Joint Data and Power Allocation in HDA Transmission with Bandwidth Compression over Fading Channels

Abstract

Recently, Hybrid Digital-Analog (HDA) transmission has been proposed to exploit both the high efficiency of digital codes and the robustness of analog codes. In HDA, power allocation has been considered as an effective way to optimize the distortion performance. However, it has not yet been well studied in fading channels with bandwidth compression. Under scenarios with time-varying channel fading, due to the absence of the exact Channel State Information (CSI) at the transmitter, the digital distortion introduced by random outage cannot be neglected and existing power strategies for error-free digital transmission fail to be optimal. In this case, data allocation, which determines the amount of data for digital transmission as well as that for analog transmission by configuring the quantizer, should be jointly considered with power allocation. To tackle these issues, we firstly derive the expression of the system distortion consisting of the digital distortion and the analog distortion. Then, the joint optimization problem on data and power allocation is formulated with the goal to minimize the system distortion over wireless fading channels. Finally, we approach the optimal solution via a two-stage iterative algorithm integrated with the gradient method. Simulations results show that the proposed solution outperforms existing schemes in terms of Signal-to-Distortion Ratio (SDR) in wireless fading channels with different conditions.