AG Technische Informatik

19554 - Technische Informatik - Network Organization and Protocols (2,3,N)

This seminar is a block seminar, i.e. at the end of the semester, there will be two days when all talks are given in a block. During the semester, there will be deadlines for status reports, but no meetings of the complete seminar group.

Students participating will give a talk and write a report (7 pages as a general rule but each advisor may specify his own requirements). Besides that, they will review one or two reports written by other students and prepare questions to be asked after the corresponding talk. This is to stimulate discussions after the talks and make the seminar more worthwhile.

 

(19554)

Typ

Seminar

Dozent/in

Prof. Dr. Mesut Günes, Stephan Schröder

Institution

Institute of Computer Science

Freie Universität Berlin

Sprache

Deutsch

Semester

SS 08

Veranstaltungsumfang

2

Leistungspunkte

4

Raum

Takustraße 9

K 60

Zeit

  • Vorbesprechung: Montag, 4.2.2008, 16:00 Uhr (s.t.)
  • Blockseminar: Mittwoch, 18.6.2008, 9:00 - 13:00 Uhr

 

Voraussetzungen

Vordiplom or B.Sc., basic knowlegde in telematics

KVV page

Schedule

  • 4.2.2008:
    • Organizational Meeting - 16:00 Uhr (s.t.), Raum K60
    • Please read the literature linked below.
  • 6.2.2008:
    • Send a list with your five prefered topics in the order of declining preference to seminar-agtech[at]inf.fu-berlin.de.
  • 8.2.2008:
    • You will receive your topic via email.
  • 10-15.3.2008:
    • Meet your advisor to discuss content and structure of your report.
    • Optional chance to hand in the first version of your report to get feedback.
  • starting 1.4.2008:
    • Meet your adviser to receive feedback.
  • 2.5.2008:
    • Your report must be handed in to your advisor and seminar-agtech[at]inf.fu-berlin.de (see below for formatting instructions).
  • 6.5.2008:
    • You will receive three paper to review them.
  • 16.5.2008:
    • Your reviews must be handed in to your advisor and seminar-agtech[at]inf.fu-berlin.de(see below for a review form).
  • 20.5.2008:
    • You will get the reviews for your paper.
  • 13.6.2008:
    • Hand in your final report and your slides to your advisor and seminar-agtech[at]inf.fu-berlin.de.
  • 18.6.2008 9-13o'clock:
    • The seminar will take place in K60. Attendance is mandatory. The talks will be given according to this schedule:
      18.6.2008:
      9:00 - 9:30 Robert Richter
      9:30 - 10:00 Svetlana Mareva
      10:00 - 10:30 Oliver Kant
      10:30 - 11:00 Thomas Pilger
      11:00 - 11:30 Pause
      11:30 - 12:00 Alexander Ende
      12:00 - 12:30 David Gutzmann
      12:30 - 13:00 Silke Rieger

Attention: Students have to meet all deadlines listed in the timetable. Otherwise, she/he will lose the right to take part in the final presentation.

Topics

Advisor: Michael Baar (Seite)

Current Approaches to Opportunistic Routing in MANETs
In mobile ad-hoc networks setups where network links are unstable, highly variable and often not available at all standard routing protocols do not work efficiently or not at all. Examples of such setups are car networks or animal monitoring. A class of routing protocols which deal explicitly with In mobile ad-hoc networks setups where network links are unstable, highly variable and often not available at all standard routing protocols do not work efficiently or not at all. Examples of such setups are car networks or animal monitoring. A class of routing protocols which deal explicitly with these setups is called "opportunistic routing". Survey current state-of-the-art approaches of opportunistic routing protocols and analyze their features, limitations, strength and weaknesses. These setups is called "opportunistic routing". Survey current state-of-the-art approaches of opportunistic routing protocols and analyze their features, limitations, strength and weaknesses. 
Assigned to:Pascal Becker (no report)

Tracking in Sensor Networks
Many different sensors provide data that can contains information on a nodes position. Most readings provide only relative information or merely hints. Examples are current GSM network cell, GPS, acceleration, heading, movement, distance estimation to moving or fixed points, proximity to other devices, common environmental readings, air-pressure, … Statistical methods, commonly used in robotics, can help to fuse all this input to a number of probable positions. Common statistical approaches include Kalman and particle filters which are limited their ability to include numerous facts into the calculation. Survey state-of-the-art approaches that overcome these limitations. 
Assigned to: Patryk Mazur (no report)

