Lucas García Cillanueva. 3D video transmission over LTE. Dissertação de Mestrado - PDF


Please download to get full document.

View again

of 60
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 20 | Pages: 60

Extension: PDF | Download: 0

Lucas García Cillanueva 3D video transmission over LTE Dissertação de Mestrado 14 de Fevereiro de 2014 FACULDADE DE CIÊNCIAS E TECNOLOGIA DA UNIVERSIDADE DE COIMBRA MESTRADO INTEGRADO EM ENGENHARIA ELECTROTÉCNICA E DE COMPUTADORES 3D Video Transmission over LTE Lucas García Cillanueva Júri Presidente: Vogal: Orientador: Professor Vítor Manuel Mendes da Silva Professor António Paulo Mendes Breda Dias Coimbra Professor Luís Alberto da Silva Cruz Coimbra, 14 de Fevereiro de 2014 Abstract This thesis presents a research work on quality of experience in 3D video transmission over LTE networks. The objective is to study the state-of-art of LTE and 3D video, described in the scientific literature, and to quantify the user quality of experience (QoE) during a simulated LTE transmission. The work will start by a study of the University of Wien LTE-A System Simulator and its capabilities. In addition, different scenarios with various users equipment (UEs) and base stations (enodebs) densities will be configured and simulated in order to obtain the frame-by-frame Block Error Rate (BLER) values experienced by different UEs. Once obtained, the Block Error Rate frames will be converted to packet level error traces, which will be used to introduce erasures and corruptions into the packetized 3D video bitstream. The corrupted encoded video stream will be decoded using an error-concealment capable video decoder and the decoded/recovered video quality (QoE) will be estimated based on the Structural Similarity Index of the recovered video. Finally, the QoE results for the different system configurations will allow classifying the severity of the QoE degradations due to transmission losses, through inferring the relationship between those system parameters and the achievable QoE. Keywords: 3D Video, Quality of Experience, Long Term Evolution, Evolved Node B, User Equipment, Block Error Rate, Structural Similarity Index. Resumo Esta dissertação apresenta um trabalho de investigação sobre a qualidade de experiência numa transmissão de vídeo 3D sobre redes LTE. O objectivo é estudar o estado-da-arte no que respeita a rede LTE e vídeo 3D, descrito na literatura científica, e obter a qualidade de experiência de usuário (QoE) durante uma simulação de transmissão LTE. O trabalho começará por um estudo do University of Wien LTE-A System Simulator e as suas capacidades. Para este efeito, vão ser configurados diferentes cenários com distintas densidades de utilizadores (UEs) e estações base (enodebs), com o fim de obter a taxa de erros do bloco (BLER) experimentada por diferentes utilizadores. Depois de obter esta taxa, as tramas da taxa de erros do bloco (BLER) serão convertidas em tramas de nível de erro de pacotes, que vão ser usadas para adicionar corrupções de bit em ficheiros de vídeo 3D. O fluxo de vídeo codificado e corrompido será descodificado usando um descodificador de vídeo e a qualidade do vídeo recuperado vai ser calculada com base no Índice de Similitude Estrutural. Finalmente, os resultados de QoE para as diferentes configurações do sistema permitirão classificar o nível das degradações de QoE devido a perdas de transmissão, por meio de inferir a relação entre os parâmetros do sistema e a QoE obtida. Palavras-chave: Vídeo 3D, Qualidade de experiência do Usuário, Evolução de Longo Prazo, utilizador LTE, Taxa de Erros de Bloco, Índice de Similitude Estrutural. Contents Abstract... 2 Resumo... 4 Contents... i List of Figures... v List of Tables... vii List of Acronyms and Abbreviations... ix 1. Introduction Context Outline of the dissertation LTE System Overview Key features of LTE Network architecture Protocol architecture Radio Link Control Medium Access Control LTE Physical Layer LTE frame structure OFDMA i SC- FDMA Multi- Antenna techniques Multiple transmit antennas Spatial multiplexing D Video transmissions State- of- art D video system D content creation D representation Delivery Visualization Objective quality metrics of 3D video Media- layer FR image quality models Methods, tools and simulated scenario The University of Vienna System Level LTE Simulator Using the simulator Evaluated scenarios Experiments and results Simulator output results Balloons ii Kendo Lovebird Newspaper Overview QoE results Balloons Kendo Lovebird Newspaper Overview Analysis of the results Conclusion References iii iv List of Figures Figure 1 - Relative subscriptions in mobile technologies... 5 Figure 2 - LTE network architecture... 7 Figure 3 - LTE Protocol stack and main functions... 7 Figure 4 - Overview of the LTE protocol architecture for downlink... 