirrors and electrodes. The combined rotation and deflection of the mirrors as a function of applied voltage is visualized and quantified. Movement of a mirror due to forces from neighbouring electrodes is also presented. The computed pull-in voltage was 127 V in good agreement with the experiments.

• Li, Shaomeng; Tørresen, Jim & Søråsen, Oddvar (2003). Exploiting Reconfigurable Hardware for Network Security. Full text in Research Archive Show summary

  Abstract� One type of network security strategy is using an Intrusion detection system (IDS). We are implementing an IDS in FPGA-based (Field Programmable Gate Array) recofigurable hardware. This is to achieve higher speed and more efficient performance of network security, as networks develop very fast with consequently more demanding constraints. This study provides novel hardware architectures for an IDS system which should be able to monitor networks with a speed up to 2.68 Gbps.

• Li, Shaomeng; Tørresen, Jim & Søråsen, Oddvar (2003). Exploiting Stateful Inspection of Network Security in Reconfigurable Hardware. Full text in Research Archive Show summary

  Abstract- One of the most important areas of a network intrusion detection system (NIDS), stateful inspection, is described in this paper. We present a novel reconfigurable hardware architecture implementing TCP stateful inspection used in NIDS. This is to achieve a more efficient and faster network intrusion detection system as todays' NIDSs show inefficiency and even fail to perform while encountering the faster Internet. The performance of the NIDS described is expected to obtain a throughput of 3.0 Gbps.

• Søråsen, Oddvar (1999). Trender og utfordringer i elektronikk system-design. Full text in Research Archive Show summary

  I foredraget vil en se på typiske trekk ved teknologi-utviklingen idag, bl.a. det økende gapet mellom kretskompleksitet og konstruktør-produktivitet og kravet til en strukturert design-metodikk. Innbakte systemer, Embedded Systems, med andre grensesnitt enn skjerm og tastatur blir stadig viktigere. Slike systemer er ofte reaktive og kan baseres på en modell med samvirkende tilstandsmaskiner (CFSMs = Co-design Finite-State-Machines). Hver CFSM består typisk av en kontrolldel som bestemmer transisjoner mellom tilstandene og en kompleks datamanipulerings-funksjon. Det er ønskelig å utsette lengst mulig i designprosessen avgjørelsen om modulene skal implementeres i hw eller sw. Gjenbruk av ferdig utviklede og testede Intellectual Properties (IPs) eller Virtual Components (VCs) fra ulike leverandører vil være viktig. Disse kan være både hardware-blokker med ulik tilpasningsmulighet og software-kode f.eks. for DSPs. Typisk kan de komplekse funksjonene i systemet utføres av disse IP-blokkene (MPEG-koding/dekoding, mønster-gjenkjenning etc.). Standardiseringsarbeid for grensesnitt mellom IPs pågår, bl.a. VSI, Virtual Socket Interface. Selv med et standard grensesnitt mellom slike primærblokker vil det være ønskelig med et overordnet synkroniserings- og kontroll-nivå for en mer strukturert og kosteffektiv system-konstruksjon. Derved unngår en å måtte modifisere importerte IP-blokker eller basere seg på ad hoc "glue" logikk for hver ny konstruksjon. Ved Institutt for informatikk, UiO, arbeider en med en 2-nivås, intelligent infrastruktur for å lette integreringen av heterogene IPs. Dette bygger videre på et flerårig arbeid i multicomputerteknikk og fokuserer det mot den nevnte problemstillingen. Foruten synkronisering og kontroll av IP-operasjoner kan det ekstra sammenkoblingsnivået i parallell overføre og bufre data mellom de ulike system-delene.

• Søråsen, Oddvar (1999). A structured system design approach based on reusing complex building blocks. Full text in Research Archive Show summary

  Abstract --- The efficiency of system design can be greatly increased by reuse of complex modules. Such Virtual Components (VCs) or Intellectual Properties (IPs) from multiple sources are integrated into an application specific system. The paper discusses this approach and introduces a secondary control level which makes system integration more cost-effective. A chosen model of computation extends the concept of Codesign Finite-State-Machines (CFSMs). The control parts of the CFSMs and data exchange are done by an intelligent infrastructure, while the complex functional parts are performed by the IPs. This facilitates the integration process since the IPs can be used without modification and a structured and modular approach substitutes the use of ad hoc glue logic. Results show that the design complexity is considerably reduced at the expense of a moderate increase in transistor overhead.

