Spanish National Research Council · University of Seville
 HOME
INTRANET
esp    ing
IMSE-CNM in Digital.CSIC


 


In all publications
Recent publications
Conversion of Synchronous Artificial Neural Network to Asynchronous Spiking Neural Network using sigma-delta quantization
A. Yousefzadeh, S. Hosseini, P. Holanda, S. Leroux, T. Werner, T. Serrano-Gotarredona and B. Linares-Barranco, B. Dhoedt and P. Simoens
Conference - IEEE International Conference on Artificial Intelligence Circuits and Systems AICAS 2019
[abstract]
Artificial Neural Networks (ANNs) show great performance in several data analysis tasks including visual and auditory applications. However, direct implementation of these algorithms without considering the sparsity of data requires high processing power, consume vast amounts of energy and suffer from scalability issues. Inspired by biology, one of the methods which can reduce power consumption and allow scalability in the implementation of neural networks is asynchronous processing and communication by means of action potentials, so-called spikes. In this work, we use the wellknown sigma-delta quantization method and introduce an easy and straightforward solution to convert an Artificial Neural Network to a Spiking Neural Network which can be implemented asynchronously in a neuromorphic platform. Briefly, we used asynchronous spikes to communicate the quantized output activations of the neurons. Despite the fact that our proposed mechanism is simple and applicable to a wide range of different ANNs, it outperforms the state-of-the-art implementations from the accuracy and energy consumption point of view. All source code for this project is available upon request for the academic purpose.

A Digital Neuromorphic Realization of the 2-D Wilson Neuron Model
M. Nouri, M. Hayati, T. Serrano-Gotarredona and D. Abbot
Journal Paper - IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 66, no. 1, 2019
IEEE    DOI: 10.1109/TCSII.2018.2852598    ISSN: 1549-7747    » doi
[abstract]
This brief presents a piecewise linear approximation of the nonlinear Wilson (NW) neuron model for the realization of an efficient digital circuit implementation. The accuracy of the proposed piecewise Wilson (PW) model is examined by calculating time domain signal shaping errors. Furthermore, bifurcation analyses demonstrate that the approximation follows the same bifurcation pattern as the NW model. As a proof of concept, both models are hardware synthesized and implemented on field programmable gate arrays, demonstrating that the PW model has a range of neuronal behaviors similar to the NW model with considerably higher computational performance and a lower hardware overhead. This approach can be used in hardware-based large scale biological neural network simulations and behavioral studies. The mean normalized root mean square error and maximum absolute error of the PW model are 6.32% and 0.31%, respectively, as compared to the NW model.

Compressive Imaging using RIP-compliant CMOS Imager Architecture and Landweber Reconstruction
M. Trevisi, A. Akbari, M. Trocan, Á. Rodríguez-Vázquez and R. Carmona-Galán
Journal Paper - IEEE Transactions on Circuits and Systems for Video Technology, first online, 2019
IEEE    DOI: 10.1109/TCSVT.2019.2892178    ISSN: 1051-8215    » doi
[abstract]
In this paper we present a new image sensor architecture for fast and accurate compressive sensing (CS) of natural images. Measurement matrices usually employed in compressive sensing CMOS image sensors (CS-CIS) are recursive pseudo-random binary matrices. We have proved that the restricted isometry property (RIP) of these matrices is limited by a low sparsity constant. The quality of these matrices is also affected by the non-idealities of pseudo-random numbers generators (PRNG). To overcome these limitations, we propose a hardware-friendly pseudo-random ternary measurement matrix generated on-chip by means of class III elementary cellular automata (ECA). These ECA present a chaotic behaviour that emulates random CS measurement matrices better than other PRNG. We have combined this new architecture with a block-based CS smoothed-projected Landweber (BCS-SPL) reconstruction algorithm. By means of single value decomposition (SVD) we have adapted this algorithm to perform fast and precise reconstruction while operating with binary and ternary matrices. Simulations are provided to qualify the approach.

Offset-calibration with Time-Domain Comparators using Inversion-mode Varactors
R. Fiorelli, M. Delgado-Restituto and A Rodríguez-Vázquez
Journal Paper - IEEE Transactions on Circuits and Systems II: Express Brief, first online, 2019
IEEE    DOI: 10.1109/TCSII.2019.2904100    ISSN: 1549-7747    » doi
[abstract]
This paper presents a differential time-domain comparator formed by two voltage controlled delay lines, one per input terminal, and a binary phase detector for comparison solving. The propagation delay through the respective lines can be adjusted with a set of digitally-controlled inversion-mode varactors. These varactors provide tuning capabilities to the comparator; feature which can be exploited for offset calibration. This is demonstrated with the implementation of a differential 10-bit SAR-ADC. The design, fabricated in a 0.18μm CMOS process, includes an automatic mechanism for adjusting the capacitance of the varactors in order to calibrate the offset of the whole converter. Correct functionality was measured in all samples.

Fast adaptive comparator offset calibration in pipeline ADC with self-repairing thermometer to binary encoder
A.J. Ginés, E. Peralías, C. Aledo and A. Rueda
Journal Paper - International Journal of Circuit Theory and Applications, vol. 47, no. 3, pp 333-349, 2019
JOHN WILEY & SONS    DOI: 10.1002/cta.2594    ISSN: 0098-9886    » doi
[abstract]
This paper presents a fast background calibration method for comparator offsets in pipeline ADCs and analyzes the practical considerations in a 1.8 V 0.18 μm 100Msps pipeline ADC with 15-bit resolution (74 dB-Signal-to-noise plus Distortion Ratio [SNDR]). A self-repairing (SR) thermometer-to-binary encoder is developed to deal with malfunctioning in presence of high comparator offsets greater than one-half least-significant bit (LSB). In this situation, the effective thresholds between two adjacent comparators could be inverted leading to a faulty behavior with conventional architectures. The proposed solution allows a dynamic assignment of the calibration code associated to each comparator improving convergence speed. As demonstrator, its application to a 15-bit pipeline ADC using a novel calibrated dynamic-latch comparator (DLC) with internal threshold reference generation and no preamplifier is presented, showing a reduction on the total power consumption of 22% with respect to a design without calibration targeting the same specifications.

