Adaptation of a comprehensive simplification method to the Adaptive Exponential Integrate and Fire Neuron and Its FPGA-based verification

Yıl 2025, Cilt: 14 Sayı: 3, 990 - 1000, 15.07.2025

Bekir Şıvga , Nimet Korkmaz

https://doi.org/10.28948/ngumuh.1573633

Öz

The preference of a comprehensive method usage is as important as less hardware usage on digital device-based implementations. The mathematical series expansions have a widespread usage in the transformation of expressions into simpler forms. The exponential, trigonometric, logarithmic, etc. functions are usually converted to simpler expressions for digital implementation easiness. In these implementations, it is an expected output that as the operands of the series increases, the revised model converges to the original one. However, the most appropriate number determination of these operands is important for hardware efficiency. In here, the exponential expression of the Adaptive Exponential Integrate and Fire (ADEX) neuron model is expanded up to the tenth operand of the Taylor series. Then, an optimum operand number is identified for getting both hardware utilization efficiency and neuronal meaningfulness. The differences between the original and revised models are compared with the error calculations and the neuronal observations. Lastly, the revised ADEX neuron model is realized by FPGA device to prove the efficiency of the proposed adaptation.

Anahtar Kelimeler

Adaptive exponential integrate and fire (ADEX) neuron model , Field Programmable Gate Array (FPGA) , digital implementation , neuromorphic engineering.

Kapsamlı bir basitleştirme yönteminin Uyarlanabilir Üstel Entegre ve Ateşlemeli Nöronuna uyarlanması ve FPGA tabanlı doğrulaması

Öz

Dijital cihaz tabanlı gerçekleştirimlerde kapsamlı bir yöntemin kullanılmasının tercihi, az donanım kullanımı kadar önemlidir. Matematiksel seri açılımları, ifadelerin daha basit biçimlere dönüştürülmesinde yaygın bir kullanıma sahiptir. Üstel, trigonometrik, logaritmik vb. işlevler genellikle dijital uygulama kolaylığı için daha basit ifadelere dönüştürülür. Bu uygulamalarda, serinin işlenenleri arttıkça, revize edilmiş modelin orijinal modele yakınsaması beklenen bir çıktıdır. Bununla birlikte, bu işlenenlerin en uygun sayısını belirlenmesi donanım verimliliği için önemlidir. Burada, Uyarlanabilir Üstel Entegre ve Ateşlemeli (ADEX) nöron modelinin üstel ifadesi, Taylor serisinin onuncu işlenenine kadar genişletilmiştir. Daha sonra, hem donanım kullanım verimliliği hem de nöronal anlamlılığı elde etmek için optimum bir işlenen sayısı belirlenmiştir. Orijinal ve revize edilmiş modeller arasındaki farklar, hata hesaplamaları ve nöronal gözlemlerle karşılaştırılmıştır. Son olarak, revize edilmiş ADEX nöron modeli, önerilen adaptasyonun verimliliğini kanıtlamak için FPGA cihazı tarafından gerçekleştirilmiştir.

Anahtar Kelimeler

Uyarlanabilir üstel entegre ve ateşlemeli (ADEX) nöron modeli , Alan Programlanabilir Kapı Dizisi (FPGA) , dijital gerçekleştirim , nöromorfik mühendislik.

