Massive Interconnection Research Articles

Multichannel activity propagation across an engineered axon network

Objective. Although substantial progress has been made in mapping the connections of the brain, less is known about how this organization translates into brain function. In particular, the massive interconnectivity of the brain has made it difficult to specifically examine data transmission between two nodes of the connectome, a central component of the ‘neural code.’ Here, we investigated the propagation of multiple streams of asynchronous neuronal activity across an isolated in vitro ‘connectome unit.’ Approach. We used the novel technique of axon stretch growth to create a model of a long-range cortico-cortical network, a modular system consisting of paired nodes of cortical neurons connected by axon tracts. Using optical stimulation and multi-electrode array recording techniques, we explored how input patterns are represented by cortical networks, how these representations shift as they are transmitted between cortical nodes and perturbed by external conditions, and how well the downstream node distinguishes different patterns. Main results. Stimulus representations included direct, synaptic, and multiplexed responses that grew in complexity as the distance between the stimulation source and recorded neuron increased. These representations collapsed into patterns with lower information content at higher stimulation frequencies. With internodal activity propagation, a hierarchy of network pathways, including latent circuits, was revealed using glutamatergic blockade. As stimulus channels were added, divergent, non-linear effects were observed in local versus distant network layers. Pairwise difference analysis of neuronal responses suggested that neuronal ensembles generally outperformed individual cells in discriminating input patterns. Significance. Our data illuminate the complexity of spiking activity propagation in cortical networks in vitro, which is characterized by the transformation of an input into myriad outputs over several network layers. These results provide insight into how the brain potentially processes information and generates the neural code and could guide the development of clinical therapies based on multichannel brain stimulation.

Seeking the -ational in derivational morphology

ABSTRACTBackground: Derived words have constituted an important bridge between aphasiology and psycholinguistics by addressing the extent to which morphology affects representation and processing in the mind/brain.Aims: Our goal was to assess how properties of whole words and their overlapping substrings affect the manner in which English derivationally suffixed words are recognised and produced.Methods & Procedures: We probed the processing of multimorphemic words containing strings of two derivational suffixes by healthy adult participants, employing both a progressive demasking naming task and a typing task. The progressive demasking paradigm that we employed integrates word recognition and production by requiring that a participant recognises a progressively demasked stimulus and then say it as quickly as possible. The typing task allowed us to focus on segment by segment aspects of processing during production by enabling us to construct per-letter typing times for each region of a word.Outcomes & Results: We found converging influences across the tasks that correspond to the properties of roots, stems, suffixes and suffix strings.Conclusions: We interpret these effects to accord with a perspective that emphasises the dynamic nature of the lexical processing system and massive interconnectivity within it. This has implications for unimpaired processing and for the word processing challenges experienced by people with aphasia.

Planning Massive Interconnects in 3-D Chips

3-D chips rely on massive interconnect structures, i.e., large groups of through-silicon vias coalesced with large multibit buses. We observe that wirelength optimization, a classical technique for floorplanning, is not effective while planning massive interconnects. This is due to the interconnects’ strong impact on multiple design criteria like wirelength, routability, and temperature. To facilitate early design progress of massively-interconnected 3-D chips, we propose a novel 3-D-floorplanning methodology which accounts for different types of interconnects in a unified manner. One key idea is to align cores/blocks simultaneously within and across dies, thus increasing the likelihood of successfully implementing complex and massive interconnects. While planning such interconnects, we also target fast, yet accurate, thermal management, routability, and fixed-outline floorplanning. Experimental results on Gigascale Systems Research Center and IBM-HB+ circuits demonstrate our tool’s capabilities for both planning massive 3-D interconnects and for multiobjective 3-D floorplanning in general.

Jade Implementation of Power Restoration in Automated Distributed System

Massive interconnection of power network has faced more operational difficulties. The multi agent based intelligent system has used to reduce the complexity, automated power distributions in power system. In this research work Jade implementation technique has used to isolate the fault line and has restored the nearest generator. The system model has developed in MATLAB and tested in JADE. It provides quick restoration without power loss will lead the power consumption in distributed automation power system.

