Querying sensor networks using ad hoc mobile devices: A two-layer networking approach

Abstract

Abstract-An interplay between mobile devices and static sensor nodes is envisioned in the near future. This will enable a heterogeneous design space that can offset the stringent resource and power constraints encountered in traditional static sensor networks by taking advantage of the more powerful mobile devices. We present a systematic framework for end-to-end query processing, using a two-layer architecture that consists of mobile devices at the upper layer and static sensor nodes at the bottom layer. The framework employs a "PULL" query model that contains staged operations including query generation, query routing, query injection, and query result routing. Each of these stages of query processing is discussed with an emphasis on techniques for energyefficient query injection and query result routing with location-ignorant sensor nodes. The techniques leverage the mobility and transmission flexibility of mobile objects at the upper layer. Numeric and simulation results are provided to support the proposed methods.Index Terms-mobile objects, opportunistic query processing, sensor networks. INTRODUCTIONStringent power constraints and limited computation and communication capabilities are key issues in the development of sensor networks. For example, a Berkeley Mote powered by two AA batteries can operate for about one year in the idle state, but only one week when fully loaded. In addition, large-scale sensor network applications impose a demand for cheap, small, lowpower sensor nodes, making it impractical to equip sensor nodes with GPS-receivers. While various localization techniques are evolving [1], localization for very large-scale sensor networks is still in the research stage, especially when there is the need to obtain accurate location information under restrained energy conditions. Thus, we focus on location-ignorant sensor networks, which impose challenges on various aspects for query processing in the network, including both query routing and results gathering.In contrast to sensor networks, mobile wireless networks have much relaxed power constraints. Various communication technologies, including Cellular, Wi-Fi, WiMAX, and Bluetooth, have been developed for connecting billions of electronic products, such as PDAs, cell phones, laptops, and cars. Moreover, GPS-receivers are reasonably affordable on such mobile devices. We observe that a heterogeneous design space consisting of both static sensor nodes and mobile devices can successfully offset the stringent resource and power constraints in traditional sensor networks. Such an interplay between mobile (ad hoc) networks and sensor networks also links the mobile devices as query requesters and data consumers directly to sensor networks that are responsible for sensing the physical world. We propose a framework for query processing that is based on the "PULL" query model [2] [3] in a two-layer network structure, including a mobile network at the upper layer and a wireless sensor network at the bottom layer. Hereafter, the term "mobile objects" refe...

Performance analysis of mobile sampling and broadcast scheduling in wireless sensor networks

“…Al., [29] has presented an efficient query processing composed of mobile objects operating within the context of a sensor rich environment.…”

“…al., [29] possesses displayed some sort of systematic construction intended for end-to-end query finalizing by utilizing some sort of two-layer construction in which comprises of cellular phones in the higher level and predetermined sensor nodes in the foundation level.…”

A novel routing recovery strategy based on particle swarm algorithm for wireless sensor networks with multiple mobile sinks

“…The SEAD as introduced in [3] is a mechanism for routing sensing data to mobile sinks. The idea is to construct a minimum Steiner tree for the mobile sink and designate some nodes on the tree as access points.…”

A particle swarm algorithm based routing recovery method for mobile sink wireless sensor networks

“…A data dissemination protocol, SEAD [8], is proposed to minimize energy consumption in building the dissemination tree and in disseminating data to multiple mobile sink. M. Gatzianas [9] presents an efficient distributed algorithm for computing the maximum lifetime of a sensor network which routes data to a mobile sink.…”