In this paper, we explore the effect of using different convolutional layers, batch normalization and the global average pooling layer upon a convolutional neural network (CNN) based gaze tracking system. A novel method is proposed to label the participant’s face images as gaze points retrieved from eye tracker while watching videos for building a training dataset that is closer to human visual behavior. The participants can swing their head freely; therefore, the most real and natural images can be obtained without too many restrictions. The labeled data are classified according to the coordinate of gaze and area of interest on the screen. Therefore, varied network architectures are applied to estimate and compare the effects including the number of convolutional layers, batch normalization (BN) and the global average pooling (GAP) layer instead of the fully connected layer. Three schemes, including the single eye image, double eyes image and facial image, with data augmentation are used to feed into neural network to train and evaluate the efficiency. The input image of the eye or face for an eye tracking system is mostly a small-sized image with relatively few features. The results show that BN and GAP are helpful in overcoming the problem to train models and in reducing the amount of network parameters. It is shown that the accuracy is significantly improved when using GAP and BN at the mean time. Overall, the face scheme has a highest accuracy of 0.883 when BN and GAP are used at the mean time. Additionally, comparing to the fully connected layer set to 512 cases, the number of parameters is reduced by less than 50% and the accuracy is improved by about 2%. A detection accuracy comparison of our model with the existing George and Routray methods shows that our proposed method achieves better prediction accuracy of more than 6%.

SUMMARYThe CAC (call admission control), which can guarantee call services to meet their QoS (Quality of Service) requirements, plays a significant role in providing QoS in wireless mobile networks. In this paper, an adaptive multiguard channel scheme‐based CAC strategy is proposed to prioritize traffic types and handoff calls. The major aim of the study is to develop the analytical model of the priority traffic and handoff calls based adaptive multiguard channel scheme and examining the performance through setting the value of the adaptive ratio parameters. Our proposed scheme tries to mediate the advantages and drawbacks of the static and dynamic CAC schemes. The proposed scheme is quite different from previous studies because multithreshold values have been considered for multiclass traffic by adaption parameters, and a closed form analytical model is developed The numerical results show that this scheme can be used to keep the targeted QoS requirement by suitably setting the adaptive ratio parameters. Copyright © 2011 John Wiley & Sons, Ltd.

SUMMARYPower control is an effective technique to reduce cochannel interference and increase capacity for cellular radio systems. The purpose of forward-link power control in a CDMA system is to reduce the amount of interference in neighbouring cells by reducing the total amount of power transmitted. In an underlaying two-tier system, microcell's capacity is limited by the forward link due to the interference from macrocell's basestation. Therefore, forward-link power control is required to enhance system capacity and reduce outage probability. In this paper, we study the effect of imperfect forward-link power control due to the limitation of power transmitted by basestation. Performance measures including capacity, outage probability and service hole area are analysed.

No abstract