A Sub-GHz Multi-ISM-Band ZigBee Receiver Using Function-Reuse and Gain-Boosted N-Path Techniques for IoT Applications

Internet of Things (IoT) represents a competitive and large market for short-range ultra-low-power (ULP) wireless connectivity [1, 2]. According to [3], by 2020 the IoT market will be close to hundreds of billion dollars (annually ~16 billions). To bring down the hardware cost of such massive inter-connections, sub-GHz ULP wireless products compliant with the existing wireless standard such as the IEEE 802.15.4c/d (ZigBee) will be of great demand, especially for those that can cover all regional ISM bands [e.g., China (433 MHz), Europe (860 MHz), North America (915 MHz) and Japan (960 MHz)]. Together with the obvious goals of small chip area, minimum external components and ultra-low-voltage (ULV) supply (for possible energy harvesting), the design of such a receiver poses significant challenges.

Definition, evolution, applications and barriers for deployment of microgrids in the energy sector

The Microgrid concept is becoming a reality since its careful design and operation could solve a variety of challenges. The massive interconnection and operation of decentralized energy resources near the loads, electric loss reductions, customized quality of electrical supply to clients, energy efficiency increase and greenhouse gas emissions reduction when electric and thermal loads are supplied by high efficiency generation units (CHP, Combined Heat and Power, or CHCP, Combined Heat, Cool and Power) or renewable energies are their major advantages. Microgrids are a common research topic in USA, Canada, Japan and Europe, where several pilot installations have been specified, designed, built and operated with some other demonstration and research facilities being planned. This paper gives an overview of definitions and evolution of the microgrid concept, the different approaches to microgrids in the three continents, specific problems solved by real microgrids and, as an example, it also analyzes the major barriers actually encountered in the Spanish legislation for their wide deployment. Finally, the role of the Energy Department of LABEIN-Tecnalia in the field of microgrids is described.

Freeman olfactory cortex model: A multiplexed KII network implementation

This paper presents the results of a CMOS-VLSI implementation of a realistic computational model proposed by Walter Freeman for the olfactory system. This model, in later years, has been studied for engineering applications such as auto-association and classification. The analogue nature of the model motivates analogue VLSI implementations. However, the dimension and complexity of such system poses many obstacles to an analogue electronic implementation; one such is the massive interconnectivity which size increases with the square of the number of inputs (channels). We suggest a multiplexing procedure that puts the burden of interconnectivity over a digital system that is simpler to design and makes the analogue system more treatable. The procedure naturally samples the signals. To avoid smoothing filters, a discrete-time solution was also employed. Although with such approach the time resolution is reduced, the advantages overcome the detriments. Previous work has shown that the model can be efficiently discretized using DSP techniques, resulting on a system that is able to predict, on sample-by-sample basis, the behaviour of the VLSI circuit, allowing for a simple and flexible way to adjust the circuit parameters. We present the measured circuit results that are further confronted with the digital implementation.

Flip chip electrical interconnection by selective electroplating and bonding

This work presents a parallel electrical interconnection process by means of flip-chip, selective electroplating and bonding. The electrical interconnection lines are built on a glass substrate made of 500/2000 Å of Cr/Au with 3150 μm in length and 10 μm in width. Two silicon chips are processed as the device chips to be electrically interconnected. It has been demonstrated that 98 out of 102 interconnects are established in parallel with a successful rate of 96% and the average resistance of the electroplating bond is 12 Ω. This process has potential applications in replacing the conventional, serial wire bonding or tape automated bonding (TAB) process for massive interconnection requirements in IC or MEMS devices. Reliability test is also performed by putting the interconnects into boiling liquid nitrogen (−195 °C) repeatedly. It is found that 100% of the interconnects survive after 2 cycles of the quenching process.

Is molybdenum necessary for the growth of single-wall carbon nanotubes from CO?

Catalytic mixtures of molybdenum and cobalt have been considered essential for the growth of single-wall carbon nanotubes (SWCNT) from carbon monoxide by the chemical vapor deposition method. Here we demonstrated extensive growth of SWCNT with only cobalt as catalyst. The typically subsequent stage of annealing the catalyst in the presence of hydrogen was also eliminated. This process resulted in a near-100% yield of essentially defect-free chiral tubes. Use of electrochemical deposition of the catalyst seed particles may open the door for the production of massive electronic interconnect assemblies of these nanotubes.

Issues of nanoelectronics: a possible roadmap.

