FORM-2 THE PATENTS ACT, 1970 (39 of 1970) & THE PATENTS RULES, 2006 PROVISIONAL SPECIFICATION (See section 10; rule 13) DYNAMIC RESOURCE ALLOCATION AND SCHEDULING METHOD FOR QOS IMPROVEMENT TATA CONSULTANCY SERVICES LIMITED, an Indian Company of Nirmal Building, 9th floor, Nariman Point, Mumbai 400 021, Maharashtra, India. THE FOLLOWING SPECIFICATION DESCRIBES THE INVENTION FIELD OF THE INVENTION This invention relates to the field of Orthogonal Frequency Division Multiple Access (OFDMA) systems. In particular, this invention relates to a way of providing optimization of guaranteeing Quality of Service (QoS) by dynamic resource allocation and scheduling in OFDMA systems. BACKGROUND OF THE INVENTION The next generation broadband wireless applications require high data rate, low latency and minimum delay, i.e., highly demanding Quality of Service (QoS), which cannot be provided unless the limited system resources are intelligently used and properly optimized. The capacity of a communication system is limited by its available resources like bandwidth and power. In a practical communication system, these quantities are finite and mostly scarce in nature. With fixed bandwidth and power, the system capacity also becomes fixed, which demands highly optimized system resource utilization to increase the spectrum efficiency. More over, in a multi-user scenario, system performance optimization cannot be guaranteed by optimizing only the individual link performance. Dynamic Resource Allocation and scheduling is a kind of cross layer optimization method mainly involving physical layer (PHY) and Media Access Control (MAC) to manage the system resources adaptively to achieve the system goal. If PHY and MAC layers are chosen to optimize the network resources, the best way to meet the objective is by exploiting the frequency and temporal dimension of the resource space. Dynamic resource 2 optimization approaches basically attempt to match the requirements of data-link connections to the available physical layer resources dynamically to maximize some system metrics. OFDMA is the standard multiple access scheme for the next generation wireless standards like WiMAX (Worldwide Interoperability for Microwave Access), LTE (Long Term Evolution), IMT-A (International Mobile Telecommunications - Advanced). The features of OFDMA including high resistance to fading and interference, multipath mitigation, high spectral efficiency, frequency diversity, scope of adaptive modulation and coding for individual user, fine granularity in temporal and frequency domain fit perfectly in fast mobile scenarios. OFDMA, also referred to as multi-user-OFDM is normally characterized by a fixed number of orthogonal sub-carriers to be allocated to the available users. The traffic for the different users is multiplexed using time-frequency mapping and is carried on the assigned sub-carriers. The randomness of the individual channel fading leads to Multi User Diversity (MUD). MUD gain takes advantage of having many users with uncorrected interference patterns. Resource allocation methods and scheduling schemes intelligently assign mutually disjoint resources to the users to meet the individual QoS requirement. When the prior knowledge of channel condition is available, resource allocation and scheduling scheme uses that information to optimize the system performance by dynamically allocating the resources following a pre-defined method to maximize MUD gain. The "already proposed optimization objectives and methods mostly consider instantaneous gain in 3 system performance, where the system is either to attain maximum aggregate throughput or to provide fairness among the users or to have a trade-off between them, but rarely considers the priority of premium users who demand privileged service. It is observed that, in order to satisfy the objective of maximizing the total capacity, each resource like a sub-carrier in OFDMA systems should be allocated to the user with the best gain on it, and the power should be allocated using the water-filling method across the sub-carriers. Large gain in throughput can be achieved through this kind of multiuser raw rate maximization scheduling scheme by properly utilizing MUD, but simultaneous fairness guarantee must also be provided for practical wireless networks. Proportional Fairness (PF) optimization provides the required fairness by transmitting the OFDMA sub-carrier when user's channel condition is good. The PF also follows up with its past throughput achieved in order to satisfy each user's instantaneous data rate requirement. PF scheduler or resource allocation tries to balance the fairness among the users in terms of outcome or throughput, while implicitly maximizing the system throughput extensively. PF optimization is a pure outcome fairness metric, which is simple to use, but does not guarantee fairness in a strict sense and it takes scheduling decision based on instantaneous channel condition. The approach of resource allocation and scheduling scheme based on instantaneous guarantee of QoS, like PF, does not consider the time-diversity gain incurred when long-term statistical QoS is guaranteed. Considering the channel dynamics of mobile wireless environment, substantial time-diversity gain can be achieved in delay-tolerant 4 applications, which can accept mean QoS