Fourier-Mellin transform (FMT) has been widely used for the extraction of rotation- and scale-invariant features. However, affine transform is a more reasonable approximation model for real viewpoint change. Due to shearing, the integral along the angular direction in the calculation of FMT cannot be used to extract the inherent features of an image undergoing affine transform. To eliminate the effect of shearing, whitening transform should be conducted on the integral along the radial direction. FMT can hardly be modified by conventional whitening-based methods with low computational cost due to additional processes. In this paper, two factors are constructed and embedded into FMT. Quasi Fourier-Mellin transform (QFMT) is proposed. The embedding of these factors is equivalent to whitening transform and can eliminate the effect of shearing in the affine transform. In particular, QFMT can also be calculated by integrating along the radial direction followed by integrating along the angular direction, as in FMT. Based on QFMT, the quasi Fourier-Mellin descriptor (QFMD) is constructed for the extraction of affine invariant features. Some experiments have also been conducted to test the performance of the proposed method.

In cognitive radio (CR) wireless mesh networks (WMNs), wireless mesh routers work as secondary users (SUs) that can opportunistically and intelligently access the idle licensed spectrum. When developing efficient routing techniques in CR networks, the aggregate interference from the SUs to the primary users (PUs) should be considered. Though the interference from a single SU that is outside the PUs' footprints is small, the aggregate interference from a great number of SUs transmitting at the same time may be significant, and this will greatly influence the PUs' performance. In this paper, a distributed routing algorithm is proposed using the network formation game to minimize the aggregate interference from the SUs to the PUs. The proposed distributed algorithm can avoid the problems in the centralized approach, such as the high cost for building the centralized coordinate nodes, high information-gathering delay, and system breakdown caused by the possible failures in the centralized nodes. From the simulation results, we can observe that the proposed algorithm can find the routes with lower interference to the PUs compared to the Dijkstra's algorithm. Compared to an upper bound, the distributed solution shows near optimum.

Cognitive radio (CR) networks have attracted many attentions recently, while the security issues are not fully studied yet. In this paper, we propose a new and powerful network layer attack, routing-toward-primary-user (RPU) attack in CR networks. In this attack, malicious nodes intentionally route a large amount of packets toward the primary users (PUs), aiming to cause interference to the PUs and to increase delay in the data transmission among the secondary users. In the RPU attack, it is difficult to detect the malicious nodes since the malicious nodes may claim that those nodes, to which they forward the packets, behave dishonestly and cause problems in the data transmission. To defend against this attack without introducing high complexity, we develop a defense strategy using belief propagation. First, an initial route is found from the source to the destination. Each node keeps a table recording the feedbacks from the other nodes on the route, exchanges feedback information and computes beliefs. Finally, the source node can detect the malicious nodes based on the final belief values. Simulation results show that the proposed defense strategy against the RPU attack is effective and efficient in terms of significant reduction in the delay and interference caused by the RPU attack.

Cognitive radio (CR) is a promising technology for future wireless spectrum allocation to improve the usage of the licensed bands. However, CR wireless networks are susceptible to various attacks and cannot offer efficient security. Primary user emulation (PUE) is one of the most serious attacks for CR networks, which can significantly increase the spectrum access failure probability. In this paper, we propose a defense strategy against the PUE attack in CR networks using belief propagation, which avoids the deployment of additional sensor networks and expensive hardware in the networks used in the existing literatures. In our proposed approach, each secondary user calculates the local function and the compatibility function, computes the messages, exchanges messages with the neighboring users, and calculates the beliefs until convergence. Then, the PUE attacker will be detected, and all the secondary users in the network will be notified in a broadcast way about the characteristics of the attacker's signal. Therefore, all SUs can avoid the PUE attacker's primary emulation signal in the future. Simulation results show that our proposed approach converges quickly, and is effective to detect the PUE attacker.

