Swaminathan Sankararaman

The resilience of WDM networks to probabilistic geographical failures

Telecommunications networks, and in particular optical WDM networks, are vulnerable to large-scale failures of their physical infrastructure, resulting from physical attacks (such as an Electromagnetic Pulse attack) or natural disasters (such as solar flares, earthquakes, and floods). Such events happen at specific geographical locations and disrupt specific parts of the network but their effects are not deterministic. Therefore, we provide a unified framework to model the network vulnerability when the event has a probabilistic nature, defined by an arbitrary probability density function. Our framework captures scenarios with a number of simultaneous attacks, in which network components consist of several dependent subcomponents, and in which either a 1+1 or a 1∶1 protection plan is in place. We use computational geometric tools to provide efficient algorithms to identify vulnerable points within the network under various metrics. Then, we obtain numerical results for specific backbone networks, thereby demonstrating the applicability of our algorithms to real-world scenarios. Our novel approach allows for identifying locations which require additional protection efforts (e.g., equipment shielding). Overall, the paper demonstrates that using computational geometric techniques can significantly contribute to our understanding of network resilience.

Network vulnerability to single, multiple, and probabilistic physical attacks

Telecommunications networks heavily rely on the physical infrastructure and, are therefore, vulnerable to natural disasters, such as earthquakes or floods, as well as to physical attacks, such as an Electromagnetic Pulse (EMP) attack. Large-scale disasters are likely to destroy network equipment and to severely affect interdependent systems such as the power-grid. In turn, long-term outage of the power-grid might cause additional failures to the telecommunication network. In this paper, we model an attack as a disk around its epicenter, and provide efficient algorithms to find vulnerable points within the network, under various metrics. In addition, we consider the case in which multiple disasters happen simultaneously and provide an approximation algorithm to find the points which cause the most significant destruction. Finally, since a network element does not always fail, even when it is close to the attacks epicenter, we consider a simple probabilistic model in which the probability of a network element failure is given. Under this model, we tackle the cases of single and multiple attacks and develop algorithms that identify potential points where an attack is likely to cause a significant damage.

Optimization schemes for protective jamming

In this paper, we study strategies for allocating and managing friendly jammers, so as to create virtual barriers that would prevent hostile eavesdroppers from tapping sensitive wireless communication. Our scheme precludes the use of any encryption technique. Applications include domains such as (i) protecting the privacy of storage locations where RFID tags are used for item identification, (ii) secure reading of RFID tags embedded in credit cards, (iii) protecting data transmitted through wireless networks, sensor networks, etc. By carefully managing jammers to produce noise, we show how to reduce the SINR of eavesdroppers to below a threshold for successful reception, without jeopardizing network performance. We present algorithms targeted towards optimizing power allocation and number of jammers needed in several settings. Experimental simulations back up our results.

Model-driven matching and segmentation of trajectories

A fundamental problem in analyzing trajectory data is to identify common patterns between pairs or among groups of trajectories. In this paper, we consider the problem of matching similar portions between a pair of trajectories, each observed as a sequence of points sampled from it. We present new measures of trajectory similarity --- both local and global --- between a pair of trajectories to distinguish between similar and dissimilar portions. We then use this model to perform segmentation of a set of trajectories into fragments, contiguous portions of trajectories shared by many of them. Our model for similarity is robust under noise and sampling rate variations. The model also yields a score which can be used to rank multiple pairs of trajectories according to similarity, e.g. in clustering applications. We present quadratic time algorithms to compute the similarity between trajectory pairs under our measures together with algorithms to identify fragments in a large set of trajectories efficiently using the similarity model. Finally, we present an extensive experimental study evaluating the effectiveness of our approach on real datasets, comparing it with earlier approaches. Our experiments show that our model for similarity is highly accurate in distinguishing similar and dissimilar portions as compared to earlier methods even with sparse sampling. Further, our segmentation algorithm is able to identify a small set of fragments capturing the common parts of trajectories in the dataset.

Graph-based iterative reconstruction of sparse signals for compressed sensing

In this paper, we consider the problem of reconstruction of sparse signals in compressed sensing. In particular, we introduce a novel iterative algorithm based on graph-based decoding of low-density parity-check codes which possesses desirable properties like good performance, low complexity and running time, and ease of implementation. In this work, we outline the reconstruction algorithm, and analyze its performance on some measurement matrices. Furthermore, we also provide some initial results on the theoretical performance limits of this algorithm.

