Optimization of silicon solar cell fabrication based on neural network and genetic programming modeling

This study describes techniques for the cascade modeling and the optimization that are required to conduct the simulator-based process optimization of solar cell fabrication. Two modeling approaches, neural networks and genetic programming, are employed to model the crucial relation for the consecutively connected two processes in solar cell fabrication. One model (Model 1) is used to map the five inputs (time, amount of nitrogen and DI water in surface texturing and temperature and time in emitter diffusion) to the two outputs (reflectance and sheet resistance) of the first process. The other model (Model 2) is used to connect the two inputs (reflectance and sheet resistance) to the one output (efficiency) of the second process. After modeling of the two processes, genetic algorithms and particle swarm optimization were applied to search for the optimal recipe. In the first optimization stage, we searched for the optimal reflectance and sheet resistance that can provide the best efficiency in the fabrication process. The optimized reflectance and sheet resistance found by the particle swarm optimization were better than those found by the genetic algorithm. In the second optimization stage, the five input parameters were searched by using the reflectance and sheet resistance values obtained in the first stage. The found five variables such as the texturing time, amount of nitrogen, DI water, diffusion time, and temperature are used as a recipe for the solar cell fabrication. The amount of nitrogen, DI water, and diffusion time in the optimized recipes showed considerable differences according to the modeling approaches. More importantly, repeated applications of particle swarm optimization yielded process conditions with smaller variations, implying greater consistency in recipe generation.     

A dispatching method for automated lifting vehicles in automated port container terminals

In automated container terminals, containers are transported from the marshalling yard to a ship and vice versa by automated vehicles. The automated vehicle type studied in this paper is an automated lifting vehicle (ALV) that is capable of lifting a container from the ground by itself. This study discusses how to dispatch ALVs by utilizing information about pickup and delivery locations and time in future delivery tasks. A mixed-integer programming model is provided for assigning optimal delivery tasks to ALVs. A procedure for converting buffer constraints into time window constraints and a heuristic algorithm for overcoming the excessive computational time required for solving the mathematical model are suggested. Numerical experiments are reported to compare the objective values and computational times by a heuristic algorithm with those by an optimizing method and to analyze the effects of dual cycle operation, number of ALVs, and buffer capacity on the performance of ALVs.     

A fuzzy-based interleaved multi-hop authentication scheme in wireless sensor networks

In sensor networks, a compromised node can either generate fabricated reports with false votes or inject false votes into real reports, which causes severe damage such as false alarms, energy drain and information loss. An interleaved hop-by-hop authentication (IHA) scheme addresses the former attack by detecting and filtering false reports in a deterministic and hop-by-hop fashion. Unfortunately, in IHA, all en-route nodes must join to verify reports while only a few are necessary to the authentication procedure. In this paper, we propose a fuzzy-based interleaved multi-hop authentication scheme based on IHA. In our scheme, the fuzzy logic system only selects some nodes for verification based on the network characteristics. Moreover, we apply a voting method and a hash-based key assignment mechanism to improve network security. Through performance evaluation, the proposed scheme is found to save up to 13% of the energy consumption and to provide more network protection compared to IHA.     

New receipt-free voting scheme using double-trapdoor commitment

It is considered to be the most suitable solution for large scale elections to design an electronic voting scheme using blind signatures and anonymous channels. Based on this framework, Okamoto first proposed a receipt-free voting scheme [30] for large scale elections. However, in the following paper, Okamoto [31] proved that the scheme [30] was not receipt-free and presented two improved schemes. One scheme requires the help of the parameter registration committee and the other needs a stronger physical assumption of the voting booth. In this paper, we utilize the double-trapdoor commitment to propose a new receipt-free voting scheme based on blind signatures for large scale elections. Neither the parameter registration committee nor the voting booth is required in our scheme. We also present a more efficient zero-knowledge proof for secret permutation. Therefore, our scheme is much more efficient than Okamoto’s schemes [30] and [31] with the weaker physical assumptions. Moreover, we prove that our scheme can achieve the desired security properties.     