Kaynakça

• E. M. Izhikevich, Simple model of spiking neurons, in IEEE Transactions on Neural Networks, 14, 6, 1569-1572, 2003, doi: 10.1109/TNN.2003.820440.
• Gerstner, Wulfram, and Werner M. Kistler. Spiking neuron models: Single neurons, populations, plasticity. Cambridge University Press, 2002. doi.org/10.1017/CBO9780511815706
• Alan L. Hodgkin and Andrew F. Huxley. A quantitative description of membrane current and its application to conduction and excitation in nerve. The Journal of physiology, 117,4, 500, 1952. doi: 10.1113/jphysiol
• C. Morris, H. Lecar, Voltage oscillations in the barnacle giant muscle fiber. Biophys. J., 35, 193–213, 1981. doi.org/10.1016/S0006-3495(81)84782-0
• A. J. Isjpeert, Central Pattern Generators for locomotion control in animals and robots: A review. Neural Networks, 21, 642-653, 2008. doi.org/10.1016/j.neunet.2008.03.014
• R. FitzHugh, Mathematical models for excitation and propagation in nerve, Biological Engineering, H.P. Schawn (Ed.), New York: McGraw-Hill, 1- 85, 1969.
• J. L. Hindmarsh, R. M. Rose, A model of neuronal bursting using three couple first order differential equations, Proc. R. Soc. Lond. Biol. Sci., 22,1222, 87-102, 1984. doi.org/10.1098/rspb.1984.0024
• H. R. Wilson, J. D. Cowan, Excitatory and inhibitory interactions in localized populations of model neurons, Biophysical Journal, 12,1, 1-24,1972. doi.org/10.1016/S0006-3495(72)86068-5
• R. Jolivet, A. Rauch, H. R. L. Lüscher, Gerstner, W. Integrate and-Fire models with adaptation are good enough: Predicting spike times under random current injection, Adv. Neural Inf. Proces. Syst., 18, 595–602, 2005. doi.org/10.1101/2024.09.03.610998
• M. G. F. Fuortes, F. Mantegazzini, Interpretation of the repetitive firing of nerve cells, J. Gen. Physiol., 45,6, 1163–1179, 1962. doi.org/10.1085/jgp.45.6.1163
• A. Treves, Mean-field analysis of neuronal spike dynamics, Network: Comput. Neural Syst., 4,3, 259–284, 1993. doi.org/10.1088/0954-898X/4/3/002
• P. E. Latham, B. J. Richmond, P. G. Nelson, S. Nirenberg, Intrinsic dynamics in neuronal networks, J. Neurophysiol, 83,2, 808–827, 2000. doi.org/10.1152/ jn.2000.83.2.808
• M. J. Richardson, N. Brunel, V. Hakim, From subthreshold to firing-rate resonance, J. Neurophysiol., 89,5, 2538–2554, 2003. doi.org/10.1152/jn.00955. 2002
• R. Naud, N. Marcille, C. Clopath, W. Gerstner, Firing patterns in the adaptive exponential integrate-and-fire model, Biol. Cybern., 99, 4–5, 335–347, 2008. doi.org/10.1007/s00422-008-0264-7
• H. Soleimani, A. Ahmadi, M. Bavandpour, Biologically inspired spiking neurons: Piecewise linear models and digital implementation, IEEE Trans. Circuits Syst. I: Regular Papers, 59,12, 2994–3001, 2012. doi: 10.1109/TCSI.2012.2206463
• Z. Jie, Y. Baoquan, Mixed signal integrated circuit design for integrate-and-fire spiking neurons, Circuits Syst. Signal Process., 42, 27–46, 2023. doi.org/10.1007/s00034-022-02131-2
• N.A. Kant, M. R. Dar, F. A. Khanday, et al. Ultra-low-voltage integrable electronic realization of integer- and fractional-order liao’s chaotic delayed neuron model, Circuits Syst. Signal Process., 36, 4844–4868, 2017. doi.org/10.1007/s00034-017-0615-5
• Y. Heo, H. Song, Circuit modeling and implementation of a biological neuron using a negative resistor for neuron chip, BioChip J., 6, 17–24, 2012. doi.org/10.1007/s13206-012-6103-x
• S. B. Furber, S. Temple, A. D. Brown, High- performance computing for systems of spiking neurons, In AISB’06 Workshop on GC5: Architecture of Brain and Mind, 2, 29–36, 2006.
• K. Yamazaki, V. K. Vo-Ho, D. Bulsara, N. Le, Spiking neural networks and their applications: A Review, Brain Sciences, 12,7, 863, 2022. doi.org/10.3390/ brainsci12070863
• Ö. Erdener, S. Ozoguz, A new neuron and synapse model suitable for low power VLSI implementation, Analog Integr. Circ. Sig. Process., 89, 749–770, 2016. doi.org/10.1007/s10470-016-0773-6
• M. Glover, A. Hamilton, L. S. Smith, Analogue VLSI leaky integrate-and-fire neurons and their use in a sound analysis system, Analog Integr. Circ. Sig. Process, 30, 91–100, 2002. doi.org/10.1023/A: 1013747426448
• E. P. Frady, S. Sanborn, S. B. Shrestha, et al., Efficient neuromorphic signal processing with resonator neurons, J. Sign. Process. Syst., 94, 917–927, 2022. doi.org/10.1007/s11265-022-01772-5
• Y. Khakipoor, H. B. Bahar, G. Karimian, An efficient analysis of FitzHugh-Nagumo circuit model, Analog Integr. Circ. Sig. Process., 110, 385–393, 2022. doi.org/10.1007/s10470-021-01947-3
• S. Millner, A. Grübl, K. Meier, J. Schemmel, M. O. Schwartz, A VLSI implementation of the adaptive exponential integrate-and-fire neuron model, Adv. Neural Inf. Process. Syst., 1642–1650, 2010. doi.org/10.4249/scholarpedia.8427