Assessing AMS-RF test quality by defect simulation
V. Gutierrez, A. Gines and G. Leger
Journal Paper - IEEE Transactions on Device and Materials Reliability, vol. 19, no. 1, pp 55-63, 2019
IEEE    DOI: 10.1109/TDMR.2019.2894534    ISSN: 1530-4388    » doi
[abstract]
In safety critical applications there is a demand for estimating defect coverage in order to meet stringent quality levels. However, defect simulation of complex AMS-RF circuits is computationally expensive since achieving good confidence interval requires sampling many defects. In this paper, we show on practical cases of study that it is beneficial to complement defect coverage with fault coverage and assess the severity of defect escapes to get a complete picture of test quality. The computational burden of defect and fault simulations is taken into account and accurate statistical estimates of defect and fault escapes are provided to allow safe early stopping of the simulations.

A Multilevel Bottom-up Optimization Methodology for the Automated Synthesis of RF Systems
F. Passos, E. Roca, J. Sieiro, R. Fiorelli, R. Castro-López, J.M. López-Villegas and F.V. Fernández
Journal Paper - IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, first online, 2019
IEEE    DOI: 10.1109/TCAD.2018.2890528    ISSN: 0278-0070    » doi
[abstract]
In recent years there has been a growing interest in electronic design automation methodologies for the optimizationbased design of radiofrequency circuits and systems. While for simple circuits several successful methodologies have been proposed, these very same methodologies exhibit significant deficiencies when the complexity of the circuit is increased. The majority of the published methodologies that can tackle radiofrequency systems are either based on high-level system specification tools or use models to estimate the system performances. Hence, such approaches do not usually provide the desired accuracy for RF systems. In this work, a methodology based on hierarchical multilevel bottom-up design approaches is presented, where multi-objective optimization algorithms are used to design an entire radiofrequency system from the passive component level up to the system level. Furthermore, each level of the hierarchy is simulated with the highest accuracy possible: electromagnetic simulation accuracy at device-level and electrical simulations at circuit/system-level.

On the implementation of asynchronous sun sensors
J.A. Leñero-Bardallo, R. Carmona-Galán and A. Rodríguez-Vázquez
Conference - IS&T International Symposium on Electronic Imaging 2019
[abstract]
Abstract not avaliable

Compact Real-Time Inter-Frame Histogram Builder for 15Bits High-Speed ToF-Imagers based on Single-Photon Detection
I. Vornicu, A. Darie, R. Carmona-Galan and A. Rodriguez-Vazquez
Journal Paper - IEEE Sensors Journal, vol. 19, no. 6, pp 2181-2190, 2019
IEEE    DOI: 10.1109/JSEN.2018.2885960    ISSN: 1530-437X    » doi
[abstract]
Time-of-flight image sensors based on single-photon detection, i.e. SPADs, require some filtering of pixel readings. Accurate depth measurements are only possible if the jitter of the detector is mitigated. Moreover, the time stamp needs to be effectively separated from uncorrelated noise such as dark counts and background illumination. A powerful tool for this is building a histogram of a number of pixel readings. Future generation of ToF imagers are seeking to increase spatial and temporal resolution along with the dynamic range and frame rate. Under these circumstances, storing the complete histogram for every pixel becomes practically impossible. Considering that most of the information contained by the histogram represents noise, we propose a highly efficient method to store just the relevant data required for ToF computation. This method makes use of the shifted inter-frame histogram (SifH). It requires a memory as low as 128 times smaller than storing the complete histogram if the pixel values are coded on up to 15 bits. Moreover, a fixed 28 words memory is enough to process histograms containing up to 215 bins. In exchange, the overall frame rate only decreases to one half. The hardware implementation of this algorithm is presented. Its remarkable robustness for a low SNR of the ToF estimation is demonstrated by Matlab simulations and FPGA implementation using input data from a SPAD camera prototype.

Power and Speed Evaluation of Hyper-FET Circuits
J. Núñez and M.J. Avedillo
Journal Paper - IEEE Access, vol. 7, pp 6724-6732, 2019
IEEE    DOI: 10.1109/ACCESS.2018.2889016    ISSN: 2169-3536    » doi
[abstract]
Many emerging devices are currently being explored as potential alternatives to complementary metal-oxide-semiconductor (CMOS) technologies for overcoming power density and energy efficiency limitations. It is now generally accepted that these emerging devices need to be evaluated at circuit level. In this paper, we investigate the speed and power performance of Hyper-Field Effect Transistor (Hyper-FET) circuits, comparing them with both high-performance (HP) and low stand-by power (LSTP) Fin-Shaped Field Effect Transistor (FinFET) designs on the same technology node. The evaluation, which was carried out at gate and circuit level, includes characterization of 8 bit ripple carry adders. Our experiments showed around 80% speed degradation and 30% power savings for a given range of operating frequencies. These power savings were much smaller than those predicted from transistor and gate level estimations. Deviations from the ideal expected behavior of the Hyper-FET circuitry are illustrated which support the obtained results.

Scopus access Wok access