SYMNET: an optical interconnection network for scalable high-performance symmetric multiprocessors

We address the primary limitation of the bandwidth to satisfy the demands for address transactions in future cache-coherent symmetric multiprocessors (SMPs). It is widely known that the bus speed and the coherence overhead limit the snoop/address bandwidth needed to broadcast address transactions to all processors. As a solution, we propose a scalable address subnetwork called symmetric multiprocessor network (SYMNET) in which address requests and snoop responses of SMPs are implemented optically. SYMNET not only has the ability to pipeline address requests, but also multiple address requests from different processors can propagate through the address subnetwork simultaneously. This is in contrast with all electrical bus-based SMPs, where only a single request is broadcast on the physical address bus at any given point in time. The simultaneous propagation of multiple address requests in SYMNET increases the available address bandwidth and lowers the latency of the network, but the preservation of cache coherence can no longer be maintained with the usual fast snooping protocols. A modified snooping cache-coherence protocol, coherence in SYMNET (COSYM) is introduced to solve the coherence problem. We evaluated SYMNET with a subset of Splash-2 benchmarks and compared it with the electrical bus-based MOESI (modified, owned, exclusive, shared, invalid) protocol. Our simulation studies have shown a 5-66% improvement in execution time for COSYM as compared with MOESI for various applications. Simulations have also shown that the average latency for a transaction to complete by use of COSYM protocol was 5-78% better than the MOESI protocol. SYMNET can scale up to hundreds of processors while still using fast snooping-based cache-coherence protocols, and additional performance gains may be attained with further improvement in optical device technology.     

Smart-pixel-based network interface chip

We present the design and experimental setup of an optically interconnected smart-pixel network interface chip designed to implement a collisionless multichannel-access control protocol. The design demonstrates the cointegration of optoelectronic pixel modules of various levels of complexity for dense, high-speed interconnection of highly functional digital logic components typical of multiprocessor network routers.     

The application of a SRAM chip as a novel neutron detector

High-density, fast digital devices, like field programmable gate arrays (FPGAs), microcontrollers, and static random access memories (SRAMs), can be produced by nanotechnology. New technologies allow the design of fast and powerful devices; however, the decreasing dimensions create new problems. Even at ground level, cosmic ray particles arriving from outer space can affect digital devices and provoke single-event effects (SEEs) due to the smaller sensitive volume (SV). In general, for decreasing feature size of memory cells the expected critical charge decreases and the expected sensitivity to radiation increases.

High-density SRAM chips were used to design a fast response, highly sensitive neutron detector. We have conducted experiments with SRAMs at the DESY Research Centre in Hamburg, Germany. Memory contents (number of SEU) were recorded as a function of neutron expose time. The chips were exposed to a neutron field from an americium-beryllium neutron source (²⁴¹AmBe). The second experiment was accomplished in the 450?MeV electron Linac (Linac II) tunnel. Another batch of SRAMs was irradiated with ⁶⁰Co gamma rays to a dose of about 60?Gy, and no SEU was registered. This shows that gamma radiation has no substantial effect on the production of SEU in the SRAM. The proposed detector could be ideal for the detection of pulsed neutron radiation produced by high-energy electron linear accelerators and synchrotron facilities, which are currently in operation and planned for the near future.     

Microfabricated polymer analysis chip for optical detection

A coupling between multimode polymer waveguides and microfluidic channels on a polymethylmethacrylate (PMMA) capillary electrophoresis (CE)-chip for optical analytical applications has been successfully realised. This technology allows the integration of polymer optical waveguides together with hermetically sealed fluidic channels. The microchannels and waveguides are made in PMMA by the approved hot-embossing technology. The technology developed for the fabrication of polymer waveguides on the microfluidic chip offers the possibility of great flexibility in the choice of core materials, design and alignment of the polymer waveguides. The integration of polymer waveguides on an analysis chip enables highly spatially resolved optical detection without the large and expensive conventionally used apparatus. The optical properties of the analytical system developed are verified by transmission and propagation loss measurements. The results of measurements prove the suitability of the presented device for optical applications between 440 and 800 nm. This was shown with absorbance measurements of the dye Sulfanilazochromotrop (SPADNS) within 50 microm fluidic channels.     

