"Wireless mobile technologies have experienced a rapid growth during the last decades. Today, mobile communications provide access and connectivity to billions of people around the world. With a steadily increasing demand for high-bandwidth multimedia applications, such as live and on-demand video streaming, virtual reality and yet-to-be-imagined applications, the need for the development of new wireless technologies becomes imperative.Although not standardized yet the Fifth Generation (5G) wireless technology promises to offer reliable, low-latency and high-density networks capable of supporting High-Definition (HD) video. Some of the supporting technologies that could enable dynamic management of network resources and could be the key components in the design of 5G wireless networks are network virtualization and Software-Defined Networking (SDN). In this thesis, we first develop a flow management framework, called Joint Slice Manager (JSM), over wireless networks. While the problem of video traffic management has received significant attention in recent years, to the best of our knowledge, this is the first work that designs an SDN-based solution over wireless networks considering that mobile devices can be connected simultaneously to multiple radio nodes of the same technology. One of the benefits of building JSM over the SDN framework is that it enables us to extend JSM over other wireless technologies like 5G. JSM does not require any modifications to the mobile device (client) or the server, facilitating a quicker implementation. Our simulation results show significant improvements not only in video quality, but also in the radio nodes' utilization.In order to deal with the network capacity constraints derived from the expected demand of video traffic mentioned above, we propose new load balancing and Device-to-Device (D2D) relaying strategies that make better use of the network resources and increases the throughput respectively.We develop three multipath load balancing algorithms that deal with the problem of how to split traffic such that the traffic load remains at the desired level. The performance evaluation focuses on uplink wireless networks transmitting User Data Protocol (UDP) video traffic. Our simulation results show significant improvements over state-of-the-art algorithms not only in performance indicators like the splitting error and power consumption, but also in Peak Signal-to-Noise Ratio (PSNR).We also develop a method to improve mobile device throughput by devising the use of mobile devices with multiple network interfaces that act as full duplex D2D relays. We model the problem as an optimization problem and propose a family of heuristic methods for selecting the relays. Our simulation results indicate that using mobile devices as relay not only helps to improve the total throughput, but also to improve fairness (throughput at the cell edge) while significantly reducing the transmission power consumption." --