Opportunistic transmission scheduling for next generation wireless communication systems with multimedia services
Department of Electrical and Computer Engineering
Doctor of Philosophy
The explosive growth of the Internet and the continued dramatic increase for all wireless services are fueling the demand for increased capacity, data rates, and support of different quality of service (QoS) requirements for different classes of services. Since in the current and future wireless communication infrastructures, the performances of the various services are strongly correlated, as the resources are shared among them, dynamic resource allocation methods should be employed. With the demand for high data rate and support of multiple QoS, the transmission scheduling plays a key role in the efficient resource allocation process in wireless systems. The fundamental problem of scheduling the users' transmissions and allocating the available resources in a realistic CDMA wireless system that supports multi-rate multimedia services, with efficiency and fairness, is investigated and analyzed in this dissertation.
Our proposed approach adopts the use of dynamically assigned data rates that match the channel capacity in order to improve the system throughput and overcome the problems associated with the location-dependent and time-dependent errors and channel conditions, the variable system capacity and the transmission power limitation. We first introduce and describe two new scheduling algorithms, namely the Channel Adaptive Rate Scheduling (CARS) and Fair Channel Adaptive Rate Scheduling (FCARS). CARS exploits the channel variations to reach high throughput, by adjusting the transmission rates according to the varying channel conditions and by performing an iterative procedure to determine the power index that a user can accept by its current channel condition and transmission power. Based on the assignment of CARS and to overcome potential unfair service allocation, FCARS implements a compensation algorithm, in which the lagging users can receive compensation service when the corresponding channel conditions improve, in order to achieve asymptotic throughput fairness, while still maintaining all the constraints imposed by the system.
Furthermore the problem of opportunistic fair scheduling in the uplink transmission of CDMA systems, with the objective of maximizing the uplink system throughput, while satisfying the users' QoS requirements and maintaining the long-term fairness among the various users despite their different varying channel conditions, is rigorously formulated, and a throughput optimal fair scheduling policy is obtained. The corresponding problem is expressed as a weighted throughput maximization problem, under certain power and QoS constraints, where the weights are the control parameters that reflect the fairness constraints. With the introduction of the power index capacity it is shown that this optimization problem can be converted into a binary knapsack problem, where all the corresponding constraints are replaced by the users' power index capacities at some certain system power index. It is then argued that the optimal solution can be obtained as a global search within a certain range, while a stochastic approximation method is presented in order to effectively identify the required control parameters. Finally, since some real-time services may demand certain amount of service within specific short span of time in order to avoid service delays, the problem of designing policies that can achieve high throughput while at the same time maintain short term fairness, is also considered and investigated. To this end a new Credit-based Short-term Fairness Scheduling (CSFS) algorithm, which achieves to provide short-term fairness to the delay-sensitive users while still schedules opportunistically the non-delay-sensitive users to obtain high system throughput, is proposed and evaluated.
njit-etd2005-130 (111 pages ~ 5,354 KB pdf)
Please complete this Feedback Form to inform us about your experience using this website. It will assist us in better serving your information needs in the future. Thank You!
Created September 8, 2008