On the Average Case Communication Complexity for Detection in Sensor Networks

The problem of sensor-network-based distributed intrusion detection in the presence of clutter is considered. It is argued that sensing is best regarded as a local phenomenon in that only sensors in the immediate vicinity of an intruder are triggered. In such a setting, lack of knowledge of intruder location gives rise to correlated sensor readings. A signal-space viewpoint is introduced in which the noise-free sensor readings associated to intruder and clutter appear as surfaces $\mathcal{S_I}$ and $\mathcal{S_C}$ and the problem reduces to one of determining in distributed fashion, whether the current noisy sensor reading is best classified as intruder or clutter. Two approaches to distributed detection are pursued. In the first, a decision surface separating $\mathcal{S_I}$ and $\mathcal{S_C}$ is identified using Neyman-Pearson criteria. Thereafter, the individual sensor nodes interactively exchange bits to determine whether the sensor readings are on one side or the other of the decision surface. Bounds on the number of bits needed to be exchanged are derived, based on communication complexity (CC) theory. A lower bound derived for the two-party average case CC of general functions is compared against the performance of a greedy algorithm. The average case CC of the relevant greater-than (GT) function is characterized within two bits. In the second approach, each sensor node broadcasts a single bit arising from appropriate two-level quantization of its own sensor reading, keeping in mind the fusion rule to be subsequently applied at a local fusion center. The optimality of a threshold test as a quantization rule is proved under simplifying assumptions. Finally, results from a QualNet simulation of the algorithms are presented that include intruder tracking using a naive polynomial-regression algorithm.

Data mining approaches to software fault diagnosis

Automatic identification of software faults has enormous practical significance. This requires characterizing program execution behavior and the use of appropriate data mining techniques on the chosen representation. In this paper, we use the sequence of system calls to characterize program execution. The data mining tasks addressed are learning to map system call streams to fault labels and automatic identification of fault causes. Spectrum kernels and SVM are used for the former while latent semantic analysis is used for the latter The techniques are demonstrated for the intrusion dataset containing system call traces. The results show that kernel techniques are as accurate as the best available results but are faster by orders of magnitude. We also show that latent semantic indexing is capable of revealing fault-specific features.

Collapse behaviour of a residual soil

Residually derived red soils occur in Bangalore District of Karnataka State, India. The porous and unsaturated nature of the red soils makes them susceptible to collapse on wetting under load. The present study analyses the collapse behaviour of an unsaturated bonded (undisturbed) red soil from Bangalore referenced to tests on samples in an unbonded (remoulded) state. A filter paper method was used to determine the matric suction of the bonded and unbonded specimens, and mercury intrusion porosimetry (MIP) was used to determine their soil structure. Analysis of the experimental results shows that bonding plays an important role in the collapse behaviour of the unsaturated residual soil. The results of the study also provide insight into the volume change behaviour of unsaturated bonded soils on wetting within and beyond the yield locus.

Role of soil structure and matric suction in collapse of a compacted clay soil

This paper examines the role of microstructure and matric suction in the collapse behavior of a compacted clay soil from Bangalore District in Karnataka State, India. The microstructure of the compacted specimens was examined by mercury intrusion porosimetry (MIP), and the ASTM Filter Paper Method was used to determine their matric suction. The microstructure and matric suction of the compacted specimens were changed by varying their compaction water content, dry density, and clay content (< 2 mum fraction). Experimental results showed that relative abundance of coarse (60 to 6 mum) pores was mainly affected by increasing the dry density of the specimens from 1.49 to 1.77 g/cm(3). The relative abundance of coarse and fine (0.01 to 0.002 mum) pores was affected by increasing the compaction water content from 10.6 to 26.4%. Variations in dry density, compaction water content, and clay contents notably affected the matric suction of the compacted specimens. The collapse behavior of the compacted specimens is explained from analysis of the MIP and matric suction results.

Model track studies on fouled ballast using ground penetrating radar and multichannel analysis of surface wave

Ballast fouling is created by the breakdown of aggregates or outside contamination by coal dust from coal trains, or from soil intrusion beneath rail track. Due to ballast fouling, the conditions of rail track can be deteriorated considerably depending on the type of fouling material and the degree of fouling. So far there is no comprehensive guideline available to identify the critical degree of fouling for different types of fouling materials. This paper presents the identification of degree of fouling and types of fouling using non-destructive testing, namely seismic surface-wave and ground penetrating radar (GPR) survey. To understand this, a model rail track with different degree of fouling has been constructed in Civil engineering laboratory, University of Wollongong, Australia. Shear wave velocity obtained from seismic survey has been employed to identify the degree of fouling and types of fouling material. It is found that shear wave velocity of fouled ballast increases initially, reaches optimum fouling point (OFP), and decreases when the fouling increases. The degree of fouling corresponding after which the shear wave velocity of fouled ballast will be smaller than that of clean ballast is called the critical fouling point (CFP). Ground penetrating radar with four different ground coupled antennas (500 MHz, 800 MHz, 1.6 GHz and 2.3 GHz) was also used to identify the ballast fouling condition. It is found that the 800 MHz ground coupled antenna gives a better signal in assessing the ballast fouling condition. Seismic survey is relatively slow when compared to GPR survey however it gives quantifiable results. In contrast, GPR survey is faster and better in estimating the depth of fouling. (C) 2011 Elsevier B.V. All rights reserved.

