H. Puttnies, Christoph Niemann, Sascha Rohde, D. Timmermann, J. Schacht
{"title":"Towards software performance estimation based on register-transfer level descriptions","authors":"H. Puttnies, Christoph Niemann, Sascha Rohde, D. Timmermann, J. Schacht","doi":"10.1109/NORCHIP.2017.8124967","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124967","url":null,"abstract":"The concept of hardware-software codesign allows to cope with the increasing complexity of very large scale integration (VLSI) systems. This concept is based on design space exploration (DSE), which means the systematic altering of design parameters (e.g., parallelization, serialisation, implementation of a functionality in software or hardware) to improve the final design. Existing DSE approaches work with high-level descriptions (e.g., SystemC) of the functionality. However, if existing systems (or submodules of those) shall be reused, only a synthesizable register-transfer level (RTL) description of these functionalities is necessarily given. Consequently, the RTL modules cannot be considered during the DSE. We propose an approach to this problem of reasonable practical relevance. To conduct the DSE, we estimate the software execution time of a functionality based on its RTL description. We show in an evaluation that this estimation is possible based on RTL code. In addition, we propose several suggestions to mitigate the problem that the software execution time of a functionality is highly dependent on the input data, which are often unknown at design time.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115298657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A dependable ASIC architecture with RT-level rollback for controller soft error recovery","authors":"Keisuke Inoue","doi":"10.1109/NORCHIP.2017.8124986","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124986","url":null,"abstract":"This paper presents a new method for soft error detection on the controller circuit and rollback to correct error. It focuses especially on the storage part (flip flops) of the controller since the transient error on the controller could cause a severe problem. It proposes a checkpoint and rollback method with a new controller system with soft error detector. It also induces design conditions in high-level synthesis to operate the proposed system. Demonstrated examples show the effectiveness of the proposed design.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122392268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Harry Weber, Gerald Alexander Koroa, W. Mathis, D. Delchev, G. Marinova
{"title":"A self-consistent Carleman linearization approach for the design of RF mixer circuits","authors":"Harry Weber, Gerald Alexander Koroa, W. Mathis, D. Delchev, G. Marinova","doi":"10.1109/NORCHIP.2017.8124950","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124950","url":null,"abstract":"An RF mixer is one of the key elements in wireless telecommunication systems, which is used for example for modulation and demodulation. Due to the fact, that such a functionality can only be achieved by a nonlinear system, it is crucial that the nonlinearity is taken into account for the analysis and design of RF mixers. For this reason, the analysis and design process is still a challenging task because linear models can only be used restrictively. Therefore, an iterative design process is performed to achieve the desired specifications, whereby a circuit simulation tool as Cadence Spectre is utilized for verification. This approach is very time consuming as a vast of numerical simulations have to be performed for each set of network parameters. In this contribution, another approach is presented based on a self-consistent Carleman linearization. This approach is used to obtain an approximate analytic solution in dependency of the network parameters.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"4 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123650346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An RNS based modular multiplier with reduced complexity","authors":"Shahzad Asif, M. Vesterbacka","doi":"10.1109/NORCHIP.2017.8124953","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124953","url":null,"abstract":"Modular multiplication (MM) based on the residue number system (RNS) is a widely researched area due to the fast arithmetic operations in the RNS. The major drawback of the RNS based MM architectures is their large area because each arithmetic operation is followed by a modular reduction. In this work, the number of modular reductions is reduced and instead the wordlength of some operations is increased to accommodate the intermediate results. The proposed scheme greatly reduces the number of multipliers and achieves a 55% reduction in the hardware complexity. Moreover the delay of the proposed architecture is also significantly lower than the reference architecture.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124432958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Subsampling phase-locked loop behavioural modelling approach for phase noise evaluation","authors":"Peco Gjurovski, M. Wei, R. Negra","doi":"10.1109/NORCHIP.2017.8124991","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124991","url":null,"abstract":"A subsampling phase-locked loop (SSPLL) model with phase noise evaluation is demonstrated. In this paper the proposed SSPLL behavioural modelling approach allows a systematic design of the complete circuit. Since phase noise is an important criterion for phase-locked loops this model has an emphasis on phase noise evaluation. Design specifications tested in this model can partially be transferred to the transistorlevel and, eventually back-annotated into the behavioural model for further verification. Using this demonstrated model a large improvement in terms of simulation speed can be achieved compared to a time consuming transient analysis on a transistorlevel from i.e. Cadence Spectre. To the authors best knowledge, this model reported here is the first SSPLL model in Simulink with a focus on phase noise evaluation.