Reconfigurable array interconnection by photorefractive correlation

Electronic parallel processors might communicate more effectively by photons sent through glass or air than by electrons sent through wires, but quickly routing thousands of optical signals remains a problem. Previous photorefractive interconnection networks have dedicated one hologram to each input channel. Instead, we compute a control image from the entire network configuration and store it as a single color-keyedvolume hologram. This lets us use hologram superposition for fast switchingbetween multiple prestored patterns. During operation, data signals from the input modulator array, powered by a shared wavelength-tunable laser, are correlated optically with one color-matched connection hologram to produce the output array. This decouples both data rate and interconnect switching speeds from the slow photorefractive response. We can display arbitrary connection weights using simple binary-phase spatial light modulators and gracefully accommodate modulator limitations by trading off control-image bandwidth for output signal-to-noise ratio. Experimental results with color-multiplexed reflection holograms in z-cut LiNbO(3) confirmed our theoretical predictions that this approach works best for densely connected networks with high fan-in to each output. We obtained an average aggregate signal-to-noise ratio of more than 200:1 for 1024 inputs and outputs.     

Design of photonic crystal-based all-optical AND gate using T-shaped waveguide

Abstract

We present a new configuration of all-optical AND gate based on two-dimensional photonic crystal composed of Si rods in air. Two AND gate structures with and without probe input are proposed. The proposed structures are designed with T-shaped waveguide without using nonlinear materials and optical amplifiers. The performance of the proposed AND gate structures is analyzed and simulated by plane-wave expansion and finite difference time domain methods. The AND gate without probe input needs only one T-shaped waveguide, whereas the AND gate with probe input needs two T-shaped waveguides. The former AND gate offers a bit rate of 6.26 Tbps with a contrast ratio of 5.74 dB, whereas the latter AND gate offers a bit rate of 3.58 Tbps whose contrast ratio is 9.66 dB. It can be expected that these small size T-shaped structures are suitable for large-scale integration and can potentially be used in on-chip photonic integrated circuits.     

Dynamic digital photorefractive memory for optoelectronic neural network learning modules

Neural network modules based on page-oriented dynamic digital photorefractive memory are described. The modules can implement two different interconnection organizations, fan-out and fan-in, depending on their target network applications. Neural network learning is realized by the real-time memory update of dynamic digital photorefractive memory. Physical separation of subvolumes in the page-oriented photorefractive memory architecture contributes to the low cross talk and high diffraction efficiency of the stored interconnection weights. Digitally encoded interconnection weights ensure high accuracy, providing superior neural network system scalability. Module scalability and feedforward throughput have been investigated based on photorefractive memory geometry and the photodetector power requirements. The following four approaches to extend module scalability are discussed: partial optical summation, semiparallel feedforward operation, time partitioning, and interconnection matrix partitioning. Learning capabilities of the system are investigated in terms of required interconnection primitives for implementing learning processes and three memory-update schemes. The experimental results of Perceptron learning network implementation with 900 input neurons with digital 6-bit accuracy are reported.     

Determination of the transition frequency of a bipolar transistorusing optoelectronic current sources

The principle is described of a measurement by a method of comparison which allows the transition frequency f_T of a bipolar transistor to be measured. The AC test signal is obtained by optoelectronic injection. The same sine-modulated monochromatic light is shone successively on two photodiodes. The transistor does not need an actual excitation current at the input or short-circuit at the output since a method of comparison is used. The frequency response of the two photoelectric signals approximates a first-order low-pass function. Optoelectronic injection allows the presence of any parasite synchronous induced signals to be detected, leading to more accurate measurements. For a signal-to-induction ratio of 40 dB and a signal-to-noise ratio of 50 dB, the accuracy over the determinations of f_T is about ±5% for both amplitude and phase shift measurement     

Symmetric operating point for increasing the extinction ratio of semiconductor based optical loop mirrors

Terahertz optical asymmetric demultiplexors (TOADs) use a semiconductor optical amplifier in an interferometer to create an all-optical switch and have potential uses in many optical networking applications. Here we demonstrate and compare experimentally a novel and simple method of dramatically increasing the extinction ratio of the device using a symmetrical configuration as compared to a ‘traditional’ configuration. The new configuration is designed to suppress the occurrence of self-switching in the device thus allowing signal pulses to be used at higher power levels. Using the proposed configuration an increase in extinction ratio of 10 dB has been measured on the transmitted port whilst benefiting from an improved input signal power handling capability.     