Advisor: Bastian Blywis (Seite)

MAC aware Routing
Routing in wireless environments usually happens without taking into consideration the properties of the shared medium. Because of the broadcast characteristic only one station at a time is allowed access and can relay its packets. Bottlenecks in meshed networks might arise due to this problem. Your task is to investigate solutions based on a cooperation of the data link and network layers. How can the medium allocation be taken into account? What routing protocols exist that change routes dynamically based on this information? How are flows detected? 
Assigned to: Alexander Ende (report)

MIMO based wireless mesh networks
The multiple-input multiple-output the radio technology exploits the usage a of more than one radio module and antenna and offers some interesting properties. Start by explaining this emerging technology. What MAC and routing protocols take advantage of these features? Try to focus your work on ad-hoc and mesh networks. Take a look at IEEE 802.11n (draft). 
Assigned to: nobody (no report)

VoIP in wireless networks
Since several years circuit switched telephony networks are on a decline and getting more and more extinct. Voice over IP (VoIP) providers on the other hand are on the rise by offering an - at the first sight - equal service to the end user. This packet switched technology is said to be common in few years in every household. Using a shared wireless medium as an alternative to cable bound solutions introduces some challenges. It's your task to investigate how the properties of the medium and the used medium access techniques affect the quality of the service. How do current approaches try to cope with the various problems in contention based and contention free environments? How can guarantees be given? Is mobility of the stations a problem? You should consider infrastucture based and ad-hoc networks. Take a look at the IEEE 802.11e standard! 
Assigned to: nobody (no report)

Multipath Routing in (mobile) mesh networks
Using multiple paths to deliver data packets from source to destination could be solution to tackle some of the problems of wireless infrastructureless networks: high bit error rate, unstable links, lower bitrate, etc. Discuss the overall performance, problems, and limitating factors of these approach(-es). Explore if node mobility can also be taken into account. How is the shared medium exploited? 
Assigned to: David Gutzmann (report)

Advisor: Freddy Lopez Villafuerte (Seite)

Tracking with RFID
There are different techniques and technologies to follow objects or persons inside buildings. Here the student will describe and classify the technology of RFID. The student will research the different methods in practical applications for different environment explaining the working way in every case. 
Assigned to: Torsten Dunkel (no report)

Vertex Cover Problem
To find node leaders for clusters in a network or to know which node have to be off to save resources in the network but at the same time covering the area of interest, is necessary to do an analysis on the connection between nodes. In this seminar topic, the student will describe different approaches to find the best Vertex Cover in a wireless sensor network using graphs theory.
Assigned to: Svetlana Mareva (report)

Localization Using Mobile Nodes
The localization problem is an open research problem in Ad hoc Wireless Sensor Nodes. There are different methods to find the node positions into a network. An interesting proposal is using mobile node to do localization. With this seminar topic you will show the problems and the advantages to do localization using this kind technique, you will look for related work and evaluate the approach for a future expectative with ubiquitous devices.
Assigned to: nobody (no report)

Advisor: Abd Al Basset Al Mamou (Seite)

Comparison of DYMO and OLSR
In Mobile Ad hoc networks (MANETs), many routing protocols have been proposed; most of them belong to even reactive or proactive routing category. The main difference between routing protocols – regardless the category they belongs to – is the way to discover the routes, maintain them and then route messages/packets.
Assigned to: nobody (no report)

Metrics for wireless routing
Many routing protocols have been proposed in wireless networks (MANETs or WSN); and every protocol has few metrics that can decide its performance. The aim of seminar is to look to the most vital common metrics which make sense in evaluating this kind of wireless routing networks.
Assigned to: nobody (no report)

Load Balancing in Wireless (Sensor, Ad hoc) Networks
With the wide use of wireless networks, a new challenge has been raised and starts to affect the network performance. The missing of load balancing in the network will affect providing the service to a large number of nodes, and then it will affect the Quality of Service guarantees. This seminar looks to the algorithms that describe the load balancing routing in wireless networks.
Assigned to: nobody (no report)