8 Figure 5 - Radio Link Control PDU Figure 6 - Flow of downlink data through all the protocol layers Figure 7 - LTE Generic Frame Structure Figure 8 - Resource blocks for uplink and downlink Figure 9 - LTE resource block vs. resource element Figure 10 - Diversity channel with an M-element transmit antenna array Figure 11 - Classical beam-forming with high mutual antennas correlation Figure 12 - Downlink transmission in MU-MIMO configuration Figure 13 - Structure of an end-to-end 3D video system Figure 14 - Image-based representation Figure 15 - DIBR procedure Figure 16 - MVD System Figure 17 - Simulated scenario Figure 18 - Schematic block diagram of the simulator v Figure 19 - CQI vs. BLER curves and CQI from the 10% BLER points Figure 20 - UEs network configured (30 UEs per enodeb) Figure 21 - Base station positions Figure 22 - UE positions for 5km/h Figure 23 - UE position for 36km/h Figure 24 - UE positions for 72km/h Figure 25 - Diagram of the evaluated scenario Figure 27 - Block Diagram of the BLER conversion Figure 28 - PLR results for Balloons Figure 29 - PLR results for Kendo Figure 30 - PLR results for Lovebird Figure 31 - PLR results for Newspaper Figure 32 - QoE results for Balloons Figure 33 - QoE results for Kendo Figure 34 - QoE results for Lovebird Figure 35 - QoE results for Newspaper vi List of Tables Table 1 - Types of Mobiles Stations... 4 Table 2 - PLCC performance comparison of VQA algorithms Table 3 - Configuration parameters of the System Simulator Table 4 - Encoder setting parameters of the videos used Table 5 - PLR results from Balloons Table 6 - PLR results from Kendo Table 7 - PLR results from Lovebird Table 8 - PLR results from Newspaper Table 9 - SSIM results from Balloons Table 10 - SSIM results from Kendo Table 11 - SSIM results from Lovebird Table 12 - SSIM results from Newspaper vii viii List of Acronyms and Abbreviations 2D 3D 3DTV 3GPP ARQ AVC BS BCH BLER CQI CRC DCCH DIBR DL-SCH DTCH EnodeB/eNB EPC EPS FVV H.264/AVC HARQ HSPA IP Two-dimensional Three-dimensional Three-dimensional Television 3 rd Generation Partnership Project Automatic Repeat Request Advance Video Coding Base Station Broadcast Channel Block Error Rate Channel Quality Identifier Cyclic Redundancy Check Dedicated Control Channel Depth-Image-Based Rendering Downlink Shared Channel Dedicated Traffic Channel Evolved Node B Evolved Packet Core Evolved Packet System Free Viewpoint Video ISO/ITU Video Coding Standard Hybrid Automatic Repeat Request High-Speed Packet Access Internet Protocol ix LTE MAC MBMS MCCH MCH MCS MIMO MME MPEG MS MTCH MVD OFDM PAPR PCCH PCH PDCP PDN PDU PHY PLCC PLR PSNR QoE Long Term Evolution Medium Access Control Multimedia Broadcast and Multicast Services Multicast Control Channel Multicast Channel Modulation and Coding Scheme Multiple-Input Multiple-Output Mobility Management Entity Moving Picture Experts Group Mobile Station Multicast Traffic Channel Multi-view Video plus Depth Orthogonal Frequency Division Multiplexing Peak-to-Average Power Ratio Paging control Channel Paging Channel Packet Data Convergence Protocol Packet Data Network Protocol Data Unit LTE Physical Layer Pearson Linear Correlation Coefficient Packet Loss Ratio Peak Signal-to-Noise Ratio Quality of Experience x RAN RB RLC ROHC RX SAE SC-FDMA SDU SGW SM SSIM SU-MIMO SVC TTI UDP UE UL UL-SCH UMTS UTRAN V+D VQA WCDMA Radio Access Network Resource Block Radio Link Control Robust Header Compression Receiver System Architecture Evolution Single-Carrier Frequency Division Multiple Access Service Data Unit Serving Gateway Spatial Multiplexing Structural Similarity Index Model Single User- MIMO Scalable Video Coding Transmission Time Interval User Data Protocol User Equipment Uplink Uplink Shared Channel Universal Mobile Telecommunications System Universal Terrestrial Radio Access Network Video plus Depth Video Quality Assessment Wideband Code Division Multiple Access xi xii 1. Introduction 1.1. Context Latest developed technologies based in 3D video and image transmission are translated into an increasing consumer demand for 3D content [6]. Recent studies have anticipated that the adaptive streaming portion of Internet video will grow at an average of 77% per year, reaching up to 51% of the network video traffic consumed by With the main purpose of delivering the best user experience, adaptive streaming has to optimize the video configurations during transmission. 3D video provides more realism of the scene, and it involves much more information needed to transmit the video than the two-dimensional (2D) representation. Thus, current and future networks should be able to dedicate a large amount of bandwidth to 3D video streaming services. Immersive and interactive multimedia applications over wireless will be enabled by the recent LTE standard, thanks to the low latencies and high data rates supported. LTE emerges as a 3GPP (Third Generation Partnership Project) standard, and enables high transmission data rates by supporting radio access with up to 100Mbit/s in full mobility wide area deployments and 1Gbps in low mobility local area deployments [24]. In terms of the high spectral efficiency, it is located between 5 and 10 b/s/hz for a single user and for 2 to 3 b/s/hz for the multiuser case. This enables reliable wireless transmission of huge content over the LTE networks. The latest broadband cellular technology, LTE, allows supporting different services with high data rates and different Quality of Service requirements. Therefore, it can be considered a very promising architecture for 3D video transmission. 