Optimal placement of protective jammers for securing wireless transmissions in a geographic domain

Wireless communication systems, such as RFIDs and wireless sensor networks, are increasingly being used in security-sensitive applications, e.g. credit card transactions or monitoring patient health in hospitals. Wireless jamming by transmitting artificial noise, which is traditionally used as an offensive technique for disrupting communication, has recently been explored as a means of protecting sensitive communication from eavesdroppers. In this paper, we consider location optimization problems related to the placement and power consumption of such friendly jammers in order to protect the privacy of wireless communications constrained within a geographic region. Under our model, we show that the problem of placing a minimum number of fixed-power jammers is NP-Hard, and we provide a PTAS ((1 + ε)-approximation scheme) for the same, where ε is a tunable parameter between 0 and 1.

Data transmission and base-station placement for optimizing the lifetime of wireless sensor networks

In this paper, we study the fundamental optimization problem in wireless sensor networks of base-station positioning such that data from the sensors may be transmitted to it in an energy-efficient manner. We primarily consider the setting where a sensor transmits all of its data directly to the base-station or relays it via one other node. This setting provides two benefits: low duty-cycling due to limited synchronization requirements between nodes and low end-to-end delay due to the limited number of hops in the routes. Given the battery limitations of the sensor nodes, our objective is to maximize the network lifetime. First, we present efficient algorithms for computing a transmission scheme for the sensors given a fixed base-station and show how to implement these in a distributed fashion with only a constant number of messages per sensor. Next, we show that the optimization problem for the setting where sensors may transmit data through more than 2 hops is NP-Hard. Finally, we present efficient algorithms for the problem of locating the base-station and simultaneously finding a transmission scheme. We compare our algorithms with linear-programming based algorithms for more general settings through extensive simulations and outline the benefits of the different approaches.

Optimization Schemes for Protective Jamming

In this paper, we study strategies for allocating and managing friendly jammers, so as to create virtual barriers that would prevent hostile eavesdroppers from tapping sensitive wireless communication. Our scheme precludes the use of any encryption technique. Applications include domains such as (i) protecting the privacy of storage locations where RFID tags are used for item identification, (ii) secure reading of RFID tags embedded in credit cards, (iii) protecting data transmitted through wireless networks, sensor networks, etc. By carefully managing jammers to produce noise, we show how to reduce the SINR of eavesdroppers to below a threshold for successful reception, without jeopardizing network performance. In this paper, we present algorithms targeted towards optimizing power consumption and number of jammers needed in several settings. Experimental simulations back up our results.

Data transmission and base-station placement for optimizing network lifetime

We study the problem of transmitting data from a set of sensors to a base-station where the data is to be gathered. Each sensor continuously generates data and has to transmit it through the network (via other sensor nodes) to the base-station. Considering the battery limitations of the sensors, our goal is to find an optimum location of the base-station and a corresponding data transmission scheme for routing the data from the sensors, such that the network is operating for the longest possible time. We focus mainly on tree networks for 2-level trees, with at most 2 hops from sensor to the base-station. For such networks we give efficient algorithms for forwarding data from sensors to the base-station and for locating the base-station optimally for maximizing network lifetime. Further, we show that determining a transmission protocol for trees with 3 or more levels is NP-hard. We demonstrate the effectiveness of our methods with experimental results on simulated data, comparing our 2-level tree algorithm with methods based on linear programming.

Secure communication through jammers jointly optimized in geography and time

Abstract Security-sensitive applications, such as patient health monitoring and credit card transactions, are increasingly utilizing wireless communication systems, RFIDs, wireless sensor networks, and other wireless communication systems. The use of interference-emitting jammers to protect such sensitive communication has been recently explored in the literature, and has shown high potential. In this paper we consider optimization problems relating to the temporal distributions of jammers’ activity, and the suitable coding regimes used for communication. Solving the joint problem optimally enables comprehensive security in space, at a low power consumption and low communication overhead. The joint optimization of jamming in space and time is driven by a new framework that uses the bit-error probability as a measure of communication quality. Under this framework, we show how to guarantee information-theoretic security within a geographic region, and with increased flexibility to tailor the coding regime to the problem’s geometry. We present efficient algorithms for different settings, and provide simulations for various scenarios using the bit-error probability functions. These simulations demonstrate the efficiency of the scheme. We believe that our scheme can lead to practical, economical and scalable solutions for providing another layer of protection of sensitive data, in cases where encryption schemes are limited or impractical.