Radio-frequency matched filtering by photorefractive optical holography

Matched filtering of megahertz-bandwidth signals by use of holograms recorded in a photorefractive crystal is demonstrated. Holographic recording of rapidly varying signals has heretofore been hampered by the relatively slow response of photorefractive crystals. For the first time, to our knowledge, synchronization between waveforms and short optical sampling pulses is used with acousto-optic electrical-to-optical conversion to build up static holograms of rf waveforms in a SBN crystal. Readout with a continuous input signal yields a time-resolved correlation with stored waveforms.     

Extending the bandwidth of magnetic tunnel junction sensors by abuffer amplifier

A simple buffer amplifier, consisting of a transistor in an emitter-follower configuration, is proposed to extend the bandwidth of magnetic tunnel junction read sensors. To maximize the bandwidth improvement, the buffer should be used on the slider or on the suspension as near to the slider as possible. The channel front-end (sensor, buffer, interconnect, and preamp) bandwidth is still somewhat limited by the RC product of the sensor resistance and shunt capacitance. However, the buffer can reduce the shunt capacitance to subpicofarad levels, allowing for bandwidths near 1 GHz. Merging the buffer with an interconnect and preamp gives a low-frequency signal-to-noise ratio (SNR) reduction at the preamp input of 0.1 dB and an SNR improvement of 5 to 10 dB over a no-buffer configuration for frequencies between 100 MHz and 1 GHz     

Highly efficient all-optical control of surface-plasmon-polariton generation based on a compact asymmetric single slit

By engaging a compact asymmetric single slit coated with a photorefractive polymer, surface-plasmon-polariton (SPP) generation was efficiently controlled by a pump beam. In the structure, the nonlinear light-matter interaction is enhanced because of the cavity effect, which increases the sensitivity of SPPs to the surrounding dielectric. By variation of the real part of the refractive index together with an interferometric configuration, high on/off switching ratios are achieved. Moreover, the SPP generation and modulation processes are integrated in the same asymmetric single slit, which makes the device ultracompact. Experimentally, a high on/off switching ratio of >20 dB and phase variation of >π were observed with the device lateral dimension of only about 2 μm.     

Secure optical data storage with random phase key codes by use of a configuration of a joint transform correlator

A secure optical storage based on a configuration of a joint transform correlator by use of a photorefractive material is presented. A key code designed through the use of an optimized algorithm so that its Fourier transform has a uniform amplitude distribution and a uniformly random phase distribution is introduced. Original two-dimensional data and the key code are placed side-by-side at the input plane. Both of them are stored in a photorefractive material as a joint power spectrum. The retrieval of the original data can be achieved with the same key code. We can record multiple two-dimensional data in the same crystal by angular multiplexing and/or key code multiplexing.     

All-optical 100-Gbit/s word packet time-division-multiplexed access node in a looped-back configuration: enabling technologies for sequential add-drop functionality

We experimentally demonstrate the adding, dropping, and passing through of 100-Gbit/s word packets in a looped-back all-optical time-division-multiplexed (TDM) access node. Packets are routed with a 17-dB contrast ratio and demultiplexed with a 20-dB contrast ratio. This node uses short 100-Gbit/s words to demonstrate its potential to process data packets from multiple sources and to perform packet switching in a multinode ring network configuration. The ability to tolerate timing jitter as well as varying input signal characteristics is essential to an all-optical access node in a multinode network. For 2-ps input pulses, the header processor has a timing window of ~5 ps, and the demultiplexer has a timing window of ~5.5 ps. This allows for tolerance to bit-to-bit timing jitters or to an increase in the pulse width of as much as 3 ps. Packet-to-packet timing jitter is detected and compensated by the technique used to synchronize the local source to each packet. The key enabling technologies of an all-optical TDM packet add-drop multiplexer are discussed, including an erbium-doped fiber laser, a nonlinear optical loop mirror logic gate, self-synchronization to incoming packets with a fast-saturation/slow-recovery gain element followed by an intensity discriminator, a two-wavelength nonlinear optical loop mirror demultiplexer, and synchronization of new packets to the network packet rate with a phase-locked loop. The local source is automatically synchronized to the incoming packet, because it uses an extracted pulse from the packet, which has a contrast ratio of >20 dB to the rest of the packet. Finally, new packets are added by use of a local laser and a synchronization method, which gives a timing jitter of ~1 ps. Using a statistical method of measuring Q value with picosecond resolution, we show that a header processor with two cascaded logic gates has a Q value of 7.1 with a 95% confidence level.     

