An integrated fuzzy-GA approach for buffer management

Abstract

This paper deals with a novel buffer management scheme based on evolutionary computing for shared-memory asynchronous transfer mode (ATM) switches. The philosophy behind it is adaptation of the threshold for each logical output queue to the real traffic conditions by means of a system of fuzzy inferences. The optimal fuzzy system is achieved using a systematic methodology, based on genetic algorithms (GAs), which allows the fuzzy system parameters to be derived for each switch size, offering a high degree of scalability to the fuzzy control system. Its performance is comparable to that of the push-out (PO) mechanism, which can be considered ideal from a performance viewpoint, and at any rate much better than that of threshold schemes based on conventional logic. In addition, the fuzzy threshold (FT) scheme is simple and cost-effective when implemented using VLSI technology.Index Terms-ATM switches, buffer management, fuzzy logic (FL), shared-memory.

Introduction to Pattern Recognition and Bioinformatics

“…The rough-fuzzy computing provides a powerful mathematical framework to capture uncertainties associated with the data. Other hybridized models for pattern recognition and data mining include neurogenetic [46, 215,251,293,362], rough-genetic [43, 317,369], fuzzy-genetic [14,61,64,116,180,233], rough-neuro-genetic [167], rough-neuro-fuzzy [11,24,243,244,262], and neuro-fuzzy-genetic [187,195,283,290,307,358] approaches.…”

Optimization of buffer overflow probability in Jackson queueing networks using Mamdani fuzzy inferecnce system

“…A novel active queue management (AQM) scheme based on Takagi Sugeno fuzzy sliding mode controlled is designed to tackle network congestion problem with modeling uncertainties, time varying parameter fluctuations and external disturbances [7].Two novel expert dynamic buffer tuners/controllers namely, the neural network controller (NNC) and fuzzy logic controller (FLC) are proposed to eliminate buffer overflow at the user/server level [8]. A fuzzy control scheme in which the value of the threshold for each logical output queue could be adapted to the real traffic conditions by means of a system of fuzzy inferences [9].A novel approach for adaptive fuzzy model-based flow control in MPLS networks is introduced to model the buffer queue size and controllable source behavior [lO].An adaptive fuzzy control traffic shaping scheme based on leaky bucket is applied to solve the traffic congestion problem over wireless networks [II].A fuzzy-neural network is used to predict the queue lengths in the multi queue single processed queuing system [12]. The energy of a sensor node in a wireless sensor networks is limited.…”

Particle swarm intelligence tuned fuzzy controller for damping modal oscillations of power system

“…The fuzzy logic and neural networks can be integrated to form a connectionist Adaptive network based Fuzzy logic controller. Literature reports tuning of fuzzy controllers using various evolutionary and intelligent techniques [13][14][15][16][17][18][19][20][21][22][23] for various applications. Besides, the swarm intelligent algorithm, which has global optimizing capacity basing upon a fitness function, is used to optimize the coefficients of the TS-fuzzy scheme as well as that of the auxiliary damping signal [19][20][21][22][23].…”