We present experimental evidence indicating the formation of dipolar holographic gratings in potassium lithium tantalate niobate at the paraelectric phase slightly above the phase transition, These gratings are formed in the dark, following the writing of space-charge-based photorefractive gratings, under an external electric field. The dipolar gratings create a spatial modulation of the low-frequency dielectric constant, and the latter induces a spatially correlated modulation of the polarization, which is transformed into birefringence gratings through the quadratic electro-optic effect. (C) 1997 Optical Society of America.

We predict spatial solitons in photorefractive centrosymmetric media driven by the dc Kerr effect.

A generic circuit, the charge controlled analog synapse (CCAS) is presented. The CCAS is designed to be the basic building block in microelectronic realizations of large scale artificial neural networks. It is based on representing the synaptic strength as a charge packet which controls the junction capacitance of a reverse biased diode. The CCAS is a synapse with three main features: (i) a small cell with few components per cell; (ii) a short term dynamic memory, and (iii) a variable accuracy which depends on the cell size. The principle of operation of the CCAS is explained. Details of the design of a first prototype are given. Experimental results which substantiate the theoretical predictions are presented. Finally, the basic properties of the CCAS are discussed. Copyright (C) 1996 Elsevier Science Ltd

We describe a generic approach for realizing networks of pulsating neurons based on charge pumping of interface states situated in the channel of MOS transistors. Two basic building blocks will be described: the pulse activated charge pumping (PSCP) synapse, and the charge sensitive oscillator (CSO). The PSCP synapse which operates as either a short or a long term memory device which produces a charge packet proportional to the number of pulses applied to its input, will be described in detail together with experimental results demonstrating its capability. The CSO circuit which is a charge controlled oscillator will be described together with simulations of its output frequency dependence on its input voltage, and the relation between the temporal dependence of output waveform on its input charge.

We describe a new approach for constructing large-scale artificial neural networks. The novelty of our approach is based on the concept of electroholography (EH), which permits interconnecting of electronic neurons by minute-volume holograms, using the voltage-controlled photorefractive effect in paraelectric crystals. Crystals of potassium lithium tantalate niobate (KLTN) in the paraelectric phase are shown to be suitable for implementing this concept. A small network composed of two KLTN crystals on which holographic connections are recorded is presented to demonstrate the EH approach.

A new approach for implementing artificial neural networks (ANNs), based ona novel concept in volume holographic storage-electroholography (EH)-is described. EH is based on the use of paraelectric photorefractive crystals in which the photorefractive mechanism is created in the paraelectric phase. Thus the quadratic electrooptic effect is used to transform the information carrying space charge fields into birefringence gratings, and therefore the diffraction efficiency is governed by an external electric field. This principle enables the construction of a page oriented architecture in which each neuron is implemented as a local electronic circuit, governing the diffraction efficiency of a small volume hologram representing the axon of the neuron. The EH approach is expected to enable the construction of networks with 105−106 neurons, each with 103−104 synapses.