Traditional Ethernet networks use the rapid spanning tree protocol (RSTP) or the enhanced RSTP (eRSTP) to ensure a loop-free topology and provide redundant links as backup paths in case an active link has failed. However, when a failure occurs, the RSTP/eRSTP require a significant amount of reconfiguration time in order to find an alternative path. The RSTP/eRSTP are also limited by the number of nodes in a ring network, and their performance degrade when the number of nodes increases; the media redundancy protocol (MRP), which is used in industrial networks, has the same issues. In this paper, we introduce a new algorithm, called loop prevention and fast recovery (LPFR), which can be applied to Ethernet ring networks to provide a loop-free topology. When a failure occurs, LPFR requires only a very short amount of time to switch to an alternative path. LPFR needs a maximum of 6.1 ms for recovery time, versus ＜ 100 ms for the RSTP/eRSTP in a ring size of 20 nodes and ＜ 200 ms for the MRP in a ring size of 50 nodes. The LPFR offers a reduction in recovery time of up to 93.9 % compared to the RSTP/eRSTP, and 96.9 % compared to the MRP. However, in most cases, LPFR requires zero recovery time for such switching. In addition, in most situations, no frames are lost when a node switches to an alternative path. Unlike the RSTP/eRSTP and MRP, LPFR provides a better way for distributing the sent traffic among network links, which in turn improves network performance, reduces the probability of bottleneck occurrences, and reduces the frame latency between the source and the destination. This is because LPFR always sends the frame through the fastest path. The LPFR algorithm can run on a wide variety of ring sizes, which is not the case for the RSTP/eRSTP and MRP. Finally, the LPFR will use the Ethernet standard frame layout without any modifications or changes. This will allow all of the standard Ethernet devices and networks to be directly integrated into the LPFR networks without any proxies.

Water controlling systems are important components of smart cities. As a system of Internet of Things, water controlling systems increasingly rely on many heterogeneous perception sensors, transportation tools, and application platforms for providing throughout services. In such environments, dynamic intrusion identification is of crucial to meet security goals and take measures on important nodes considering resources for quality assurance limited by time and by cost. In this paper, we introduce a Bayesian network model for intrusion identification in IoT and propose an importance index to identify important nodes for further security management when intrusion occurs. An experiment is carried out to illustrate how this model works. After formalizing a Bayesian network on a water controlling system, Bayesian inference can be performed based on conditional probability tables of nodes with parents and prior probabilities without parents, which can be acquired statistically based on historical data. This model has good potential applications on Internet of things due to its great capability of coping with thousands of sensors, tools, and platforms with Bayesian inference, and ability of dynamically identifying important nodes to improve the efficiency of security management.

This paper addresses the secure outsourcing problem for large-scale matrix computation to a public cloud. We propose a novel public-key weave ElGamal encryption (WEE) scheme for encrypting a matrix over the field Zp. The scheme has the echelon transformation property. We can apply a series of elementary row/column operations to transform an encrypted matrix under our WEE scheme into the row/column echelon form. The decrypted result matches the result of the corresponding operations performed on the original matrix. For security, our WEE scheme is shown to be entry irrecoverable for non-zero entries under the computational Diffie-Hellman assumption. By using our WEE scheme, we propose five secure outsourcing protocols of Gaussian elimination, Gaussian-Jordan elimination, matrix determinant, linear system solver, and matrix inversion. Each of these protocols preserves data privacy for clients (data owners). Furthermore, the linear system solver and matrix inversion protocols provide a cheating-resistant mechanism to verify correctness of computation results. Our experimental result shows that our protocols gain efficiency significantly for an outsourcer. Our outsourcing protocol solves a linear system of n = 1, 000 equations and m = 1, 000 unknown variables about 472 times faster than a non-outsourced version. The efficiency gain is more substantial when (n,m) gets larger. For example, when n = 10, 000 and m = 10, 000, the protocol can solve it about 56, 274 times faster. Our protocols can also be easily implemented in parallel computation architecture to get more efficiency improvement.

In photorealistic rendering, refraction is a vital optical phenomenon for transparent objects. This study proposes a real-time image-space method for approximate multiple refraction. First, a depth-peeling technique is used to disassemble a 3D polygonal model into four layers of normal and depth textures. Next, a modified ray-height-field intersection algorithm is employed to resolve discontinuities between the layers and compute the intersection, refraction, and total internal reflection. Finally, an environment map with a refraction vector is used to compute the final shading. The image space and shader technique enables generating a maximum of four bounces to increase the refraction detail and handle more complex scenes compared with those examined in previous methods. We performed several experiments with various scenes and the results showed the feasibility and quality of the proposed multi-refraction method. In addition, the rendering efficiency was demonstrated by comparing the proposed method with GPU-based ray tracing.

Internet is an important part in the daily life of people in many ways, which allows the people in any corner of the world to share all types of information. Image encryption is used to transmit sensitive multimedia information in unsecured networks to provide high degree of confidentiality. In this paper, a new image encryption method is proposed based on novel implementation of pixel scan, utilizing the Knight’s Travel Path, and true random number. The Knight’s travel path is a pattern in which the path of a Knight around a chess board is taken without revisiting any node. This travel path pattern is used to permute the pixel positions of the original image to obtain the scrambled image. The scrambled image is further XORed with the random key numbers to get the cipher image. To change the pixel values of the scrambled image, true random numbers are generated from the amplitude values of a chosen noise audio file. Decryption is performed to confirm the reception of the sent image. The proposed method resists the statistical, differential, and entropy attacks significantly which have been demonstrated with various standard images.

Network coding provides a new perspective for anonymous communication. Wang et al. propose an anonymous communication scheme named ALNCode, in which the linear relation of the global encoding vectors can be hidden without the participation of encryption/decryption operations. But the effectiveness of the scheme needs to be further analyzed. In this paper, we present an exact formula of the effectiveness of ALNCode scheme, which goes a step further than the lower bound given by Wang et al.. Based on the exact formula, the influential factors of effectiveness of the scheme are taken into account in details. Moreover, a comparison between the exact effectiveness and the lower bound shows that the lower bound is not true. The reason of the error is analyzed finally.

The run-time memory of a program may be described with a directed graph in which nodes represent chunks of memory and edges represent references. We define a ＂closed cluster induced by a node n＂, denoted as CC(n), as the largest set of nodes that are reachable from n but are unreachable from nodes outside the closed cluster. Based on closed clusters, there is a Brouwerian structure under the run-time memory. We present the Brouwerian model and discuss its properties, transformations, and applications. We also propose a two-counter algorithm for calculating CC(n). The two-counter algorithm is never slower than a traditional one-counter algorithm. Our study of the Brouwerian structure is motivated by work on garbage collection.