Adaptive Resource Allocation in Multiuser OFDM Systems Literature Survey Multidimensional Digital Signal Processing Malik Meherali Saleh The University of Texas at Austin [email protected] Spring 2005 Abstract Orthogonal Frequency Division Multiplexing (OFDM) has successfully prevented ISI in a frequency selective wireless environment. An interesting application of OFDM is in Multiuser OFDM systems in which multiple users share the same channel and the total transmit power. Due to independent multipath fading characteristic for each user, channel diversity is also created. This survey studies the methods that have been proposed in the literature to allocate the resources and achieve better performances in data rates and Bit error rates and minimum transmit power levels.I. Introduction Next Generation wireless communication systems will support wireless multimedia and wireless internet access which require high data rate and complex designs. High data rate communication over wideband channels are significantly limited by inter-symbol interference (ISI) due to frequency selective or time dispersive nature of the channels. In a multiuser systems such as cellular systems, users experience ISI [1] as a result of multiple copies (multipath) of the transmitted signal created by the objects (such building, cars, etc) around them. To combat ISI, multicarrier modulation techniques, including Orthogonal Frequency Division Multiplexing (OFDM) are among the possible solutions that have been suggested. OFDM [2] divides a broadband channel into narrow subcarrier (of the same width) such that the channel response on a particular subcarrier seems flat. Adding a guard band or cyclic prefix (CP), whose length equals the dispersion time of the channel, to the transmitted symbols, makes each of the subcarriers parallel independent additive white Gaussian noise channels. This setup allows the received signal to be ISI free. II. Background For tutorial purposes the allocation of bits for a single user OFDM system is briefly explained. It is assumed that the channel information for the subcarriers is known. Under the total power constraint a greedy algorithm, also known as the water-filling [3] algorithm is applied to maximize the total bit rate. It could also be run for a fixed data rate constraint while minimizing total power. Effectively, the algorithm assigns bit rates to the subcarriers depending on their channel gain, giving higher bit rates to higher gain channels. A subcarrier may be assigned no bits if it is in deep fade or low gain.Users of multiuser OFDM systems observe multipath fading but have independent fading parameters due to their different locations. The probability that a subcarrier appearing to be in deep fade for one user may not be in deep fade for other users is quite high. Hence, multiuser system creates channel diversity which increases with the number of users. Therefore, in multiuser OFDM environment, the system needs to allocate bits as well as subcarriers to the users. There are two approaches to allocate these resources; fixed and adaptive allocation. Fixed allocations use time division multiple access (TDMA) or frequency division multiple access (FDMA) as multi-access schemes to allocate each user a predetermined time slot or frequency band for transmission. While applying fixed allocation the system neglects the channel diversity and does not use the deep faded subcarriers for other users which do not seem as deep faded to them. [4] discusses and compares these two fixed allocation schemes in much detail. To exploit the channel diversity and achieve higher bit rates authors of [5, 6, 7, 8, 9, and 10] have suggest adaptive allocation of resources. They use the instantaneous knowledge of the channel for each user to allocate the subcarriers accordingly and then subsequently allocate the bits and transmit power for each subcarrier. The rest of this report is divided into the following sections. Section III describes the adaptive scheme proposed in [5] which minimizes the total transmitted power with the fixed user data rate. Section IV discuses the scheme which maximizes the total capacity under fixed total power mentioned in [6]. Section V mentions the Quality of Service (QoS) which is neglected by [6] and discusses the solution proposed by [7]. In Section VI, results of simulations performed by the authors using each of these methods are compared. Finally conclusions and future plans for project are presented in section VII.III. Margin Adaptive [5] presents a multiuser subcarrier, bit and power allocation scheme where all users transmit in all the time slots. The authors use the given set of user data rates and attempt to minimize total transmit power, which is also know as Margin Adaptive. Following is the problem formulated in the paper ( )nkNnKkknkDCTcfPnk,1 12,*1min,∑∑= =∈=α Where *TP is the total power, nkc,is the bite rate for kth user on the nth subcarrier, 2,nkαis the channel gain squared for nth subcarrier for the kth user and kf is the required received power. In the single user case, a subcarrier gets the additional bit if it requires the minimum power to transmit it. This is known as the greedy algorithm and the authors argue that a multiuser allocation problem is more complicated than a single user bit allocation. To make the above problem tractable the authors relax the single user per subcarrier constraint by allowing multiple users to share any subcarrier and also send non integer number of bits. With a new optimization problem and constraints, the paper applies the Lagrangian optimization technique [11]. The authors simplify the optimization problem by splitting the subcarrier allocation from the bit allocation. The subcarrier allocation is performed first and then the single user bit allocation is applied on each user, using the assigned subcarriers. IV. Rate Adaptive The authors of [6] present an algorithm that maximizes the total data rate of the multiuser OFDM system by adapting the transmit power for each user and each subcarrier. The total transmit power for the system is fixed and represented by ∑∑= ==KkNnnkSs1 1, where S is the total transmit power and nks,is
View Full Document
Unlocking...