Towards estimating the lifetime of wireless sensor networks
Wireless sensor networks have very constrained energy resources. The existing protocols laid on the network stack should be energy aware and energy efficient. This seminar aims to do an analytical study about energy awareness in wireless sensor networks for various protocols at the routing layer in wireless sensor network stack.
Assigned to: nobody (no report)

Advisor: Kirsten Terfloth (Seite)

Self-x properties and emergent behavior - Hype or Potential?
Organic or bio-inspired computing has been proposed in the past to obtain global behavior by means of spontaneous local interaction of participants e.g. sensor nodes in a sensornetwork. The idea is to enable robust services in dynamically changing environments, to control and manage complex and dynamic networks by relying on local intelligence of devices that implement so called self-x-properties (self-awareness, self-organisation, self-healing, etc. to name but a few). The goal of this topic is to examine the current state of the art in self-organization systems, evaluate the quality of metrics that have been proposed for quantitative analysis and to clearly judge the potential of this domain.
Assigned to: Thomas Pilger (report)

Domain Specific Languages for Wireless Sensor Networks
Due to the distribution of nodes in the environment and the embedded nature of devices, application development on top of wireless sensor networks (WSN) is still a time-consuming and cumbersome task. Lately, several so called 'domain specific languages' (DSL) have been proposed to tackle this problem at the language level by encapsulation of typical WSN actions as language primitives. The goal of this topic is to examine some representative languages introduced for WSNs, categorize the level of abstraction they provide and critically evaluate their applicability, advantages and shortcomings.
Assigned to: nobody (no report)

Next-generation middleware for Wireless Sensor Networks
A tremendous amount of solutions to provide a middleware abstraction layer for wireless sensor networks have been proposed in the past years. Nowadays, it is extremely interesting to see how these approaches may be grouped by the different core elements they propose a middleware should supply for application developers. The goal of this work is two-fold: First of all, an overview of representative approaches along with their view on what characteristic features WSN applications have, shall be provided. Secondly, a catalogue of common metrics to evaluate middleware approaches, both qualitatively and quantitatively, has to be established and discussed thoroughly.
Assigned to: Silke Rieger (report)

Advisor: Nico Wanner (Seite)

Comparison of platforms in the area of telephony and mobile computing
Compare representative examples of all PDA- and Smartphone platforms at the market in reference to the used operating system, the functional range and the possibilities of application development. The goal of this topic is to introduce state of the art in telephony and mobile computing platforms and to examine the key differences between the two approaches. At the end we will discuss the open research question about future changes in mobile platforms.
Assigned to: Oliver Kant

Advisor: Georg Wittenburg (Seite)

Distributed Pattern Recognition in WSNs
Due to the processing power of the individual sensor nodes that form a Wireless Sensor Network (WSN), the nodes can do better than simply forwarding sampled data to a base station. In fact, this is wasteful for some applications, e.g. area surveillance, because the users are interested in events rather than raw data. In order to support these applications in an energy-efficient way, sensor nodes can be programmed to recognize patterns in raw data and only notify the base station after a pattern has been detected. The algorithms used for this end go well beyond classic AI research because sensor nodes may cooperate to recognize a pattern. The goal of this project is to summarize recent approaches to distributed pattern recognition in WSNs, thoroughly evaluate the proposed architectures for their strengths and weaknesses, and point out open questions given the current state of the art.
Assigned to: Robert Richter (report)

Practical Approaches to Service Placement in MANETs
Service placement deals with selecting which node in a network is most suitable for hosting a service that responds to queries from other nodes. Optimally placing services reduces network traffic and improves connectivity between clients and servers. Practical approaches to service placement commonly employ heuristics based on information gathered from nodes in the neighborhood of the current host. Some heuristics are tailored to the specific application, e.g. coverage, topology or group mobility, and are thus not applicable to general-purpose service placement. The goal of this project is to summarize recent practical approaches to service placement in MANETs, evaluate in detail their strengths and weaknesses, and discuss the question in how far they are applicable to general-purpose service placement.
Assigned to: nobody (no report)

Literature