06-09-2016, 10:22 AM
Design of fuzzy logic controller for QoS optimization in Mobile Ad-hoc Networks (MANETs)
1453122917-srm.final.docx (Size: 129.79 KB / Downloads: 9)
ABSTRACT
Data transmission in Mobile ad-hoc networks need high Bandwidth and secure transfer because of larger data size and frequent path break due to mobility. When the Quality of service (QoS) in mobile ad-hoc networks is improved, the consistency of the network increases. In order to minimize the packet loss during data transmission, selection of suitable microcontroller and the wireless transceivers play an important role in which the data control should be designed. In this paper, series queue with blocking model has been used in OSI network layers to analyze packets blocking. Packets arrival rate is taken as trapezoidal fuzzy number and service rates are fuzzy variable. Fuzzy logic controller has been designed to reduce packets blocking and improve QoS such as speed of the network, latency and throughput.
Keywords
Fuzzy sets, trapezoidal fuzzy number, series queue with blocking, robust ranking technique, fuzzy logic controller
1. INTRODUCTION
When QoS in ad-hoc networks is improved, longevity of the network increases [1]. Available QoS metrics based on Queuing or buffer management in wired and other wireless networks don’t applicable in MANETs because of its unique characteristics. Thus, the simplest QoS model buffer-less system has been proposed. Selecting suitable microcontroller and wireless transceiver chips play an important role in assembling ad-hoc mobile devices, in which data control should be designed, so that packet loss is minimized. In ad-hoc networks, data dropping play a vital role during data transmission. In real time, delay sensitive applications such as Tsunami monitoring, nuclear plant radiation monitoring, border surveillance, forest fire monitoring like disaster prone areas become more vulnerable due to packet dropping [2]. To overcome these kinds of problems, Quality of Service (QoS) metrics have been used in fuzzy queuing models. As fuzzy values are much more realistic than the crisp values, we assume that the packet arrival rate is a trapezoidal fuzzy number. During data transmission, packets are blocked due to congestion. To reduce this congestion, fuzzy blocking probability has been found and it will be reduced by suitable increasing the service rate in MAC layer. As blocking probability is reduced, packets dropping will be considerably reduced and hence energy consumption will be reduced due to reduced retransmission rate. In this paper, a series queue with blocking model has been considered and the objective is to improve the Quality of Service (QoS) in Mobile Ad hoc Networks (MANETs). As buffer overflows lead to re-transmission of same data packets several times, there is a need of additional energy consumption. Thus queuing management is directly dependent to energy consumption in network communication. To analyze energy consumption, probability of blocking of the arrival packets has been found for different fuzzy arrival rates and service rates. Fuzzy logic controller has been designed to reduce data loss.
1.1 MOBILE AD-HOC NETWORKS (MANETs)
Mobile ad-hoc Networks (MANETs) is defined as a set of mobile nodes that moved freely, connected among each other without any infrastructure and intercommunicate using single-hop and multi-hop paths . MANETs consist of limited power, limited memory and wirelessly connected devices such as cell phones, laptops, Personal Digital Assistants (PDA), wearable or handheld digital devices. These devices act both as hosts and routers, dynamically self organized mobile nodes deployed in a distributed fashion with no pre existing infrastructure. In MANETs, routing paths are created and deleted due to the nodal mobility. MANETs come together as needed, not necessarily with any support from the existing infrastructure or any other kind of fixed stations. This statement can be formalized by defining an ad hoc network is an autonomous system of Mobile Hosts(MH) (also serving as routers) connected by wireless links, the union of which forms a communication network modeled in the form of an arbitrary graph. This is in contrast to the well-known single hop cellular network model that supports the needs of wireless communication by installing base stations as access points.
In the cellular networks, communications between two mobile nodes completely rely on the wired backbone and the fixed base stations. In MANETs, no such infrastructure exists and the network topology may dynamically change in an unpredictable manner, since nodes are free to move. As for the mode of operation, ad-hoc networks are basically peer-to-peer multi-hop mobile wireless networks where information packets are transmitted in a stored-and-forward manner from a source to an arbitrary destination, via intermediate nodes.
Series Queues with Blocking
A series queue model or a tandem queue model is, in general, one in which
Customers may arrive from outside the system at initial node and may leave the system from terminal node.
Customers may enter the system at initial node, traverse from node to node in the system and leave the system from terminal node, not necessarily following the same order and
Customers may return to any node and proceed further.