• O. Sharifipoorand, A. Ahmadi, An analog implementation of biologically plausible neurons using CCII building blocks, Neural Network, 36, 129-135, 2012. doi.org/10.1016/j.neunet.2012.08.017
• Z. T. Njitacke, T. F. Fozin, S. S. Muni, J. Awrejcewicz, J. Kengne, Energy computation, infinitely coexisting patterns and their control from a Hindmarsh–Rose neuron with memristive autapse, Circuit implementation, AEU- Int. J. Electron. Commun., 155, 154361, 2022. doi.org/10.1016/j.aeue.2022.154361
• T. Matsubara, H. Torikai, T. Hishiki, A generalized rotate-and-fire digital spiking neuron model and its on-FPGA learning, IEEE Trans. Circuits Syst. II: Express Briefs, 58,10, 677–681, 2011. doi.org/10.1109/ TCSII.2011.2161705
• A. Grübl, S. Billaudelle, B. Cramer, et al., Verification and design methods for the brainscales neuromorphic hardware system, J. Sign. Process. Syst., 92, 1277–1292, 2020. doi.org/10.1007/s11265-020-01558-7
• S. Majidifar, M. Hayati, M. R. Malekshahi, D. Abbott, FPGA implementation of memristive Hindmarsh–Rose neuron model: Low cost and high-performing through hybrid approximation, AEU- Int. J. Electron. Commun., 154968, 2023. doi.org/10.1016/j.aeue. 2023.154968
• J. Toubouland, R. Brette, Dynamics and bifurcations of the adaptive exponential integrate-and-fire model, Biol. Cybern., 99,4–5, 319–334, 2008. doi.org/10.1007/ s00422-008-0267-4
• H. Soleimani, A. Ahmadi, M. Bavandpour, Biologically inspired spiking neurons: piecewise linear models and digital implementation, IEEE Trans. Circuits. Syst. I Regul. Pap., 59,12, 2991-3004, 2012. doi.org/10.1109/TCSI.2012.2206463
• N. Korkmaz, İ. Öztürk, A. Kalınlı, R. Kılıç, A Comparative study on determining nonlinear function parameters of the Izhikevich neuron model, J. Circuits, Syst. Comput., 27,10, 2018. doi.org/10.1109/TCSI. 2012.2206463
• M. Hayati, M. Nouri, D. Abbott, S. Haghiri, Digital multiplierless realization of two-coupled biological hindmarsh–rose neuron model, IEEE Trans. Circuits Syst. II Express Briefs, 63,5, 463-467, 2016. doi.org/10.1109/TCSII.2015.2505258
• N. Korkmaz, İ. Öztürk, R. Kılıç, The investigation of chemical coupling in a HR neuron model with reconfigurable implementations, Nonlinear Dyn., 86,3, 1841-1854, 2016. doi.org/10.1007/s11071-016-2996-6
• S. Gomar, A. Ahmadi, Digital multiplierless implementation of biological Adaptive-Exponential neuron model, IEEE Trans. Circuits. Syst. I Regul. Pap., 61,4, 1206-1219, 2014. doi.org/10.1109/TCSI. 2013.2286030
• S. Haghiri, A. Ahmadi, A novel digital realization of ADEX neuron model. IEEE Trans. Circuits Syst. II Express Briefs, 67,8, 1444-1448, 2020. doi.org/10.1109/TCSII.2019.2938180
• E. Jokar, H. Abolfathi, A. Ahmadi, M. Ahmadi, An efficient uniform-segmented neuron model for large-scale neuromorphic circuit design: simulation and FPGA synthesis results, IEEE Trans, Circuits Syst. I Regul. Pap., 66,6, 2336-2349, 2019. doi.org/10.1109/ TCSI.2018.2889974
• J. Touboul, R. Brette, Dynamics and bifurcations of the adaptive exponential integrate-and-fire model, Biol. Cybern., 99,4, 319-334, 2008. doi.org/10.1007/s00422-008-0267-4
• A. Başargan, A synaptic coupling for the adaptive exponential integrate and fire (ADEXI&F) neuron model with circuit simulations, M. Sc. Thesis, Istanbul Technical University, 65, 2013.
• J. J. Duistermaat, J. A. C. Kolk, Taylor expansion in several variables. in: distributions, Cornerstones, Birkhäuser Boston, 2010. doi.org/10.1007/978-0-8176-4675-2_6
• M. Tuna, A novel secure chaos-based pseudo random number generator based on ANN-based chaotic and ring oscillator: design and its FPGA implementation, Analog Integr. Circ. Sig. Process., 105,167–181, 2020. doi.org/10.1007/s10470-020-01703-z
• M. Greenberg, Advanced engineering mathematics. 2nd ed., Prentice Hall, ISBN 0-13-321431-1, 1998.
• C. Willmott, K. Matsuura, On the use of dimensioned measures of error to evaluate the performance of spatial interpolators, Int. J. Geogr. Inf. Sci., 20, 89–102, 2006. doi.org/10.1080/13658810500286976
• Z. Li, Exponential stability of synchronization in asymmetrically coupled dynamical networks, Chaos Interdiscip. J. Nonlinear. Sci., 18,2, 023124, 2008. doi.org/10.1063/1.2931332
• J. W. Shuai, D. M. Durand, Phase synchronization in two coupled chaotic neurons, Phys. Lett. A, 264,4, 289–297, 1999. doi.org/10.1016/S0375-9601(99) 00816-6
• N. J. Gogtay, U. M. Thatte, Principles of correlation analysis, Journal of the Association of Physicians of India, 65,3, 78-81, 2017.
• www.xilinx.com , Accessed 24.10.2024