1 1.2. Outline of the dissertation Firstly, an overview of LTE technology is presented, with emphasis on the main goals and key features of LTE. Moreover, the network and protocol architectures will be introduced. Section 3 talks about the LTE Physical Layer, the main LTE layer studied in this work. In section 4, the state-of-art of the 3D technology is explained. For this purpose, the structure of an end-to-end 3D video system and the different types of the 3D video representations are mentioned. Additionally, the main quality metrics of 3D video are enumerated. Section 5 provides information about the tools used in this scope, with emphasis on the University of Wien LTE-A System Simulator, the main tool employed in this investigation. In section 6, the experimental results are described. This section is divided into two parts: the simulator output results (Packet Loss Ratio results) and the QoE results (Structural Similarity Index results). This thesis concludes with section 7, by summarizing the results obtained and drawing some conclusions. 2 2. LTE System Overview Long Term Evolution (LTE) has been designed to support only packet-switched services, contrary to the prior cellular systems, based on circuit-switched models. The main LTE purpose is to provide IP connectivity between UE (User Equipment) and the PDN (Packet Data Network) without occurring interruptions of user applications during connection. In LTE there is a big change compared to previous mobile technologies, UMTS (Universal Mobile Telecommunications System) and HSPA (High Speed Packet Access) due to the introduction of a novel physical layer and the core network reform. The main reasons for these Radio Access Network (RAN) system design developments are the requirement to provide higher spectral efficiency, lower delay and more multiuser flexibility and secure service than the existing deployed networks. Thus, while LTE includes the evolution of the Universal Mobile Telecommunications System (UMTS) Radio Access Network by designing the Evolved UTRAN (E- UTRAN), there is also a development of other aspects, like the System Architecture Evolution (SAE), which covers the Evolved Packet Core (EPC) network. The Evolved Packet System (EPS) is included by the LTE and SAE. 3 2.1 Key features of LTE LTE aims to achieve a peak data rate of 100 Mbit/s to 326,4 Mbit/s (with ideal conditions) in the downlink and for 50 Mbit/s to 86,4 Mbit/s in the uplink (UL), with a 20 MHz spectrum allocation for each of the downlink and uplink. Thus, it is required a spectral efficiency of 5 for the downlink and 2.5 bit/s/hz for the uplink [3]. Due to the wide range of applications and requirements, LTE defines different types of User Equipment (UE), depending on the antenna configuration and the modulation chosen [4]: UE category Peak downlink data rate (Mbit/s) Downlink antenna configuration (enodeb transmit x UE receive) Peak uplink data rate (Mbit/s) Support for 64QAM in uplink Category x No Category x No Category x No Category x No Category x Yes Table 1 - Types of Mobiles Stations [4] For latency, the goals distinguish between: Control-plane latency (defined as the time for a handset to transition from various non-active states to active states), which are between 50 and 100 ms, depending on the state in which the UE originally was. Furthermore, at least 400 active UEs per cell should be supported. User-plane latency (defined as the time required to transmit a small Internet Protocol (IP) packet to the edge node of the Radio Access Network, RAN), which should not exceed 5ms in a network with a single UE (i.e., no congestion problems). LTE defined performance requirements into the WCDMA systems, for operation under realistic circumstances. Basically, the main requirement is relative to the user 4 throughput, that it should improve from 2 to 4 times. Since the main usage, especially for data services, is expected to be for mobile terminals, the LTE system is intended to be optimized for low speeds (to about 15 km/h). Relative to higher speeds, is allowed light performance degeneration for speeds up to 120 km/h, while for really high-speed applications (up to 500 km/h), only basic connectivity needs to be kept. Due to the need of coexistence of the WCDMA and LTE systems for a considerable number of years, usually in the same frequency band, the transition from both systems should be made as seamless as possible. Transitions/handovers from one system to the other will be frequently required, especially during the initial deployment of LTE, when only parts of the service area will be covered by LTE Base Stations (BSs). Relative to the transition times, for real-time applications it should be less