The IEEE 802.16 or WiMAX webs came into being in the twelvemonth 2001 to make full the spread between Wireless Local Area Network and the planetary web ( cyberspace ) . Admission Control and Scheduling plays an of import function in supplying the Quality of Service ( QoS ) that guarantees to heterogenous categories of traffic with different QoS demands. Admission control is the QoS mechanism that describes whether a new connexion can be admitted by the base station.
Existing CAC mechanisms were unable to supply both bandwidth and hold warrants for the admitted connexions. The end of this research is to suggest an efficient bandwidth and detain guaranteed Call Admission Control ( CAC ) mechanism for QoS purveying in IEEE 802.16 webs. A new CAC mechanism that can supply both bandwidth and hold warrants for admitted connexions, called Non-Reservation based Prioritized CAC with Degradation has been proposed for Fixed and Mobile WiMAX and the public presentation is evaluated by comparing it with the bing CAC mechanisms. An analytical theoretical account based on Markov Chain Model has been developed to measure the public presentation of the CAC mechanisms.
The public presentation of the proposed strategy is evaluated in footings of barricading chances of the connexions and bandwidth use of the system. We besides propose an algorithm to plan the optimal QoS parametric quantities so that the barricading chances of service flows are less than the given threshold. This is peculiarly utile for web operators when they need to repair the QoS parametric quantities, so that they can assure both QOS and public presentation warrants to their clients.
1. Statement of job and significance
1.1 General Literature Review
Keywords: WiMAX, Quality of Service ( QoS ) , Broadband Wireless Access Network ( BWA ) , Scheduling, Packet programming, Token Bucket, Uplink, Downlink, Call Admission Control ( CAC ) , Medium Access Control ( MAC ) , Mobile Station ( MS ) , Base Station ( BS ) , Voice/Data, Multiplexing, Markov Chain theoretical account, Fixed WiMAX, Mobile WiMAX.
The IEEE 802.16 or Worldwide Interoperability for Microwave Access ( WiMAX ) [ 1 ] webs came into being in the twelvemonth 2001 to make full the spread between Wireless Local Area Network ( WLAN ) and the planetary web ( cyberspace ) [ 2 ] . The term Quality of Service ( QoS ) [ 1 ] refers to the ability of service operator to back up user ‘s application demands with regard to service classs like bandwidth, latency ( hold ) , jitter, and traffic ( package ) loss [ 3 ] .
Admission Control and Scheduling plays an of import function in supplying the Quality of Service ( QoS ) guarantees to heterogenous categories of traffic with different QoS demands [ 4 ] . “ Admission control is the QoS mechanism that describes whether a new connexion can be admitted by the base station [ 5 ] ” . The IEEE 802.16 criterion specifies five Unsolicited Grant Service ( UGS ) , real-time Polling Service ( rtPS ) , non real-time Polling Service ( nrtPS ) , Best Effort Service ( BE ) , and extended real-time Polling Service ( ertPS ) [ 6,7 ] .There are assorted applications supported by the WiMAX programming services like Voice over Internet Protocol ( VoIP ) , Traveling Pictures Expert Group ( MPEG ) picture, File Transfer Protocol ( FTP ) [ 6 ] .
1.2 Needs country
Admission Control and Scheduling plays an of import function in supplying the QoS warrants to heterogenous categories of traffic with different QoS demands [ 4 ] . Bandwidth warrant for heterogenous categories of traffic consequences in bandwidth use of the system and bandwidth use represents the degree of web operator satisfaction. Delay warrant helps in cut downing the blocking chances. The Blocking Probabilities represents the degree of user satisfaction. Lesser the Blocking Probabilities of connexions, the more will be the user satisfaction [ 8 ] .
The bing CAC mechanisms were unable to supply both bandwidth and hold warrants for the admitted connexions. Therefore there is a definite demand for a more efficient CAC mechanism that can supply both bandwidth and hold warrants for admitted connexions for heterogenous categories of traffic in fixed and nomadic WiMAX when compared to the bing CAC mechanisms.
1.3 Technology spreads
Gap 1: Understanding the MAC protocol, informations transportation between the different beds to develop the QoS confidence mechanisms for the QoS in IEEE 802.16 BWA webs.