Intruder detection over sensor placement in a hexagonal lattice

The problem of intrusion detection and location identification in the presence of clutter is considered for a hexagonal sensor-node geometry. It is noted that in any practical application,for a given fixed intruder or clutter location, only a small number of neighboring sensor nodes will register a significant reading. Thus sensing may be regarded as a local phenomenon and performance is strongly dependent on the local geometry of the sensor nodes. We focus on the case when the sensor nodes form a hexagonal lattice. The optimality of the hexagonal lattice with respect to density of packing and covering and largeness of the kissing number suggest that this is the best possible arrangement from a sensor network viewpoint. The results presented here are clearly relevant when the particular sensing application permits a deterministic placement of sensors. The results also serve as a performance benchmark for the case of a random deployment of sensors. A novel feature of our analysis of the hexagonal sensor grid is a signal-space viewpoint which sheds light on achievable performance.Under this viewpoint, the problem of intruder detection is reduced to one of determining in a distributed manner, the optimal decision boundary that separates the signal spaces SI and SC associated to intruder and clutter respectively. Given the difficulty of implementing the optimal detector, we present a low-complexity distributive algorithm under which the surfaces SI and SC are separated by a wellchosen hyperplane. The algorithm is designed to be efficient in terms of communication cost by minimizing the expected number of bits transmitted by a sensor.

Wireless sensor networks forhuman intruder detection

In this paper we report on the outcomes of a research and demonstration project on human intrusion detection in a large secure space using an ad hoc wireless sensor network. This project has been a unique experience in collaborative research, involving ten investigators (with expertise in areas such as sensors, circuits, computer systems,communication and networking, signal processing and security) to execute a large funded project that spanned three to four years. In this paper we report on the specific engineering solution that was developed: the various architectural choices and the associated specific designs. In addition to developing a demonstrable system, the various problems that arose have given rise to a large amount of basic research in areas such as geographical packet routing, distributed statistical detection, sensors and associated circuits, a low power adaptive micro-radio, and power optimising embedded systems software. We provide an overview of the research results obtained.

Two timescale convergent Q-learning for sleep-scheduling in wireless sensor networks

In this paper, we consider an intrusion detection application for Wireless Sensor Networks. We study the problem of scheduling the sleep times of the individual sensors, where the objective is to maximize the network lifetime while keeping the tracking error to a minimum. We formulate this problem as a partially-observable Markov decision process (POMDP) with continuous stateaction spaces, in a manner similar to Fuemmeler and Veeravalli (IEEE Trans Signal Process 56(5), 2091-2101, 2008). However, unlike their formulation, we consider infinite horizon discounted and average cost objectives as performance criteria. For each criterion, we propose a convergent on-policy Q-learning algorithm that operates on two timescales, while employing function approximation. Feature-based representations and function approximation is necessary to handle the curse of dimensionality associated with the underlying POMDP. Our proposed algorithm incorporates a policy gradient update using a one-simulation simultaneous perturbation stochastic approximation estimate on the faster timescale, while the Q-value parameter (arising from a linear function approximation architecture for the Q-values) is updated in an on-policy temporal difference algorithm-like fashion on the slower timescale. The feature selection scheme employed in each of our algorithms manages the energy and tracking components in a manner that assists the search for the optimal sleep-scheduling policy. For the sake of comparison, in both discounted and average settings, we also develop a function approximation analogue of the Q-learning algorithm. This algorithm, unlike the two-timescale variant, does not possess theoretical convergence guarantees. Finally, we also adapt our algorithms to include a stochastic iterative estimation scheme for the intruder's mobility model and this is useful in settings where the latter is not known. Our simulation results on a synthetic 2-dimensional network setting suggest that our algorithms result in better tracking accuracy at the cost of only a few additional sensors, in comparison to a recent prior work.

