Showing papers on "Neuromorphic engineering published in 2000"

Point-to-point connectivity between neuromorphic chips using address events

Abstract: This paper discusses connectivity between neuromorphic chips, which use the timing of fixed-height fixed-width pulses to encode information. Address-events (log/sub 2/(N)-bit packets that uniquely identify one of N neurons) are used to transmit these pulses in real time on a random-access time-multiplexed communication channel. Activity is assumed to consist of neuronal ensembles-spikes clustered in space and in time. This paper quantifies tradeoffs faced in allocating bandwidth, granting access, and queuing, as well as throughput requirements, and concludes that an arbitered channel design is the best choice. The arbitered channel is implemented with a formal design methodology for asynchronous digital VLSI CMOS systems, after introducing the reader to this top-down synthesis technique. Following the evolution of three generations of designs, it is shown how the overhead of arbitrating, and encoding and decoding, can be reduced in area (from N to /spl radic/N) by organizing neurons into rows and columns, and reduced in time (from log/sub 2/(N) to 2) by exploiting locality in the arbiter tree and in the row-column architecture, and clustered activity. Throughput is boosted by pipelining and by reading spikes in parallel. Simple techniques that reduce crosstalk in these mixed analog-digital systems are described.

Introduction to Artificial Neural Network

Abstract: Artificial neural networks or simply “neural nets” go by many names such as connectionist models, parallel distributed processing models, and neuromorphic systems. Whatever terminology it may be, they all attempt to borrow the structure and running way of the biological nervous system based on our present understanding of it. Instead of performing a program consisting of instructions sequentially as in a von Neumann computer, artificial neural nets have their structures in dense interconnection of simple computational elements— the artificial neurons or simply “neurons”, and operate the massive computational elements in parallel to achieve high performance speed.

Neuromorphic Vision Sensors

Abstract: Vision is one of the most useful sensory functions, but the real-time processing of the continuous, high-dimensional input signals provided by vision sensors is a major challenge in robot design. Conventional digital vision sensors tend to have excessive power consumption, size, and cost for useful applications. In this TechView, Indiveri and Douglas discuss an alternative strategy, namely neuromorphic vision sensors that are based on biological vision systems. In these sensors, specialized sensory processing functions inspired by biological systems such as fly eyes are integrated in parallel, asynchronous circuits that respond in real time. These sensors offer significant advantages over conventional vision sensors.

Modeling Selective Attention Using a Neuromorphic Analog VLSI Device

Abstract: Attentional mechanisms are required to overcome the problem of flooding a limited processing capacity system with information. They are present in biological sensory systems and can be a useful engineering tool for artificial visual systems. In this article we present a hardware model of a selective attention mechanism implemented on a very large-scale integration (VLSI) chip, using analog neuromorphic circuits. The chip exploits a spike-based representation to receive, process, and transmit signals. It can be used as a transceiver module for building multichip neuromorphic vision systems. We describe the circuits that carry out the main processing stages of the selective attention mechanism and provide experimental data for each circuit. We demonstrate the expected behavior of the model at the system level by stimulating the chip with both artificially generated control signals and signals obtained from a saliency map, computed from an image containing several salient features.

A neuromorphic aVLSI model of global motion processing in the fly

Abstract: Flies orientate themselves quickly in an unstructured environment through motion information computed from their low-resolution compound eyes. The fly visual system is an example of a robust motion system that works in a natural environment. In this paper, the author describes a low-power analog very large scale integration (aVLSI) chip that models motion computation in the fly. The architecture of this motion chip closely follows the anatomical layout of of the fly visual layers. The output of the chip corresponds to the responses of the wide-field direction-selective cells in the final layer of the visual system. The silicon chip has a one-dimensional array of 37 elementary motion detectors (EMDs) each of which provides local motion information. The EMD outputs are aggregated in a nonlinear way to produce a motion output that is independent of the stimulus size and contrast. The author employed various circuit techniques to ensure robust motion computation in each processing stage. Results from the circuit fabricated in a 1.2 /spl mu/m CMOS technology are compared with the responses of the direction-selective cells.