Supplying Quality of Service ( QoS ) warrants for heterogenous categories of traffic with different QoS demands in IEEE 802.16 Broadband Wireless Access ( BWA ) system is really of import and ambitious job, since the criterion has non specified any QoS faculty and left it for the makers for their ain execution [ 2 ] . Although IEEE 802.16 criterion defined QoS parametric quantities for different types of services like UGS, rtPS, it has non provided the existent values of the QoS parametric quantities and has left it for the web operators to make up one’s mind on that. Therefore if we can understand and happen out the bandwidth values that can be assigned to each service flow in order to maintain the blocking chances below the given threshold we can develop a suited algorithm.
Gap2: Develop a CAC mechanism for QoS purveying in WiMAX for all categories of traffic.
The chief map of CAC mechanism is to supply bandwidth and hold warrant. CAC mechanisms proposed by Wongthavarawat and Ganz [ 5 ] , Jiang and Tsai [ 9 ] supply both bandwidth and hold warrants to the admitted connexions in their CAC mechanisms. But the CAC mechanisms proposed by them were simple and they did non take the belongingss of assorted types of services into consideration. In that the CAC strategy provides both bandwidth warrants to all service flows and hold warrants to rtPS service flows. Wang et Al. [ 10 ] , Yin Ge and Geng-Sheng ( G.S. ) Kuo [ 11 ] , proposed a CAC strategy for IEEE 802.16e Mobile WiMAX. The drawbacks of these strategies are that they do non supply any hold warrants to admitted connexions [ 10, 11 ] .
This proposal addresses the Gap 2. To the best of our cognition there is no proposed CAC mechanism that can supply both bandwidth and hold warrants to the connexions. The solution is to suggest an efficient CAC mechanism for both fixed every bit good as Mobile WiMAX in such a manner that QoS warrants in footings of both bandwidth and hold are satisfied in the proposed CAC mechanisms.
1.4 Aims and methods
The chief aim of this research is to suggest efficient CAC mechanisms for both fixed every bit good as Mobile WiMAX for all categories of traffic. An efficient CAC mechanism is one which decreases the Blocking Probabilities of the connexions and increases the bandwidth use of the system. A new mechanism called Non-Reservation based Prioritized CAC with Degradation ( NRP-CAC-Deg ) is proposed for QoS purveying in WiMAX. This mechanism does non reserve any bandwidth to UGS connexions but provides precedence to the UGS connexions. Supplying the precedence to UGS connexions benefit both users and web operators because the lesser the blocking probabilities the more satisfied are the clients and the more he uses the service. This will automatically better the grosss of the service supplier.
2. Introduction
2.1 Overview of IEEE 802.16 Networks
Wireless Communication came into being when Guglielmo Marconi successfully transmitted a message across the Bristol Channel in 1896, utilizing the alleged radio telegraph [ 3 ] . The full development of “ wireless experience ” requires a connexion to the cyberspace, which is going more and more critical due to the countless services upon which we all rely, such as electronic mail, instant messaging, and Voice over IP ( VoIP ) . This deficiency of engineering to bridge the spread between local ( Wireless Local Area Network ( WLAN ) and Wireless Personal Area Network ( WPAN ) ) and planetary ( i.e. , cyberspace ) , wireless engineerings drove the development of the first version of the IEEE 802.16 [ 2 ] , which was published in 2001.
Initially the IEEE 802.16 was aimed at supplying high velocity cyberspace entree in the Point-to-Multipoint ( PMP ) mode merely. The support of Quality of Service ( QoS ) [ 12 ] was embedded, since the first release which clearly stated the function of IEEE 802.16 as a taking engineering for the support to progress multimedia applications. During the subsequent old ages the criterion has been
Amended and published in 2004 as IEEE 802.16d [ 1 ] , so as to include support to non-LOS ( Line Of Sight ) deployment. The support to mobile terminuss was merely added in the 2005 release of the IEEE 802.16 criterion, i.e. IEEE 802.16e [ 13 ] .
