In the Embedded Signal Processing Laboratory, we have completed research in the following topics in our work in OFDM basesation design has focused on the following: Broadband wireless system standards, e.g. IEEE 802.16e and Third Generation Partnership Project (3GPP) Long Term Evolution (LTE), consider Orthogonal Frequency Division Multiple Access (OFDMA) as the preferred physical layer multiple access scheme, esp. for the downlink. Due to the limited resources available at the basestation, e.g. bandwidth and power, intelligent allocation of these resources to the users is crucial for delivering the best possible quality of service to the consumer with the least cost.

The problem of allocating subcarriers, rates, and powers to the different users in an OFDMA system has been an area of active research. Due to the discrete decisions required in assigning only one user to each subcarrier, this problem is combinatorial in nature and is very difficult to solve optimally. Thus, previous research efforts in OFDMA resource allocation have typically focused on:

  1. Maximizing instantaneous performance: The allocation decisions are performed for the current time instant subject to the current resource constraints. Hence, the time-varying nature of the wireless channel is not exploited to improve the performance of the system.
  2. Developing heuristic sub-optimal algorithms: A popular approach in previous works to achieve near-optimal performance requires solving a large constrained convex optimization problem, which is too complex for real-time implementation. Hence, the focus has been on developing sub-optimal heuristic algorithms typically with quadratic complexity and no performance guarantees.
  3. Assuming the availability of perfect channel state information (CSI): Since allocation decisions are made based on CSI feedback from the users, this assumption is unrealistic due to channel estimation errors and feedback delay.
Our work, on the other hand, addresses the aforementioned shortcomings by:
  1. Maximizing average performance: We perform OFDMA resource allocation considering the expected performance by using the statistics of the wireless channel, allowing us to exploit temporal diversity, in addition to frequency and multi-user diversity. We propose algorithms for both continuous and discrete rate systems.
  2. Developing algorithms based on dual optimization techniques: We develop algorithms based on a dual optimization framework that have complexities that are linear in the number of subcarriers and users, and that achieve solutions that are guaranteed to be within 99.9999% of the optimal solution in typical scenarios.
  3. Assuming the availability of imperfect CSI: We assume that only imperfect CSI acquired through channel estimation and prediction is available. Thus, our allocation decisions are made while explicitly considering the error statistics of the imperfect CSI.
In summary, our work attempts to dispel the myth that OFDMA resource allocation is too complex to solve optimally, even when considering average performance and assuming imperfect CSI.

Mail comments about this page to