Semi-supervised annotation of faces in image collection

The objective of this work is to correctly detect and recognize faces in an image collection using a database of known faces. This has applications in photo-tagging, video indexing, surveillance and recognition in wearable computers. We propose a two-stage approach for both detection and recognition tasks. In the first stage, we generate a seed set from the given image collection using off-the-shelf face detection and recognition algorithms. In the second stage, the obtained seed set is used to improve the performance of these algorithms by adapting them to the domain at hand. We propose an exemplar-based semi-supervised framework for improving the detections. For recognition of images, we use sparse representation classifier and generate seed images based on a confidence measure. The labels of the seed set are then propagated to other faces using label propagation framework by imposing appropriate constraints. Unlike traditional approaches, our approach exploits the similarities among the faces in collection to obtain improved performance. We conduct extensive experiments on two real-world photo-album and video collections. Our approach consistently provides an improvement of \({\sim } 4\)% for detection and \(5{-}9\)% for recognition on all these datasets.     

Characterization of Secure Multiparty Computation Without Broadcast

A major challenge in the study of cryptography is characterizing the necessary and sufficient assumptions required to carry out a given cryptographic task. The focus of this work is the necessity of a broadcast channel for securely computing symmetric functionalities (where all the parties receive the same output) when one third of the parties, or more, might be corrupted. Assuming all parties are connected via a point-to-point network, but no broadcast channel (nor a secure setup phase) is available, we prove the following characterization:

• A symmetric n-party functionality can be securely computed facing \(n/3\le t<n/2\) corruptions (i.e., honest majority), if and only if it is \((n-2t)\) -dominated; a functionality is k-dominated, if any k-size subset of its input variables can be set to determine its output to some predetermined value.
• Assuming the existence of one-way functions, a symmetric n-party functionality can be securely computed facing \(t\ge n/2\) corruptions (i.e., no honest majority), if and only if it is 1-dominated and can be securely computed with broadcast.

It follows that, in case a third of the parties might be corrupted, broadcast is necessary for securely computing non-dominated functionalities (in which “small” subsets of the inputs cannot determine the output), including, as interesting special cases, the Boolean XOR and coin-flipping functionalities.

     

Adaptive coordination function for IEEE 802.11 wireless LANs

Multiple access control (MAC) protocols play a significant role in wireless LANs. The IEEE 802.11 MAC protocol specifies two coordination functions that are Distributed Coordination Function (DCF) and Point Coordination Function (PCF). While both DCF and PCF are available in a wireless cell, we propose a novel access mechanism called Adaptive Coordination Function (ACF) to support various classes of traffic. The ACF superframe comprises two periods, one TDMA period designed for real-time traffic and followed by an adaptive period which adaptively employs DCF or PCF to support non-real-time traffic. In this paper, we apply the theory of M/G/1 queues to analyze the performance of adaptive period in terms of queuing delay, end-to-end delay, and saturation throughput. With our analytic model, DCF or PCF can be invoked appropriately according to the number of stations, packet arrival rate, packet payload size, and effective channel bit rate. Analytical results are derived for an extensive throughput and delay performance evaluation of both DCF and PCF.     

A class of generalized models for shadowed fading channels

The commonly used models for shadowed fading channels arise by compounding the Nakagami distribution with others such as the log-normal and gamma distributions. A distribution that is more flexible than the Nakagami distribution is the well known Weibull distribution. In this paper, we derive a collection of generalized models for shadowed fading channels by compounding the Weibull distribution with other distributions belong to some fifteen flexible families. We also illustrate the superior performance of these models over the standard Nakagami models.     

Energy-efficient Rate Adaptation MAC Protocol for Ad Hoc Wireless Networks

In this paper, two novel energy-efficient rate adaptation schemes are presented. The proposed protocols use the Distributed Power Control (DPC) algorithm to predict the channel state and determine the necessary transmission power which optimizes the energy consumption. The first proposed rate adaptation scheme heuristically alters the modulation rate to balance the energy-efficiency and the required throughput which is estimated with queue fill ratio. Moreover, the back-off scheme is incorporated to mitigate congestion and reduce packet losses due to buffer overflows thus minimizing energy consumption. Consequently, the nodes will conserve energy when the traffic is low, offer higher throughput when needed and save energy during congestion by limiting transmission rates. The second rate adaptation scheme employs dynamic programming (DP) principle to analytically select modulation rate and a burst size to be used during transmission. The proposed quadratic cost-function minimizes the energy consumption while alleviating network congestions and buffer overflows. The proposed DP solution renders a Riccati equation ultimately providing an optimal rate selection. The simulation results indicate that an increase in throughput by 96% and energy-efficiency by 131% is observed when compared to other available protocols, for example Receiver Based AutoRate (RBAR).

Outage Probability with Correlated Lognormal Interferers using Log Shifted Gamma Approximation

The computation of outage probability in cellular radio system has been extensively studied. The Signal-to-Interference-plus-Noise Ratio (SINR) distribution involves the sum of lognormal distributions due to dominant effect of shadowing in both the signal and interference components. Since no closed-form expression can be found for the sum of lognormal distributions, many approximation methods and bounds were proposed in the past. In this paper, Log Shifted Gamma (LSG) approximation is proposed to calculate the sum of correlated lognormal random variables (RVs), hence the outage probability, accurately with a wide range of dB spreads, number of interferers M, correlation coefficients r among interference components, and noise power N. Overall, LSG approximation shows consistent accuracy due to its flexibility over the classical lognormal approximation, especially with small correlation coefficients r and/or large dB spreads.     

Isolation of Failing Scan Cells through Convolutional Test Response Compaction

This paper describes a non-recursive fault diagnosis technique for scan-based designs with convolutional test response compaction. The proposed approach allows a time-efficient and accurate identification of failing scan cells using Gauss–Jordan elimination method.

一种新的综合安全管理架构的设计与实现

传统的安全管理框架缺少对底层数据的整合分析,不能形成有效的网络安全态势信息和风险评估,也不能满足用户对其扩展性和健壮性的要求。文中利用TMN(电信管理网)管理与受管资源分开的思路启发,提出了一种新的综合安全管理框架。新的框架在J2EE分布式架构平台加以实现,能够满足用户对综合安管系统扩展性、易用性和健壮性的需求。     

辐射发射测量天线的校准方法

文章介绍了辐射发射测量所用宽带天线天线系数的校准方法,并对影响测试结果的因素进行分析。     

异构融合网络认证机制研究

移动通信和无线局域网的快速发展使两者的融合成为热点。描述了融合网络的场景和耦合方案,对现有融合网的认证机制缺点进行了分析,提出把传输层安全协议TLS引入到融合网认证中,并对该认证机制对融合网络安全的增强进行了详细描述和分析。最后搭建了融合网络的试验平台,测量了新的认证机制在执行时延、吞吐量以及能量耗费方面的性能。结果表明,新认证机制在大大增强了融合网络安全性的同时提高了网络的性能和效率。