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132254125","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Implementation of a fault-tolerant, globally-asynchronous-locally-synchronous, inter-chip NoC communication bridge on FPGAs","authors":"Eleftherios Kyriakakis, Kalle Ngo, Johnny Öberg","doi":"10.1109/NORCHIP.2017.8124972","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124972","url":null,"abstract":"Network-on-Chip (NoC) architectures were introduced to help mitigate the bottleneck and scalability issues faced by the traditional bus interconnect in Multi-Processor System-On Chip (MPSoC). Nowadays, many embedded systems host a significant number of micro-controllers and processors (i.e. vehicles, airplanes, satellites, etc.) and as this number continues to increase, traditional bus solutions will start to fail on those platforms as well. NoCs not only offer a scalable solution for MPSoC interconnects but they can also provide a uniform platform of communication to embedded systems with multiple off-chip, often heterogeneous, processors. This leads to the need for investigation on inter-chip communication bridges suitable for transmitting flits/packets across chips and possibly across clock domains. This paper investigates an inter-chip communication link, of an MPSoC NoC architecture which is extended with an off-chip, heterogeneous processor (node) and proposes a scalable, fault-tolerant, globally asynchronous locally synchronous bridge for inter-chip communication. The proposed bridge is implemented on a prototype board of the SEUD KTH experiment where it successfully enables the communication of a NoC distributed over two FPGAs. The inter-chip bridge is verified in-circuit achieving transfer speeds up to 24 MByte/s (≈ 1.5 Mflit/s) and its ability to correct single bit errors is demonstrated in simulation.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123018685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A highly compact, 16.8 dBm Pgat Ka-band power amplifier in 250 nm SiGe:C BiCMOS","authors":"Iancu Somesanu, H. Schumacher","doi":"10.1109/NORCHIP.2017.8124959","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124959","url":null,"abstract":"This paper presents a highly compact (0.24 mm2) power amplifier operating at Ka band, capable of delivering a saturated output power of 16.8 dBm with an output 1 dB compression point of up to 15 dBm. The small signal gain is higher than 15 dB with a 3 dB bandwidth of almost 10 GHz centered at 29 GHz. At 15 dBm output power the amplifier consumes 210 mW with a measured power added efficiency (PAE) of 15%.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117008376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xuefei You, A. Zjajo, Sumeet S. Kumar, R. V. Leuken
{"title":"Energy-efficient neuromorphic receptors for wide-range temporal patterns of post-synaptic responses","authors":"Xuefei You, A. Zjajo, Sumeet S. Kumar, R. V. Leuken","doi":"10.1109/NORCHIP.2017.8124951","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124951","url":null,"abstract":"In a neuromorphic integrated circuit synaptic dynamics are of great importance to capture accurate neural behaviors. In this paper, we propose a current-based synapse design mediated with multiple receptor types, namely AMPA, NMDA and GABAa, and a weight-dependent learning algorithm. Due to various biological conducting mechanisms, the receptors demonstrate different kinetics in response to stimulus. The designed circuit offers distinctive features of receptors as well as the joint synaptic function. An increased computation ability is verified through synchrony detection in a two-layer recurrent network of synapse clusters. The design implemented in TSMC 65 nm CMOS technology consumes 1.92, 3.36, 1.11 and 35.22 pJ per spike event of energy for AMPA, NMDA, GABAa receptors and the advanced learning circuit, respectively.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125354075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
W. SteveNgueya, Julien Mellier, Stephane Ricard, J. Portal, H. Aziza
{"title":"Power efficiency optimization of charge pumps in embedded low voltage NOR flash memory","authors":"W. SteveNgueya, Julien Mellier, Stephane Ricard, J. Portal, H. Aziza","doi":"10.1109/NORCHIP.2017.8124949","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124949","url":null,"abstract":"A new concept for improving efficiency of a charge pump system is presented. The concept is based on a set of elementary charge pumps connected in parallel in addition to the detection of the number of memory cells to be programmed at the same time. The proposed solution is able to supply current to a memory cell, while operating with maximum efficiency by activating one, two or more charge pumps. The proposed system is implemented using a 55 nm UMC High Voltage CMOS technology with a power supply of 1.2V. In the worst case where only one cell is to be programmed, simulation results show an improvement of power efficiency by a factor 2. The area penalty of the proposed system is only 3.8%.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"511 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130356980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Martin Nielsen-Lönn, Pavel Angelov, J. Wikner, A. Alvandpour
{"title":"Self-oscillating multilevel switched-capacitor DC/DC converter for energy harvesting","authors":"Martin Nielsen-Lönn, Pavel Angelov, J. Wikner, A. Alvandpour","doi":"10.1109/NORCHIP.2017.8124977","DOIUrl":"https://doi.org/10.1109/NORCHIP.2017.8124977","url":null,"abstract":"This paper presents three self-oscillating multilevel time-interleaved switched-capacitor DC/DC converters implemented and taped-out in 0.18-μm CMOS targeting microwatt power levels. Two of the converters are step-up with ratios of 1:2, 1:3, and 1:4, and one is a step-down with ratios of 2:1, 3:1, and 4:1. They all regulate the output voltage towards a targeted reference removing the need for a separate regulator. Aimed for use in vibration energy harvesting systems, the converters have a wide combined input voltage range of 450 mV to 20 V. The low voltage step-up converter operates from an input voltage of 475 mV and has a peak measured power efficiency of 82.2% with an area of 0.62 mm2. The medium voltage step-up converter operates from an input voltage of 700 mV and has a peak power efficiency of 74.5% and an area of 0.53 mm2. Lastly, the step-down converter works with input voltages up to 20 V and achieves a peak power efficiency of 68.7% with an area of 0.55 mm2.","PeriodicalId":373686,"journal":{"name":"2017 IEEE Nordic Circuits and Systems Conference (NORCAS): NORCHIP and International Symposium of System-on-Chip (SoC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130679961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}