Phytoplankton size structure in the southern Bay of Bengal modified by the Summer Monsoon Current and associated eddies: Implications on the vertical biogenic flux

The present study combines field and satellite observations to investigate how hydrographical transformations influence phytoplankton size structure in the southern Bay of Bengal during the peak Southwest Monsoon/Summer Monsoon (July-August). The intrusion of the Summer Monsoon Current (SMC) into the Bay of Bengal and associated changes in sea surface chemistry, traceable eastward up to 90 degrees E along 8 degrees N, seems to influence biology of the region significantly. Both in situ and satellite (MODIS) data revealed low surface chlorophyll except in the area influenced by the SMC During the study period, two well-developed cydonic eddies (north) and an anti-cyclonic eddy (south), closely linked to the main eastward flow of the SMC, were sampled. Considering the capping effect of the low-saline surface water that is characteristic of the Bay of Bengal, the impact of the cyclonic eddy, estimated in terms of enhanced nutrients and chlorophyll, was mostly restricted to the subsurface waters (below 20 m depth). Conversely, the anti-cyclonic eddy aided by the SMC was characterized by considerably higher nutrient concentration and chlorophyll in the upper water column (upper 60 m), which was contrary to the general characteristic of such eddies. Albeit smaller phytoplankton predominated the southern Bay of Bengal (60-95% of the total chlorophyll), the contribution of large phytoplankton was double in the regions influenced by the SMC and associated eddies. Multivariate analysis revealed the extent to which SMC-associated eddies spatially influence phytoplankton community structure. The study presents the first direct quantification of the size structure of phytoplankton from the southern Bay of Bengal and demonstrates that the SMC-associated hydrographical ramifications significantly increase the phytoplankton biomass contributed by larger phytoplankton and thereby influence the vertical opal and organic carbon flux in the region. (C) 2014 Elsevier B.V. All rights reserved.

Adaptive Sleep-Wake Control using Reinforcement Learning in Sensor Networks

The aim in this paper is to allocate the `sleep time' of the individual sensors in an intrusion detection application so that the energy consumption from the sensors is reduced, while keeping the tracking error to a minimum. We propose two novel reinforcement learning (RL) based algorithms that attempt to minimize a certain long-run average cost objective. Both our algorithms incorporate feature-based representations to handle the curse of dimensionality associated with the underlying partially-observable Markov decision process (POMDP). Further, the feature selection scheme used in our algorithms intelligently manages the energy cost and tracking cost factors, which in turn assists the search for the optimal sleeping policy. We also extend these algorithms to a setting where the intruder's mobility model is not known by incorporating a stochastic iterative scheme for estimating the mobility model. The simulation results on a synthetic 2-d network setting are encouraging.

Secure Routing with an Integrated Localized Key Management Protocol in MANETs

A routing protocol in a mobile ad hoc network (MANET) should be secure against both the outside attackers which do not hold valid security credentials and the inside attackers which are the compromised nodes in the network. The outside attackers can be prevented with the help of an efficient key management protocol and cryptography. However, to prevent inside attackers, it should be accompanied with an intrusion detection system (IDS). In this paper, we propose a novel secure routing with an integrated localized key management (SR-LKM) protocol, which is aimed to prevent both inside and outside attackers. The localized key management mechanism is not dependent on any routing protocol. Thus, unlike many other existing schemes, the protocol does not suffer from the key management - secure routing interdependency problem. The key management mechanism is lightweight as it optimizes the use of public key cryptography with the help of a novel neighbor based handshaking and Least Common Multiple (LCM) based broadcast key distribution mechanism. The protocol is storage scalable and its efficiency is confirmed by the results obtained from simulation experiments.

Appraisal of groundwater resource in Holocene soil deposits by resistivity, hydrochemical and granulomerial studies in the Gulf of Mannar Coast from Southern India

This article is aimed to delineate groundwater sources in Holocene deposits area in the Gulf of Mannar Coast from Southern India. For this purpose 2-D electrical resistivity tomography (ERT), hydrochemical and granulomerical studies were carried out and integrated to identify hydrogeological structures and portable groundwater resource in shallow depths which in general appears in the coastal tracts. The 2-D ERT was used to determine the two-dimensional subsurface geological formations by multicore cable with Wenner array. Low resistivity of 1-5 Omega m for saline water appeared due to calcite at the depth of about 5 m below the ground level (bgl). Sea water intrusion was observed around the maximum resistivity as 5 Omega m at the 8 m depth, bgl in the calcite environs, but the calcareous sandstone layer shows around 15-64 Omega m at the 6 m depth, bgl. The hydrochemical variation of TDS, HCO3-, Cl-, Na+, K+, Ca2+, and Mg2+ concentrations was observed for the saline and sea water intrusion in the groundwater system. The granulometic analysis shows that the study area was under the sea between 5400 and 3000 year ago. The events of ice melting an unnatural ice-stone rain/hail among 5000-4000 years ago resulted in the inundation of sea over the area and deposits of late Holocene marine transgression formation up to Puthukottai quartzite region for a stretch of around 17 km.