In this review, we will discuss a possible roadmap in scaling a nanoelectronic device from today's CMOS technology to the ultimate limit when the device fails. In other words, at the limit, CMOS will have a severe short channel effect, significant power dissipation in its quiescent (standby) state, and problems related to other essential characteristics. Efforts to use structures such as the double gate, vertical surround gate, and SOI to improve the gate control have continually been made. Other types of structures using SiGe source/drain, asymmetric Schottky source/drain, and the like will be investigated as viable structures to achieve ultimate CMOS. In reaching its scaling limit, tunneling will be an issue for CMOS. The tunneling current through the gate oxide and between the source and drain will limit the device operation. When tunneling becomes significant, circuits may incorporate tunneling devices with CMOS to further increase the functionality per device count. We will discuss both the top-down and bottom-up approaches in attaining the nanometer scale and eventually the atomic scale. Self-assembly is used as a bottom-up approach. The state of the art is reviewed, and the challenges of the multiple-step processing in using the self-assembly approach are outlined. Another facet of the scaling trend is to decrease the number of electrons in devices, ultimately leading to single electrons. If the size of a single-electron device is scaled in such a way that the Coulomb self-energy is higher than the thermal energy (at room temperature), a single-electron device will be able to operate at room temperature. In principle, the speed of the device will be fast as long as the capacitance of the load is also scaled accordingly. The single-electron device will have a small drive current, and thus the load capacitance, including those of interconnects and fanouts, must be small to achieve a reasonable speed. However, because the increase in the density (and/or functionality) of integrated circuits is the principal driver, the wiring or interconnects will increase and become the bottleneck for the design of future high-density and high-functionality circuits, particularly for single-electron devices. Furthermore, the massive interconnects needed in the architecture used today will result in an increase in load capacitance. Thus for single-electron device circuits, it is critical to have minimal interconnect loads. And new types of architectures with minimal numbers of global interconnects will be needed. Cellular automata, which need only nearest-neighbor interconnects, are discussed as a plausible example. Other architectures such as neural networks are also possible. Examples of signal processing using cellular automata are discussed. Quantum computing and information processing are based on quantum mechanical descriptions of individual particles correlated among each other. A quantum bit or qubit is described as a linear superposition of the wave functions of a two-state system, for example, the spin of a particle. With the interaction of two qubits, they are connected in a "wireless fashion" using wave functions via quantum mechanical interaction, referred to as entanglement. The interconnection by the nonlocality of wave functions affords a massive parallel nature for computing or so-called quantum parallelism. We will describe the potential and solid-state implementations of quantum computing and information, using electron spin and/or nuclear spin in Si and Ge. Group IV elements have a long coherent time and other advantages. The example of using SiGe for g factor engineering will be described.

Sensors in Intelligent Buildings

We've all heard of `sick-building' syndrome and the misery this can inflict in the workplace in terms of poor health and lost production. The notion of the Intelligent Building is the modern civil engineer's Big Idea in tackling these and other such deficiencies. The intelligent building can adapt itself to maintain an optimized environment. This ability relies on sensors as a front-line technology - the subject of this volume. Gassmann and Meixner tackle the subject matter by using five categories of intelligent building technology: energy and HVAC (heating, ventilation and air conditioning), information and transportation, safety and security, maintenance, and facility management. These categories of home and building technology are intended to encompass domestic as well as workplace and public environments, but as the introduction states, the breakthrough into the domestic market is not quite here yet. They have targeted, successfully in this reviewer's opinion, the researcher and designer in the field who has his or her own specific interest in sensor issues. Each section of the book contains a number of articles contributed from predominantly European, particularly German and Swiss, research institutions. They cover subjects as diverse as inflatable buildings and a combination of a sensing chair and sensing floor that allows the exact interaction of a human user with his/her immediate surroundings to be integrated within an information system. A fascinating item on biometric access security systems brings to life the `spy-thriller' world of automatic iris and fingerprint scanners as a means of secure access to key-less buildings. The discussion extends to threats to such systems whereby `replay' attacks, for example presenting a face recognition system with a photograph of an individual, reveal the necessary further steps if the technology is to be considered safe. Inevitably though, it is the massive strides in communications and processing technology that are visible in the contributions to this volume that point to where most of the advances will come. The massive interconnection issue of numerous sensors, the `nervous system' of the building is approached in items covering the complexities of fieldbus systems and the use of `wireless' in-building networks. This well presented volume with good quality illustration ends in an investigation of system technologies for the automated home of the near future. Presence-controlled heating and air-conditioning, telemonitoring of the sick or elderly, gas and fire detection and of course a high data rate communication backbone to link everything: all these could feature in the future household. We shall see. Peter Foote