2.2 Objective of Work
Supplying Quality of Service ( QoS ) warrants for heterogenous categories of traffic with different QoS demands in IEEE 802.16 Broadband Wireless Access ( BWA ) system is really of import and ambitious job, since the criterion has non specified any QoS faculty and left it for the makers for their ain execution. Admission Control and Scheduling plays an of import function in supplying the QoS warrants to heterogenous categories of traffic with different QoS demands [ 5 ] . There are three conditions for constitution of a new connexion. They are viz. “ Presence of sufficient bandwidth to suit it, bandwidth and hold warrant received, care of Quality of Service of already bing 1s [ 5 ] ” .
Present province Desired StateInefficient CAC mechanisms do non supply bandwidth and hold warrants for heterogenous categories of traffic.To design an efficient algorithm for CAC mechanism that provides both bandwidth and hold warrants for all heterogenous categories of traffic for both fixed and nomadic WiMAX.Although IEEE 802.16 criterion defined QoS parametric quantities for different types of services, it has non provided the existent values to the QoS parametric quantities and left it for the web operator to make up one’s mind on that. Therefore in this thesis, an algorithm to happen the optimal bandwidth values required for each type service, while keeping the barricading chances of each type of service below a given threshold has been proposed. This helps the web operators to assure both the QoS warrants and Performance warrants to their clients.
Fig.1: Aim of work utilizing Present State/Desired State Technique [ 17 ]
2.3 Related Work
This proposal addresses the spread 2 in which an efficient CAC mechanism is to be developed for QoS purveying in WiMAX for all categories of traffic. There are different attacks by assorted writers but they have their disadvantages. These different attacks are discussed in this related work.
2.3.1 CAC for QoS Provisioning in Fixed WiMAX
1. Leong et Al. [ 14 ] proposed a Call Admission Control ( CAC ) policy for the cellular system incorporating on/off voice and best-effort informations. It besides develops a theoretical account to qualify the interaction of the voice and informations traffic flows which leads to better resource use and QoS. The proposed CAC is complex as it consists of two sub-parts, one handling admittances for voice calls and another for information calls.
2. An attack to blend pricing as one of the parametric quantities for planing an admittance control policy was proposed by Hou et Al. [ 15 ] . It uses dynamic web conditions to administer inducements and therefore determining the sum traffic in the web. The public presentation of the system is analyzed in footings of congestion bar, accomplishable user public-service corporation, and earned grosss.
3. Niyato and Hossian [ 16 ] discussed general theoretical accounts and basic admittance control attacks and so extended them into 4G wireless webs. But these bing strategies are non straight applicable to IEEE 802.16 BWA webs.
4. Wang et Al. [ 11 ] proposed a dynamic CAC strategy for IEEE 802.16d Fixed WiMAX. The proposed CAC scheme utilizations bandwidth reserve and debasement policies. Bandwidth reserve policy is used to prioritise the UGS connexions. Degradation is the method of diminishing the bandwidth allocated to the admitted connexions in order to suit more figure of connexions. Two QoS parametric quantities Maximum Sustained Traffic Rate ( MSTR ) and Minimum Reserved Traffic Rate ( MRTR ) specified in the criterion allows us to degrade the bandwidth. Initially when there is few Numberss of connexions in the web, the connexions are given maximal sum of bandwidth ( which is equal to the MSTR ) . In this paper authors shows that debasement theoretical account improves the bandwidth use of the system every bit good as lessenings the barricading chances of the service flows than that of the changeless rate CAC strategy. But this strategy does non supply any hold warrants to the admitted connexions.
5. Wongthavarawat and Ganz [ 5 ] , Jiang and Tsai [ 9 ] , proposed a programming algorithm and CAC policy for QoS in IEEE 802.16 Fixed WiMAX. In that the CAC strategy provides both bandwidth warrants to all service flows and hold warrants to rtPS service flows. Both the CAC strategies differ in the manner they provide hold warrants to rtPS connexions. Both the CAC strategies were token pail based CAC strategies, but the CAC strategies proposed were simple, in which the BS admits or rejects a connexion petition merely based on the available bandwidth. The distinction of the supported services of different precedences is considered merely in the programming algorithm alternatively of in the CAC policy.
