Loading...

Energy Efficient Policies for TDMA based Wireless Body Area Network (WBAN)

Karimzadeh Farshbafan, Mohammad | 2015

687 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47246 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Nasiri-Kenari, Masoumeh; Ashtiani, Farid
  7. Abstract:
  8. Recent advancements in wireless communication along with the urgent need for new care systems for the patients, has attracted the attention of many communication researchers to Wireless Body Area Networks (WBANs), which has led to IEEE 802.15.6 and IEEE 802.15.4j standards, accordingly. The main goal of such networks, is to send the vital data of the associate, through the over/in-body-implanted sensors, to one central node. One of the key features of these networks is the limited energy sources of the sensors, transmit the data. Meanwhile, on-time error-free transmission of the data is also of great importance. Therefore, designing a proper algorithm, satisfying the aforementioned challenges, to handle the MAC sublayer, is of crucial importance. Although, considering the unique features of WBANs, applying the convenient sensor networks algorithms, is not feasible. One of such features, is the absence of perfect channel knowledge at the decision maker's side; which in turn leads to applying partially observable Markov decision process (POMDP) to schedule the transmission policy of the nodes. The decision-making in this algorithm at each superframe, is based on considering the overall effects of all the possible decision- makings, on the upcoming superframes' decisions. This results in an exponential computational complexity in terms of the number of active nodes of the network. In this thesis, we try to come up with an algorithm to schedule the transmission of the nodes, satisfying not only the required performance of the network, but also outperforming the existing algorithms in terms of computational complexity that could also have a practically efficient implementation.As we will observe, according to the specialfeatures of the channel state transition matrix along with the modifications applied in the structure of the cost function, myopic algorithm could easily be applied to schedule the transmissions. This algorithm relies on the strictly limited memory usage. In fact, in our proposed method, the structure of the cost function is modified such that the decision making at each superframe is performed regardless of its overall effect on the upcoming superframes' decisions; this, in turn, drastically reduces the computational complexity. Simulation results demonstrate that, in most of the packet arrival scenarios, the proposed algorithm has a negligible difference in terms of energy consumption and reliability with those of the optimal algorithms. Moreover, in more practical scenarios such as heterogeneous WBAN, in which some sensors have their priority of energy consumption whereas some have their priority of sending data, sensors can easily achieve a considerable amount of energy savings and also reliability for their transmission
  9. Keywords:
  10. Wireless Body Area Network ; Data Transmission ; Media Access Control (MAC)Layer ; Partially Observable Markov Decision Process ; Suboptimal Energy Consumption Algorithms ; MYOPIC Algorithm

 Digital Object List

 Bookmark

No TOC