Future mobile communication technology will be used to build an integrated global coverage network. Unmanned aerial vehicles (UAVs) are the first choice for low-altitude networks due their low cost, flexibility, and ease of operation. The characteristics UAVs also bring new challenges networks, such as short flight time, complex networking, unstable quality. Therefore, it has become urgent problem reasonably plan location UAV Base Stations (UAV-BSs), reduce power consumption, optimize...

Source noise affects the security of continuous-variable quantum key distribution (CV QKD) and is difficult to analyze. We propose a model characterize Gaussian source through introducing neutral party (Fred) who induces with general unitary transformation. Without knowing Fred's exact state, we derive bounds for both reverse direct reconciliations show that bound reconciliation tight.

The original two-way continuous-variable quantum-key-distribution (CV-QKD) protocols [S. Pirandola, S. Mancini, Lloyd and L. Braunstein, Nat. Phys. 4 (2008) 726] give the security against collective attack on condition of tomography quantum channels. We propose a family new CV-QKD prove their entangling cloner attacks without simulation result indicates that maintain same advantage as whose tolerable excess noise surpasses one-way protocol. also show all sub-protocols within have higher...

In this paper, a new method of human detection based on depth map from 3D sensor Kinect is proposed. First, the pixel filtering and context are employed to roughly repair defects due information inaccuracy captured by Kinect. Second, dataset consisting maps with various indoor poses constructed as benchmark. Finally, introducing Kirsch mask three-value codes Local Binary Pattern, novel Ternary Direction Pattern (LTDP) feature descriptor extracted used for SVM classifier. The performance...

An improved face detection and recognition method based on information of skin color depth obtained by binocular vision system was proposed in this paper. With method, the area detected firstly using Adaboost algorithm. Afterwards, real distinguished from fake one data. Then, PCA algorithm, a specific can be recognized comparing principal components current to those known individuals facial database built advance. This applied service robot equipped with camera for real-time experiment,...

Depth map contains the space information of objects and is almost free from influence light, it attracts many research interests in field machine vision used for human detection. Therefore, hunting a suitable image feature detection on depth rather attractive. In this paper, we evaluate performance typical features map. A dataset containing various indoor scenes with constructed by using Microsoft's Kinect camera as quantitative benchmark study methods The smoothed pixel filtering context so...

3D reconstruction has been recognised as an important cue for tasks like active computer vision and image understanding. However, most existing techniques need to calibrate the intrinsic extrinsic parameters of projector, making calculation very complex. This paper presents method multicolour pseudo–random encoded projected recovering shape surface from one calibrated image, based on projection ray intersecting location. combines geometric photometric information in order reconstruct only...

A binocular vision system and its application to face detection tracking in robot is introduced this paper. With the system, can do detection, identification, recognition tracking. The area detected realtime by using AdaBoost algorithm. And a method proposed with which real be distinguished from picture one skin color information depth data. specific recognized comparing principal components of current those known individuals database built advance. Finally, track specified according...

Abstract With the continuous development of wireless network technology, heterogeneous networks with multiple fusions and overlapping coverage have been formed. In process vehicles passing through different types areas at high speed, there are frequent inefficient switching behaviors. Aiming this situation, a linkage selection strategy based on attractor profit prediction is proposed in vehicle network. By analyzing correlation between transmission rate position, location update was found to...

The post-distortion algorithm is necessary for visible light communication (VLC) systems, which aiming at reducing the nonlinear interference caused by LED transmitter. In this paper, BP neural network investigated to remove nonlinearity receiver, improved genetic (GA) aided particle swarm optimization (PSO) enhancing training process. As compared with conventional BP, GA-PSO-BP has been proved be faster convergence and high approximate accuracy. Computer simulations are carried out...

Machine learning is a important part of the field artificial intelligence. Principal component analysis (PCA) classical linear dimensionality reduction algorithm in machine learning, which usually solved by gradient ascending method. As traditional method requires convexity objective function, it easy to be trapped into local maximum value. Therefore, order solve limitation being optimal solution, meaningful study performance optimization for principal analysis. In this paper, novel hybrid...