Series queue with blocking is a sequential queue model consisting of two service points s_1and s_2, at each of which there is only one server and where no queue is allowed to form at either point. An entering customer will first go to s_1, After he gets the service completed at s_1, he will go to s_2 if it is empty or will wait in s_1 until s_2 becomes empty. This means that a potential customer will enter the system only when s_1 is empty, irrespective of whether s_2 is empty or not. Since the model is a sequential model, all the customers require service at s_1 and then at s_2.The possible states of the system are given below with their interpretation:
1.2.2 Series Queue with Blocking in the Processor Memory
The above image shows that the processor memory is divided into two parts as separate service channels [8]. The MAC sub layer has been determined as one server with service rate µ_1 and rest of the network layers were taken as one server with service rateµ_2. Arrival packets enter the MAC sub layer through the physical medium (physical layer) and waits in the queue if the system is busy. In this model, buffer-less system has been taken and QoS metric has been derived.
2. PRELIMINARIES
Zadeh [3] in 1965 first introduced Fuzzy set as a mathematical way of representing impreciseness or indistinctness in everyday life.
2.1 Definition(fuzzy set)
A fuzzy set is characterized by a membership function of mapping elements of a domain space or universe of discourse X to the unit interval [0,1].(i,e)A = {(x, µ_A(x); xЄ X}, Here
µ_A: X →[0,1] is a mapping called the degree of membership function of the fuzzy set A and
µ_A(x )is called the membership value of xεX in the fuzzy set A.These membership grades are often represented by real numbers ranging from [0,1]
2.2 Definition (Trapezoidal fuzzy number)
For a trapezoidal number A(x), it can be represented by A(a,b,c,d;1) with membership function μ(x) is given as[3],
μ(x) ={█((x-a)/(b-a),a≤x≤b@1 ,b≤x≤c@(d-x)/(d-c) ,c≤x≤d@0,otherwise)┤
2.3 Function principle
Function principles proposed to be as the fuzzy arithmetical operations by trapezoidal fuzzy numbers. We describe some fuzzy arithmetical operations under Function Principle as follows [4],[5].
Suppose A ̃ = (a_1,b_1,c_1,d_1) and B ̃ = (a_2,b_2,c_2,d_2) are two trapezoidal fuzzy numbers. Then
(i) Addition of two fuzzy numbers A ̃ and B ̃ is defined as
A ̃⊕B ̃=(a_1+a_2,b_1+b_2,c_1+c_2,d_1+d_2)
Where a_1,b_1,c_1,d_1,a_2,b_2,c_2,d_2 are any real numbers.
(ii) Multiplication of two fuzzy numbers A ̃ and B ̃ is defined as
A ̃⊗B ̃=(a_3,b_3,c_3,d_3 )
WhereX=(a_1 a_2,a_1 d_2,d_1 a_2,d_1 d_2 )
Y=(b_1 b_2,b_1 c_2,c_1 b_2,c_1 c_2)
a_3=minX,b_3=minY,c_3=maxY,d_3=maxX
wherea_1,b_1,c_1,d_1,a_2,b_2,c_2,d_2 are any positive real numbers.
(iii)Division of two fuzzy numbers A ̃ and B ̃ is defined as
A ̃ѲB ̃=(a_1/d_2 ,b_1/c_2 ,c_1/b_2 ,d_1/a_2 )
wherea_1,b_1,c_1,d_1,a_2,b_2,c_2,d_2 are any positive real numbers.
(iv) Scalar Multiplication
Take αbe any real number. Then for α≥0, αA ̃=(αa_1,〖αb〗_1,αc_1,〖αd〗_1)
α<0, αA ̃=(αd_1,αc_1,〖αb〗_1,αa_1)
(v) The inverse of a fuzzy number A ̃ = (a_1,b_1,c_1,d_1) is defined as A ̃^(-1)= (1/d_1 ,1/c_1 ,1/b_1 ,1/a_1 )
wherea_1,b_1,c_1,d_1 are any positive real numbers.
3. Robust Ranking Technique – Algorithm
Using robust ranking technique fuzzy numbers can be converted into crisp ones. Robust ranking technique satisfies compensation, linearity, and additive properties and provides results which are consistent with human intuition. Give a convex fuzzy number ã, the Robust Ranking index is defined by
R(ã) =∫_0^1▒〖0.5(〖a_α〗^L 〗+〖a_α〗^U)dα
Where(〖a_α〗^L,〖a_α〗^U)is the α –level cut of the fuzzy number ã [7].
In this paper we use this method for ranking the fuzzy numbers values. The Robust ranking index R(ã) gives the representative value of the fuzzy number ã . It satisfies the linearity and additive property.