Reflection-grating self-pumped phase conjugation with optical feedback free of input conditions

A reflection-grating self-pumped phase-conjugate mirror is demonstrated with a photorefractive KNbO(3) crystal in a novel ring configuration. The feedback beam in the ring configuration is made to track the input probe beam automatically so that the phase conjugation is sustained without readjustment of the optical feedback loop when the angle or the position of the input probe beam is changed. The tracking feedback is realized by means of a simple three-lens erect imaging system. The angular and the positional acceptances for a tightly focused input beam of 50-μm diameter are Δθ(x). = 55 mrad, Δθ(y), = 67 mrad, and Δ(x) = 2.1 mm, Δ(y) = 4.2 mm, respectively.     

All-optical compact encoder assembled by polymer fibres

An all-optical encoder based on a 3 × 3 crossed structure was assembled by stacking three polymer fibres. The encoding functions were predicated theoretically and demonstrated experimentally with very good agreement. The total insertion loss of the device is about 1 dB and the extinction ratio is larger than 10 dB. The output state can be controlled by launching optical signals to different input port, and the input state can be identified according to the output state. We believe that the all-optical compact encoder would be used in high-density PICs.     

External phase-modulation interferometry

We analyze and test a laboratory benchtop version of a compound interferometric phase sensor, a Michelson interferometer whose output is combined coherently with a phase-modulated local oscillator beam tapped off the Michelson input beam. This configuration models a whole class of external-modulation interferometers designed to shift signals, obscured by low-frequency intensity noise of the light source, into a shot-noise-limited region of the photocurrent spectrum. We find analytically that the shot-noise-limited sensitivity achievable with this system is comparable with that obtained by using internal phase modulation, with both schemes suffering (for different reasons) approximately a 22% sensitivity penalty compared with ideal shot-noise-limited direct detection. Experimentally we achieve true shot-noise-limited sensitivity, and we investigate trade-offs necessitated by commonly encountered nonideal features in any external-modulation system. Our analytic model, which specifically accounts for Michelson fringe contrast, electronic receiver noise, phase-modulation depth, and the local oscillator tap-off fraction, is sufficiently accurate to predict the absolute sensitivity of our benchtop instrument to within 0.5 dB.     

Spatial-Frequency Response of Photorefractive Phase Conjugators with Ce:BaTiO(3)

We propose and demonstrate an interferometric method to measure thespatial-frequency response of photorefractive phase conjugators withCe:BaTiO(3). Two coherent beams are incident on a crystaland form an interference pattern inside the crystal. The two beamsundergo stimulated photorefractive backscattering, which creates theircorresponding phase conjugations. Then the four waves interactwithin the crystal. The spatial-frequency resolution of the phaseconjugators is measured to be as high as 3750 line pairs/mm by use ofthe interferometric method. There are several factors that limitthe measured spatial resolution when using a U.S. Air Force ResolutionChart. The output modulation deviates from the input modulation forhigh spatial frequencies. In the presence of a strong additionalpump beam, the output modulation of the phase conjugators is almost thesame as the input modulation for a wide range of input spatialfrequencies. The phase conjugator exhibits a large dynamic range ofintensity. We analyze theoretically the modulation transferfunction of photorefractive phase conjugators withCe:BaTiO(3) for two mutually coherent beams. Thetheoretical analysis is in agreement with the experimental resultswithin a small incident-angle region.     

Tracking novelty filter at 780 nm based on a photorefractive polymer in a two-beam coupling geometry

We have constructed an all-optical tracking novelty filter based on the dynamic holographic properties of an efficient and fast infrared-sensitive photorefractive polymer. The photorefractive polymer was used in a two-beam coupling geometry. The polymer had a gain coefficient of 175 cm(-1) at a wavelength of 780 nm and an applied field of 72 V/mum. In contrast to what has been observed in photorefractive crystals, the gain coefficient and the filter contrast are largely independent of the writing beam's intensity ratio. We show images of a swinging pendulum observed through the novelty filter.     

