Sharanya Eswaran

Adaptive In-Network Processing for Bandwidth and Energy Constrained Mission-Oriented Multihop Wireless Networks

In-network Processing, involving operations such as filtering, compression, and fusion is a technique widely used in wireless sensor and ad hoc networks for reducing the communication overhead. In many tactical stream-oriented applications, especially in military scenarios, both link bandwidth and node energy are critically constrained resources. For such applications, in-network processing itself imposes nonnegligible computing cost. In this work, we have developed a unified, utility-based closed-loop control framework that permits distributed convergence to both 1) the optimal level of compression performed by a forwarding node on streams, and 2) the best set of nodes where the operators of the stream processing graph should be deployed. We also show how the generalized model can be adapted to more realistic cases, where the in-network operator may be varied only in discrete steps, and where a fusion operation cannot be fractionally distributed across multiple nodes. Finally, we provide a real-time implementation of the protocol on an 802.11b network with a video application and show that the performance of the network is improved significantly in terms of the packet loss, node lifetime, and quality of video received.

Adaptive In-Network Processing for Bandwidth and Energy Constrained Mission-Oriented Multi-hop Wireless Networks

In-network processing, involving operations such as filtering, compression and fusion, is widely used in sensor networks to reduce the communication overhead. In many tactical and stream-oriented wireless network applications, both link bandwidth and node energy are critically constrained resources and in-network processing itself imposes non-negligible computing cost. In this work, we have developed a unified and distributed closed-loop control framework that computes both a) the optimal level of sensor stream compression performed by a forwarding node, and b) the best set of nodes where the stream processing operators should be deployed. Our framework extends the Network Utility Maximization (NUM) paradigm, where resource sharing among competing applications is modeled as a form of distributed utility maximization. We also show how our model can be adapted to more realistic cases, where in-network compression may be varied only discretely, and where a fusion operation cannot be fractionally distributed across multiple nodes.

Utility-based bandwidth adaptation in mission-oriented wireless sensor networks

This article develops a utility-based optimization framework for resource sharing by multiple competing missions in a mission-oriented wireless sensor network (WSN) environment. Prior work on network utility maximization (NUM) based optimization has focused on unicast flows with sender-based utilities in either wireline or wireless networks. In this work, we develop a generalized NUM model to consider three key new features observed in mission-centric WSN environments: i) the definition of the utility of an individual mission (receiver) as a joint function of data from multiple sensor sources; ii) the consumption of each senders (sensor) data by multiple missions; and iii) the multicast-tree-based dissemination of each sensors data flow, using link-layer broadcasts to exploit the “wireless broadcast advantage” in data forwarding. We show how a price-based, distributed protocol (WSN-NUM) can ensure optimal and proportionally fair rate allocation across multiple missions, without requiring any coordination among missions or sensors. We also discuss techniques to improve the speed of convergence of the protocol, which is essential in an environment as dynamic as the WSN. Further, we analyze the impact of various network and protocol parameters on the bandwidth utilization of the network, using a discrete-event simulation of a stationary wireless network. Finally, we corroborate our simulation-based performance results of the WSN-NUM protocol with an implementation of an 802.11b network.

Strategy and modeling for building DR optimization

While it is well recognized that renewable microgrid generation and intelligent storage can significantly reduce a buildings need for grid power and its peak loading, there is currently no sound and generalized approach to combine these two technologies. Further, loads are becoming increasingly responsive, in terms of both sheddability and shiftability. In this paper, we formulate the building energy management problem based on a demand-response (DR) adaptation framework that extends the traditional “supply-demand matching” to a “supply-store-demand-time-shift-utility adaptation” paradigm. Stochastic modeling of distributed-energy resources (DER) and measurement-based stochastic models of loads are used to approximately optimize building DR actions. Compared to systems that have no DR, the proposed optimization achieves savings in the range of approximately 35–70%, depending on the amount of energy storage, the flexibility of the loads, and the accuracy of the stochastic models.

IRIS: Tapping wearable sensing to capture in-store retail insights on shoppers

We investigate the possibility of using a combination of a smartphone and a smartwatch, carried by a shopper, to get insights into the shoppers behavior inside a retail store. The proposed IRIS framework uses standard locomotive and gestural micro-activities as building blocks to define novel composite features that help classify different facets of a shoppers interaction/experience with individual items, as well as attributes of the overall shopping episode or the store. Besides defining such novel features, IRIS builds a novel segmentation algorithm, which partitions the duration of an entire shopping episode into atomic item-level interactions, by using a combination of feature-based landmarking, change point detection and variable-order HMM-based sequence prediction. Experiments with 50 real-life grocery shopping episodes, collected from 25 shoppers, we show that IRIS can demarcate item-level interactions with an accuracy of approx. 91%, and subsequently characterize item-and-episode level shopper behavior with accuracies of over 90%.

