CAREER: Data Management for Ad-Hoc Geosensor Networks

Environmental phenomena, such as fires, poisonous gases, and oil spills, can be detected by wireless sensor networks (WSNs) that cover the geographical area of the phenomena. These sensors collaboratively monitor the area to detect the sensors' readings that deviate from normal reading patterns after which a phenomena is declared. This research proposes a distributed algorithm to detect dynamic phenomena using mobile WSNs under the assumption that there is no centralized server to collect and aggregate the sensors data. Therefore, the sensors self-organize into disjoint groups by first electing a few sensors to be group heads (GHs) and then the rest of the sensors group themselves with the nearest GH. Each group of sensors detect phenomena locally. Then, the GHs communicate the detected local phenomena information among themselves to aggregate the information and detect the global phenomena.

In biological research domain like "Study of animal social behavior" and "Wild life migration" object tracking sensor networks are used. In this application concentrating on finding the group of object with similar movement pattern using distributed mining techniques. But generally WSN concentrating on finding moving patterns of single object or all objects. Tracking moving objects having two phases i.e. mining phase and cluster ensembling phase. In first phase of algorithm we find movement patterns based on local data then we are identifying new term of similarity measure of to computing the similarity of moving objects and find relationship between them. In second phase algorithm combine the local grouping results to derive the group relationship from global view. We hope that the final output shows that the proposed mining algorithm achieves good grouping quality and the mining technique helps reduce the energy consumption by reducing the amount of data to be transmitted.

Lecture Notes in Computer Science, 2015

Festivals and big scale events are becoming more and more popular, they can attract thousands of spectators. Ensuring the safety of the crowd has become a top priority to many organisers after the multitude of dramatic accidents that resulted in losses in human lives. Monitoring the crowd via smartphones is a relatively new technique that emerged recently with the capabilities of mobile phones to transmit their GPS location data. We present a novel approach, based on the local crowd pressure, combined with the detection of groups in a crowd, to detect critical situations and propose evacuation plans that does not separate groups of people that are together. Groups were detected using DBSCAN clustering algorithm with 80 % accuracy. Location acquisition was tested during the Campus Fever event, and 87 % of the collected data had an accuracy lower than 10 m while 29 % of the total data had 5 m of accuracy. During 2 h of monitoring, activity of the application, reduced the battery of 20 %.

Big Data Analytics, 2019

Increasing availability of location-based applications and sensor devices have necessitated quicker analysis of moving object data streams in order to identify patterns. The efficiency of currently available algorithms used in pattern detection is not adequate to handle large scale data streams that are increasingly available. We focus on the particular problem of flock detection in moving object data and our goal is to detect flocks quickly and using fast algorithms. Firstly, we employ a triangular grid to reduce the search space of clustering algorithms which has a significant effect in case of dense objects. As a second step, we implement a modified flock membership function and pipeline creation that ensures better memory and time performance during cluster detection. We show that this refinement also improves the rate of flock detection. Finally, we parallelize our algorithm to further enhance the handling of massive data streams. Based on an extensive empirical evaluation of these algorithms across a variety of moving object data sets, we show that our method is significantly faster than the existing comparable methods over sliding windows. In particular, it requires lesser time to identify flocks and is 2-4 times faster thus confirming the efficiency and effectiveness of our approach.

Geosensor networks comprise small electro-mechanical devices that communicate over a wireless network. These devices collect environmental measures and send them to a base station. Energy consumption and data routing are critical factors for efficient geosensor networks. The usual cluster-based data routing protocols for sensor networks group the nodes based on their geographical closeness and aggregate their data to save energy. However, this clustering procedure does not produce the best data summaries. We propose to group the nodes into spatially homogeneous clusters, which consider both the geographical distance and the similarity of measurements between the nodes. Through simulated experiments, we have concluded that spatially homogeneous clusters produce better spatial zones identification and data summaries with a higher statistical quality if compared with the usual clustering methods. Besides, spatially homogeneous clustering can be seen as a tool for spatial sensor data mining, since their clusters represent the partition of the sensor field that has maximum internal homogeneity regarding the values of monitored variable. To make possible the use of our data-aware clustering proposal to collect the sensors' data, we present a design guideline for a cluster-based data routing protocol, the HR-DASH.

2001

This paper describes several ways sensor networks can benefit from geospatial information and identifies two research directions. First, better models of localization error, logical location, and communications costs are required to understand the interactions between spatial information and control and communications algorithms in sensor networks. Second, wider use of spatial information in densely deployed sensor networks will move sensor networking applications from simple tracking to object counting and area monitoring, and can enable data mining techniques sensor networks to accomplish "spatial sensor mining".

We present distributed system architecture for smartphone-based participatory sensing applications, where the computational load is distributed between the participating devices. The system allows using of static data to reduce the computational load further. Secondly, we present a conceptual participatory sensing application for detecting pedestrian flocks moving into the same direction in certain area, using this architecture. The flock detection is based on selecting an energy efficient set of sensors in the smartphone and available location-based static data, provided by previous computations by the smartphones or by external Web services. Initially, we believe this method reduces energy consumption in the participating devices beyond the previous approaches.

IEEE Systems Journal, 2018

Knowledge about people density and mobility patterns is the key element towards efficient urban development in smart cities. The main challenges in large-scale people tracking are the recognition of people density in a specific area and tracking the people flow path. To address these challenges, we present SenseFlow, a lightweight people tracking system for smart cities. SenseFlow utilizes off-the-shelf sensors which sniff probe requests periodically polled by user's smartphones in a passive manner. We demonstrate the feasibility of SenseFlow by building a proof-of-concept prototype and undertaking extensive evaluations in real-world settings. We deploy the system in one laboratory to study office hours of researchers, a crowded public area in a city to evaluate the scalability and performance ''in the wild'', and four classrooms in the university to monitor the number of students. We also evaluate SenseFlow with varying walking speeds and different models of smartphones to investigate the people flow tracking performance. This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.

2007

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

At present, research on mobile geosensor networks gains more and more attention in the geographic information science. Such networks may be deployed to monitor and capture the spatio-temporal variance of phenomena appearing in the geographic space. Each node in a network is either self-propelled or carried in space by agents. In this paper we take a closer look at three exemplary mobility strategies for self-propelled geosensors. We hypothesize that the ability of individual geosensors to adapt their mobility to stimuli in the physical world improves the model of the phenomenon being captured by a mobile geosensor network over a finite time period. An executable model provides empirical evidence for our assumption. However, the scientific inquiry and the resulting findings are at an early stage. There is a need to cope with more complex models to come around with grounded theories. Before we conclude our work, we formulate four directions for continuing the research presented herein.