2.3.2 CAC for QoS Provisioning in Mobile WiMAX
The CAC mechanisms for Mobile WiMAX are different from that of Fixed WiMAX because in Mobile WiMAX BS receives petitions from the Handover connexions from the neighbouring BS besides. Therefore a mechanism for managing handover petitions should besides be present in the Mobile WiMAX BS. Besides the handover connexions should be given higher precedence than the freshly originated connexions because it is more raging to drop an on-going connexion than barricading a freshly originated connexion.
1. Liping Wang et Al. [ 10 ] proposed a CAC strategy based on the IEEE 802.16e Mobile WiMAX criterion. In that Handoff connexions are given higher precedence than freshly originated connexions, because it is more raging to drop an on-going connexion instead than barricading a freshly originated connexion. The precedence to Handoff connexions is given by entirely reserving a fixed sum of bandwidth to Handover connexions. The writers besides use debasement for increasing Bandwidth Utilization and diminishing Blocking Probabilities and Dropping Probabilities. But the chief drawback of this strategy is that it does non supply any hold warrants to the admitted connexions.
2. Ge and Kuo [ 11 ] proposed a CAC strategy for IEEE 802.16e Mobile WiMAX. In that the writers gave precedence to handoff connexions by leting them to utilize the debauched bandwidth of the admitted connexions, contrary to the above strategy where bandwidth reserve is used. The drawback of this strategy is that it besides does non supply any hold warrants to admitted connexions.
To the best of our cognition there is no proposed CAC mechanism that can supply both bandwidth and hold warrants to the connexions. This proposal addresses an efficient CAC strategy for both Fixed and Mobile WiMAX which provides both bandwidth and hold warrants to the admitted connexions for all categories of traffic.
3. Research Plan
3.1 Overview of Research Plan
Planning is a cardinal factor in any research undertaking. Planning is done sing the aims, procedures involved, resources available like figure of people involved and the equipment, clip and budget. For this undertaking the basic program can be given by the undermentioned stairss represented in a block diagram as follows.
Roll uping and analyzing the relevant literature.
Analyzing QoS support in WiMAX webs.
.
Availability and Preliminary probe of stuffs to be used.
Developing CAC mechanism for Fixed WiMAX for given QoS demands.
Developing CAC mechanism for Mobile WiMAX for given QoS demands.
Analyzing and proving the developed CAC mechanism for fixed WiMAX fulfilling the given QoS demands.
Analyzing and proving the consequences.
Analyzing and proving the developed CAC mechanism for Mobile WiMAX fulfilling the given QoS demands.
Planing of algorithm for obtaining optimal QoS parametric quantity values.
Fig. 2: Research Plan Overview
3.2 Research Objectives, Hypothesis, Materials and Methods
3.2.1 Aims
Average Term Objective: To develop an efficient CAC mechanism that can supply both bandwidth and hold warrants for admitted connexions for heterogenous categories of traffic in fixed and nomadic WiMAX when compared to the bing CAC mechanisms.
Long term aim: To plan 100 % efficient VoIP ( Voice over Internet Phone ) without silence suppression [ 1 ] .
WiMAX defines five programming services that should be supported by the BS MAC scheduler for informations conveyance over a connexion. VoIP uses the Unsolicited Grant Service ( UGS ) of WiMAX [ 5 ] . The bing methods developed based on the confidence mechanisms were unable to supply 100 % efficiency. The mensurable QoS service flow parametric quantities for this programming service are Minimum Reserved Traffic Rate, Maximum Request/Transmission Policy, Maximum Sustained Traffic Rate, Maximum Latency, and Tolerated Jitter [ 1 ] . When the efficient programming and CAC mechanisms are developed, so this long term aim is achieved. This proposal addresses a portion of the long term aim where an efficient CAC mechanism is developed.
3.2.2 Working Hypothesis
Planing Non-Reservation based Prioritized CAC with Degradation for Quality of Service ( QoS ) provisioning in IEEE 802.16 Network provides more efficient bandwidth and hold warrant for all categories of traffic for bandwidth runing from 1MHz to 20MHz ( WiMAX scope of frequences ) than antecedently reported CAC mechanisms [ 5, 9, 10, 11, 14, 15, 16 ] .
