Applied Mechanics and Materials Vols. 135-136

Abstract: Capacity sharing is efficient for private L2 caches to utilize cache resources in chip multiprocessors. We propose a probabilistic sharing mechanism using reuse replacement strategy. This mechanism adopts decoupled tag and data arrays, and partitions the data arrays into private and shared regions. Probability is introduced to control the capability of each core to compete shared data resources. We assign high probabilities to cores with stress memory demands and dynamically adjust these probabilities corresponding to the monitored run-time memory demands. Simulation results of PARSEC benchmarks show that our mechanism exceeds conventional LRU managed private cache. Compared with reuse replacement managed private cache without sharing among cores, our mechanism achieves an average L2 miss rate reduction of 8.70%.

Abstract: The current self-organizing feature map (SOFM) neural network algorithm used for image compression, of which a large amount of network training time and the blocking effect in the reconstructed image existed in codebook design vector calculation. Based on the above issue, this paper proposed an improved SOFM. The new SOFM introduced normalized distance between the sum of input vectors and the sum of the codeword vectors as a constraint in the process of searching for the winning neuron, which can remove redundant Euclidean distance calculation in the competitive process. Furthermore, this paper has done image compression by combining wavelet transform with the improved SOFM (WT & improved SOFM). The method firstly conducted wavelet decomposition for the image, retained low-frequency sub-band, then put the high-frequency sub-band into improved SOFM network, and achieved the purpose of compression. Experimental results showed that this algorithm can greatly reduce the network training time and enhance the learning efficiency of neural network, while effectively improve the PSNR (increased 0.6dB) of reconstructed.

Abstract: Instability is the inherent limitation of downward continuation. Based on integral iteration idea and regularization theory, a regularization method is proposed for downward continuation of potential fields. We also present an L-curve method for the selection of regularization parameter. The improved method is tested on model data both with and without noise. The results show that the new method is much more stable than the iteration method, especially in dealing with low signal-to-noise data, and that it also has a large downward continuation distance.

Abstract: A reverse driving control method for one type of hydrostatic mobile machinery is devised in this paper. Based on the analysis of dynamics of stopping process, the relationship between engine speed and pump displacement is established. To restrict the peak value of engine overspeed within the region of the allowable value, an offline parameter determination approach complied to the requirements of average stopping deceleration is developed. The online control method using the selected parameter is presented. A flow diagram is given to illustrate the control method.

Abstract: This paper, based on the analysis of the innate properties of productive activities of logistics service industry, constructs input and output index based on SUPER-SBM DEA model. The paper empirically selects provincial data of China in 2008, evaluates the score of productive efficiency of logistics service industry in various regions and then explores the approaches to promote productive efficiency according to related slack variables analysis. According to the analysis, the overall efficiency in logistics service industry in China is comparatively low and there is a remarkable feature in its regional distribution. The efficiency in eastern regions is comparatively higher with the western regions following it and that in central regions is the worst. There exists larger input redundancy in the process of developing the logistics service industry. Thus, regional coordination of the logistics service industry should be emphasized and basic infrastructure investment should also be strengthened. Personnel quality should be improved and meanwhile redundancy should be reduced. Marketization degree of logistics service industry should be improved and capital-intensive industry of logistics service industry should be upgraded.

Abstract: A time-programmable soft-start circuit with hiccup control for switching regulators is presented. By using a novel soft-start voltage generating circuit, the soft-start time can be set internally and externally by one pin which enhances the flexibility for applications. A hiccup control has also been integrated into the circuit to deal with the problems of current inrush and voltage overshot in short circuit case which can effectively improves the system reliability. The proposed soft-start circuit has been fabricated with UMC 0.35μm CDMOS process successfully, and the detailed experimental results are shown at the end of the article.

Abstract: with the rapid development of information technology, the control mechanism of traffic is continuous improved towards automatic and intelligent. Traditional pre-setting control method based on data statistical can not meet the complex needs of the current traffic control. Aiming at this issue, we propose traffic control mechanism based on fuzzy logic. The technology real-time gets the situation of traffic according to a series of fuzzy inference rules that is pre-setting to achieve optimal control of the traffic situation. Experiments show that this method not only can improve the passage efficiency of intersection, but also provide more predicted result as well compared to the network-based control methods.

Abstract: This paper presents a word sense disambiguation method by reconstructing the context using the correlation between words. Firstly, we figure out the relevance between words though the statistical quantity(co-occurrence frequency , the average distance and the information entropy) from the corpus. Secondly, we see the words that have lager correlation value between ambiguous word than other words in the context as the important words, and use this kind of words to reconstruct the context, then we use the reconstructed context as the new context of the ambiguous words .In the end, we use the method of the sememe co-occurrence data[10] for word sense disambiguation. The experimental results have proved the feasibility of this method.

Abstract: Inspired by Accumulate-Repeat-Accumulate (ARA) codes, in this letter we propose a new channel coding scheme called Accumulate-Repeat-Delay-Partial-Delay (ARDPD) codes. Contrary to ARA codes, ARDPD codes simultaneously achieve good threshold and linear minimum distance. Based on EXIT for LDPC codes, we show that for maximum variable node degree 5, the threshold of ARDPD codes outperforms the best known unstructured irregular LDPC codes. Iterative decoding simulation results show that the proposed codes perform as predicted.

Abstract: As a powerful branch of error-control coding, Turbo codes have been paid a great attention since they are proposed. Due to the powerful error correcting capability, Turbo codes are very attractive for mobile wireless systems to combat channel fading. However, there are many kinds of decoding algorithms for Turbo codes, any of them have merits and demerits. In this paper, we present a new method to decode Turbo codes over cdma2000 mobile system, that is introducing a controller to achieve the purposer of adaptive decoding. And this method make the system more reliable.