Analysis of a digital microstrip optical switch: a novel method

A time domain analysis of an optically controlled digital microstrip switch for microwave integrated circuits on Si substrates is studied. A new model for high-frequency pulse propagation on a microstrip optical switch for different optical parameters is presented. A frequency-dependent macromodel for a microstrip line with a gap is implemented in Spice 3, taking into consideration high-frequency pulse dispersion, conductor and dielectric losses, metallization thickness, gap length, and different optical parameters such as optical energy, surface recombination velocities, and diffusion of generated carriers. In addition, the developed model has been used to optimize the switching frequency, gap length, level of optical power, and suitable substrate material parameters.     

Characterization of diffraction gratings in a rigorous domain with optical scatterometry: hierarchical neural-network model

Characterization of microstructures with features from submicrometers to hundreds of micrometers requires versatile methods. Profilometry and optical microscopy cannot cope with submicrometer features, and atomic-force microscopy, scanning-electron microscopy, and near-field microscopy are inherently slow, off-line methods. In optical scatterometry, the laser light scattered from a sample is measured and the sample profile is subsequently characterized. We propose the use of a two-stage model based on neural networks: rough categorization followed by refinement, thus reducing the need for prior information on the sample. We simulate the method for a submicrometer diffraction grating characterized by five parameters. It is shown that intensity measurements of few diffraction orders by use only of one wavelength are enough to yield rms errors of less than 2 nm for the parameters (approximately 2-3% of the optimal values of the parameters).     

Frequency-multiplexed logic circuit based on a coherent optical neural network

We propose an adaptive logic circuit whose function can be controlled by optical carrier frequency modulation. The circuit learns the desired functions by adjusting the delay time at a spatial light modulator with a complex-valued Hebbian learning rule. After the learning, the circuit can switch its function all at once. A high degree of mechanical stability is achieved by spatial phase-difference coding. Two orthogonal phase components are detected in parallel spatially. Experiments demonstrate that the system works as an AND circuit at a certain frequency and as an XOR at another. The proposal will enhance the design of optical plastic cell architectures.     

A dc microplasma on a chip employed as an optical emission detector for gas chromatography

A micromachined plasma chip is coupled to a conventional gas chromatograph to investigate its performance as an optical emission detector. The device employs a 180-nL plasma chamber in which an atmospheric pressure dc glow discharge is generated in helium. Applied power is 9 mW (770 V, 12 microA) and helium flow rate 320 nL s(-1). A number of carbon-containing compounds are detected in the column effluent by recording the emission at 519 nm. For hexane, the detector has a linear dynamic range of over two decades and a minimum detectability of 10(-12) g s(-1) (800 ppb). The detector signal shows a marked peak broadening and tailing when compared with the signal of a flame ionization detector. This is mainly attributed to dead volumes and chromatographic processes introduced by the connecting tubing and the chip glass channels. The device was operated for more than 24 h without a significant change in performance. Operation is stable and instrumental requirements are simple. Future use of the detector chip in conventional gas chromatography or as an integrated detector in on-chip gas chromatography is discussed.     

基于遗传算法的层间等级网络编码优化

面向单信源异构信宿网络,研究了层间等级网络编码的编码类型优化。基于遗传算法,提出了一种最优编码类型的快速搜索方案。该方案充分考虑了信源输出链路上进行的层间等级网络编码的编码类型对整个网络传输性能的影响,将网络总吞吐量作为评价编码类型优劣的标准,设计了符合层间等级网络编码本质特性的遗传操作。实验结果表明,与分层组播网络编码和基于现有启发式算法的层间等级网络编码相比,基于本文方案实现的层间等级网络编码能够为单信源异构信宿网络获得更高的网络总吞吐量。     

