Abstract: In this paper, we study the factors which determine the capacity of a Convolutional Neural Network (CNN) model and propose the ways to evaluate and adjust the capacity of a CNN model for best matching to a specific pattern recognition task. Firstly, a scheme is proposed to adjust the number of independent functional units within a CNN model to make it be better fitted to a task. Secondly, the number of independent functional units in the capsule network is adjusted to fit it to the training dataset. Thirdly, a method based on Bayesian GAN is proposed to enrich the variances in the current dataset to increase its complexity. Experimental results on the PASCAL VOC 2010 Person Part dataset and the MNIST dataset show that, in both conventional CNN models and capsule networks, the number of independent functional units is an important factor that determines the capacity of a network model. By adjusting the number of functional units, the capacity of a model can better match the complexity of a dataset.
Abstract: Cooperative spectrum sensing is a crucial challenge in
cognitive radio networks. Cooperative sensing can increase the
reliability of spectrum hole detection, optimize sensing time and
reduce delay in cooperative networks. In this paper, an efficient
central capacity optimization algorithm is proposed to minimize
cooperative sensing time in a homogenous sensor network using OR
decision rule subject to the detection and false alarm probabilities
constraints. The evaluation results reveal significant improvement in
the sensing time and normalized capacity of the cognitive sensors.
Abstract: Modeling of a manufacturing system enables one to
identify the effects of key design parameters on the system performance and as a result to make correct decision. This paper
proposes a manufacturing system modeling approach using a spreadsheet model based on queuing network theory, in which a
static capacity planning model and stochastic queuing model are integrated. The model was used to improve the existing system utilization in relation to product design. The model incorporates few
parameters such as utilization, cycle time, throughput, and batch size.
The study also showed that the validity of developed model is good enough to apply and the maximum value of relative error is 10%, far
below the limit value 32%. Therefore, the model developed in this
study is a valuable alternative model in evaluating a manufacturing system
Abstract: The dynamic spectrum allocation solutions such as
cognitive radio networks have been proposed as a key technology to
exploit the frequency segments that are spectrally underutilized.
Cognitive radio users work as secondary users who need to
constantly and rapidly sense the presence of primary users or
licensees to utilize their frequency bands if they are inactive. Short
sensing cycles should be run by the secondary users to achieve
higher throughput rates as well as to provide low level of interference
to the primary users by immediately vacating their channels once
they have been detected. In this paper, the throughput-sensing time
relationship in local and cooperative spectrum sensing has been
investigated under two distinct scenarios, namely, constant primary
user protection (CPUP) and constant secondary user spectrum
usability (CSUSU) scenarios. The simulation results show that the
design of sensing slot duration is very critical and depends on the
number of cooperating users under CPUP scenario whereas under
CSUSU, cooperating more users has no effect if the sensing time
used exceeds 5% of the total frame duration.