Loading...

The Information Theory Approach to Communication over Deletion Channel with Hidden Markov Codebook

Molavipour, Sina | 2015

580 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47278 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Aminzadeh Gohari, Amin
  7. Abstract:
  8. One of the main challenges in data transmission is synchronization error. In practice there are solutions to this issue, but this incurs a cost that can not be neglected. Synchronization error consists of deletion, insertion and substitution. Investigation of such errors in a communication channel has been of a great concern in information theory. Furthermore, in many applications such as biology and data storage on disks, we observe synchronization errors. Deletion channel is defined as a channel in which input symbols are deleted with a probability d independently of each other, such that the order of symbols remains unchanged. In spite of attempts to find a closed form for the capacity of deletion channel, few bounds have been derived.
    Deletion channels and their capacity regions also find applications in the problem of recovery from random samples. In sampling of a data, with a probability some symbols could be missed in the output. these symbols need to be recovered to reconstruct the original signal. We can model such random sampling with a deletion channel.The error probability of decoding will give us the recovery error. Hidden Markov model is a comprehensive model to express different types of data such as in bioinformatics and speech. Our goal in this thesis is to investigate capacity of deletion channel with hidden Markov distribution, which is also applicable to the problem of recovery of randomly deleted samples for a hidden Markov dataset
  9. Keywords:
  10. Deletion Channel ; Hidden Markov Model ; Channel Capacity ; Synchronization Error ; Recovery From Random Samples

 Digital Object List

 Bookmark

No TOC