In this paper, we introduce a distributed dynamic routing algorithm in multi-hop cognitive radio (CR) networks, in which secondary users (SUs) want to minimize their interference to the primary users (PUs) while keeping the delay along the route low. We employ a cognitive pilot channel (CPC) for SUs to be able to access the information about PUs, including PUs' locations and channel conditions. Medial axis with a relaxation factor is used as a reference path for the routing, along which we develop a hierarchical structure for multiple sources to reach their destinations. We introduce a temporal and spatial dynamic non-cooperative game to model the interactions among the SUs as well as their influences on the PUs, and obtain by backward induction a set of mixed (behavioral) Nash equilibrium strategies. We also employ a multi-stage fictitious play learning algorithm for distributed routing, which minimizes the overall interference from the SUs to the PUs, as well as the average packet delay along the route from the SU nodes to their destinations. Simulation results show that our proposed algorithm can avoid congestion in the CR network and minimize delay while keeping the interference level low.

Cognitive radio (CR) is a promising technology for future wireless spectrum allocation to improve the usage of the licensed bands. However, CR wireless networks are susceptible to various attacks and cannot offer efficient security. Primary user emulation (PUE) is one of the discovered attacks for CR networks, which can significantly increase the spectrum access failure probability. In this paper, we propose a defense strategy against the PUE attack in CR networks using belief propagation (BP), which avoids the deployment of additional sensor networks and expensive hardware in the networks used in existing literatures. In our proposed approach, each secondary user calculates the local functions based on location information, computes the messages, exchanges messages with the neighboring users, and calculates the beliefs until convergence. Then, the PUE attacker will be detected according to the mean of the final beliefs. After that, all the secondary users in the network will be notified in a broadcast way about the characteristics of the attacker's signal, and avoid the attacker's primary emulation signal in the future. Simulation results show that our proposed technique converges quickly, and is effective to detect the PUE attacker.

In this paper, we introduce a distributed dynamic routing algorithm for secondary users (SUs) to minimize their interference with the primary users (PUs) in multi-hop cognitive radio (CR) networks. We use the medial axis with a relaxation factor as a reference path which is contingent on the states of the PUs. Along the axis, we construct a hierarchical structure for multiple sources to reach cognitive pilot channel (CPC) base stations. We use a temporal and spatial dynamic non-cooperative game to model the interactions among SUs as well as their influences from PUs in the multi-hop structure of the network. A multi-stage fictitious play learning is used for distributed routing in multi-hop CR networks. We obtain a set of mixed (behavioral) Nash equilibrium strategies of the dynamic game in closed form by backward induction. The proposed algorithm minimizes the overall interference and the average packet delay along the routing path from SU nodes to CPC base stations in an optimal and distributed manner.

Cognitive radio is a new communication paradigm that is able to observe, learn, optimize, and intelligently adapt to achieve optimal frequency band usage. Cognitive radio (CR)-based wireless mesh networks (WMNS) provide infrastructures for secondary users' transmission, and it is necessary to develop techniques with low interference to primary users in such networks. In this paper, we propose an interference-minimization routing and scheduling algorithm for CR-WMNs. The flow control, route selection, time sharing, and opportunistic spectrum access are jointly formulated to minimize the aggregate interference to primary users. The optimization problem can be solved in a centralized way, and a distributed algorithm is also developed by adaptively modifying the individual link cost by the interference to the primary user and flow conditions. Simulation results show that the proposed scheme finds better routes in terms of interference to the primary users compared to the shortest path scheme, and the distributed solution shows near optimum compared with the centralized solution.

With the aim to provide an inexpensive but high performance real-time head mounted eye tracker, named EyeSecret, we present a novel system derived from openEyes which share the common idea to integrated Eye tracker to everyday life and reach its full potential. To obtain the robustness, intrusiveness and accuracy from relative low-cost and off-the-shelf components, we developed the system by parallel hardware and software design. We employed reverse engineering and the rapid prototyping design and manufacturing as a measurement of head gear designing; moreover, two sets of auto-calibration plans were introduced to automatically acquire the coordinates of calibration markers in real scene, and we also improved infrared (IR) source and implemented laser to get qualified image and facilitates the eye tracker. Our system can reliably estimate eye position with normal head motion indoors or outdoors and reach an accuracy of approximately 1° visual angle, ±30° horizontal FOV (Field of View) and ±25° vertical FOV.