Testability:
The hypothesis can be tested by measuring the public presentation of the Call Admission Control ( CAC ) mechanism by utilizing the Markov Chain Model. In this each BS is represented in the signifier of an analytical theoretical account. The obtained analytical theoretical account is simulated utilizing the MATLAB tool. The end product generated is compared with the end products of the CAC mechanisms reported by the old writers [ 5, 9, 10, 11, 14, 15, and 16 ] . If the consequences show that the proposed CAC outperforms the already reports 1s, it is a good hypothesis.
Disproof of hypothesis:
a ) There are five types of service categories “ Unsolicited grant service ( UGS ) , real-time polling service ( rtPS ) , extended real-time polling service ( ertPS ) , non-real-time polling service ( nrtPS ) , and best attempt service ( BE ) [ 7 ] ” . If we can turn out that the CAC mechanisms do n’t supply both efficient bandwidth and hold warrant for at least one category of traffic so the hypothesis is falsified.
B ) If we can turn out that there is at least one frequence between 1MHz to 20MHz [ 1 ] where the CAC mechanisms do n’t supply both efficient bandwidth and hold warrant for all categories or at least one category of traffic, so the hypothesis is falsified.
degree Celsius ) There are two types of WiMAX, viz. the Fixed WiMAX and Mobile WiMAX. The nomadic WiMAX requires an extra parametric quantity in the CAC mechanism viz. the handoff parametric quantity coming from the Handover mechanism. If we can turn out that CAC mechanisms do n’t supply both efficient bandwidth and hold warrant for at least one category of traffic in nomadic WiMAX, so the hypothesis is falsified.
Cardinal Footing:
The above hypothesis generated depends on cardinal footing. The rules used are extraneous frequence division multiplexing ( OFDM ) and extraneous frequence division multiple entree ( OFDMA ) [ 1 ] . “ OFDM is based on the thought of spliting a given high-bit-rate informations watercourse into several analogues lower bit-rate watercourses and modulating each watercourse on separate carriers-called subcarriers, or tones [ 1 ] ” . OFDMA is a rule that allows “ Sub channelisation in both the uplink and the downlink [ 1 ] ” .The sub channels can be allotted to assorted users. Fixed WiMAX ( IEEE 802.16d ) is based on OFDM and Mobile WiMAX ( IEEE 802.16e ) is based on OFDMA. These communicating rules help in planing the CAC algorithms for Quality of Service ( QoS ) provisioning in IEEE 802.16 with respect to bandwidth and detain warrant.
3.2.3 Materials and Methods to prove the Hypothesis
Materials Used:
1. TreeAge Pro ( TreeAge Software, Inc. ) is the package used for the development and analysis of the Markov theoretical accounts [ 18 ] .
2. MATLAB7.1 ( The MathWorksTM ) is the package used for the simulation of the Markov theoretical accounts [ 19, 20 ] .
3. Hidden Markov theoretical account tool is the package that supports the simulation of the Markov theoretical account by the MATLAB. This tool can be found in the MATLAB statistics tool chest.
4. A computing machine with the minimal constellation required and the demands are:
a. Operating System such as XP, Vista. b. Intel Pentium IV processor. c. 2GB free Hard disc infinite. d. Monitor that supports 256 Color Graphics. e. 2 GB RAM.
Method Used:
The public presentation rating of the CAC mechanism is obtained by utilizing the Markov Chain Model [ 5 ] which is a stochastic procedure. A Markov Chain is used to happen out the distribution for the hereafter province. The distribution for the hereafter province Xn+1, depends merely on the present province Xn and is independent of the past provinces X0, X1, — – , Xn. This analytical theoretical account resembles an ergodic Markov Chain [ 20 ] . We consider a individual Fixed WiMAX BS in isolation. This BS will have the bandwidth petitions from the SSs within the coverage country of a BS. Three types of services UGS, rtPS, nrtPS need QoS warrants and petition for connexion admittance. See the simple CAC mechanism in which the BS will acknowledge the connexion whenever there is sufficient sum of bandwidth left at the BS, to suit the connexion. Otherwise the BS will reject the connexion.
