Asynchronous Data Aggregation for Real-Time Monitoring in Sensor Networks

Eager, Derek L.; Makaroff, Dwight

doi:10.1007/978-3-540-72606-7_7

Cited by 6 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous work shows that for asynchronous unicast-based aggregation, all parameters except R can be fixed to certain values that yield good performance for a broad range of network configurations [3]. Through experimentation, we find that the same conclusion holds for asynchronous broadcast-based aggregation as well.…”

Section: Principal Performance Comparisonsupporting

confidence: 62%

“…The unicast-based synchronous protocol is a variant proposed in previous work [3] of the original synchronous aggregation protocol as used in TAG [1]. Aggregation is performed over a tree rooted at the sink.…”

Section: Unicast-basedmentioning

confidence: 99%

“…Here it has been assumed that all nodes agree on the same base time T 0 defining the beginning of the first round. It has been found that randomizing transmissions over λI yields better performance than when all nodes at the same tree level attempt to transmit at approximately the same time, and that setting λ to 0.8 yields good performance over a wide variety of network configurations [3].…”

Section: Unicast-basedmentioning

confidence: 99%

“…In contrast, in the asynchronous aggregation protocols considered here each node adaptively determines when to transmit based on its history of past packet receptions from its children in the aggregation tree. We use the unicast-based "adaptive asynchronous" aggregation protocol proposed in previous work [3] as a basis for our new broadcast-based asynchronous aggregation protocol. The former protocol is reviewed in Section 3.1, while Section 3.2 describes the design of the new broadcast-based protocol.…”

Section: Asynchronous Aggregationmentioning

confidence: 99%

See 3 more Smart Citations

Aggregation Protocols for High Rate, Low Delay Data Collection in Sensor Networks

Eager

Makaroff

2009

Networking 2009

Self Cite

View full text Add to dashboard Cite

Abstract. Sensor network applications commonly require sensor data to be periodically collected. Aggregation protocols can make this process considerably more efficient. This paper considers the problem of devising aggregation protocols for applications that must achieve as "real-time" a view of the monitored area as possible, entailing a high sampling rate and a low data collection delay, at the possible cost of some modest amount of data loss. We examine in particular broadcast-based protocols that minimize the number of packet transmissions, relying on multipath delivery rather than ARQ for reliability, and consider the question of whether such protocols can achieve lower collection delays and support higher sampling rates than conventional aggregation protocols. Our results suggest that broadcast-based protocols can yield significantly improved performance in some scenarios, when sensor data can be aggregated into packets of size that is independent (or largely independent) of the number of values being aggregated.

show abstract

Section: Principal Performance Comparisonsupporting

confidence: 62%

Section: Unicast-basedmentioning

confidence: 99%

Section: Unicast-basedmentioning

confidence: 99%

Section: Asynchronous Aggregationmentioning

confidence: 99%

See 2 more Smart Citations

Aggregation Protocols for High Rate, Low Delay Data Collection in Sensor Networks

Eager

Makaroff

2009

Networking 2009

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the unicast-based protocol, an aggregation tree is formed among all nodes, with the sink as the root [12], [16]. Each round, the interior tree nodes aggregate the data received from their child nodes, and forward their aggregate packets towards the sink.…”

Section: Introductionmentioning

confidence: 99%

Coverage preserving aggregation protocols for dense sensor networks

Eager

Makaroff

2010

IEEE Local Computer Network Conference

Self Cite

View full text Add to dashboard Cite

Abstract-Sensor networks are often deployed more densely than would be minimally required. In such cases, node scheduling protocols can be used to determine which nodes are active, and which nodes sleep so as to conserve energy and prolong network lifetime. A drawback of node scheduling approaches, however, is delay due to node or communication failure(s), and subsequent wake-up of replacement node(s), during which monitoring coverage of some sub-region may be lost. This paper proposes an alternative approach for use in contexts in which the objective is to periodically collect sensing data that completely covers a region of interest. In the proposed approach, nodes dynamically determine during each round of data collection whether they should transmit their data, or whether their area is covered by neighbouring nodes that have already transmitted. Both unicast and broadcast-based data collection protocols are designed, and their performance compared using simulation to that of data collection protocols relying on node scheduling. Our results suggest that the coverage-preserving broadcast-based protocol can greatly improve reliability at the potential cost of increased traffic volume owing to non-minimal selection of transmitting nodes.

show abstract

An Efficient Data Aggregation Method for Event-Driven WSNs: A Modeling and Evaluation Approach

Kamarei

Hajimohammadi

Patooghy

et al. 2015

Wireless Pers Commun

View full text Add to dashboard Cite

This paper proposes and models an efficient data aggregation method for wireless sensor networks (WSNs). In the proposed data aggregation method, every cluster head (CH) node incorporates a local forwarding history to decide whether to forward or to drop a recently received packet. When a new packet arrives at a CH node, a threshold value is calculated based on the information of the forwarding history; then, a random number is generated and compared with the threshold value to determine whether the packet should be dropped or not. In fact, the CH node forwards the new packet with the probability of 1 -p and drops it with probability of p where the parameter p is determined based on the forwarding history. In order The submitted manuscript is an extended version of the work previously proposed in [1]. The extensions of the current version include: (1) a wide range of simulation experiments added to the current version. To do this the proposed data aggregation method is compared with previously proposed methods i.e., RAG, SPEED, DASDR, and RDAG is terms of network performance, data aggregation rate, event miss rate. The previous version of the paper does not offer any comparison experiments, (2) an analytical performance model is proposed and validated in the paper. The proposed model estimates network performance, data aggregation rate, event miss rate of WSN using the proposed data aggregation method with an acceptable prediction error. The model provides a remarkable speed up in the evaluation process of data aggregation methods in the field of wireless sensor networks.to evaluate the proposed data aggregation method, two approaches consisting of simulation and analytical modeling are used. Various scenarios are considered in simulations conducted with NS2 software to compare the proposed data aggregation method with four previously proposed methods. Results reveal that the proposed method (1) aggregates more than 70 % of redundant packets, (2) reduces the network end-to-end delay by at least 22 %, and (3) reduces the missed event rate compared with the other methods. The proposed method is also evaluated by means of an analytical model based on queuing networks. The model accurately estimates the network performance utilizing the proposed data aggregation method. Comparisons of the results obtained by the proposed model and simulations confirm that the proposed model has at most 7 % prediction error. The proposed model allows WSN designers to easily achieve useful information about their networks before the establishment and manufacturing of the networks.

show abstract

Asynchronous Data Aggregation for Real-Time Monitoring in Sensor Networks

Cited by 6 publications

References 15 publications

Aggregation Protocols for High Rate, Low Delay Data Collection in Sensor Networks

Aggregation Protocols for High Rate, Low Delay Data Collection in Sensor Networks

Coverage preserving aggregation protocols for dense sensor networks

An Efficient Data Aggregation Method for Event-Driven WSNs: A Modeling and Evaluation Approach

Contact Info

Product

Resources

About