Date of Award

June 2018

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical Engineering and Computer Science

Advisor(s)

Mustafa C. Gursoy

Second Advisor

Senem Velipasalar

Subject Categories

Engineering

Abstract

In recent years, multimedia applications such as video telephony, teleconferencing, and video streaming, which are delay sensitive and bandwidth intensive, have started to account for a significant portion of the data traffic in wireless networks. Such multimedia applications require certain quality of service (QoS) guarantees in terms of delay, packet loss, buffer underflows and overflows, and received multimedia quality. It is also important to note that such requirements need to be satisfied in the presence of limited wireless resources, such as power and bandwidth. Therefore, it is critical to conduct a rigorous performance analysis of multimedia transmissions over wireless networks and identify efficient resource allocation strategies.

Motivated by these considerations, in the first part of the thesis, performance of hierarchical modulation-based multimedia transmissions is analyzed. Unequal error protection (UEP) of data transmission using hierarchical quadrature amplitude modulation (HQAM) is considered in which high priority (HP) data is protected more than low priority (LP) data. In this setting, two different types of wireless networks are considered. Specifically, multimedia transmission over cognitive radio networks and device-to-device (D2D) cellular wireless networks is addressed. Closed-form bit error rate (BER) expressions are derived and optimal power control strategies are determined.

Next, throughput and optimal resource allocation strategies are studied for multimedia transmission under delay QoS and energy efficiency (EE) constraints. A Quality-Rate (QR) distortion model is employed to measure the quality of received video in terms of peak signal-to-noise ratio (PSNR) as a function of video source rate. Effective capacity (EC) is used as the throughput metric under delay QoS constraints. In this analysis, four different wireless networks are taken into consideration:

First, D2D underlaid wireless networks are addressed. Efficient transmission mode selection and resource allocation strategies are analyzed with the goal of maximizing the quality of the received video at the receiver in a frequency-division duplexed (FDD) cellular network with a pair of cellular users, one base station and a pair of D2D users under delay QoS and EE constraints.

A full-duplex communication scenario with a pair of users and multiple subchannels in which users can have different delay requirements is addressed. Since the optimization problem is not concave or convex due to the presence of interference, optimal power allocation policies that maximize the weighted sum video quality subject to total transmission power level constraint are derived by using monotonic optimization theory. The optimal scheme is compared with two suboptimal strategies.

A full-duplex communication scenario with multiple pairs of users in which different users have different delay requirements is addressed. EC is used as the throughput metric in the presence of statistical delay constraints since deterministic delay bounds are difficult to guarantee due to the time-varying nature of wireless fading channels. Optimal resource allocation strategies are determined under bandwidth, power and minimum video quality constraints again using the monotonic optimization framework.

A broadcast scenario in which a single transmitter sends multimedia data to multiple receivers is considered. The optimal bandwidth allocation and the optimal power allocation/power control policies that maximize the sum video quality subject to total bandwidth and minimum EE constraints are derived. Five different resource allocation strategies are investigated, and the joint optimization of the bandwidth allocation and power control is shown to provide the best performance. Tradeoff between EE and video quality is also demonstrated.

In the final part of the thesis, power control policies are investigated for streaming variable bit rate (VBR) video over wireless links. A deterministic traffic model for stored VBR video, taking into account the frame size, frame rate, and playout buffers is considered. Power control and the transmission mode selection with the goal of maximizing the sum transmission rate while avoiding buffer underflows and overflows under transmit power constraints is exploited in a D2D wireless network. Another system model involving a transmitter (e.g., a base station (BS)) that sends VBR video data to a mobile user equipped with a playout buffer is also adopted. In this setting, both offline and online power control policies are considered in order to minimize the transmission power without playout buffer underflows and overflows. Both dynamic programming and reinforcement learning based algorithms are developed.

Access

Open Access

Download

Included in

Engineering Commons

Share

COinS