In this scenario the BS can be unambiguously represented in the signifier of a State S = ( nu, nr, nn ) based on the figure of admitted connexions of each type. State S = ( nu, nr, nn ) represents that the BS has presently admitted nu, nr, and nn figure of UGS, rtPS and nrtPS connexions severally into the web. The BS will be in a peculiar State S = ( nu, nr, nn ) until a new connexion of either type ( UGS, rtPS, nrtPS ) is admitted into the web or an on-going connexion of either type ( UGS, rtPS, nrtPS ) is terminated.
For computation of steady province chances the undermentioned stairss are followed:
1. Specify the proposed CAC mechanism.
2. Based on the proposed CAC mechanism, obtain the province infinite S of the Markov Chain. The province infinite will be different for different CAC mechanisms.
3. Obtain the province balance equation for each province S in the province infinite S of the Markov Chain.
4. Obtain the Steady State Probabilities of the all the provinces in the province infinite S of the Markov Chain by work outing the province balance equations. The steady province chance of the province S represents the chance of the BS being in that peculiar province s at any given blink of an eye of clip.
a. Calculation of new connexion barricading chance:
Using the steady province chances the barricading chances of the connexions can be calculated. The new connexion barricading chance is the chance of rejecting a new petition for connexion.
B. Calculation of Bandwidth Utilization
The bandwidth use is the mean ratio of the used bandwidth to the entire bandwidth.
The simulation of this Markov theoretical account can be done by assorted simulators like the ns-2, Omnet++ , OPNET, MATLAB. We prefer to imitate the Markov theoretical account utilizing the familiar MATLAB.
3.3 Analysis and Interpretation of Expected Results
Markov theoretical account developed utilizing the TreeAge Pro ( TreeAge Software, Inc. ) is analyzed foremost. We are traveling to utilize MATLAB7.1 ( The MathWorksTM ) [ 19, 20 ] package in this proposal to imitate our Markov theoretical account. The frequence of operation considered is 1-20 MHz. The arrival procedure of new connexion petitions for UGS, rtPS and nrtPS is considered Poisson with arrival rates I»UGS, I»rtPS, I»nrtPS and the arrival rates of all the connexions are assumed to be same. The service clip for UGS, rtPS, and nrtPS connexions are exponentially distributed. Maximal Sustained Rate ( in Kbps ) , Minimum Reserved Rate ( Kbps ) , Bucket Size ( bi spots ) , Delay ( MS ) are taken as the input QoS parametric quantities. The MATLAB plan is given the above input parametric quantities in the numerical signifier and the assorted graphs are plotted.
We will foremost choose the best theoretical account of the already bing CAC theoretical accounts and compare them with the proposed theoretical account. The graphs give the clear visual image of the consequences for all scope of frequences. The XY-Plot we analyze for holes WiMAX includes Blocking Probability vs. Total Arrival Rate ( 1/sec ) , Bandwidth Utilization of system vs. Entire Arrival Rate ( 1/sec ) . A new Non-Reservation based Prioritized CAC with Degradation ( NRP-CAC-Deg ) mechanism which provides Bandwidth Guarantees and besides satisfies the hold guarantees to all connexions is proposed and this is compared with the bing 1s [ 5, 9, 10, 11, 14, 15, and 16 ] .
The CAC mechanisms for Mobile WiMAX are different from that of Fixed WiMAX because in Mobile WiMAX Base Station receives petitions from the Handover connexions from the neighbouring BS besides. The NRP-CAC-Deg mechanism proposed for Fixed WiMAX can be extended for Mobile WiMAX by including the mechanism for managing handover connexion petitions besides [ 1 ] . The XY-Plot we analyze for holes WiMAX includes Blocking Probability and dropping chance vs. Entire Arrival Rate ( 1/sec ) , Bandwidth Utilization of system vs. Entire Arrival Rate ( 1/sec ) .
The Blocking Probabilities represents the degree of user satisfaction. Lesser the Blocking Probabilities of connexions, the more will be the user satisfaction. Bandwidth Utilization represents the degree of web operator satisfaction. The more the Bandwidth Utilization, the more the grosss the web operator will acquire. If the graphs of the proposed CAC mechanism shows the best sum of lessening for Blocking Probability vs. Total Arrival Rate ( 1/sec ) and best sum of addition for Bandwidth Utilization of system vs. Entire Arrival Rate ( 1/sec ) , so the hypothesis tested is verified.