Disposable optical sensor chip for medical diagnostics: new ways in bioanalysis

An optical sensor system is described which is particularly well suited for medical point-of-care diagnostics. The system allows for all kinds of immunochemical assay formats and consists of a disposable sensor chip and an optical readout device. The chip is built up from a ground and cover plate with in- and outlet and, between, of an adhesive film with a capillary aperture of 50 microns. The ground plate serves as a solid phase for the immobilization of biocomponents. In the readout device, an evanescent field is generated at the surface of the ground plate by total internal reflection of a laser beam. This field is used for the excitation of fluorophor markers. The generated fluorescence light is detected by a simple optical setup using a photomultiplier tube. Because of the evanescent field excitation, washing or separation steps can be avoided. With this system the pregnancy hormone chorionic gonadotropin (hCG) could be determined in human serum with a detection limit of 1 ng/mL. Recovery values were 86, 106, and 102% for 5, 50, and 100 ng/mL hCG, respectively. The SD in repeated measurements (n = 10) was 5.6%. Furthermore, the feasibility of the system in competitive-type immunoassays was demonstrated for serum theophylline. A linear calibration curve of signal vs theophylline between 1 and 50 mg/L was obtained. Recovery values varied between 118% (10 mg/L) and 81.0% (20 mg/L).     

Design of hierarchical buffer structure for silicon/carbon composite as a high-performance Li-ion batteries anode

Silicon-based materials are used as anode material for lithium-ion batteries, due to ultra-high theoretical specific capacity. However, large volume changes, continuous formation of unstable solid electrolyte interface film and low conductivity greatly restricted its large-scale development and application. In this case, a composite with hierarchical buffer structure coated Si nanoparticles (Si@RF@MP) was designed and manufactured by the surfactant template and emulsification method in this study. The resorcinol–formaldehyde resin acts as the structural buffer and the conductive layer to accommodate the volume change of silicon and provide fast channels for electron transfer and lithium-ion diffusion. The unique turbostratic structure of mesophase pitch can effectively improve the integral conductivity and the structural stability of the electrode. As a result, the Si@RF@MP composite exhibited an excellent reversible discharge capacity of 389 mA h g^?1 after 200 cycles at 200 mA g^?1, and retained a discharge capacity of 345 mA h g^?1 after 300 cycles at a high current density of 1000 mA g^?1. In addition, the Si@RF@MP composite delivered reversible capacities of about 546 mA h g^?1, 495 mA h g^?1, and 437 mA h g^?1 in current densities of 500 mA g^?1, 1000 mA g^?1, and 2000 mA g^?1, respectively, indicating good rate performance. Hence, this strategy provides a new method and idea for the further development of silicon/carbon composites and a strategy to achieve high value and green utilization of pitch.

     

A nano- and micro- integrated protein chip based on quantum dot probes and a microfluidic network

A novel nano- and micro-integrated protein chip (NMIPC) that can detect proteins with ultrahigh sensitivity has been fabricated. A microfluidic network (μFN) was used to construct the protein chips, which allowed facile patterning of proteins and subsequent biomolecular recognition. Aqueous phase-synthesized, water-soluble fluorescent CdTe/CdS core-shell quantum dots (aqQDs), having high quantum yield and high photostability, were used as the signaling probe. Importantly, it was found that aqQDs were compatible with microfluidic format assays, which afforded highly sensitive protein chips for cancer biomarker assays. Electronic Supplementary Material  Supplementary material is available for this article at and is accessible for authorized users.     

Synthesis and optical properties of hierarchical pure ZnO nanostructures

We report the catalyst-free synthesis of hierarchical pure ZnO nanostructures with 6-fold structural symmetry by two-step thermal evaporation process. At the first step, the hexagonal-shaped nanowires consisting of a great deal of Zn and little oxide were prepared via the layer-by-layer growth mechanism; and at the second step, hierarchical pure ZnO nanostructures were synthesized by evaporating the Zn source on the basis of the step-one made substrate. Scanning electron microscopy, transmission electron microscope images, and the corresponding selected area electron diffraction pattern have been utilized to reveal the screw dislocation growth mechanism, through which the single crystal ZnO nanorods are epitaxially grown from the side-wall of central axial nanowires. Raman and photoluminescence spectra further indicate that, for the hierarchical ZnO nanostructures, the ultraviolet peak is related to the free exciton recombination, while the oxygen vacancies and high surface-to-volume ratio are responsible for the strong green peak emission.     

