dc.description.abstract | In the third generation of wireless cellular system, adaptive dynamic channel borrowing is presented to maximize the number of served calls in wireless cellular networks. This has led to an intensive research in mobile computing to provide mobile users accesses to an internet. Efficient use of limited radio channels with a simultaneous increase of traffic capacity requires proper channel assignment. The uneven traffic load may create hot-spot cells and possibly causes a high blocking rate in hot-spot cells. Most conventional methods use load indices with a threshold value to determine the load status of a cell. However, those exists a ping-pong effect, as loads are around the threshold value. This result causes an unstable system and unnecessary message passing overhead. In addition, the estimation of traffic load is difficult and time-consuming. Thus, an intelligent prediction mechanism is needed. In this dissertation, a fuzzy-based dynamic channel-borrowing scheme (FDCBS) is presented to maximize the number of served calls in a distributed wireless cellular network. We develop a method to predict the cell load and solve the channel-borrowing problem based on the fuzzy logic control. A new channel borrowing algorithm with multi-channels borrowing is also presented in this dissertation. A borrowing mechanism supporting the present facility has been built on the application-level of wireless cellular networks. Performance results show that the FDCBS has better adaptability, robustness, and fault-tolerant capability thus yielding better performance compared with other algorithms.
Fuzzy logic control can deal with perceptual uncertainties in classification problems. In order to design an adaptive channel borrowing in wireless cellular networks, it is an important task to construct the fuzzy membership function for each attribute and generate fuzzy rules from training instances for handling a specific classification problem in wireless cellular networks. However, there are some drawbacks in the proposed DFCBS: (1) Need human experts to predefine initial membership functions, i.e., these methods can not construct membership functions from the training data set fully automatically. (2) Too complicated and need a lot of computation time, and generate too many fuzzy rules.
Neural-fuzzy controllers, based on a fusion of ideas from fuzzy control and neural networks, possess the advantages of both neural networks (e.g., learning abilities, optimization abilities, and connectionist structures) and fuzzy control systems (e.g., human- like IF-THEN rule thinking and ease of incorporating expert knowledge). In this way, we can bring the low-level learning and computational power of neural networks to fuzzy control systems and also provide the high-level, human-like IF-THEN rule thinking and reasoning of fuzzy control systems to neural networks. In brief, neural networks can improve their transparency, making them closer to fuzzy control systems, while fuzzy control systems can self-adapt, making them closer to neural networks. We shall focus on structure learning and parameter learning in neural-fuzzy controllers, and examine cases when they can take place sequentially and separately in two phases, when only parameter learning, or only structure learning is necessary, when they can occur simultaneously in a single phase, and when they can be performed with reinforcement signals. All these variations of structure and parameter learning of neural-fuzzy controllers are discussed with simulation examples.
In order to adaptive for channel-borrowing in wireless cellular networks, a neural-fuzzy controller for the dynamic channel-borrowing scheme (NFDCBS) is presented to provide multimedia services and to support increasing number of users. In a cellular network, the call arrival rate, the call duration and the communication overhead between the base stations and the control center are vague and uncertain. Therefore, we propose a new efficient dynamic-channel borrowing for load balancing in distributed cellular networks based on NFDCBS. The proposed scheme exhibits better learning abilities, optimization abilities, robustness, and fault-tolerant capability thus yielding a better performance than other algorithms. It aims to efficiently satisfy their diverse quality-of-service (QoS) requirements of multimedia traffic. The results show that our algorithm has lower blocking rate, lower dropping rate, less update overhead, and shorter channel-acquisition delays than previous methods. Through simulations, we evaluate the blocking rate, update overhead, and channel acquisition delay time of the proposed method. The simulation results demonstrate that our algorithm has lower blocking rate, less updated overhead, and shorter channel acquisition delays. | en_US |