guarantee. Another degree of freedom in QoS guarantee is the priority of the user. A good resource allocator or scheduler generally depends on the business model of the operator, where QoS guarantee is an important feature. Mostly, user satisfaction of a well-served user increases only marginally by increasing the QoS level even further. Resource allocation and scheduling play an important role in QoS provision. The scheduling schemes and methods for wire-line networks cannot be directly applied to the wireless networks because they consider only traffic and queuing status, but not the channel state dependent capacity variation. Even if large bandwidth is allocated to a certain connection, delay-bound or throughput performance requirement may not be satisfied, and the allocated bandwidth is wasted when the wireless channel experiences deep fades. In order to overcome this, an invariable over provisioning of bandwidth is made to satisfy the strict QoS requirements of the high priority users, which in turn, severely compromises on the optimization issue. To schedule wireless resources (such as bandwidth, time-slot and power) efficiently for satisfying diverse QoS parameters like throughput, buffer-size and delay bound behaviour induced by heterogeneous, elastic or non elastic traffic should be considered along with the dynamic variation of wireless channel. Hence, proper QoS provisioning plays a major role in deciding resource allocation and scheduling strategy. There are several scheduling methods existing as described in the prior art documents given below: 5 For instance, the United States Patent Application No. 20080225742 titled 'Scheduling method and system for guaranteeing real-time service quality of wibro cpe' published on 18.09.2008 discloses a scheduling method and apparatus for guaranteeing real-time service quality of Wireless Broadband (WiBro) customer premises equipment (CPE). The scheduling apparatus includes a real-time protocol (RTP) packet monitoring unit for monitoring an RTP packet passing through a local area network (LAN) section and detecting a bandwidth of real-time service and a queue managing unit for determining a window size corresponding to the bandwidth of a real-time service checked by the RTP packet monitoring unit and generating/changing a real-time service queue. The scheduling method and apparatus monitor an RTP packet and adjust a real-time service queue, thereby ensuring real-time service quality of terminals. Similarly, United States Patent Application No. 20080205532 titled 'Carrier allocation method in ofdm system and transmitting apparatus using the method' published on 28.08.2008 discloses a method for allocating sub carriers in an orthogonal frequency division multiplex (OFDM) system, and a transmitter thereof. In this OFDM system, the sub carrier allocation is performed in a buffer before a modulation mapping operation is performed according to a modulation method such that the delay may not be generated in the sub channel formed on the symbol axis. The delay corresponding to the symbols is prevented without using any additional hardware for eliminating the delay generated when the sub carriers are allocated to the sub channel formed on the symbol axis. 6 Further, United States Patent Application No. 20080232320 titled 'Dynamic resource allocation method based on frequency reuse partitioning for ofdma/fdd system, and frame transmission method therefore' published on 25.09.2008 discloses a dynamic resource allocation method based on OFDMA/Frequency Division Duplex (OFDMA/FDD) system for acquiring a multi-user diversity gain by allocating a good channel to each mobile station in consideration of fairness thereof based on channel circumstance obtained as a feedback from each mobile station to which each base station services, and a frame transmission method for the dynamic resource allocation. Still further, United States Patent Application No. 2008004029 titled 'Optimizing of channel allocation in a wireless communications system' published on 03.01.2008 discloses Orthogonal Frequency Division Multiple Access OFDMA and Orthogonal Frequency Division Multiplexing OFDM techniques involving transmitting the map messages in order to inform user terminals about the slots allocated to them. The idea is to use semi-static allocation maps, wherein a location, size, and modulation and coding method of allocated slots are predefined for a connection. Then, the validity information indicating which of the predefined allocations are to be used for a current frame is defined. Still further, United States Patent Application No. 2007211616 titled 'Resource allocation for shared signaling channels' published on 13.09.2007 discloses a shared signaling channel which can be used in an Orthogonal Frequency Division Multiple Access (OFDMA) communication system to provide signaling, acknowledgement, and power control messages to access terminals within the system. The shared signaling channel may comprise reserved logical resources that can be assigned to sub carriers, OFDM symbols, or combinations thereof. Also, the patent documents numbered US2008165766, US2007255839, US2009028095, US7324434, US7379741, US2005286408, US2008256272, 3004/CHENP/2005, 915/CHENP/2004 and 751/KOL/2005 disclose different resource allocation and scheduling schemes. The