Context-aware and personalized event filtering for low-overhead continuous remote health monitoring

Distrib Parallel Databases

Misra

2012

Self Cite

There is a growing interest in applications that utilize continuous sensing of individual activity or context, via sensors embedded or associated with personal mobile devices (e.g., smartphones). Reducing the energy overheads of sensor data acquisition and processing is essential to ensure the successful continuous operation of such applications, especially on battery-limited mobile devices. To achieve this goal, this paper presents a framework, called ACQUA, for 'acquisition-cost' aware continuous query processing. ACQUA replaces the current paradigm, where the data is typically streamed (pushed) from the sensors to the one or more smartphones, with a pull-based asynchronous model, where a smartphone retrieves appropriate blocks of relevant sensor data from individual sensors, as an integral part of the query evaluation process. We describe algorithms that dynamically optimize the sequence (for complex stream queries with conjunctive and disjunctive predicates) in which such sensor data streams are retrieved by the query evaluation component, based on a combination of a) the communication cost & selectivity properties of individual sensor streams, and b) the occurrence of the stream predicates in multiple concurrently executing queries. We also show how a transformation of a group of stream queries into a disjunctive normal form provides us with significantly greater degrees of freedom in choosing this sequence, in which individual sensor streams are retrieved and evaluated. While the algorithms can apply to a broad category of sensor-based applications, we specifically demon- strate their application to a scenario where multiple stream processing queries execute on a single smartphone, with the sensors transferring their data over an appropriate PAN technology, such as Bluetooth or IEEE 802.11. Extensive simulation experiments indicate that ACQUA's intelligent batch-oriented data acquisition process can result in as much as 80% reduction in the energy overhead of continuous query processing, without any loss in the fidelity of the processing logic.

Section: Introductionmentioning

confidence: 99%

Adaptive data acquisition strategies for energy-efficient, smartphone-based, continuous processing of sensor streams

Distrib Parallel Databases

Misra

2012

Self Cite

“…The use of complex event processing of sensor data streams on a smartphone for detecting context on a smartphone has been previously explored in system prototypes such as Harmoni [3] (which used such context to dynamically change the definition of anomalous medical states) and CenceMe [2] (which applied rich operators on audio and accelerometer sensor streams to identify pre-defined human activities). To further reduce the energy overheads, the MediAlly prototype [8] used such inferred context to dynamically activate the collection of data from other external sensors.…”

Section: Related and Prior Workmentioning

confidence: 99%

“…Suppose each stream is associated with a buffer and the buffers are initially empty. At time t = 7, suppose the evaluation of Q3 requires the retrieval of the data elements A : (2,7], B : (3,7], C : (6,7]. Now, at time t = 12, suppose MAX(B, 4) > 100 is evaluated first and is false; we need proceed to evaluate the right subtree of the top OR node (in Fig.…”

Section: The Stream-oriented Query Modelmentioning

confidence: 99%

“…This computing paradigm is broadly referred to as a "3-tier" monitoring architecture, with the smartphone acting as a pervasive, mobile gateway that makes the information captured by the PAN-connected sensors available to the backend "cloud" logic. A major role of the smartphone in this paradigm is to perform embedded event processing on the sensor data streams to extract appropriate individual context in near-real time, for use in a variety of applications, ranging from automatic activity updates for social networking applications (e.g., [2]) to dynamic adaptation of reporting thresholds for adaptive remote health monitoring (e.g., [3]). …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimizing Sensor Data Acquisition for Energy-Efficient Smartphone-Based Continuous Event Processing

Misra

2011 IEEE 12th International Conference on Mobile Data Management

2011

Self Cite

Abstract-Many pervasive applications, such as activity recognition or remote wellness monitoring, utilize a personal mobile device (aka smartphone) to perform continuous processing of data streams acquired from locally-connected, wearable, sensors. To ensure the continuous operation of such applications on a battery-limited mobile device, it is essential to dramatically reduce the energy overhead associated with the process of sensor data acquisition and processing. To achieve this goal, this paper introduces a technique of 'acquisition-cost' aware continuous query processing, as part of the Acquisition Cost-Aware Query Adaptation (ACQUA) framework. ACQUA replaces the current paradigm, where the data is typically streamed (pushed) from the sensors to the smartphone, with a pull-based asynchronous model, where the phone retrieves appropriate blocks of sensor data from individual sensors, only when the stream elements are judged to be relevant to the query being processed. We describe algorithms that dynamically optimize the sequence (for complex stream queries with conjunctive and disjunctive predicates) in which such sensor data streams are retrieved by the phone, based on a combination of the communication cost and selectivity properties of individual sensor streams. Simulation experiments indicate that this approach can result in 70% reduction in the energy overhead of continuous query processing, without affecting the fidelity of the processing logic.

“…For instance, smartphones are equipped with increasingly sophisticated sensors (e.g., GPS, accelerometer, gyroscope, microphone) that enable near real-time sensing of an individual's activity or environmental context. A smartphone can then perform embedded query processing on the sensor data streams, e.g., for social networking [1], remote health monitoring [2]. The continuous processing of streams, even when data rates are moderate (such as for GPS or accelerometer data), can cause commercial smartphone batteries to be depleted in a few hours [3].…”

Section: Introductionmentioning

confidence: 99%

Cost-Optimal Execution of Boolean Query Trees with Shared Streams

Casanova

2014 IEEE 28th International Parallel and Distributed Processing Symposium

Robert

et al. 2014

Abstract-The processing of queries expressed as trees of boolean operators applied to predicates on sensor data streams has several applications in mobile computing. Sensor data must be retrieved from the sensors, which incurs a cost, e.g., an energy expense that depletes the battery of a mobile query processing device. The objective is to determine the order in which predicates should be evaluated so as to shortcut part of the query evaluation and minimize the expected cost. This problem has been studied assuming that each data stream occurs at a single predicate. In this work we remove this assumption since it does not necessarily hold in practice. Our main results are an optimal algorithm for single-level trees and a proof of NP-completeness for DNF trees. For DNF trees, however, we show that there is an optimal predicate evaluation order that corresponds to a depth-first traversal. This result provides inspiration for a class of heuristics. We show that one of these heuristics largely outperforms other sensible heuristics, including a heuristic proposed in previous work.