Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Lane Department of Computer Science and Electrical Engineering

Committee Chair

Daryl S. Reynolds

Committee Co-Chair

Matthew C. Valenti

Committee Member

Matthew C. Valenti

Committee Member

Natalia A. Schmid

Committee Member

Vinod K. Kulathumani

Committee Member

Erdogan Gunel


In cooperative communication networks, the source node transmits its data to the destination either directly or cooperatively with a cooperating node. When using energy harvesting technology, where nodes collect their energy from the environment, the energy availability at the nodes becomes unpredictable due to the stochastic nature of energy harvesting processes. As a result, when the source has a transmission, it cannot immediately transmit its data cooperatively with the cooperating node. It first needs to determine whether the cooperating node has sufficient energy to forward its transmission or not. Otherwise, its transmitted data may get lost. Therefore, when using energy harvesting, the challenge is for the source to schedule its transmissions whether directly or cooperatively, such that the fraction of its events (sensed data) that are successfully reported to the destination is maximized.

Hence, in this dissertation, we address the problem of cooperating node scheduling in energy harvesting sensor networks. We consider the problem for the case of a single cooperating node and the case of multiple cooperating nodes, as well as the scenarios of one-way and two-way cooperative communications. We propose a simple scheduling scheme, called feedback scheme, which enables the source to optimally schedule its transmissions whether directly or cooperatively. We show that the feedback scheme maximizes the system performance, but does not require auxiliary parameter optimization as does the-state-of-the-art scheme, i.e., the threshold-based scheme. However, the feedback scheme has the problem of overhead caused by transmitting the energy status of the cooperating node to the source. To overcome this burden, we introduce a statistical model that enables the source to estimate the energy status of the cooperating node. Because cooperation may result in the cooperating node performing worse than the source, we address this problem through fairness in the performance between the nodes in the network. In addition, we address the problem of scheduling for throughput maximization in a wireless energy harvesting uplink. We propose centralized and distributed algorithms that find the optimal solution, and we address complexity issues. Our algorithms are shown to have a linear or quadratic complexity compared to the exponential complexity of the brute force approach. Compared with cooperative transmission, our approach maximizes the network throughput such that no node's throughput is adversely affected.