requirement for a method 'which is very much suitable and feasible for all wireless broadband systems including WiMAX, LTE, LTE-Advanced, IMT-A and which allocates the resources like OFDMA sub-carriers to the users to converge to their QoS parameters of data rate, delay requirement with a degree of biasing to the higher priority users, still persisted which were not met in the prior art documents. Also, the priority has to be computed from the QoS class and degree of satisfaction demand in heterogeneous traffic condition, to reach the global optima of throughput maximization and outage minimization and local optima of individual packet loss minimization. Therefore there is felt a need for a resource allocation and scheduling method which: • is suitable for all wireless broadband systems; • allocates the resources to the users to meet individual user's QoS parameters in a converging sense; • has a biasing towards higher priority users, i.e. providing privileged service to the high priority users ; • achieves overall throughput maximization and outage minimization; 8 • maximizes time diversity gain; • provides substantial fairness among all users ; and • achieves individual packet loss minimization. OBJECTS OF THE INVENTION It is an object of the present invention to provide a resource allocation and scheduling method which is suitable for all wireless broadband systems. It is another object of the present invention to provide a resource allocation and scheduling method which allocates the resources to the users to meet individual user's QoS parameters in a converging sense. It is still another object of the present invention to provide a resource allocation and scheduling method which has a biasing towards higher priority users. It is yet another object of the present invention to provide a resource allocation and scheduling method which achieves overall throughput maximization and outage minimization as well as provides considerable fairness among all the users and maximizes time diversity gain. It is still another object of the present invention to provide a resource allocation and scheduling method which achieves individual packet loss minimization. SUMMARY OF THE INVENTION In accordance with the present invention, a Priority Fair Scheduling (PFS) scheme and method in dynamic OFDMA system is provided, which gives 9 priority based optimum resource allocation and scheduling with individual user QoS provisioning based on the QoS class it belongs and the QoS requirement it demands. PFS method attempts to meet user's long-term mean QoS requirement for a non premium user class and short-term or instantaneous QoS requirement for a premium user class by dynamically estimating individual user's weighting factor, which in turn is generated from user's estimated QoS defined priority profile. User's priority profile is based on his measured priority-index, which is derived from the QoS class it belongs to and the degree of service satisfaction it demands. Priority index estimation aims to achieve the global optima of maximizing aggregate system throughput and fairness as well as the local optima of minimizing the individual packet loss probability. PFS optimization also exploits the time diversity gain of wireless mobile networks and performs incremental optimization of proportional fairness and overall system raw-rate maximization, particularly for the delay-tolerant applications. This method dynamically assigns system resources like OFDMA sub-carriers to heterogeneous mobile users with diversified QoS requirements to deliver QoS guarantee in statistical or deterministic sense based on the delay-bound characteristics of the application. The method is composed of three modules: scheduler, resource allocator and priority-index estimator. QoS parameters across the different service classes include data rate and delay, where the parameters' values vary as per the user class and the application running. QoS guarantee to a non premium user is assumed in long term average basis and the average is computed over few frame 10 duration and for premium classes QoS guarantee is satisfied in quasi-deterministic way. Constant Bit Error Rate (BER) and fixed QoS are considered within the period of averaging, which is equal to few frame duration. Frequency re-use factor of 1 is considered, so that all the OFDMA sub-carriers are available to be assigned to the users. However, unlike most of the current resource allocation frameworks, this invention does not assume homogeneity in QoS requirement. A heterogeneous traffic model, with diversified QoS demand by the users is considered. This invention involves a very realistic assumption in the current and next generation broadband wireless networks such as WiMAX, LTE, LTE-A and IMT-A. The present invention involves a resource allocation and scheduling scheme and method which provides delay-differentiated allocation of system resources to satisfy the users' requirement based on their QoS class and demand. In contrast to the methods which provide proportional fair optimization and raw-rate maximization independently to achieve the global optima of system throughput maximization and fairness, the method in accordance with the present invention provides both kinds of optimizations simultaneously and also attampts to maximize time diversity gain. This