Loading...
- Type of Document: Ph.D. Dissertation
- Language: Farsi
- Document No: 55906 (05)
- University: Sharif University of Technology
- Department: Electrical Engineering
- Advisor(s): Aminzadeh Gohari, Amin; Ashtiani, Farid; Nasiri Kenari, Masoumeh
- Abstract:
- Molecular communication (MC) is a relatively new branch in telecommunication science. This branch is designed to transmit information in small environments (about nanometers). Applications of molecular communication include identification of abnormalities, detection of cancers, etc. The smallness of the transmission environment has caused new challenges and areas in this field. One of the unique features of MC with no parallel in classical wireless communication is chemical reactions: different types of molecules can react with each other and form new types of molecules in the medium. This feature of MC poses a challenge in macroscale MC since equations describing chemical reaction with diffusing particles are nonlinear partial differential equations with no closed-form solutions. This makes it difficult to design optimal modulation schemes and optimal encoder. while there are many numerical methods to solve these equations, these methods are not provable. Unprovability means that the proof of optimality of a given modulation is impossible using numerical methods. On the other hand, numerical methods are dependent on parameters and input functions. Changes in parameters or input functions will lead to re-simulation. Therefore, proving and designing the optimal modulation requires a solution of equations even in an approximate way. Another important and challenging topic in molecular communication is the problem of Blind Tracking and Localization. Definition of this problem is: tracking a path or being in a certain place by a nano device without any knowledge about path or position of target point. Finding the target location and moving towards it is very difficult. Because the concentration of the released molecules in the medium is a time-variant function and non-convex with respect to the distance. Also due to the presence of counting noise, the measurements are discrete random variables. This thesis consists of two parts. In the first part, modeling channels with chemical reactions and designing optimal modulation schemes for these channels are provided. First, we will introduce molecular communication scenarios in which chemical reactions occur, and then we will design modulation scheme for these models locally and globally. In the local modulation design, we will first provide a method to obtain an approximate solution of equations, and then we will design the modulation based on this approximation. In the global modulation design, we will solve the optimal modulation without considering any approximation. Solutions for the blind tracking and localization problem are proposed at the second part. These solutions include algorithms to solve blind tracking and localization problem and calculate their complexity
- Keywords:
- Molecular Communication ; Chemical Reactions ; Anomaly Detection ; Cancer Diagnosis ; Blind Tracking and Localization ; Optimal Modulation Scheme