Detta är ett uppsatsförslag hämtat från Nationella Exjobb-poolen. Klicka här för att komma tillbaka till samtliga exjobbsförslag.
Latency and Error Control in Energy Constrained Wireless Mesh Network for Automatic Control Applications
In many wireless sensor networks, and in particular in industrial wireless sensor networks, the requirement on battery lifetime is extreme. To achieve long lifetime, it is necessary to put the node’s radio to sleep for a significant fraction of the time. In conflict with sleeping radios stands the goal of low latency in the network, especially in industrial applications such as wireless control. With
constraints on latency and therefore on sleep schedules, it becomes increasingly important to use the radios’ on-time energy-efficiently and choose transmission technique judiciously. Today there is no forward error correction implemented in industrial mesh networks, resulting
in packet losses and retransmissions. Retransmissions improve the throughput but also introduce undesired latency in time-critical control applications. The energy efficiency compared to other techniques of error control, such as forward error correcting codes or transmit-power control, depends on hardware, application and radio channel constraints. There is an overall trade-off between overhead (extra bits for coding, feedback for power control, and retransmission of packets), energy efficiency
(energy consumption per correctly received bit, say) and latency. It is at present not clear how this
trade-off should be handled in industrial wireless networks in control applications.
The questions in this work are, in order of priority (all are not necessarily addressed):
? Should industrial wireless networks use error correcting codes or retransmissions (or both in
combination)? If coding is attractive, which coding technique is most appropriate, and how is
it to be combined with retransmission strategies?
? How should transmit power control be employed (amount of feedback, rate and coarseness of
adjustment)? How does it relate to error correcting codes?
? Is the choice between coding and retransmission affected by other diversity schemes (spatial,
polarization) that reduce error rates in fading environments?
The constraints/utilities of energy consumption, latency, complexity and throughput imposed by the
industrial control application underpin the approach to answering the questions. This includes the use
of relevant time constants in control loops and channel fading respectively.
Tasks and tools
The thesis work should comprise the following tasks:
? Theoretical study of the approaches, the industry-specific constraints and standards
? Preliminary evaluation through simulation and computations (Matlab or similar)
? Implementation of the studied algorithms on Tmote/Telos platform (or similar)
? Performance evaluation in terms of latency, packet error rate, complexity, etc.
? Final report describing the work and results
Students in MSc programs in Electrical Engineering, Engineering Physics, and Computer Engineering
or similar is suited for this job. Candidates are expected to have a strong background in communication
theory. Good mathematical and programming skills are desired. It is also important that the applicant
have good writing and communication skills. Please indicate a list of courses with marks and a CV in
Mikael Gidlund, +4621345163, [email protected]
Erik Björnemo, +46184713068, [email protected]
Informationen om uppsatsförslag är hämtad från Nationella Exjobb-poolen.