method also achieves the local optima of individual requirement satisfaction based on the service or QoS class the user is assigned to. The present invention consists of three steps which are invoked sequentially. These steps are: (1) initialization of resource allocation to the users at the start of the allocation process, (2) estimating the priority index of individual user based on the service class it is assigned to by considering the QoS class, 11 delay, buffer size and data rate requirement, and (3) assigning resource allocation to provide the QoS related instantaneous minimum or average data rate mainly depending on the user QoS class, by minimizing packet loss or outage by considering the instantaneous or predefined delay tolerance value and queue size of the user's traffic and maximizing the aggregate system throughput and fairness. In the initialization process of resource allocation to the users, i.e., at the start of the allocation or scheduling process, QoS^aware pure proprtional fair method is implemented in order to fulfill the initial data rate requirement of each user irrespective of his QoS service class, in a reasonable way. The initialization phase is mainly based on preserving a balance between two mutually exclusive interests of wireless system optimization; viz., maximization of overall system throughput and fairness among the users. Assignment of initial resources to the users is based on the philosophy of simultaneously providing system throughput maximization as well as fairness in an optimum and QoS-aware way. Priority index of individual user is dynamically estimated at each frame or scheduling instant; based on the service class, delay requirement, remaining queue length and data rate requirement of the individual user. This is an important step of the overall scheme, as it provides the necessary convergence of achieving, local optima of minimizing packet-loss and outage probability, while satisfying the QoS requirement^ of the users. Based on the estimated priority index of the individual, optimum resource allocation and scheduling process of the PFS method is initiated. This part ensures the convergence of global optima of niaximization of aggregate 12 system throughput and overall fairness. Here the main objective is to maximize the time diversity gain of delay-tolerant applications, which is achieved by defining the proportional fairness metrics as a maximization of the ratio of instantaneous channel condition to the achieved mean data rate up to the time instant of computation and QoS metric like throughput is considered in an average sense over that computation time window, which is considered to be larger than the coherence time of the wireless channel. However, the delay-sensitive high priority users will enjoy their satisfaction of instantaneous QoS requirement. Users of delay-tolerant applications with achieved mean data rate less than the required QoS value are only allowed to participate in the scheme of optimization. If low priority user applications or the user applications with low delay sensitive value achieves higher average throughput than its mean QoS requirement at the point of allocation, it will be ignored at that instant of allocation, which helps the system to control overachieving of QoS requirement. This idea of optimization in PFS makes the resource allocation and scheduling scheme follow the QoS profile of users with delay-tolerant applications within a defined long term duration, in the most optimum way. This is achieved due to the fact that with the progression of time cycles or frame numbers, ergodicity of the system and time diversity gain are maximized. BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS The invention will now be described with reference to the accompanying drawings, in which: Figure 1 shows a typical downlink multiuser wireless system; Figure 2 shows a conceptual system of cross-layer optimized resource allocation and scheduling; 13 Figure 3 shows a MAC-centric Cross-layer optimization reference system; Figure 4 shows an OFDMA Priority Fair Scheduler (PFS) system; Figure 5 shows a comparison of user achievable throughput by PFS and PF method, when number of users = 20; Figure 6 shows a comparison of user achievable throughput by PFS and PF method, when number of users = 30; Figure 7 shows a bar-chart comparison between PFS and PF; and Figure 8 shows a Cumulative Distribution Function Plot in accordance with Figure 7 DETAILED DESCRIPTION OF INVENTION The drawings and the description thereto are merely illustrative of a resource allocation and scheduling method for QoS improvement and optimization and only exemplify the invention and in no way limit the scope thereof. The present invention is a resource allocation and scheduling scheme and method that optimize the performance of dynamic OFDMA systems in a QoS aware way through cross-layer optimization involving physical (PHY) and medium access control (MAC) layers. The method is named as Priority Fair Scheduler (PFS). PFS is composed of three important and well defined modules: scheduler, resource allocator and priority-index estimator. In OFDMA systems, resource allocator maps OFDMA sub-carriers to the user at each frame, scheduler assigns the time-slots to transmit data from each user's buffer and priority-index estimator estimate