Santashil PalChaudhuri

Safari: A self-organizing, hierarchical architecture for scalable ad hoc networking

As wireless devices become more pervasive, mobile ad hoc networks are gaining importance, motivating the development of highly scalable ad hoc networking techniques. In this paper, we give an overview of the Safari architecture for highly scalable ad hoc network routing, and we present the design and evaluation of a specific realization of the Safari architecture, which we call Masai. We focus in this work on the scalability of learning and maintaining the routing state necessary for a large ad hoc network. The Safari architecture provides scalable ad hoc network routing, the seamless integration of infrastructure networks when and where they are available, and the support of self-organizing, decentralized network applications. Safaris architecture is based on (1) a self-organizing network hierarchy that recursively groups participating nodes into an adaptive, locality-based hierarchy of cells; (2) a routing protocol that uses a hybrid of proactive and reactive routing information in the cells and scales to much larger numbers of nodes than previous ad hoc network routing protocols; and (3) a distributed hash table grounded in the network hierarchy, which supports decentralized network services on top of Safari. We evaluate the Masai realization of the Safari architecture through analysis and simulations, under varying network sizes, fraction of mobile nodes, and offered traffic loads. Compared to both the DSR and the L+ routing protocols, our results show that the Masai realization of the Safari architecture is significantly more scalable, with much higher packet delivery ratio and lower overhead.

Power mode scheduling for ad hoc networks

An ad hoc network is a group of mobile wireless nodes that cooperatively form a network among themselves without any fixed infrastructure. Increasingly, power consumption within ad hoc networks is becoming a core issue for these low-power mobile devices. This paper focuses on a novel approach for energy conservation within the routing protocol of the ad hoc network. A wireless network interface in sleep mode expends an order of magnitude less power than in idle mode, but no packets can be sent or received while in sleep mode. We propose two probabilistic algorithms for scheduling transition from idle mode to sleep mode. Performance evaluation of these strategies show a substantial reduction in power usage, with only a slight decrease in performance.

Design of adaptive overlays for multi-scale communication in sensor networks

In wireless sensor networks, energy and communication bandwidth are precious resources. Traditionally, layering has been used as a design principle for network stacks; hence routing protocols assume no knowledge of the application behavior in the sensor node. In resource-constrained sensor-nodes, there is simultaneously a need and an opportunity to optimize the protocol to match the application. In this paper, we design a network architecture that efficiently supports multi-scale communication and collaboration among sensors. The architecture complements the previously proposed Abstract Regions architecture for local communication and collaboration. We design a self-organizing hierarchical overlay that scales to a large number of sensors and enables multi-resolution collaboration. We design effective Network Programming Interfaces to simplify the development of applications on top of the architecture; these interfaces are efficiently implemented in the network layer. The overlay hierarchy can adapt to match the collaboration requirements of the application and data both temporally and spatially. We present an initial evaluation of our design under simulation to show that it leads to reduced communication overhead, thereby saving energy. We are currently building our architecture in the TinyOS environment to demonstrate its effectiveness.

System Support for Cross-Layering in Sensor Network Stack

Wireless Sensor Networks are deployed in demanding environments, where application requirements as well as network conditions may change dynamically. Thus the protocol stack in each node of the sensor network has to be able to adapt to these changing conditions. Historically, protocol stacks have been designed with strict layering and strong interface between the layers leading to a robust design. However, cross-layer information sharing could help the protocol modules to make informed decisions and adapt to changing environmental conditions. There have been ad hoc approaches to facilitating cross-layer cooperation for adaptability. However, there has been no concerted effort at providing a uniform framework for cross-layer adaptability that preserves the modularity of a conventional protocol stack. This paper presents a novel service, information exchange service (IES), as a framework for cross-module information exchange. IES is a centrally controlled bulletin-board where different modules can post available data, or request for useful information, and get notified when the information becomes available. IES is integrated into the proposed SensorStack architecture that preserves the benefits of layering while facilitating adaptability. IES has been implemented in TinyOS and Linux, to show both the feasibility of the design as well as demonstrate the utility of cross-layering to increase application longevity.

Design and Performance of PRAN: A System for Physical Implementation of Ad Hoc Network Routing Protocols

Simulation and physical implementation are both valuable tools in evaluating ad hoc network routing protocols, but neither alone is sufficient. In this paper, we present the design and performance of PRAN, a new system for the physical implementation of ad hoc network routing protocols that unifies these two types of evaluation methodologies. PRAN (physical realization of ad hoc networks) allows existing simulation models of ad hoc network routing protocols to be used - without modification - to create a physical implementation of the same protocol. We have evaluated the simplicity and portability of our approach across multiple protocols and multiple operating systems through example implementations in PRAN of the DSR and AODV routing protocols in FreeBSD and Linux using the standard existing, unmodified ns-2 simulation model of each. We illustrate the ability of the resulting protocol implementations to handle real, demanding applications by describing a demonstration with this DSR implementation transmitting real-time video streams over a multihop mobile ad hoc network; the demonstration features mobile robots being remotely operated based on the real-time video stream transmitted from the robot over the network. We also present a detailed performance evaluation of PRAN to show the feasibility of our architecture

An adaptive scheduling protocol for multi-scale sensor network architecture

In self-organizing networks of battery-powered wireless sensors that can sense, process, and communicate, energy is the most crucial and scarce resource. However, since sensor network applications generally exhibit specific limited behaviors, there is both a need and an opportunity for adapting the network architecture to match the application in order to optimize resource utilization. Many applications-such as large-scale collaborative sensing, distributed signal processing, and distributed data assimilation-require sensor data to be available at multiple resolutions, or allow fidelity to be traded-off for energy efficiency. We believe that cross-layering and application-specific adaptability are the primary design principles needed to build sensor networking protocols. In previous work, we proposed an adaptive cross-layered self-organizing hierarchical data service under COMPASS architecture, that enables multi-scale collaboration and communication. In this paper we propose a time division multiplexing medium scheduling protocol tailored for this hierarchical data service, to take advantage of the communication and routing characteristics to achieve close to optimal latency and energy usage. We present an analytical proof on the bounds achieved by the protocol and analyze the performance via detailed simulations.

Conferences - Mobile computing at the beach

The Fifth IEEE Workshop on Mobile Computing Systems and Applications descended on the Beach Resort in Monterey, California, for a series of technical discussions of new research in pervasive and mobile computing.