Governmentality and the Sociology of Education: Media, educational policy and the politics of resentment

This paper argues that theorizations of the state which are sensitive to both its durability and its permeability, and theorizations which can account for the massive interconnections between local and global forces as well as different material and discursive sites are missing from contemporary work in the sociology of education. Drawing on Foucault's notion of 'governmentality' as a key resource for addressing this impasse, the authors highlight the constant fabrication of racial identity through the production of the pure space of racial origins or 'resentment'—the process of defining one's identity through the negation of the other. This dynamic, the article maintains, now informs key discourses both in popular culture and education. The authors conclude that these processes operate in tandem in the prosecution of the politics of racial exclusion in our times, informing key policy debates, including those around affirmative action and bilingual education.

Opto-Electronic Block-Cipher Based on Iteration of the 2-D Toggle Cellular Automata: Algorithm

Parallel architectures and algorithms may offer a solution to the system bottleneck arising from the need to encrypt a very large amount of data without compromising security. In this respect the use of cellular automata with their parallel, simple, regular and modular structure is very promising. We extend the blockcipher algorithm, based on the iterations of so called “toggle” cellular automata rules to two dimensions.The advantages are higher complexity of the crypt-analytical attacks and substantial increase in the speed of the algorithm. Due to its massive parallelism and interconnectivity, the algorithm is very suitable for opto-electronic implementation.

Dinosaurs and white elephants: the science center in the twenty-first century

This paper argues that science centers are expensive to create as capital projects, expensive to maintain with professional staff, and, given the high costs of exhibit development, expensive to change. Lacking a fixed collection of unique artifacts with which to attract visitors, the science center is at risk when it cannot change quickly enough to meet the demands of its users. In the past, temporary exhibitions have been used as a means of creating more frequent change. Now, however, given the exponential increase of the availability of new electronic media, coupled with their massive interconnection via the Internet, informal learning can be had at home and in other sites, rendering the science center unwieldy, expensive, irrelevant, and obsolete. Threats to the science center cannot be lightly shrugged off, and a real transformation of the institution is required. The paper concludes that the science center, as an institution and as a building project, is doomed to extinction as a consequence of two factors—ecology and economy. It argues for the need to develop a new kind of institution of informal learning in its place.

Implementation of Art1 and Art2 Artificial Neural Networks on Ring and Mesh Architectures

The Artificial Neural Networks (ANNs) are being used to solve a variety of problems in pattern recognition, robotic control, VLSI CAD and other areas. In most of these applications, a speedy response from the ANNs is imperative. However, ANNs comprise a large number of artificial neurons, and a massive interconnection network among them. Hence, implementation of these ANNs involves execution of computer-intensive operations. The usage of multiprocessor systems therefore becomes necessary. In this article, we have presented the implementation of ART1 and ART2 ANNs on ring and mesh architectures. The overall system design and implementation aspects are presented. The performance of the algorithm on ring, 2-dimensional mesh and n-dimensional mesh topologies is presented. The parallel algorithm presented for implementation of ART1 is not specific to any particular architecture. The parallel algorithm for ARTE is more suitable for a ring architecture.

Neural networks' power

Neural networks (NNs) are effective systems for learning pattern discriminants from a body of examples. Artificial neural networks (ANNs) have been developed in a wide variety of configurations with some common underlying characteristics. All ANNs attempt to achieve good performance via massive interconnection of simple computational elements. Neural networks are characterized by the model of their neurons, the connections among them and the methods used to train them to do specific tasks. The author describes multi-layer neural networks and Kohonen neural networks. The author then discusses how they are used in electric load forecasting and power system security assessment.

