Hierarchical DWT based optimal diversity power allocation for video transmission in 4G OFDMA wireless systems

In this paper we propose novel algorithms for optimal subcarrier power allocation towards video transmission in Orthogonal Frequency Division Multiple Access (OFDMA) based 4G wireless systems. A unique feature of the proposed schemes is that they employ diversity dependent subcarrier power distribution relying on a discrete wavelet transform (DWT) based hierarchical video decomposition. Another key aspect of the schemes is that they exclusively employ partial channel state information (CSI) feedback based on the order statistics of the allotted subcarriers in the OFDMA system. This significantly reduces the communication overhead required on the reverse link, thus reducing complexity and enhancing throughput. It is demonstrated that the paradigm of ordered subcarrier diversity based video distortion minimization can be formulated as a constrained convex cost minimization problem. Further, we illustrate that this cost minimization reduces to a polynomial root computation under a suitable approximation. Closed form expressions are derived for the diversity based optimal power allocation scheme. Simulation results employing several video sequences for an OFDMA system demonstrate superior performance of the proposed optimal subcarrier power allocation schemes over equal power allocation.