Abstract
The application of an electronic real time emulator for biology-inspired pulse processing neural networks (BPN) to recognition and temporal tracking of discrete impulse patterns via delay adaptation is demonstrated. The electronic emulation includes biologically plausible features, such as asynchronous impulses, membrane potentials and adaptive weights, as well as a mechanism to modify signal delays. The rule for the adaptation of impulse propagation delays is as follows: ‘error neurons” detect temporal differences between single impulses of other neurons and adjust corresponding signal delay parameters. In the application presented BPN adapts its time delays in order to form a finely tuned match with a given sequence of three discrete impulses. After learning, BPN is capable not only of highly selective recognition of the learned impulse pattern but also of tracking a gradually changing impulse pattern. Tracking is achieved by continuously re-adjusting the delay profile. Delay adaptation (rather than weight adaptation) appears to be the more effective mechanism for such applications.
Similar content being viewed by others
References
Beerhold JR, Jansen M, Eckmiller R (1990) Pulse-processing neural net hardware with selectable topology and adaptive weights and delays. In: Proc IEEE Int Joint Conf Neural Networks, San Diego, vol. II, pp 569–574.
Canditt S, Eckmiller R (1990) Pulse coding hardware neurons that can learn boolean functions. In: Proc IEEE Int Joint Conf Neural Networks, Washington, vol. II, pp 102–105.
Carr CE, Konishi M (1990) A circuit for detection of interaural time differences in the brain stem of the barn owl. J Neurosci 10: 3227–3246.
De Yoe EA, Van Essen DC (1988) Concurrent processing streams in monkey visual cortex. Trends Neurosci 11: 219–226.
Eckhorn R, Reitböck HJ, Arndt M, Dicke P (1990) Feature linking via sychronisation among distruibuted assemblies: simulations and results of cat visual cortex. Neural Comp 2: 293–307.
Eckmiller R (1975) Electronic simulation of the vertebrate retina. IEEE Trans Biomed Eng 22: 305–311.
French AS, Stein RB (1970) A flexible neural analog using integrated circuits. IEEE Trans Biomed Eng 17: 248–253.
Gerstner W, Ritz R, Hemmen JL van (1993) Why spikes? Hebbian learning and retrieval of time-resolved excitation patterns. Biol Cybern 69: 503–515.
Hartmann G (1992) Hierarchical neural representations by synchronised activity: a concept for visual pattern recognition. In: Taylor JG, Caianello ER, Cotterill RMJ, Clark JW (eds) Neural network dynamics. Springer, Berlin Heidelberg New York, pp 356–370.
Jansen M, Bluhm M, Napp-Zinn H, Eckmiller R (1991) Asynchronous pulse-processing neural net hardware for dynamic functions based on frequency and phase information. In: Ramacher U, Rückert U, Nossek JA (eds) Proc 2nd Int Conf Microelectronics for Neural Networks. Kyrill & Method, Munich, pp 359–365.
Lang KJ, Waibel AH (1990) A time-delay neural network architecture for isolated word recognition. Neural Networks 3: 23–43.
Prange SJ, Klar H (1991) Architectures for a biology-oriented neuroemulator. In: Ramacher U, Rückert U (eds) VLSI design of neural networks. Kluwer, Boston, pp 83–102.
Richert P, Hosticka BJ, Kesper M, Scholles M, Schwarz M (1993) An emulator for biologically-inspired neural networks. In: Proc IEEE Int Joint Conf Neural Networks, Nagoya, pp 841–844.
Schwarz M, Hosticka B, Kesper M, Richert P, Scholles M (1991) A CMOSarray-computer with on-chip communication hardware developed for massively parallel applications. In: Proc IEEE Int Joint Conf Neural Networks, Singapore, pp 89–94.
Widrow W, Winter R (1988) Neural nets for adaptive filtering and adaptive pattern recognition. Computer 21(3): 25–39.
Yamauchi K, Fukuda M, Fukushima K (1995) Speed invariant speech recognition using variable velocity delay lines. Neural Networks 8: 167–177.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Napp-Zinn, H., Jansen, M. & Eckmiller, R. Recognition and tracking of impulse patterns with delay adaptation in biology-inspired pulse processing neural net (BPN) hardware. Biol. Cybern. 74, 449–453 (1996). https://doi.org/10.1007/BF00206711
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF00206711