| In the field of Wireless Multimedia Networking, both Wireless Multimedia Sensor Networks (WMSNs) and Wireless Multimedia Mesh Networks (WMMNs) have drawn the tremendous attention of the research community in the last few years. On the one hand, the features of WMSNs ensure a wide range of applications; on the other hand, WMMNs play an important role in mobile Internet. These networks are characterized by limited resources, dynamic changes of network condition and scalability. These characteristics identify significant challenges for effective and efficient delivery of multimedia data with QoS guarantees.This thesis focuses on studying QoS communication protocols for WMSNs and WMMNs. Our research work begins with a thorough investigation and understanding of the performance bottleneck of existing protocols when supporting multimedia applications. On this basis, we propose a series of novel communication protocols, including adaptive opportunistic routing protocols in consideration of energy efficiency and QoS provisioning, network-coding-based communication protocols for multimedia transmission and QoS supports, cloud-assistated and cross-layer-based network architecture for multimedia applications.The main contributions regarding WMSNs are as follows.1. A QoS-aware Multi-sink Opportunistic Routing (QMOR) protocol is proposed to deliver multimedia information efficiently under QoS constraints for WMSNs. We begin with differential coding-based relay node selection problem in a multi-sink scenario, to reduce redundant multimedia data with maximum energy gain. We then investigate how to select and prioritize forwarder list efficiently, including transmission cost calculation, forwarder priority assignment, and QoS guarantee in timeliness and reliability. Our simulation results demonstrate that the QMOR performs better than typical WMSN QoS routing algorithm, in terms of the video transmission quality and energy efficiency.2. A Low-cost Low-complexity Opportunistic Routing (L20R) protocol is proposed to achieve an energy-efficient data delivery over lossy WSNs. We begin with a thorough investigation of properties of forwarder list under nonideal link-layer conditions. A greedy rule is designed for candidate relay selection, followed by an approximation priority assignment rule for the chosen candidates. Based on these rules, a low-complexity heuristic algorithm is designed to efficiently find a reasonable choice of forwarder list. We conduct theoretical analysis and simulation experiment to study the performance of our L2OR protocol. It is shown that L2OR can achieve performance improvement in terms of data transmission efficiency as well as energy consumption.3. A Cloud-assisted Complex Event Monitoring architecture (C2EM) is proposed to alleviate bottlenecks effectively in computing power and storage space of sensor nodes while providing reliable and stable monitoring service. C2EM involves scalar sensors, video sensors, cloudlets and clouds to leverage computation offloading reliability, service response time, coverage efficiency and energy efficiency. Through computation division, most complex computations and network environment profilers are executed on cloudlets or clouds. Sensors only need to carry out very simple operations. The simulation results demonstrate that C2EM can enhance complex event monitoring performance with potential adaptability to dynamics of complex events.The main contributions in WMSNs areas are as follows.4. A Network-Coding-based Opportunistic Forwarding protocol (NCOF) is proposed improve the transmission quality of video stream. First, a novel asymmetric coding method is designed to improve exchange gain. Second, we present a traffic-aware data scheduling algorithm, working together with the above network coding mechanism, to reduce the loss of potential coding opportunities. Finally, we design a coding-gain-based opportunistic forwarding strategy to enhance throughput. Our simulation results demonstrate that NCOF can greatly enhance video transmission quality, and efficiently utilize bandwidth and energy resources.5. A network-coding-based QoS broadcast protocol is proposed for multirate wireless networks, with a focus on maximizing throughput while providing QoS guarantees. We begin with a multirate coding graph model for coding decisions, building upon which the optimal broadcast scheduling problem is formulated as an integer linear programming model. A clique-based heuristic algorithm is then designed to get a suboptimal solution. We provide probabilistic guarantee for QoS constraint values to limit coding graph size, thus reducing computational complexity, without compromising performance. In addition, a controlled computational complexity is achieved by limiting the maximum number of data packets that participate in coding. The simulation results demonstrate the availability and flexibility of the proposed solution, as well as provide a powerful basis for rate-adaptive coding algorithm design in complex network environment.6. A Cross-layer Optimized Control Protocol (COCP) is proposed to provide efficient end-to-end QoS support for video streaming in wireless networks. A new COCP sub-layer, relying on cross-layer parameters, is introduced into the network layer. Source node notifies video synchronization information to deploy a time-aware scheme for each node. On this basis, a distributed video streaming control is deployed for each node. We also investigate a GOP-based protect strategy, working together with video streaming control mechanism. The purpose is to protect video packets with high-priority. Our simulation results show that COCP improves the quality of real-time video streaming, reduces average end-to-end video frame delay, as well as effectively supports various delay requirements.We hope that our work could contribute in research, design and development of the coming multimedia Internet of Things.
|
PDF Full Text Request