Information Technology Reference
In-Depth Information
best channel gain [6]. However, if the channel gain differences among users are large,
users with higher channel gain will occupy nearly all the resource causing other users
to receive no data. The max-min problem was studied in [7] to ensure that all users
generate a similar data rate. However, this study ignored the fact that different users
may require different data rates. The authors in [8] proposed a suboptimal allocation
scheme whereby a sub-channel is first allocated to the most suitable user based on a
proportional rate parameter by assuming equal power distribution, and then power
allocation is employed to maintain fairness. However, the sub-channel scheme algo-
rithm has large computational complexity.
In this paper, we focus on a new optimization problem to achieve both proportional
fairness and maximum system capacity. As before, the sub-channel is first allocated
using a rough rate constraint to maximize capacity under the premise of equal power
allocation. Then, the power allocation algorithm is employed to maintain maximum
total capacity and ensure that each user has the required proportional rate. To reduce
the computational complexity, we introduce genetic algorithms to both the sub-
channel allocation and power allocation.
The rest of this paper is organized as follows. Section 2 introduces the multiuser
OFDM system model and derives the optimization objective function. Sub-channel
allocation and power allocation based on genetic algorithms are discussed in
Section 3 with simulation results presented in Section 4. Conclusions are drawn in
Section 5.
2
System Model
Fig. 1 depicts a multiuser OFDM system. It is assumed that instantaneous channel
information is available at the base station, and all channel information is sent to the
resource allocation algorithm through feedback channels from all the users. Then, the
resource allocation algorithm formulates the related allocation schemes and forwards
these to the OFDM transmitter. The transmitter selects a different number of bits from
each user to form an OFDM symbol. The resource allocation schemes change adap-
tively according to variation in the wireless channel.
Fig. 1.
Block diagram of multiuser OFDM system
