Published in IET Communications Received on 24th April 2010 Revised on 15th November 2010 doi: 10.1049/iet-com.2010.0319 ISSN 1751-8628 Radio resource allocation in OFDM-based cooperative relaying networks for a mixture of elastic and streaming traffic N. Mokari K. Navaie Wireless Innovation Laboratory (WIL), Department of Electrical and Computer Engineering, Tarbiat Modares University, P.O. Box 14115-349, Tehran, Iran E-mail: navaie@modares.ac.ir Abstract: In this study, the authors formulate the optimal radio resource allocation for OFDM-based relay networks with heterogeneous delay requirements. The authors consider streaming traffic that requires a maximum guaranteed average delay and elastic traffic with flexible rate requirements. The main objective is to maximise the total transmission rate of elastic users, while average delay constraint for streaming traffic is satisfied. In the proposed formulation we also consider transmit power constraint for the base station as well as each relay. The authors then propose an algorithm based on dual approach to find the optimum subcarrier, relay assignment and power allocation. Using simulations, the authors evaluate the impact of streaming traffic on the total rate of elastic users for different number of relays and various power constraints. 1 Introduction Cooperative relaying is a promising technology for the next- generation communications such as WiMAX-2 and IMT- advanced systems because of its potential to increase the coverage area, channel capacity and communication reliability [1]. Generally, there are two basic relaying schemes including amplify and forward (AF), and decode and forward (DF) [2]. Since the radio resources are limited in wireless systems, optimal radio resource allocation is a challenging task. In OFDM-based relaying wireless systems, relay selection, subcarrier assignment and power allocation could be managed in a way to optimise a system performance metric, for example, total transmit power and/or total throughput. In single hop (non-cooperative) OFDM systems [3] proposes an adaptive resource allocation with the objective of minimising the total transmission power. In their proposed method, subcarrier assignment as well as power allocation over each subcarrier is performed based on subcarrier channel state information (CSI) with the objective of minimising the total transmission power subject to a rate constraint for each user. In [4], OFDM system with multiple types of traffic is considered and a method is introduced in which the delay requirements of real-time streaming traffic is satisfied. Also in [5], resource allocation in OFDMA systems with elastic and streaming traffic is considered. In cooperative OFDM systems the base station (BS) can also communicate with users through relays that makes the radio resource allocation problem different from non- cooperative OFDM networks. In [6, 7], resource allocation in OFDMA-based relay systems is considered. In [6], it is assumed that there is only one relay and no direct link is considered in the modelling. Also in [7], direct links are considered; however, transmission power of relays assumed to be fixed. The resource allocation for OFDM cooperative networks is studied in [8] where the objective is to maximise the total throughput of network subject to the total power constraint for BS and relays. However, the proposed method in [8] does not consider the transmit power constraint of each relay and only considers the total transmission power and proposes equal power allocation to all active subcarriers. In [9] unlike [8], one subcarrier can be used in the link between the BS and the relay and at the same time the same subcarrier can be used in the link between the relay and user. It is also assumed that the transmit power of both relay and BS are constrained. In [10], adaptive resource allocation in OFDM-based cooperative cellular networks is investigated. In addition, fairness is handled through the problem formulation whose objective is to maximise the minimum achievable throughput of users. The algorithm in the first step for allocating subcarriers and power of BS aims at exploiting multi-user diversity of the OFDM network, while greedy relay selection and iterative relay power allocation algorithms in the second step try to exploit cooperative diversity as much as possible. Note that in the above-mentioned works only one type of traffic is considered. In this paper, we formulate the optimal radio resource allocation for OFDM-based relay networks with heterogeneous delay requirements. We consider the problem of resource allocation in an OFDM relay system with ‘streaming traffic’, which requires a maximum guaranteed average delay. We also consider ‘elastic traffic’ with IET Commun., 2011, Vol. 5, Iss. 8, pp. 1083–1089 1083 doi: 10.1049/iet-com.2010.0319 & The Institution of Engineering and Technology 2011 www.ietdl.org