A Modular Multi-Chip Neuromorphic Architecture for Real-Time Visual Motion Processing

Abstract: The extent of pixel-parallel focal plane image processing is limited by pixel area and imager fill factor. In this paper, we describe a novel multi-chip neuromorphic VLSI visual motion processing system which combines analog circuitry with an asynchronous digital interchip communications protocol to allow more complex pixel-parallel motion processing than is possible in the focal plane. This multi-chip system retains the primary advantages of focal plane neuromorphic image processors: low-power consumption, continuous-time operation, and small size. The two basic VLSI building blocks are a photosensitive sender chip which incorporates a 2D imager array and transmits the position of moving spatial edges, and a receiver chip which computes a 2D optical flow vector field from the edge information. The elementary two-chip motion processing system consisting of a single sender and receiver is first characterized. Subsequently, two three-chip motion processing systems are described. The first three-chip system uses two sender chips to compute the presence of motion only at a particular stereoscopic depth from the imagers. The second three-chip system uses two receivers to simultaneously compute a linear and polar topographic mapping of the image plane, resulting in information about image translation, rotation, and expansion. These three-chip systems demonstrate the modularity and flexibility of the multi-chip neuromorphic approach.

Four-legged Walking Gait Control Using a Neuromorphic Chip Interfaced to a Support Vector Learning Algorithm

Abstract: To control the walking gaits of a four-legged robot we present a novel neuromorphic VLSI chip that coordinates the relative phasing of the robot's legs similar to how spinal Central Pattern Generators are believed to control vertebrate locomotion [3]. The chip controls the leg movements by driving motors with time varying voltages which are the outputs of a small network of coupled oscillators. The characteristics of the chip's output voltages depend on a set of input parameters. The relationship between input parameters and output voltages can be computed analytically for an idealized system. In practice, however, this ideal relationship is only approximately true due to transistor mismatch and offsets. Fine tuning of the chip's input parameters is done automatically by the robotic system, using an unsupervised Support Vector (SV) learning algorithm introduced recently [7]. The learning requires only that the description of the desired output is given. The machine learns from (unlabeled) examples how to set the parameters to the chip in order to obtain a desired motor behavior.

A 2D neuromorphic VLSI architecture for modeling selective attention

Abstract: Selective attention is a mechanism used to sequentially select the spatial locations of salient regions in the sensor's field of view. This mechanism overcomes the problem of flooding limited processing capacity systems with sensory information. It is found in many biological sensory systems and can be a useful engineering tool for artificial visual systems. We present a hardware model of a selective attention mechanism implemented on a VLSI chip, using analog neuromorphic circuits. The chip makes use of a spike based representation for receiving input signals, transmitting output signals and for shifting the selection of the attended input stimulus over time. The chip can be interfaced to neuromorphic sensors and actuators, for implementing multi-chip selective attention systems. We describe the characteristics of the circuits used in the architecture, and present experimental data measured from the system.

Foundations for a Circuit Complexity Theory of Sensory Processing

Abstract: We introduce total wire length as salient complexity measure for an analysis of the circuit complexity of sensory processing in biological neural systems and neuromorphic engineering. This new complexity measure is applied to a set of basic computational problems that apparently need to be solved by circuits for translation- and scale-invariant sensory processing. We exhibit new circuit design strategies for these new benchmark functions that can be implemented within realistic complexity bounds, in particular with linear or almost linear total wire length.

Smart Vision Chip Fabricated Using Three Dimensional Integration Technology

Abstract: The smart vision chip has a large potential for application in general purpose high speed image processing systems. In order to fabricate smart vision chips including photo detector compactly, we have proposed the application of three dimensional LSI technology for smart vision chips. Three dimensional technology has great potential to realize new neuromorphic systems inspired by not only the biological function but also the biological structure. In this paper, we describe our three dimensional LSI technology for neuromorphic circuits and the design of smart vision chips.

Toward a digital neuromorphic pitch extraction system

Abstract: Presents the design of a biologically based signal processing system implemented using standard digital inferior colliculus (IC) technology. The four-stage AM detection system is a step toward a full-pitch detection system and based on known mammalian physiology. The system is operational and has been successfully realized in field programmable grid array technology. Details of the system architecture, its operating principles, and the design decisions necessary to realize successfully neuromorphic systems in digital technology are given.

A qualitative model-framework for spatio-temporal effects in vertebrate retinas

Abstract: Retinal models based on the cellular neural network (CNN) paradigm have been widely used. These neuromorphic models are based on retinal anatomy and physiology. In this paper a framework is proposed for qualitative spatio-temporal studies in vertebrate retinas, the underlying retinal anatomy is followed as closely as possible, the characteristics of the physiological models, however, are kept simple. The goal is to model the qualitative effects, since the developed models are simple, compared to a fully neuromorphic one, we have a good chance to implement them on CNN Universal Machine chips using multi-layer technology.

