Patrik Moravek

Received signal strength uncertainty in energy-aware localization in wireless sensor networks

In wireless sensor networks where the nodes are equipped with limited energy sources, energy consumption is a crucial parameter of each part of a node activity. This paper deals with node localization based on received signal strength (RSS). The signal power at receiving node is influenced by several stochastic factors. To eliminate so called RSS uncertainty, several measurements of signal strength must be performed to get the value to be used in distance calculation. Multiple packet transmission means bigger energy consumption during distance estimation. This paper presents the analysis of energy consumption during the signal strength measurement with respect to the accuracy of RSS estimation.

Investigation of radio channel uncertainty in distance estimation in wireless sensor networks

The distance estimation between nodes is a crucial requirement for localization and object tracking. Received signal strength (RSS) measurement is one of the used methods for the distance estimation in wireless networks. Its main advantage is that there are no additional hardware requirements. This paper describes a lateration approach for localization and distance estimation using RSS. For the purpose of investigation of RSS uncertainty, several scenarios were designed for both indoor and outdoor measurements. The first set of RSS measurement scenarios was proposed with the intention of hardware independent investigation of radio channel. For the second set of measurements, we employed IRIS sensor nodes to evaluate the distance estimation with certain devices. The experiments considered also obstacles in the radio channel. The results obtained in the proposed scenarios present usability of the method under different conditions. There is also a signal propagation model constructed from measured data at a node, which subsequently serves for distance determination.

Study and performance of localization methods in IP based networks: Vivaldi algorithm

This paper deals with station localization in distributed applications in the Internet. Several localization techniques are described with their advantages and disadvantages. This paper is focused on the sophisticated Vivaldi algorithm and its variations. A special simulation tool was developed in order to simulate the influence of configuration parameters and setting to algorithm performance. Several tests were performed to examine how both convergence and accuracy of localization process are affected by different settings of algorithm constants and by the number of reference points.

Measurement of LowPAN Network Coexistence with Home Microwave Appliances in Laboratory and Home Environments

Due to a low installation cost of communication infrastructure, Low Power Wireless Area Networks (low PAN) presents an efficient solution for home and building automation. The low PANs operate in a microwave radio band that is often shared by several home appliances. Laptops, wireless routers, cordless phones and microwave ovens occupy the 2.4 GHz band without any channel allocation and thus interoperability with the low PAN systems presents a significant challenge. Therefore, we have performed several laboratory experiments to study the coexistence of low PAN devices with WiFi equipped laptop, WiFi router and microwave oven. The performance evaluation was based on a measurement of Packet Delivery Ratio (PDR) metric. Results shows that presence of the home appliances in the low PAN environment significantly affects the low PAN communication efficiency. Furthermore, we have deployed Zigbee based wireless network in the home environment to study the communication performance in area perturbed by common home radio interferences and obstacles. The lessons learned from network deployment are also presented within the paper.

Energy demands of 802.15.4/ZigBee communication with IRIS sensor motes

Limited energy sources of nodes in wireless sensor networks require a careful consideration of energy consumption of all processes during the sensors deployment. To analyze energy consumption and to predict a lifetime of a network, a comprehensive energy model based on commercial products is necessary. Therefore, we have focused on the analysis of energy consumption during RF communication with particular 802.15.4 compliant nodes. We have designed experimental testbed and explored the scenario of node association and data transmission. For the measurement we used IRIS sensor nodes with 802.15.4/Zigbee protocol stack and shunt connection. The results show that energy consumption cannot be calculated only from datasheet values of current drain and length of a packet but intervals of listening and waiting between transmissions play important role as well.

Feedback transmission in large-scale IPTV sessions

In this paper, we introduce the Tree Transmission Protocol (TTP), which we use for the organization of a tree structure consisting of IPTV nodes for hierarchical feedback aggregation in large-scale IPTV (Internet Protocol Television) systems. The feedback transmission is based on the hierarchical feedback aggregation algorithm, which gives a lower feedback reporting interval compared with the standard feedback transmission in the RTP/RTCP protocol stack. We use the TTP protocol to establish a tree structure by grouping IPTV nodes into groups using a defined set of parameters. One of the parameters we use is the logical position of the IPTV nodes. The motivation for utilizing this parameter in the TTP protocol is to optimize communication in the created overlay network for feedback transmission. We use the Vivaldi algorithm for the localization of the IPTV nodes. In this paper, we also describe the implementation of the TTP protocol on the PlanetLab experimental network as well as the application, we used for the visualization of the established tree.

Centralized boundary discovery algorithms for anchor-free localization in wireless sensor networks

A study of localization techniques in the wireless sensor network are a subject of the many research team in the world. Most of the solutions propose the optimization in term of the accuracy of the coordinates assignment improvement. However, the other very important metric considering the energy cost of the localization process also exists, but nowadays there is a lack of this work regarding to the study of localization in the energy consumption terms. This paper deals with the investigation of the anchor-free localization technique AFL and introduces the results from the simulations, where the energy cost of the several initial localization processes was investigated. We have focused on the work, with the centralized boundary discovery CBD algorithm to study its portion on the entire energy consumption.

Open-packet analyser platform for wireless sensor networks based on IEEE 802.15.4

Wireless sensor networks consist of dozens of tiny constrained devices, which transfer data encapsulated into the packets over the air. It is crucial to have a packet analyser at the hand for real development and debug purposes. This article gives an overview of the available tools for packet analysis and ongoing research focused on open packet analyser development. We proposed two platforms one with Ethernet and second one with a serial interface. The latter has been evaluated. The results show that the analyser is able to work with the maximum throughput for IEEE 802.15.4.

Optimization of boundary recognition algorithms for wireless sensor network applications

Recognition of network boundaries in wireless sensor network (WSN) applications is crucial from two reasons. First, a knowledge of boundary nodes can help to increase a security performance of sensor network. And second it can significantly contributes to the energy conservation of inner nodes that can be switched into the sleeping modes during no activity is presented on the network boundaries. Furthermore, the knowledge of boundary nodes is necessary condition for some localization algorithms applied into the WSN environment. We proposed two novel algorithms referred to as BRC and BRBC and compare their recognition efficiency with three know related algorithms. The results of simulations showed that all algorithms achieved high ratio of success detections, however it is shown that only the novel algorithms achieved the low ratio of the false detections.

Establishment of reference localization system based on boundary recognition of wireless sensor network

Understanding the energy consumption of individual tasks in wireless sensor networks (WSN) is an important aspect for a network deployment. The fundamental task of WSNs is to gather data in the long-term manner. All additive processes such as routing and localization should not significantly affect the life time of the WSN. It was already proved that concurrent anchor free localization algorithms have great localization accuracy but also suffer from the high communication cost. We have performed a detailed analysis of a well-known AFL (Anchor-Free Localization) algorithm to show its excessive energy dissipation. We have proposed a novel BRL (Boundary Recognition aided localization) that significantly optimizes a communication cost of the selection of the reference nodes which are used for subsequent nodes localization. The main idea of the algorithm is to successfully identify the nodes lying on the boundary. Only these nodes subsequently cooperate in the reference nodes selection process and thus decrease the energy load of the network.