Joint Clustering and Scheduling for Correlated Data Gathering in Machine-to-Machine Wireless Networks
Date Issued
2014
Date
2014
Author(s)
Tsai, Yun-Da
Abstract
Clustering and transmission power control have been proposed as an e ective
way to support massive access in wireless machine-to-machine (M2M) networks.
For M2M service, the quality of gathered information is a more realistic factor to
evaluate the system performance than the link quality of each machine. However,
most of the recent work focuses on enhancing the service quality of individual
machines but ignore the nature of data correlation among machines (sensors). In
this thesis, we rst formulate a problem for 2-tier minimum power data gather-
ing in M2M networks. Then, we decompose the problem into two sub-problems:
minimal power consumption data-centric clustering and minimal power consump-
tion scheduling. We apply Constrained Simulated Annealing (CSA) to solve the
cluster formation problem. After the cluster structure is determined, the minimal
power consumption scheduling problem can be transformed into a mixed-integer
linear programming (MILP) problem. We then proposed a 0-1 branch-and-bound
algorithm to obtain the optimal solution. To evaluate our proposed transmission
scheme, we consider both Gaussian data source and real image data. Evaluation
results show that the proposed scheme achieves better performance than conven-
tional clustering algorithms.
Subjects
無線網路
物聯網
感知無線網路
最佳化
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-103-R01942053-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):84e3ce43d6989b468734f86c8014b1a1