Neuromorphic CNN models for spatio-temporal effects measured in the inner and outer retina of tiger salamander

Abstract: A vertebrate retina model is described based on a cellular neural network (CNN) architecture. Though largely built on the experience of previous studies the CNN computational framework is considerably simplified: first order RC cells are used with space-invariant nearest neighbor interactions only. All nonlinear synaptic connections are monotonic continuous functions of the pre-synaptic voltage. Time delays in the interactions are continuous represented by additional first order cells. The modeling approach is neuromorphic in its spirit relying on both morphological and pharmacological information. However, the primary motivation lies in fitting the spatio-temporal output of the model to the data recorded from biological cells (tiger salamander). In order to meet a low complexity (VLSI) implementation framework some structural simplifications have been made and large neighborhood interaction (neurons with large processes), furthermore the inter-layer signal propagation are modeled through diffusion and wave phenomena. This work presents novel CNN models for the outer and some partial models for the inner (light adopted) retina.

Neuromorphic cochlea implants

Abstract: A novel approach to signal processing in cochlea implants using bio-inspired models is proposed. The novel architecture uses spike-based signal processing eliminating the need for DSPs. The programmability is achieved using redundancy with digital control avoiding ADCs altogether. The neuromorphic cochlea implant should reduce both power and size to a level of an implantable system.

A current-mode spike-based overrange-subrange analog-to-digital converter

Abstract: We present a novel technique for performing current-mode analog-to-digital conversion using two integrate-and-fire spiking neurons. The conversion is performed in two steps. The first step is an overranging step that extracts the integer part of the ratio between the input current and a reference current; the second step is a subranging step that extracts the fractional part of this ratio. The conversion is performed by having analog spike-time information in one neuron, the timing neuron, generating digital spike-count information in another neuron, the counting neuron. The roles of the timing neuron and the counting neuron are reversed as we transition between the steps. Operations in the converter are coordinated via a spike-triggered asynchronous finite state machine, obviating the need for a clock to synchronize the internal operations of the converter. We present experimental data from a proof-of-concept VLSI implementation. Because spike number is discrete while interspike intervals are analog, spikes are naturally suited for hybrid computation, i.e., computation that is not purely analog or purely digital but that is a mixture of both forms. Our converter provides an example for the use of such spike-based hybrid computation. The converter is suited for compact, low-power neuromorphic applications.

Neuromorphic Pattern Recognition Using Arrays of Quantum Dots

Abstract: Revolutionary advances in computing, communication, detection, and sensing using nanoscale devices subsume a pro found understanding of the complex dynamics of small arrays of quantum structures. Such arrays produce bi-stable and multi-stable robust behavior, which can be harnessed for unconventional, yet powerful computational concepts. In this paper, we propose a n ovel approach to signal pattern analysis using an array of quantum dots (QD). Our methodology combines an ultrafast neuromorphic learning algorithm with photon-assisted tunneling in the QD array. The latter enables emulation of the plasticity of neural synapses.

CMOS mixed digital analog reconfigurable neural network with Gaussian synapses

Abstract: In the neuromorphic arena, different engineering problems require different neural network topologies. In view of this concept, the motivation was to build a reconfigurable neural network chip. The distributed Gaussian-neuron synapse is introduced as a new type of synapses. Also a new improved resolution current-mode winner-takes-all circuit is added to realize a self-organizing topology. The chip is organized into 4 partially connected tiles with 4/spl times/3 fully connected neurons per tile. The chip was fabricated through MOSIS in 1.2 /spl mu/m AMI CMOS process occupying an area of 2 mm /spl times/2 mm.

Current mode elementary motion perception neural cell

Abstract: The paper proposes biologically inspired NN architecture that employs neural correlation of motion detection via a two-channel mechanism similar to the visual system of invertebrates that is capable of detecting the direction of motion. The correlation-based scheme with mutual lateral inhibition is adopted to implement current mode VLSI neuromorphic cells with the dynamics characterized by a first order nonlinear differential equation. Such cells act as elementary motion detectors (EMDs) sensitive to the direction of motion and can be implemented as a front-end section of a sophisticated object tracking system. Spatio-temporal responses of the VLSI model are investigated and simulation results are given.

An analog vision chip applicable to real-time image processings in indoor illumination

Abstract: We have fabricated an analog VLSI circuit with 1D /spl nabla//sup 2/G-like receptive field. Into the chip are incorporated active pixel sensors and sample/hold circuits to achieve high gain and low noise. A real time edge detection was carried out under indoor fluorescence with the chip, indicating that the neuromorphic retina fabricated in the present study is usable for industrial applications.