Compact holographic optical neural network system for real‐time pattern recognition

One of the important characteristics of artificial neural networks is their capability for massive interconnection and parallel processing. Recently, specialized electronic neural network processors and VLSI neural chips have been introduced in the commercial market. The number of parallel channels they can handle is limited because of the limited parallel interconnections that can be implemented with onedimensional electronic wires. High-resolution pattern recognition problems can require a large number of neurons for parallel processing of an image. This paper describes a holographic optical neural network (HONN) that is based on high-resolution volume holographic materials and is capable of performing massive 3-D parallel interconnection of tens of thousands of neurons. A HONN with more than 16,000 neurons packaged in an attache case has been developed. Rotation-shift-scaleinvariant pattern recognition operations have been demonstrated with this system. System parameters such as the signal-to-noise ratio, dynamic range, and processing speed are discussed.

Collective computational activity in self-assembled arrays of quantum dots: a novel neuromorphic architecture for nanoelectronics

We describe a new class of nanoelectronic circuits which exploits the charging behavior in resistively/capacitively linked arrays of nanometer-sized metallic islands (quantum dots), self-assembled on a resonant tunneling diode, to perform neuromorphic computation. These circuits produce associative memory effects and realize the additive short-term memory (STM) or content addressable memory (CAM) models of neural networks without requiring either large-area/high-power operational amplifiers, or massive interconnectivity between devices. Both these requirements had seriously hindered the application of neural networks in the past. Additionally, the circuits can solve NP-complete optimization problems (such as the traveling salesman problem) using single electron charge dynamics, exhibit rudimentary image-processing capability, and operate at room temperature unlike most quantum devices. Two-dimensional (2D) processors, with a 100/spl times/100 pixel capacity, can be fabricated in an area of 10/sup -8/ cm/sup 2/ leading to unprecedented functional density. Possible routes to synthesizing these circuits, employing self-assembly, are also discussed.

Mapping of artificial neural networks onto message passing systems

Various Artificial Neural Networks (ANNs) have been proposed in recent years to mimic the human brain in solving problems involving human-like intelligence. Efficient mapping of ANNs comprising of large number of neurons onto various distributed MIMD architectures is discussed in this paper. The massive interconnection among neurons demands a communication efficient architecture. Issues related to the suitability of MIMD architectures for simulating neural networks are discussed. Performance analysis of ring, torus, binary tree, hypercube, and extended hypercube for simulating artificial neural networks is presented. Our studies reveal that the performance of the extended hypercube is better than those of ring, torus, binary tree, and hypercube topologies.

Optimization of groundwater remediation using artificial neural networks with parallel solute transport modeling

A new approach to nonlinear groundwater management methodology is presented which optimizes aquifer remediation with the aid of artificial neural networks (ANNs). The methodology allows solute transport simulations, usually the main computational component of management models, to be run in parallel. The ANN technology, inspired by neurobiological theories of massive interconnection and parallelism, has been successfully applied to a variety of optimization problems. In this new approach, optimal management solutions are found by (1) first training an ANN to predict the outcome of the flow and transport code, and (2) then using the trained ANN to search through many pumping realizations to find an optimal one for successful remediation. The behavior of complex groundwater scenarios with spatially variable transport parameters and multiple contaminant plumes is simulated with a two‐dimensional hybrid finite‐difference/finite‐element flow and transport code. The flow and transport code develops the set of examples upon which the network is trained. The input of the ANN characterizes the different realizations of pumping, with each input indicating the pumping level of a well. The output is capable of characterizing the objectives and constraints of the optimization, such as attainment of regulatory goals, value of cost functions and cleanup time, and mass of contaminant removal. The supervised learning algorithm of back propagation was used to train the network. The conjugate gradient method and weight elimination procedures are used to speed convergence and improve performance, respectively. Once trained, the ANN begins a search through various realizations of pumping patterns to determine whether or not they will be successful. The search is directed by a simple genetic algorithm. The resulting management solutions are consistent with those resulting from a more conventional optimization technique, which combines solute transport modeling and nonlinear programming with a quasi‐Newton search. The results suggest that the ANN approach has the following advantages over the conventional technique for the test remediations: more independence of the flow and transport code from the optimization, greater influence of hydrogeologic insight, and less computational burden due to the potential for parallel processing of the flow and transport simulations and the ability to “recycle” these simulations. The ANN performance was observed upon variation of the problem formulation, network architecture, and learning algorithm.