• Berg, Yngvar; Lande, Tor Sverre & Søråsen, Oddvar (1991). A simulation framework for artificial neural networks. Full text in Research Archive
• Hasanbegovic, Amir; Siem, Sunniva; Søråsen, Oddvar & Aunet, Snorre (2017). Exploring the SEU Dependence on Supply Voltage scaling in 90 nm and 65 nm CMOS Flip-flops. Universitetet i Oslo. Full text in Research Archive Show summary

  Down-scaling of the supply voltage is considered as the most effective means of reducing the power- and energy consumption of integrated circuits (ICs). Reduction in the power- and energy consumption is highly beneficial in aerospace and defense applications that have a constrained power budget. These applications include, but are not necessarily limited to, payloads in solar powered spacecraft and rovers. The benefits that can be harvested from reducing the powerand energy consumption in such applications are reduced weight, reduced mass and/or increased functionality for a given power budget. Although supply voltage scaling can improve the energy efficiency of ICs, radiation induced errors also tend to increase with decreasing supply voltage. In order to enable reliable operation in radiation-rich environments, radiation induced errors must be mitigated, preferably with minimum area, power and performance penalties. In this thesis, the single event upset (SEU) dependence on supply voltage scaling is investigated for data flip-flops (DFFs) designed in 90 nm and 65 nm CMOS technology nodes. The radiation tolerance of the DFFs was characterized at supply voltages between 0.18 V and 1 V, and heavy ion radiation testing was performed using ions with linear energy transfer (LET) between 5:8 MeV-cm2=mg and 68:8 MeV-cm2=mg. Both temporal and spatial hardening techniques are utilized as a means of mitigating SEUs, and the impact of drive strength and sensitive node separation is evaluated. The examined circuit-level hardening techniques include triple modular redundancy (TMR), dual interlocked storage cell (DICE) and temporal dual-feedback (TDF), as well as inverter-based and current starved delay elements for SET filtering purposes. This study shows that radiation tolerant DFFs can offer soft error rate (SER) improvements of up to 55x, 121x and 600x, compared to a standard non-radiation tolerant DFF, when scaling the supply voltage down to 0.18 V, 0.25 V and 0.5 V, respectively. Simultaneously, by scaling the supply voltage down to 0.5 V and 0.25 V, radiation tolerant DFFs can achieve ~3.9x and ~12x higher energy efficiency, compared to when operating at a supply voltage of 1 V. Selective placement of high drive strength components showed to reduce the SEU sensitivity in DFFs by up to 112x, compared to DFFs utilizing standard drive strength. The impact of charge sharing was, on the other hand, increasingly challenging to mitigate with decreasing supply voltage. Nevertheless, based on the findings in this work, radiation tolerant DFFs operated at reduced supply voltage offer a clear advantage over standard non-radiation tolerant DFFs, and may therefore be suited for implementation in low power payloads, depending on the error rate requirements of the application. In addition to investigating the SEU dependence on supply voltage scaling, this thesis also presents the design and performance of subthreshold to above threshold level shifters, and the characterization of the proton beam properties at the Oslo Cyclotron Laboratory (OCL).

• Berge, Hans Kristian Otnes; Aunet, Snorre; Lande, Tor Sverre & Søråsen, Oddvar (2015). Improvements towards Optimal Design of Reliable Subthreshold Digital CMOS with applications in Logic and Memory. Universitetet i Oslo. Full text in Research Archive Show summary

  This dissertation is organized as a collection of papers, where each paper represents original research contributions relating to the design and analysis of ultra low power CMOS, with a particular emphasis on ultra low voltage and subthreshold operation. The individual papers represent advancements particularily within methods and practices related to the design of both digital logic and memory circuits in the presence of severe process variation. At the device level it is demonstrated how the use of multiple minimum-width gates can exploit the inverse narrow-width subthreshold device effect to improve performance and power-delay products. Measurement results from a 90 nm prototype confirm the effect. Multi-objective optimization strategies are developed and applied to allow exploration of the Pareto optimal design space for reliable logic at 150mV. Targeting operation at 300mV, the design of a 9-transistor SRAM memory cell employing multi-Vt and virtual power techniques is presented. A multi-objective optimization strategy is developed and applied to achieve an optimal trade-off for an efficient and reliable sizing of the SRAM cell. Based on the 9-transistor cell, measured results from an ultra low voltage 64 × 32 SRAM module operating down to 273mV in a 65 nm technology indicate good yield and competitive performance metrics (17.8 fJ/access/bit at averages of 761 kHz @ 321mV supply). Finally, the behavior of subthreshold logic circuits under the influence of adverse fluctuations in the transistor threshold voltages is treated analytically, with specific emphasis on minimum-energy operation and yield constraints. The analysis can suggest optimal choices for supply voltage and device sizing, prior to simulation.

View all works in NVA