the priority-index of the users dynamically 14 based on the QoS requirement of data rate, delay of the user and the current status of channel condition and user's queue length. OFDMA is the standard multiple access scheme for next generation wireless standards like WiMAX, LTE, IMT-A. OFD]vlA sub-carrier allocation methods dynamically assign mutually disjoint sub-carriers to the users by taking the advantage of MUD to meet the1 some specified system performance objective with the help of knowledge of the channel condition available from channel quality indicator (CQI). A typical downlink mobile wireless system is shown in Figure 1. There are K number of users and N number of resource elements to be managed by the Base Station (BS). Considering an OFDMA system, the resource elements can be assumed as OFDMA sub-carriers. The role of the BS is like a nodal decision-maker to allocate the resources to the user under some constraints to achieve the system objective. The present invention utilizes a MAC-centri optimization, which is relatively simpler and easy to implement. The MAC-initiated system is based on the information like minimum data rate for the user, maximum packet-delay and CQI from higher and lower layers. The last two parameters (delay and CQI) come to MAC directly from adjacent layers and the minimum data rate requirement parameter comes from a remote application layer, which gives the true impression of cross-layer optimization. Based on these parameters, MAC takes the scheduling decision with the help of an in¬built method. A Conceptual system of MAC-cetric cross-layer optimized resource allocation is shown in Figure 2 and MAC-centnc cross-layer optimization reference system is shown in Figure 3. The interference from 15 adjacent cells is treated as the background noise. The total available bandwidth B is partitioned into N equal narrowband mutually exclusive OFDMA sub-carriers; the sub-carriers are denoted as Si, S2 ....Sjxij where sn = B_ N Hz and in order to overcome frequency selective fading, sn is chosen to be sufficiently smaller than the coherence bandwidth of the channel, i.e. Sn ~ c, where c is the coherence bandwidth of the channel. Each OFDMA sub-carrier n, belonging to a user k, is subject to flat fading, path loss and shadowing with channel gains H = In addition, the signals suffer from additive white Gaussian noise (AWGN), which has Gaussian distributed with zero mean with power spectral density (PSD) N0 and total noise power density is assumed as "T including the background noise. Sub-carrier allocation is performed with the periodicity of frame-duration. Each sub-carrier allocation instant is called an epoch, which is assumed to be less than the channel coherence time, i.e. where Tf is the frame-duration and Tc is the channel coherence time. Perfect channel characteristic is assumed in the form of channel state information (CSI). The QoS of the kth user is described by the minimum individual rate requirement = [yu y2, .... YK] in bits/sec. Total available transmitter power is Fr, PFS method is simplified by equally distributing the total available transmitted power as performance can hardly be deteriorated by equal power allocation to each sub Carrier. fa is the transmitted power for the nth sub-carrier when transmitted to the kth user, which is equal to Let tof = f( fa")be the instantaneous achievable rate for the kth user when nth sub- 16 carrier is allocated at the t time instant and is the achieved data rate of the k user at allocation epoch t. is expressed as, equation (1) According to Shannon's capacity theorem, f {hkm) can be expressed as: where Pknl is the sub-carrier assignment matrix at allocation epoch t, if the nth 17 sub-carrier is assigned to the k111 user at the tm time instant, then Phu equals to 1, otherwise 0. Equation (3) reflects the mutual exclusivity of the sub-carrier allocation among the users, which states that no sub-carrier will be allocated to more than one user in a cell in one allocation epoch. In practice the achieved data rate is less than that of what equation (2) suggests, as there exists a few dB SNR gap. SNR gap in simplified term can be approximated By long term optimization, it is meant that sub-carrier allocation will be done in each epoch but the performance metric like fairness, QoS guarantee, and system throughput will be computed in long term by statistical mean-value sense. By long term, few allocation epochs or frames is meant. Let be the long-term unit, then: equation (5) where r' is the frame duration, which is taken as the allocation epoch. In order to get time-diversity gain the following condition should be met, which becomes the lower delay-bound of the application-running: equation (6) To avoid the frequency selectivity of the mobile tireless channel, Now, in the system-equations (1-4), long-term notion needs to be incorporated: OFDMA Priority Fair Scheduler (PFS) system architecture is shown in Figure 4, which consists of three components, namely, Resource Allocation 18 Module, Priority Index Estimator (PIE) and Scheduler. PIE calculates the unique priority of each of the user based on QoS parameter like maximum buffer size, maximum tolerable delay, QoS class, current queue length of the user and assigns each user with an unique priority index. Let the available QoS classes be , where l