Testbed results of an opportunistic routing for multi-robot wireless networks

Opportunistic routing is a candidate for multihop wireless routing where the network topology and radio channels vary rapidly. However, there are not many opportunistic routing algorithms that can be implemented in a real multihop wireless network while exploiting the node mobility. It motivates us to implement an opportunistic routing, random basketball routing (BR), in a real multi-robot network to see if it can enhance the capacity of the multihop network as mobility increases. For implementation purposes, we added some features, such as destination RSSI measuring, a loop-free procedure and distributed relay probability updating, to the original BR. We carried out the experiments on a real multi-robot network and compared BR with AODV combined with CSMA/CA (routing + MAC protocol). We considered both static and dynamic scenarios. Our experiments are encouraging in that BR outperforms AODV + CSMA/CA, particularly in dynamic cases; the throughput of BR is 6.6 times higher than that of AODV + CSMA/CA. BR with dynamic networks shows 1.4 times higher throughput performance than BR with static networks. We investigate the performance of BR in the large-scale network using NS-2 simulation. We verify the effect of node density, speed, destination beacon signal and loop-free procedure. According to the large-scale simulation, the end-to-end throughput grows with the node speed.     

Quasi-rocking real-time pushdown automata

A pushdown automaton (PDA) is quasi-rocking if it preserves the stack height for no more than a bounded number of consecutive moves. Every PDA can be transformed into an equivalent one that is quasi-rocking and real-time and every finite-turn (one-turn) PDA can be transformed into an equivalent one that is quasi-rocking or real-time. The quasi-rocking [quasi-rocking in the increasing mode, and quasi-rocking in the decreasing mode] real-time restriction in finite-turn (one-turn) PDAs coincides with the double Greibach [reverse Greibach, and Greibach] form in nonterminal-bounded (linear) context-free grammars. This provides complete grammatical characterizations of quasi-rocking and/or real-time (finite-turn and one-turn) PDAs and, together with known relations and other relations proved in the present paper, yields an extended hierarchy of PDA languages. Basic decision properties for PDAs can be stated in stronger forms by using the quasi-rocking and real-time restrictions and their undecidability/decidability status rests on the way PDAs quasi-rock.     

On the use of energy minimization for CA based analysis in elasticity

There has been recent interest in exploring alternative computational models for structural analysis that are better suited for a design environment requiring repetitive analysis. The need for such models is brought about by significant increases in computer processing speeds, realized primarily through parallel processing. To take full advantage of such parallel machines, however, the computational approach itself must be revisited from a totally different perspective; parallelization of inherently serial paradigms is subject to limitations introduced by a requirement of information coordination. The cellular automata (CA) model of decentralized computations provides one such approach which is ideally tailored for parallel computers. The proposed paper examines the applicability of the cellular automata model in problems of 2-D elasticity. The focus of the paper is in the use of a genetic algorithm based optimization process to derive the rules for local interaction required in evolving the cellular automata. Received August 28, 2000     

An efficient state space generation for the analysis of real-timesystems

State explosion is a well-known problem that impedes analysis and testing based on state-space exploration. This problem is particularly serious in real time systems because unbounded time values cause the state space to be infinite even for simple systems. The author presents an algorithm that produces a compact representation of the reachable state space of a real time system. The algorithm yields a small state space, but still retains enough information for analysis. To avoid the state explosion which can be caused by simply adding time values to states, our algorithm uses history equivalence and transition bisimulation to collapse states into equivalent classes. Through history equivalence, states are merged into an equivalence class with the same untimed executions up to the states. Using transition bisimulation, the states that have the same future behaviors are further collapsed. The resultant state space is finite and can be used to analyze real time properties. To show the effectiveness of our algorithm, we have implemented the algorithm and have analyzed several example applications     

Data cache and direct memory access in programming mediaprocessors

Mediaprocessors provide high performance by using both instruction- and data-level parallelism. Because of the increased computing power, transferring data between off- and on-chip memories without slowing down the core processor's performance is challenging. Two methods, data cache and direct memory access, address this problem in different ways