In the early stages of wireless sensor networks (WSNs), low data rate traffic patterns are assumed as applications have a single purpose with simple sensing task and data packets are generated at a rate of minutes or hours. As such, most of the proposed communication protocols focus on energy efficiency rather than high throughput. Emerging high data rate applications motivate bulk data transfer protocols to achieve high throughput. The basic idea is to enable nodes to transmit a sequence of packets in burst once they obtain a medium. However, due to the low-power, low-cost nature, the transceiver used in wireless sensor networks is prone to packet loss. Especially when the transmitters are mobile, packet loss becomes worse. To reduce the energy expenditure caused by packet loss and retransmission, a burst transmission scheme is required that can adapt to the link dynamics and estimate the number of packets to transmit in burst. As the mobile node is moving within the network, it cannot always maintain a stable link with one specific stationary node. When link deterioration is constantly detected, the mobile node has to initiate a handover process to seamlessly transfer the communication to a new relay node before the current link breaks. For this reason, it is vital for a mobile node to (1) determine whether a fluctuation in link quality eventually results in a disconnection, (2) foresee potential disconnection well ahead of time and establish an alternative link before the disconnection occurs, and (3) seamlessly transfer communication to the new link.
In this dissertation, we focus on dealing with burst transmission and handover issues in low power mobile wireless sensor networks. To this end, we begin with designing a novel mobility enabled testing framework as the evaluation testbed for all our remaining studies. We then perform an empirical study to investigate the link characteristics in mobile environments. Using these observations as guidelines, we propose three algorithms related to mobility that will improve network performance in terms of latency and throughput:
i) Mobility Enabled Testing Framework (MobiLab). Considering the high fluctuation of link quality during mobility, protocols supporting mobile wireless sensor nodes should be rigorously tested to ensure that they produce predictable outcomes before actual deployment. Furthermore, considering the typical size of wireless sensor networks and the number of parameters that can be configured or tuned, conducting repeated and reproducible experiments can be both time consuming and costly. The conventional method for evaluating the performance of different protocols and algorithms under different network configurations is to change the source code and reprogram the testbed, which requires considerable effort. To this end, we present a mobility enabled testbed for carrying out repeated and reproducible experiments, independent of the application or protocol types which should be tested. The testbed consists of, among others, a server side control station and a client side traffic ow controller which coordinates inter- and intra-experiment activities.
ii) Adaptive Burst Transmission Scheme for Dynamic Environment. Emerging high data rate applications motivate bulk data transfer protocol to achieve high throughput. The basic idea is to enable nodes to transmit a sequence of packets in burst once they obtain a medium. Due to the low-power and low-cost nature, the transceiver used in wireless sensor networks is prone to packet loss. When the transmitter is mobile, packet loss becomes even worse. The existing bulk data transfer protocols are not energy efficient since they keep their radios on even while a large number of consecutive packet losses occur. To address this challenge, we propose an adaptive burst transmission scheme (ABTS). In the design of the ABTS, we estimate the expected duration in which the quality of a specific link remains stable using the conditional distribution function of the signal-to-noise ratio (SNR) of received acknowledgment packets. We exploit the expected duration to determine the number of packets to transmit in burst and the duration of the sleeping period.
iii) Kalman Filter Based Handover Triggering Algorithm (KMF). Maintaining a stable link in mobile wireless sensor network is challenging. In the design of the KMF, we utilized combined link quality metrics in physical and link layers, such as Received Signal Strength Indicator (RSSI) and packet success rate (PSR), to estimate link quality fluctuation online. Then Kalman filter is adopted to predict link dynamics ahead of time. If a predicted link quality fulfills handover trigger criterion, a handover process will be initiated to discover alternative relay nodes and establish a new link before the disconnection occurs.
iv) Mobile Sender Initiated MAC Protocol (MSI-MAC). In cellular networks, mobile stations are always associated with the nearest base station through intra- and inter-cellular handover. The underlying process is that the quality of an established link is continually evaluated and handover decisions are made by resource rich base stations. In wireless sensor networks, should a seamless handover be carried out, the task has to be accomplished by energy-constraint, resource-limited, and low-power wireless sensor nodes in a distributed manner. To this end, we present MSI-MAC, a mobile sender initiated MAC protocol to enable seamless handover.
Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:32057 |
Date | 29 October 2018 |
Creators | Wen, Jianjun |
Contributors | Schill, Alexander, Hogrefe, Dieter, Jorswieck, Eduard A., Technische Universität Dresden |
Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
Language | English |
Detected Language | English |
Type | doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0014 seconds