203 results found.
Year: 2013, Volume: 13, Number: 1
... Molva, "Privacy Preserving TABLE I. FUNCTIONALITY COMPARISON WITH [1 ], [4] -[ 6 ], [9]. Products Tracking in Clustered Supply Chain", Proc. Functionality Ours [1] [5] [6] [9] [4] SENSORCOMM'11, pp. 399-404, ...
Terms matched: 1 - Score: 3 - 629k
Year: 2013, Volume: 13, Number: 1
... model by starting from whole process, a parallel architecture can be easily a single MIMO NN is frequently adopted in literature, the implemented since each node of the cluster is used to train proposed method introduces three other steps: 1) a further SISO NNs separately (see [17]). The optimization of decomposition ...
Terms matched: 1 - Score: 8 - 689k
Year: 2012, Volume: 12, Number: 4
... interconnected neurons percent of the images were used as the training sample for is a mode of network connection with the environment. In each type of images. This clusterization can lead to the this context we may speak about an environment model- situation that classes formed, to which input patterns fall, related training paradigm. ...
Terms matched: 1 - Score: 19 - 701k
Year: 2012, Volume: 12, Number: 4
... [25] O., Younis, S., Fahmy, "Distributed Clustering in Ad-Hoc Sensor 10.1109/MWC.2008.4599225. Networks: A Hybrid, Energy-Efficient Approach ... optimized energy RTC/TCXO/Crystal, 2010. efficiency and minimum delay in clustered wireless sensor networks", [38] ** *, 8-bit Atmel ...
Terms matched: 2 - Score: 5 - 529k
Year: 2012, Volume: 12, Number: 3
... shape hypothesis is similarity measurement between points is defined to cluster prohibitively large. 3D HT is time-consuming and high ... work [22] was used to III. SIMILARITY-BASED CLUSTERING segment the trees in the scene, which is ... on k nearest neighboring clustering. The raw point cloud data P is Inspired by the graph-based image segmentation [ ...
Terms matched: 7 - Score: 550 - 838k
Year: 2012, Volume: 12, Number: 3
... neural networks KNN with a accuracy rate of compared the cluster position with the statistical parameters 65% (normal ... methods is more robust, accurate and noise-resistant. The clustering algorithms divide the input sets based on The formants ... number unsupervised and does not know the real number of clusters M=3 clusters and C=2 specified ...
Terms matched: 4 - Score: 109 - 855k
Year: 2012, Volume: 12, Number: 3
... Volume 12, Number 3, 2012 The main block, in contrast to coherent demodulator, is profile. Therefore, the channel is composed by "L" clusters, the squaring device, which removes any phase/polarity [7]. either having its propagation parameters as the gain and the It is followed by ...
Terms matched: 1 - Score: 3 - 692k
Year: 2012, Volume: 12, Number: 2
... 1.2], for example, representing a+/- 20% error. The eigenvalues of (15) will generally be in clouds in the complex plane clustered around the desired positions.  0 I   ï€K ï€ D ï€ BQB T ï€ BQΔBT  (15)   Figure 15 shows the clouds ...
Terms matched: 1 - Score: 3 - 1,177k
Year: 2012, Volume: 12, Number: 2
... 89/1.5/S /57649, Project ID 57649 (PERFORM-ERA), co- financed by the European Social Fund – Investing in People, within the clustering type. [21]. This is able to determine the car Sectoral Operational Programme Human Resources Development 2007- number by analyzing the images provided by a ...
Terms matched: 1 - Score: 3 - 769k
Year: 2012, Volume: 12, Number: 2
... end technical architecture, Celtic [6] D. G. Andersen, A. C. Snoeren, and H. Balakrishnan, "Best-path Project – Eureka Cluster CP5-013 – MARCH, Review December versus multi-path overlay routing," in Proc. 3rd ACM SIGCOMM 2010. Available: http://projects.celtic- Conf ...
Terms matched: 1 - Score: 5 - 634k