Design of high-performance photodiode receivers for optical tomography

The design of instrumentation hardware for tomographic systems must take careful account of measurement noise. This is especially true in near-infrared absorption tomography, where the signal of interest is typically only a few percent of the total signal at the detector, and the available optical power may have to be shared among many measurement channels. In this paper, the monitoring of photodiodes in near-IR absorption tomography is examined in detail, but much of the material is applicable at wavelengths ranging from the UV to beyond 2.5 /spl mu/m. The authors' application involves the frequency region 50 kHz to 2 MHz, which lies above that utilized in the majority of radiometric sensing systems, yet substantially below telecoms bit rates. The problem is further distinguished by the use of phase-sensitive detection schemes, which make local noise density more relevant than wideband noise performance and relax the requirement for dc precision. Alternative transimpedance circuit configurations, including both single-ended and differential topologies, are analyzed with a view to optimization of the signal-to-noise ratio. Typical values of photodiode capacitance and shunt resistance are shown to result in significant noise gain, greatly increasing the importance of amplifier voltage noise relative to other intrinsic noise sources. It is shown that for applications of this type, viable alternatives to the traditionally dominant FET amplifier do exist. The relative susceptibility to coupled interference is also considered. The results of practical tests, involving class-leading operational amplifiers, are presented to support the analyses. These results also underline the need for careful circuit layout and shielding if the capabilities of these devices are to be fully exploited.     

Testbed for a scalable terabit optical local area network

The design of a fiber-optic local area network (LAN) demonstration system is described. A complete LAN system would consist of an array of 16 personal computers (PC's), where each PC has a network interface card (NIC) with a parallel fiber-optic datalink to a centralized optoelectronic switch core. The centralized core switches the data generated by 16 NIC's, up to 128 Gbit/s of bandwidth. The demonstrator is designed to scale to terabits of bandwidth by use of an emerging optoelectronic technology, i.e., integrated complementary metal-oxide semiconductor (CMOS) substrates with vertical-cavity surface-emitting laser (VCSEL) and photodetector optical input and output. A subset of the complete system was constructed and is operational. A prototype NIC card, with Motorola Optobus VCSEL transceivers for the optical datalinks, was constructed and is described. A prototype high-speed bipolar switch core, with statically configurable electrical positive-emitter coupled-logic 16 x 16 crossbar switches, CMOS field-programmable gate arrays, and Motorola Optobus transceivers, was constructed and is described. We successfully demonstrated the transmission of high-speed packetized data from one NIC card, through 10 m of parallel fiber ribbon and the centralized switch core, and back to the NIC. We summarize our experiences on the design and testing of our first demonstration system and our development toward a terabit switch core.     

Road network safety evaluation using Bayesian hierarchical joint model

Safety and efficiency are commonly regarded as two significant performance indicators of transportation systems. In practice, road network planning has focused on road capacity and transport efficiency whereas the safety level of a road network has received little attention in the planning stage. This study develops a Bayesian hierarchical joint model for road network safety evaluation to help planners take traffic safety into account when planning a road network. The proposed model establishes relationships between road network risk and micro-level variables related to road entities and traffic volume, as well as socioeconomic, trip generation and network density variables at macro level which are generally used for long term transportation plans. In addition, network spatial correlation between intersections and their connected road segments is also considered in the model.A road network is elaborately selected in order to compare the proposed hierarchical joint model with a previous joint model and a negative binomial model. According to the results of the model comparison, the hierarchical joint model outperforms the joint model and negative binomial model in terms of the goodness-of-fit and predictive performance, which indicates the reasonableness of considering the hierarchical data structure in crash prediction and analysis. Moreover, both random effects at the TAZ level and the spatial correlation between intersections and their adjacent segments are found to be significant, supporting the employment of the hierarchical joint model as an alternative in road-network-level safety modeling as well.     

Enhanced fuzzy-connective-based hierarchical aggregation network using particle swarm optimization

The fuzzy-connective-based aggregation network is similar to the human decision-making process. It is capable of aggregating and propagating degrees of satisfaction of a set of criteria in a hierarchical manner. Its interpreting ability and transparency make it especially desirable. To enhance its effectiveness and further applicability, a learning approach is successfully developed based on particle swarm optimization to determine the weights and parameters of the connectives in the network. By experimenting on eight datasets with different characteristics and conducting further statistical tests, it has been found to outperform the gradient- and genetic algorithm-based learning approaches proposed in the literature; furthermore, it is capable of generating more accurate estimates. The present approach retains the original benefits of fuzzy-connective-based aggregation networks and is widely applicable. The characteristics of the learning approaches are also discussed and summarized, providing better understanding of the similarities and differences among these three approaches.