PHI: Path-Hidden Lightweight Anonymity Protocol at Network Layer

Abstract We identify two vulnerabilities for existing highspeed network-layer anonymity protocols, such as LAP and Dovetail. First, the header formats of LAP and Dovetail leak path information, reducing the anonymity-set size when an adversary launches topological attacks. Second, ASes can launch session hijacking attacks to deanonymize destinations. HORNET addresses these problems but incurs additional bandwidth overhead and latency. In this paper, we propose PHI, a Path-HIdden lightweight anonymity protocol that solves both challenges while maintaining the same level of efficiency as LAP and Dovetail. We present an efficient packet header format that hides path information and a new back-off setup method that is compatible with current and future network architectures. Our experiments demonstrate that PHI expands anonymity sets of LAP and Dovetail by over 30x and reaches 120 Gbps forwarding speed on a commodity software router.

Download Full-text

Speeding Up Back-Propagation Neural Networks

10.28945/2931 ◽

2005 ◽

Cited By ~ 4

Author(s):

Mohammed A. Otair ◽

Walid A. Salameh

Keyword(s):

Neural Networks ◽

Back Propagation ◽

Network Size ◽

Learning Rate ◽

Network Architectures ◽

Local Minima ◽

Set Size ◽

Long Time ◽

Multilayer Neural Networks ◽

Optical Time

There are many successful applications of Backpropagation (BP) for training multilayer neural networks. However, it has many shortcomings. Learning often takes long time to converge, and it may fall into local minima. One of the possible remedies to escape from local minima is by using a very small learning rate, which slows down the learning process. The proposed algorithm presented in this study used for training depends on a multilayer neural network with a very small learning rate, especially when using a large training set size. It can be applied in a generic manner for any network size that uses a backpropgation algorithm through an optical time (seen time). The paper describes the proposed algorithm, and how it can improve the performance of back-propagation (BP). The feasibility of proposed algorithm is shown through out number of experiments on different network architectures.

Download Full-text

Network Layer Modeling of WDM Fiber Optic Network Architectures for Aerospace Platforms

2007 IEEE Avionics, Fiber-Optics and Photonics Technology Conference ◽

10.1109/avfop.2007.4365748 ◽

2007 ◽

Cited By ~ 1

Author(s):

Henrik N. Poulsen ◽

Dwight H. Richards ◽

Anil Ramapanicker ◽

Daniel J. Blumenthal

Keyword(s):

Fiber Optic ◽

Network Architectures ◽

Network Layer ◽

Fiber Optic Network ◽

Optic Network

Download Full-text

A Novel Method for the Performance Control of a Gas Transmission Compressor

Volume 2: Turbo Expo 2002, Parts A and B ◽

10.1115/gt2002-30539 ◽

2002 ◽

Author(s):

W. N. Pearson ◽

A. F. Armitage ◽

D. S. Henderson

Keyword(s):

Neural Network ◽

Network Architectures ◽

Fuel Gas ◽

Performance Control ◽

Set Size ◽

The Neural Network ◽

Feed Forward Neural Networks ◽

Instrument Noise ◽

Novel Method ◽

Process Constraints

This paper presents the application of feed forward neural networks to the performance control of a gas transmission compressor. It is estimated that a global saving in compressor fuel gas of 1% could prevent the production of 6 million tonnes of CO2 per year, [1]. Results of compressor model testing suggest that compressor speed can be estimated to within ± 2.5%. The neural network property of function approximation is used to predict compressor speed for given process constraints and instrument input sets. The effects of training set size, instrument noise, reduced input sets and extrapolation from the training domain, are quantified. Various neural network architectures and training schema were examined. The embedding of a neural network into an expert system is also discussed.

Download Full-text

Face and Word Recognition in Patients with Left and Right Hemispheric Lesions: Evidence from Reaction Times and ERPs

Zeitschrift für Neuropsychologie ◽

10.1024//1016-264x.13.1.69 ◽

2002 ◽

Vol 13 (1) ◽

pp. 69-83 ◽

Cited By ~ 1

Author(s):

Stefan R. Schweinberger ◽

Thomas Klos ◽

Werner Sommer

Keyword(s):

Slow Wave ◽

Reaction Times ◽

Memory Search ◽

Event Related Potentials ◽

Performance Deficits ◽

Set Size ◽

Left Brain ◽

Related Potentials ◽

Abstract Words ◽

Left Brain Damage

Abstract: We recorded reaction times (RTs) and event-related potentials (ERPs) in patients with unilateral lesions during a memory search task. Participants memorized faces or abstract words, which were then recognized among new ones. The RT deficit found in patients with left brain damage (LBD) for words increased with memory set size, suggesting that their problem relates to memory search. In contrast, the RT deficit found in patients with RBD for faces was apparently related to perceptual encoding, a conclusion also supported by their reduced P100 ERP component. A late slow wave (720-1720 ms) was enhanced in patients, particularly to words in patients with LBD, and to faces in patients with RBD. Thus, the slow wave was largest in the conditions with most pronounced performance deficits, suggesting that it reflects deficit-related resource recruitment.

Download Full-text

Extraverts Have Larger Social Network Layers

Journal of Individual Differences ◽

10.1027/1614-0001/a000048 ◽

2011 ◽

Vol 32 (3) ◽

pp. 161-169 ◽

Cited By ~ 74

Author(s):

Thomas V. Pollet ◽

Sam G. B. Roberts ◽

Robin I. M. Dunbar

Keyword(s):

Social Network ◽

Social Relationships ◽

Outer Layer ◽

Network Size ◽

Emotional Intensity ◽

Emotional Closeness ◽

Network Layer ◽

Network Layers ◽

Social Network Size

Previous studies showed that extraversion influences social network size. However, it is unclear how extraversion affects the size of different layers of the network, and how extraversion relates to the emotional intensity of social relationships. We examined the relationships between extraversion, network size, and emotional closeness for 117 individuals. The results demonstrated that extraverts had larger networks at every layer (support clique, sympathy group, outer layer). The results were robust and were not attributable to potential confounds such as sex, though they were modest in size (raw correlations between extraversion and size of network layer, .20 < r < .23). However, extraverts were not emotionally closer to individuals in their network, even after controlling for network size. These results highlight the importance of considering not just social network size in relation to personality, but also the quality of relationships with network members.

Download Full-text