Yong Yao

The cougar approach to in-network query processing in sensor networks

The widespread distribution and availability of small-scale sensors, actuators, and embedded processors is transforming the physical world into a computing platform. One such example is a sensor network consisting of a large number of sensor nodes that combine physical sensing capabilities such as temperature, light, or seismic sensors with networking and computation capabilities. Applications range from environmental control, warehouse inventory, and health care to military environments. Existing sensor networks assume that the sensors are preprogrammed and send data to a central frontend where the data is aggregated and stored for offline querying and analysis. This approach has two major drawbacks. First, the user cannot change the behavior of the system on the fly. Second, conservation of battery power is a major design factor, but a central system cannot make use of in-network programming, which trades costly communication for cheap local computation.In this paper, we introduce the Cougar approach to tasking sensor networks through declarative queries. Given a user query, a query optimizer generates an efficient query plan for in-network query processing, which can vastly reduce resource usage and thus extend the lifetime of a sensor network. In addition, since queries are asked in a declarative language, the user is shielded from the physical characteristics of the network. We give a short overview of sensor networks, propose a natural architecture for a data management system for sensor networks, and describe open research problems in this area.

Multi-query optimization for sensor networks

The widespread dissemination of small-scale sensor nodes has sparked interest in a powerful new database abstraction for sensor networks: Clients “program” the sensors through queries in a high-level declarative language permitting the system to perform the low-level optimizations necessary for energy-efficient query processing. In this paper we consider multi-query optimization for aggregate queries on sensor networks. We develop a set of distributed algorithms for processing multiple queries that incur minimum communication while observing the computational limitations of the sensor nodes. Our algorithms support incremental changes to the set of active queries and allow for local repairs to routes in response to node failures. A thorough experimental analysis shows that our approach results in significant energy savings, compared to previous work.

The Cougar Project: a work-in-progress report

We present an update on the status of the Cougar Sensor Database Project, in which we are investigating a database approach to sensor networks: Clients program the sensors through queries in a high-level declarative language (such as a variant of SQL). In this paper, we give an overview of our activities on energy-efficient data dissemination and query processing. Due to space constraints, we cannot present a full menu of results; instead, we decided to only whet the readers appetite with some problems in energy-efficient routing and in-network aggregation and some thoughts on how to approach them.

WaveScheduling: energy-efficient data dissemination for sensor networks

Sensor networks are being increasingly deployed for diverse monitoring applications. Event data are collected at various sensors and sent to selected storage nodes for further in-network processing. Since sensor nodes have strong constraints on their energy usage, this data transfer needs to be energy-efficient to maximize network lifetime. In this paper, we propose a novel methodology for trading energy versus latency in sensor database systems. We propose a new protocol that carefully schedules message transmissions so as to avoid collisions at the MAC layer. Since all nodes adhere to the schedule, their radios can be off most of the time and they only wake up during well-defined time intervals. We show how routing protocols can be optimized to interact symbiotically with the scheduling decisions, resulting in significant energy savings at the cost of higher latency. We demonstrate the effectiveness of our approach by means of a thorough simulation study.

Wave scheduling and routing in sensor networks

Sensor networks are being increasingly deployed for diverse monitoring applications. Event data are collected at various sensors and sent to selected storage nodes for further in-network processing. Since sensor nodes have strong constraints on their energy usage, this data transfer needs to be energy-efficient to maximize network lifetime. In this article, we propose a novel methodology for trading energy versus latency in sensor database systems. We propose a new protocol that carefully schedules message transmissions so as to avoid collisions at the MAC layer. Since all nodes adhere to the schedule, their radios can be off most of the time and only wake up during well-defined time intervals. We show how routing protocols can be optimized to interact symbiotically with scheduling decisions, resulting in significant energy savings at the cost of higher latency. We demonstrate the effectiveness of our approach by means of a thorough simulation study, using synthetic data as well as real-world traffic workloads.

Network scheduling for data archiving applications in sensor networks

Since data archiving in sensor networks is a communication intensive application, a careful power management of communication is of critical importance for such networks. An example is FPS, an adaptive power scheduling algorithm that combines slotted scheduling with a CSMA MAC [7]. In this paper, we first propose a new global power scheduling protocol called Multi-Flow Power Scheduling (MPS) that delivers more data and consumes less energy than existing power scheduling protocols. MPS sets up a transmission schedule through standard data aggregation and dissemination operations, however since it creates a global schedule it does not scale to large networks. We then present a new power scheduling protocol called Hybrid Power Scheduling (HPS) that retains the scalability of FPS while maintaining the energy efficiency and high data delivery rate of MPS. In a thorough simulation study, we compare HPS and MPS, and our results show the efficacy of HPS.

Directions in Multi−Query Optimization for Sensor Networks

The widespread dissemination of small-scale sensor nodes has sparked interest in a powerful new database abstraction for sensor networks: Clients “program” the sensors through queries in a high-level declarative language (such as a variant of SQL), and catalog management and query processing techniques abstract the user from the physical details of tasking the sensors. We call the resulting system a sensor data management system (SDMS). Sensor networks have important constraints on communication, computation and power consumption. Energy is the most valuable resource for unattended battery-powered nodes. Since radio communication consumes most of the available node power, our goal is to identify strategies that reduce network traffic. We give an overview of three distinct approaches to reducing the cost of processing aggregate queries in sensor networks: i) selection of suitable routes for collecting results of multiple queries, ii) data reduction techniques that exploit query commonalities and iii) a hybrid pull-push communication paradigm for query and result propagation. We pay particular attention to the third approach and present in detail an algorithm for finding a pull-push configuration that minimizes on expectation the network traffic. Experimental analysis shows that our algorithm offers significant energy savings.