Ultra wideband inphase power divider for multilayer technology

A multilayer inphase power divider with an ultra wideband behaviour is presented. The proposed divider exploits broadside coupling via a multilayer microstrip/slot configuration. The design method utilised for the device is based on the conformal mapping techniques. The developed device has a compact size with an overall dimension of 20 mm x 30 mm. The simulated and measured results show that the proposed device has equal power division between the two output ports with <0.2 dB amplitude imbalance between them, better than 10 dB return loss and isolation and < 2degrees phase difference between the two output signals across the frequency band 3.1-10.6 GHz.     

ADC characterization based on singular value decomposition

A new method for ADC characterization, based on singular value decomposition, is introduced in this paper. Here, the singular values of the sampled data matrix, directly derived from the measured input data, are used to characterize the signal-to-noise ratio (SNR), and further to estimate the number of effective bits. Various input signals, such as single-tone, dual-tone, or multitone, can be used to obtain accurate estimation results. In addition, the new method avoids the difficulties and problems of the earlier characterization methods such as the sensitivity to the applied sinewave frequency and sampled data sizes, and the spectral leakage. Extensive simulations indicate that the proposed method provides excellent performance with single-tone, dual-tone, and multitone test signals. The proposed method also shows remarkable robustness over a truly wide SNR range: from 5 dB to 200 dB     

Optoelectronic implementation of a 256-channel sonar adaptive-array processor

We present an optoelectronic implementation of an adaptive-array processor that is capable of performing beam forming and jammer nulling in signals of wide fractional bandwidth that are detected by an array of arbitrary topology. The optical system makes use of a two-dimensional scrolling spatial light modulator to represent an array of input signals in 256 tapped delay lines, two acousto-optic modulators for modulating the feedback error signal, and a photorefractive crystal for representing the adaptive weights as holographic gratings. Gradient-descent learning is used to dynamically adapt the holographic weights to optimally form multiple beams and to null out multiple interference sources, either in the near field or in the far field. Space-integration followed by differential heterodyne detection is used for generating the system's output. The processor is analyzed to show the effects of exponential weight decay on the optimum solution and on the convergence conditions. Several experimental results are presented that validate the system's capacity for broadband beam forming and jammer nulling for linear and circular arrays.     

Contrast imaging with chirped excitation

Coded excitation has been successfully used in imaging to increase the signal-to-noise ratio (SNR) and penetration depth. With a contrast agent, wideband signals have been hypothesized to increase the contrast-to-tissue ratio (CTR). However, nonlinear properties of contrast agents make decoding difficult when applying coded excitation to contrast imaging. We propose two chirped excitation methods to image contrast agents, with a mechanical index (MI) ranging from 0.05 to 0.34. In the single chirp method, one chirp is transmitted, followed by a clutter filter to reject tissue echoes, then a matched filter is used to recover range resolution. In the chirp sequence method, an increasing and decreasing chirp sequence is transmitted followed by subtraction of the compressed echoes to reject tissue echoes (assuming tissue is a linear scatterer at low MI). Ten independent acoustic experiments were performed to evaluate the CTR for chirp and tone burst insonation, with the same spatial peak temporal averaged intensity (I(SPTA)). A significant increase in CTR, ranging from 4 dB to 8 dB, is observed for chirped excitation as compared with tone burst insonation, at an I(SPTA) of 0.1 and 0.3 mW/cm2 (P < or = 5e-3). To achieve the same CTR of 15 dB, the spatial peak pulse averaged intensity (I(SPPA)) can be decreased by 6 dB for chirp insonation as compared with tone burst insonation (P < 1e-5). Additionally, an increase of more than 10 dB in tissue rejection ratio (TRR) is observed for a chirp sequence insonation compared to tone burst phase inversion for this set of parameters (P < or = 1e-9). Deconvolution of the linear microbubble response from the received echoes is proposed as a method to recover spatial resolution. The difference in the axial resolution resulting from chirp and three-cycle tone burst insonation is approximately 220 microm. The difference in the mainlobe width between experimental and predicted compressed echoes is less than 20%. The side-lobe amplitude is 9 dB to 16 dB below the mainlobe with a transmitted I(SPTA) from 0.1 to 6.6 mW/cm2.     

High sensitivity noise measurement with a correlation spectrumanalyzer

A spectrum analyzer with enhanced sensitivity has been built and used in noise measurements. It is based on the processing of the input signal by two independent channels in parallel and takes advantage of the incoherent property of the noise in each of the two input stages. The instrument has demonstrated an improvement in sensitivity of at least 50 dB with respect to a traditional system, and therefore can measure low input signals down to the hundred pV/√Hz range