Utility-based joint sensor selection and congestion control for task-oriented WSNs

Task-centric wireless sensor network environments are often characterized by the simultaneous operation of multiple tasks. Individual tasks compete for constrained resources and thus need resource mediation algorithms at two levels. First, different sensors must be allocated to different tasks based on the combination of sensor attributes and task requirements. Subsequently, sensor data rates on various data routes must be dynamically adapted to share the available wireless bandwidth, especially when links experience traffic congestion. In this paper we investigate heuristics for incrementally modifying the sensor-task matching process to incorporate changes in the transport capacity constraints or feasible task utility values.

Janayuja: A People-centric Platform to Generate Reliable and Actionable Insights for Civic Agencies

With the proliferation of smartphone apps, social media, and online forums, modern citizens are actively discussing and expressing opinions about city related issues in open public forums on the web. This paper presents a people-centric platform, Janayuja, that can act as an effective conduit between residents and civic agencies, by collecting timely information from different online sources and providing actionable insights to the agencies on pressing city issues. In particular, we elaborate on four major components of the platform: (i) curation of reports from heterogeneous data sources; (ii) categorization of individual reports into relevant issues (e.g. bus breakdown) using suitable classification techniques; (iii) aggregation of related reports (in the spatio-temporal sense) to identify specific issues pivoted to distinct city locations; and (iv) verification of issues to ensure reliability of information being provided to the agencies. Janayuja is deployed for the largest public transport agency in Bangalore, India. Based on insights generated by Janayuja, the agency can better manage their current operations, understand and anticipate new requirements from commuters (e.g. for new bus routes) and, in turn, encourage greater usage of a public transportation system.

Control-theoretic Optimization of Utility over Mission Lifetimes in Multi-hop Wireless Networks

Both bandwidth and energy become important resource constraints when multi-hop wireless networks are used to transport relatively high data rate sensor flows. A particularly challenging problem involves the selection of flow data rates that maximize application (or mission) utilities over a time horizon, especially when different missions are active over different time intervals. Prior works on utility driven adaptation of flow data rates typically focus only on instantaneous utility maximization and are unable to address this temporal variation in mission durations. In this work, we derive an optimal control-based Network Utility Maximization (NUM) framework that is able to maximize the system utility over a lifetime that is known either deterministically or statistically. We first consider a static setup in which all the missions are continuously active for a deterministic duration, and show how the rates can be optimally adapted, via a distributed protocol, to maximize the total utility. Next, we develop adaptive protocols for the dynamic cases when we have (i) complete knowledge about the mission utilities and their arrivals and departures, and (ii) a varying amount of statistical information about the missions. Our simulation results indicate that our protocols are robust, efficient and close to the optimal.

Enhancing application performance with network awareness in Tactical Networks

This paper presents Telcordias Network Awareness Service (NAS), a suite of network sensing technologies that allows applications to adapt to medium-to-long time scale performance variations in mobile tactical wireless networks. NAS is a working software prototype developed for the purpose of measuring, estimating, and inferring network metrics that impact application performance such as congestion level, packet transmission loss rate, and available path capacity. The prototype operates in a distributed manner, publishing its metrics to tactical network applications operating on protected (so-called “red-side”) computing systems, and is compatible with communication security (COMSEC) considerations that prohibit direct monitoring of network elements on the encrypted side (so-called “black-side”) of tactical radio networks. Each NAS instance performs the following key high-level functions: (1) Collection/measurement of raw packet performance data, (2) inference of network performance metrics and (3) reporting of network performance to other NAS instances and tactical applications. NAS has been successfully integrated and demonstrated as part of the U.S. Office of Naval Research Limited Technology Experiment (LTE) at SSC-Charleston in the summer of 2010. Experimental results reported here for NAS provide proof-of-concept validation of its ability to afford timely, accurate network awareness for tactical applications, and illustrate the benefits of network awareness to command and control applications in Naval settings.

Information utility in mission-oriented networks

In recent years, utility-oriented resource allocation for wired and wireless networks has been extensively studied. A key goal of utility based analysis is to provide evaluation criteria for efficient network operation based on subjective user assessments such as usefulness and value of data. Yet the vast majority of the prior work has focused on topics like mathematical functions (concave or otherwise) of network metrics such as bandwidth, delay, packet loss, etc., or information entropy, or user-perceived quality (e.g., MOS) for interpreting utility. In this work1, we propose an alternative mission-oriented definition and metric for utility that is based on the accuracy and speed at which tasks are completed, which we believe is closer to the intent of utility-based analysis. Like previous work, our definition supports the design and engineering of networks by mapping utility metrics to the typical network design metrics (bandwidth, delay, loss, etc). Unlike previous work, it also permits simple solutions to important questions such as the joint utility or usefulness of different data streams, the impact on the utility of one data stream by another, and cross-sensory utility (e.g., the impact of a side audio channel on an image processing task, or a side video channel on an audio processing task). We present a novel experimental approach to the design of such experiments, and provide measurement results. The results quantify the effects of information encoding and the impairments incurred during transmission through imperfect networks on the informations usefulness to end-users in terms of being able to complete tasks correctly and on time.