than 300 ms, and for non-real time applications should be less than 500 ms. Figure 1 represents the evolution of relative subscriptions in mobile technologies since 1990: Figure 1 - Relative subscriptions in mobile technologies [26] 5 2.2. Network architecture In principle, the LTE network structure is quite simple. Actually, it is simplified with respect to the GSM and WCDMA structure: there is only a single type of access point, the enodeb (or Base Station, BS). Each BS can support one or more cells, providing the following functionalities: Air interface communications and PHYsical layer (PHY) functions Radio resource allocation/scheduling Retransmission control The X2 interface is the interface between different BSs. Important information is exchanged through this interface for the coordination of transmissions in adjacent cells (e.g. for reduction of the intercell interference). The S1 interface connects each BS to the core network. The LTE developed core network, called System Architecture Evolution (SAE) or Enhanced Packet Core (EPC), is based on packet-switched transmission. It consists of a Mobility Management Entity (MME), the serving gateway (connecting the network to the RAN), and the packet data network gateway, which connects the network to the Internet. In addition, the Home Subscriber Server is defined as a separate entity. The core network complies the following functionalities: Subscriber management and charging Quality of service provisioning, and policy control of user data flows Connection to external networks The network must also provide enough user security and privacy and network protection against fraudulent use. Figure 2 represents the LTE network architecture: 6 Figure 2 - LTE network architecture [27] 2.3. Protocol architecture In this section the LTE protocol architecture is explained. Figure 3 shows the different protocol layer that structure the processing specified for LTE: Figure 3 - LTE Protocol stack and main functions [25] 7 In figure 4 is illustrated a general overview of the LTE protocol architecture for the downlink. Related to the LTE protocol structure of the uplink transmission, it is quite similar to the downlink. However, with respect to multi-antenna transmission and transport format selection there are some differences between them [1]. Prior to transmission over the radio interface, incoming IP packets are passed through multiple protocol entities, enumerated bellow [1]: Figure 4 - Overview of the LTE protocol architecture for downlink [1] Packet Data Convergence Protocol (PDCP) ciphers user and signaling traffic over the radio interface. It ensures the integrity protection of the transmitted data by protecting against attack scenarios. At the receiver side, PDCP performs the 8 deciphering and decompression operations. PDCP is also responsible of the IP header compression. It is necessary to reduce the number of bits necessary to transmit over the radio interface. The header compression is based on the ROHC algorithm, used in WCDMA as well as other mobile-communication standards. Radio Link Control (RLC) is responsible for segmentation and reassembly, retransmission handling, and in-sequence delivery for higher layers. The higher layer packets need to be adapted to packet sizes that can be sent over the radio interface. This protocol is located in the enodeb since there is only a single type of node in the LTE radio-access-network architecture. The RLC offers services to the PDCP in the form of radio bearers. Medium Access Control (MAC) handles hybrid-arq retransmissions and uplink and downlink scheduling. The scheduling functionality is located in the enodeb, which has one MAC entity per cell, for both uplink and downlink. The MAC offers services to the RLC in the form of logical channels. Physical Layer (PHY) handles coding/decoding, modulation/demodulation, multi-antenna mapping, and other physical layer functions. The physical layer offers services to the MAC layer in the form of transport channels. Now the RLC and MAC entities will be explained in more detail Radio Link Control The Service Data Units (SDUs) are segmented and concatenated by the RLC into available packets for transmission across the radio channel, named as Protocol Data Units (PDUs). The PDU size can be adjusted dynamically, due to the dynamic changes of the transmission data rates. The RLC also ensures that all PDUs arrive at the receiver (RX) (and arrange for retransmission if they do not), and delivers them to the PDCP in their correct order. The scheduler decides the amount of data from the RLC SDU buffer should be selected for transmission, and in order to create the RLC PDU, the SDUs are 9 segmented/concatenated. Hence, during a LTE transmission the PDU size varies dynamically [2]. Figure 5 represents the RLC PDU creation. It is important to note that the large PDU size resulted of high data rates means to a smaller overhead. On the other hand, for low data rates is required a small PDU s
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks