|Year : 2012 | Volume
| Issue : 3 | Page : 229-247
Survey of Cross-layer Designs for Video Transmission Over Wireless Networks
Ali Abdulqader Bin-Salem1, Tat Chee Wan2
1 National Advanced IPv6 Center, Universiti Sains Malaysia, 11800, Pulau Pinang, Malaysia
2 School of Computer Sciences and National Advanced IPv6 Center, Universiti Sains Malaysia, 11800, Pulau Pinang, Malaysia
|Date of Web Publication||23-Jul-2012|
Ali Abdulqader Bin-Salem
National Advanced IPv6 Center, Universiti Sains Malaysia, 11800, Pulau Pinang
| Abstract|| |
Wireless networks are designed to enable a variety of existing and emerging multimedia streaming applications. Multimedia streaming applications demand high quality of service (QoS), especially from wireless networks, because of the problems that exist in these types of environments, including interference, packet loss, delay, reduction of wireless link utilization, and so on. QoS necessitates the utilization of available resources at different layers of the multimedia system components, such as network, terminal, and content. Thus, a new design called cross-layer has been developed and is now considered beneficial by the research community. The aim of the current paper is to present different cross-layer designs or approaches that are classified into many categories. A cross-layer design based on the direction of information (i.e., downward, upward, hybrid, MAC-centric, and joint adaptation) and on the information itself (i.e., channel state information, QoS-related parameters, resources information, and application-based data) are also presented. In addition, a number of the video-streaming applications and their issues that cross-layer approaches attempt to solve, such as Pre-stored video streaming, real-time streaming, and congestion loss on high-definition three-dimensional television, are presented. As a result, a comparative study among the different approaches in cross-layer design is presented based on cost, QoS parameters (i.e., packet loss, delay, throughput), and complexity factors. This comparison can help in the selection of an appropriate design for the problems desired to be solved.
Keywords: Cross-layer design, Cross-layer optimization, Packet loss, QoS, Video transmission over wireless
|How to cite this article:|
Bin-Salem AA, Wan T. Survey of Cross-layer Designs for Video Transmission Over Wireless Networks. IETE Tech Rev 2012;29:229-47
| 1. Introduction|| |
The IEEE 802.11 wireless local area networks (WLANs) play an important role in offering connectivity to the internet. Multimedia, such as voice, video, gaming, and conferences, are becoming widespread. Hence, the demand for quality of service (QoS) support in WLANs is increasing  . At present, data carried by wireless network devices are merged with the multimedia system, and they have received a great deal of attention. Furthermore, these applications require guaranteed QoS. Video transmission over wireless networks, for example, has become a challenging issue  . Higher data bit rates, minimum packet losses, and stringent delay requirements for QoS have necessitated the search for optimum performance from different layers of the multimedia system, such as network, terminal, and content. Thus, a new design paradigm, called the cross-layer approach  , should be adopted in integrating and extending the achieved gains in individual components.
As previously mentioned, video transmission over a wireless network demands guaranteed QoS for good performance, and such a condition may be achieved through the use of cross-layer designs. In the present study, various cross-layer designs are compared. A summary of the interactions between layers is also provided to identify the parameters exposed by some layers to the others.
Section 2 describes the motivation of the cross-layer design. Issues of video streaming over wireless networks are presented in section 3. The cross-layer design and its importance are described in Section 4. Section 5 presents the types of cross-layer design based on the flow of information and on the information itself including their comparative advantages and disadvantages. The video-streaming applications and their issues solved by cross-layer optimization, such as Pre-stored video streaming, real-time streaming, and congestion loss on high-definition three-dimensional television (HD 3DTV), are described in Section 6. Section 7 presents a summary of the interactions between layers. In Section 8, a comparison between the types of cross-layer design based on the flow of information and evaluation is also presented. The comparison is based on cost, QoS parameters, and complexity factors.
| 2. Motivation of Cross-layer|| |
The cross-layer allows communication between layers by permitting one layer to access the data of another layer, thereby facilitating the exchange of information. The problem that the current paper attempts to solve is the selection of the appropriate cross-layer design based on a problem that it is desired to be solved, thus helping users choose the efficient parameters to be adopted by the proposed cross-layer design.
Cross-layer design plays an important role for the next-generation of wireless systems, featured by all IP-based protocol stack, heterogeneous access networks, and multimedia data traffic. The motivation for cross-layering in communications can be viewed from two perspectives: from a general communications viewpoint and from a more targeted wireless viewpoint.
2.1 General Communications Viewpoint
Reference  investigates how traditional layering Open Systems Interconnection (OSI) and Transmission Control Protocol/Internet Protocol (TCP/IP) in a network may not be optimal for accessing and sharing information that may be available to next-generation networks because shared layer information is a prerequisite for numerous forms of performance optimization. Cross-layer systems shift the research away from optimizing the performance of individual layers, and instead treat optimization as a problem of the entire stack. The technique considers the information available from different layers to create a system that is more sensitive to its environment, as well as load.
OSI and TCP/IP support an upward approach driven by physical and network constraints, which make capturing and responding to downward user demands or requirements difficult. Cross-layer design can help capture these concerns by providing a uniform framework at different semantic levels.
Introducing a single co-located layer for various adaptation tasks would be extremely complex, heavy, and inadequate, especially because QoS adaptation requires that the QoS is addressed at all network layers  . A cooperative solution involving the coordination among the individual adaptations of multiple layers would result in a more flexible approach, although such a solution also introduces instability. In addition, cross-layering allows the design of new kinds of applications, specially distributed applications, and applications sensitive to changing network conditions, such as QoS-sensitive multimedia applications  .
2.2 Wireless Network Viewpoint
The assumptions in the wired IP stack are inadequate for wireless networking, and TCP is known to suffer from performance degradation in mobile wireless environments attributable to the characteristics of such wireless networks, including interference, mobility resulting in a high bit error rate, and packet losses. TCP considers packet losses as an indication of congestion, consequently invoking congestion control mechanisms that result in poor performance. Cross-layering can mitigate such a problem  .
The combination of radio resource and limited power necessitates the optimization of network performance, but such optimization can hardly be met in the sub-optimal wired architecture with strict layering. Using cross-layering, better optimization is possible, as shown in  .
Wireless networks offer several possibilities for opportunistic communications that cannot be exploited sufficiently in a strictly layered design. Furthermore, the new modalities of communication offered by the wireless medium is not accommodated by layered architectures, for example, making the physical layer capable of receiving multiple packets simultaneously  .
| 3. Issues of Video Streaming Over Wireless Networks|| |
Video streaming is a multimedia transaction that is constantly received and presented by an end-user while being delivered by a streaming provider. The client can start displaying data before the entire file has been transmitted. Video streaming is a very useful feature for devices with low-storage capacity, such as mobile phones and personal digital assistants (PDAs), which use cellular, WiFi, or WiMax wireless communication technologies. Video-streaming applications can be classified into Pre-stored video streams (video-on-demand) or Real-time streams (live broadcast, online gaming, and video conferencing).
Video streaming faces various issues, especially in delivering video over wireless networks. Researchers are currently using a cross-layer design to solve video-streaming issues, which are throughput, delay, jitter, and packet loss. These issues commonly occur because of the huge amount of video-streaming data and the nature of wireless networks, such as interference and mobility. These issues are collectively called QoS. The details of each issue are as follows.
Throughput is the actual measurement of how fast data are sent  . The increase in amount and complexity of video data and coding increase throughput transmission, and thus, improvements in wireless network designs and techniques are necessary. Users are currently looking for breakthroughs in audio/video networks at the 1 Gbit/s level. An assessment on the development of user data rates reveal various emerging wireless networks, such as global system for mobile (GSM) to global packet radio service (GPRS), universal mobile telecommunication system (UMTS), high-speed downlink packet access (HSDPA), etc. Increasing the throughput until it is close enough to the maximum capacity of the bandwidth is the main challenge in throughput. Moreover, Signal-to-noise ratio (SNR) plays an important role in maximizing throughput, and a cross-layer optimization is proposed to meet this challenge.
3.2 Latency (Delay)
Delay is defined as the time it takes for an entire message to arrive completely at the destination from the time the first bit is sent out from the source  . Delay consists of four components, namely, propagation time, transmission time, queuing time, and processing time. Various OSI layers contain many parameters, such as delay tolerance in the application layer, routing path in the network layer, and appropriate scheduling algorithm in the link/MAC layer, which can be used in a cross-layer solution to avoid delay.
Jitter is related to delay. Jitter occurs when different packets of data encounter different delays and the applications using the data at the receiver side is time-sensitive  . Delay and jitter are important issues that are taken into account in real-time streaming.
3.4 Packet Loss
Packet loss occurs when one or more packets of data transferred across a computer network fail to reach their destinations. Packet loss is a critical and common problem in multimedia wireless networks caused by interference and mobility in wireless environments, resulting in a high Bit Error Rate (BER) (e.g., BER reaches up to 10-1 in 16 QAM with 4 dB bit-to-noise ratio). Appropriate modulation and channel coding schemes could be used to improve BER. Investigating the improvement of modulation and channel coding in the cross-layer technique could facilitate an easy solution for packet loss.
The video-streaming applications including their issues and the prospective solutions through the cross-layer technique to solve these issues will be presented later in this study.
| 4. Cross-layer Design for QoS in a Wireless Network|| |
In the original OSI networking model, strict boundaries between layers are enforced, where data are kept strictly within a given layer. Cross-layer optimization removes such strict boundaries to allow communication between layers by permitting one layer to access the data of another layer to exchange information and enable interaction  . For example, having knowledge of the current physical state will help a channel allocation scheme or automatic repeat request (ARQ) strategy at the MAC layer in optimizing tradeoffs and achieving throughput maximization  .
In the layered architectures of OSI, protocols at different layers are designed independently. For example, application layer adaptation schemes, such as source coding, can improve multimedia QoS, but such schemes only address performance optimization in the upper layer. However, extensive layering results in high latency and complexity in responding to changes in the QoS of the system. In contrast, adaptation schemes, such as modulation and coding in the physical layer, can rapidly respond to the change in circumstances to optimize throughput and transmission reliability. However, although multimedia applications require specific qualities in terms of throughput, delay, and packet loss rates (PLRs), the achievement of lower layer QoS targets may not satisfy the application layer QoS sufficiently. The said condition introduces the need for a cross-layer design that actively exploits the dependence between protocol layers to obtain performance gains, and better QoS guarantees to optimize multimedia QoS in WLANs  .
[Figure 1] and [Figure 2] show the OSI model and the possible cross-layer interactions that can be considered in a system design.
|Figure 1: Possible cross layer interaction between layers from down to top.|
Click here to view
|Figure 2: Possible cross layer interaction between layers from top to down.|
Click here to view
| 5. Cross-layer Design Types and Classifications|| |
In the definition of cross-layer, which focuses on the transfer of information between layers and the information itself, the classification and discussion of cross-layer designs are based on the direction of information flow between layers and the cross-layer information used. The existing cross-layer solutions are also presented based on video-streaming applications and their issues such as pre-stored streaming, real-time streaming, and congestion loss distinguished in HD 3DTV. However, the examples mentioned for each type are representatives and are not exhaustive.
5.1 Cross-layer Design based on the Direction of Information Flow
A cross-layer design based on the direction of information flow defines the interaction between layers or the layer where the adaptation occurs. Such a design is divided into five approaches based on what is performed, namely, downward, upward, hybrid, MAC-centric, and joint adaptation.
5.1.1 Downward Approach
In this approach, the lower layer acquires the information from higher layers to perform optimal adaptation , . For example, the transmission power at the physical layer can be fine-tuned by the MAC layer to increase the transmission range. In other words, downward information flow provides hints to the lower layers on how the application data should be processed  . The "cross-layer interpreter (xQoS-Interpreter)" provides a downward approach by allowing any layer of the stack to access Application Data Units (ADU) conveyed information via an Application Programming Interface (API). This approach utilizes the information included within the ADU headers to provide a generic representation of the QoS properties, thus hiding the complexity  . The "cross-layer packet scheduling scheme"  introduces schemes for audio-video transmission using IEEE 802.11e HCCA to improve QoS and quality of experience. This approach is based on the initial source of the information sent to the lower layers.
The parameters available at application layer through the application of QoS needs are Delay Tolerance, Acceptable Delay Variation, Required Throughput, Acceptable PLR, Frame Video Resolution, and Frame Rate. The approach is classified according to the target layers of these parameters:
- Application to transport
Applications may indicate their QoS requirements to TCP. Based on this information, TCP may manipulate the receiver windows for various applications to improve throughput  . In  , a part of the cross-layer solution uses Unequal Error Protection (UEP) codes at the transport layer, which is an efficient method to protect video data generated by the application layer by exploiting their fundamental properties.
- Application to network
Application Network Interface (ANI) is defined by ITU-T Y.2012 as a channel for interactions and exchanges between applications and Next Generation Network (NGN) elements. Such an interface is similar to the cross-layer optimization interface that provides location mapping information for servers and clients from a network perspective, flowing from the application layer to the network layer using this specific interface. This interface enables the network to provide some performance estimates on the routes associated with the given location mapping information .
- Application to link/MAC
This direction is clarified in Reference  . Such an application allocates the resources on the link/MAC layer based on the data rate requirements of the applications, as well as the choice of the application data rates in accordance with the available resources on the link/MAC layer.
- Application to physical
If bandwidth is low, a lower quality video coding, which requires less bandwidth, may be used. Similarly, an email application could defer downloading file attachments in an email when the channel conditions are poor  . In  , a service-driven cross-layer routing protocol for mobile ad hoc network based on Ad hoc On Demand Distance Vector (AODV) and transmit power control was introduced. This model used different service types in the application layer to drive the transmit power to the physical layer as part of the solution. The PLR is decreased, the average delay time is reduced, and average transmit power is controlled reasonably in this model.
The parameters available in this layer are Round-Trip Time (RTT), Retransmission Time Out, Maximum Transmission Unit (MTU), Receiver Window, Congestion Window, Number of Packets Lost, Actual Throughput, Forward error correction (FEC), ARQ, UEP, and quantization parameter (QP). The approach is classified according to the target layers of parameters:
- Transport to network
The Explicit Congestion Notification (ECN) field in the IP header is used to carry ECN-capable information from the transport protocol to the routers .
- Transport to link/MAC
Link/MAC layer retransmission is controlled using the TCP RTT. A TCP-aware dynamic ARQ algorithm  has been introduced. This algorithm dynamically adapted the maximum allowed number and priority of transmission reattempts in reference to two parameters, namely, the TCP retransmission time-out and the number of retransmissions that have been sent. A packet priority weighting function helps in the retransmission process by receiving information via the cross-layer interactions between TCP and the link/MAC layer, extracting this information on TCP retransmission time-outs, and passing the data to the aforementioned algorithm that operates at the link/MAC layer. In  , a cross-layer framework for MPEG-4 video streaming over MANET networks is proposed. The MAC and transport layer are jointly optimized in this study to regulate data rate from transport layer based on congestion information of Access Category queue.
- Transport to physical
To date, no related research involving this type of interaction has been conducted. Existing work on cross-layer wireless network designs involving both the transport and physical layers can be classified either as an upward approach or joint optimization approaches involving multiple layers .
The parameters available in the network layer are Mobile-IP Hand-off Initiation/Completion Events, Network Interface, Routing Path, and Type of Service (ToS). The approach is classified according to the target layers of these parameters:
- Network to link/MAC
A new cross-layer QoS provisioning framework for Wireless Multimedia Sensor Networks is proposed. Although the objective of the network layer is to obtain QoS-routes with the application-specific QoS requirements, the link/MAC uses this routing information for actual packet classification and delivery  . Additionally, in  , a cross-layer framework in a WIMAX downlink is proposed. This framework requires a classifier and QoS mapping at the base station (BS) for resource management in link/MAC layer admission control, indicating that the BS must be capable of inspecting network layer headers, such as ToS field, to provide better support for QoS.
- Network to physical
To date, no study has been conducted on this subject.
The parameters available at the link/MAC layer are FEC, ARQ, Frame Length, Number of Frame Transmit, Hand-off Initiate/Completion Event, and Received signal strength indication (RSSI) values. These parameters are sent to the physical layer:
- Link/MAC to physical
The error control mechanism at the link/MAC layer may be adapted to reduce transmission errors based on channel conditions, as shown in  , wherein an MTU for a particular BER results in improved transmission range and goodput. Moreover, using channel condition information, frame length and link error recovery can be adapted to improve channel throughput and reduce waste of valuable radio resources, as shown in  . Directional and smart antennae can significantly reduce interference among closely spaced nodes to increase network capacity using upper layer algorithms, such as the straightforward algorithm in the MAC layer, to implement MAC to Physical layer controls .
The downward approach may include multiple layer interactions, such as iterative joint source-channel fountain code (ISCFC)  , which interact among application, transport, and physical layers. The transport layer can interleave the video symbol sequences (V) from the application layer into (`V) using interleave encoder FEC, subsequently passing such sequences to the physical layer for processing by the recursive systematic convolutional encoder. This process provides strong bit error correction capability, as well as overcomes packet losses. Furthermore, a presented multi-layer protection scheme based on an Application/Transport/PHY joint layer design is proposed in  . This scheme uses FGS/DA-UEP at the transport layer to achieve high-efficiency and forwarders (n) using the parayer to achieve high throughput.
5.1.2 Upward Approach
In this approach, the information from the lower layers is passed on to a higher layer to perform a condition-dependent adaptation at the higher layer , . For example, the obtained information on the channel conditions at the physical layer may be used to update and feedback to the link layer for adapting its error-control mechanisms or to the application layer for adapting its sending rate. This approach is not the optimal solution due to the incurred delays and throughput reductions when delays occur and during throughput reduction because the propagation of messages to the transport layer ends in a specific socket. Moreover, the propagation of notifications to the application layer ends in a particular application that introduces a small overhead problem, thereby resulting in delays. In this approach, the higher layers are notified of the underlying network conditions  . In the "rate and quality adaptation schemes"  , the rate adaptation module in the data link and physical layers regulates the data transmission rate and informs the quality adaptation module in the application layer, reporting the current rate limit [Figure 3]. In "intra-layer and inter-layer optimizations"  , a cost-efficient equivalence class of operating points is communicated to the higher layers, which aims at maximizing the QoS.
This approach is further categorized based on the initial source of the information sent to the upper layers.
The parameters available at the physical layer are Transmission Power, Coding, BER, SNR, and Modulation type. The approach is classified according to the target layers of these parameters:
- Physical to link/MAC
An optimal power transmit can be introduced in this interaction. In addition, varying the packet lengths according to the BER also helps reduce energy consumption, as shown in  . Moreover, modulation type can be adapted with coding FEC, called Adaptive Modulation and Coding (AMC), to reduce packet loss .
- Physical to network
BER and data rate in the physical layer may be used as a guide by network layers to change the interface. In "Cross-Layer Routing in Wireless Mesh Networks"  , three physical layer parameters, namely, interference, packet success rate, and data rate, are exploited to make the routing layer aware of issues arising from the underlying layers. This process aims to find paths with the lowest levels of generated interference, increased reliability in terms of packet transmission success rates, and the highest achievable transmission rates.
- Physical to transport
This direction uses physical layer information to differentiate the sources of packet loss, whether attributable to congestion or link quality-related issues in the given link. Furthermore, this direction can help in terms of end-to-end flow control and radio resource management. For example, in  , distributed end-to-end flow control and radio resource management problems in multi-hop wireless networks are solved using a cross-layer design involving the physical and transport layers. The design uses the Shannon capacity to optimize the physical constraints for improving the TCP Vegas/Reno performance in the Transport layer.
- Physical to application
Video application adapts the coding rate based on the channel condition information from the physical layer  . The physical layer parameters can be tuned and passed on to the application layer to improve throughput, such as the application-driven modular cross-layer optimization, for coded orthogonal frequency division multiple access (OFDMA) system in .
The parameters are identified in Section (22.214.171.124). The approach is classified according to the target layers of these parameters:
- Link/MAC to network
Mobile IP is used for hand-off when mobile devices have to change their subnets. This information may be not available as quickly as the signal changes monitored at the link layer. Thus, the hand-off latency of mobile IP can be reduced using link layer hand-off information  . In "An L3-Driven Fast Handover Mechanism"  , an abstraction of the link layer information to achieve fast handover in IPv6 is presented.
- Link/MAC to transport
When channel conditions are poor, delays in retransmission at the link layer result in increased TCP retransmission, thus reducing throughput  . The exchange of retransmission information between transport and link/MAC layers can be avoided through the use of a "cross-layer error detection scheme"  , where FEC is adapted based on channel conditions and network traffic, resulting in increased video quality and decreased packet loss. Furthermore, in "TCP Venoplus"  , the RSSI in the link/MAC layer can compute the random frequency of losses to allow the TCP layer to learn the RSSI for every received packet from the Link/MAC layer.
- Link/MAC to application
At the link/MAC layer, frames from different applications are treated differently. For instance, frames with a low delay requirement may be transmitted on priority. Likewise, FEC/ARQ may be introduced for applications with a high reliability requirement, which is based on multi-service link layer, such as in  . However, this process may amplify processing overhead, delay, and power consumption. Another feature of this linkage is that it can improve the user-perceived quality of multimedia applications via good utilization of the wireless resources, such as "Analysis of Cross-Layer Optimization between Application and Link Layer"  , which allocates the resources on the Link/MAC layer based on the data rate requirements of the applications.
The parameters are identified in Section (126.96.36.199). The approach is classified according to the target layers of these parameters:
- Network to transport
Mobile IP delay may result in reduced throughput attributable to the TCP retransmission timer. The use of mobile IP hand-off information conditions with a fast retransmission reduces retransmission timeout  , resulting in an improved throughput, as shown in  . In addition, based on the routing path packet recovery protection scheme that adaptively controls parity packets with I and P frames in the transport layer, reliability is improved, overhead is reduced, and the quality of received video is increased .
- Network to application
The sending rate of the application can be controlled based on mobile IP hand-off. The sending rate is also dependent on the application needs of the network layer choosing the appropriate interface. This condition is demonstrated through a devise with multiple wireless network interfaces providing different services, such as a WLAN interface that offers less delays and higher throughput compared with a GPRS interface on the same device. Moreover, in "A Cross-Layering and Autonomic Approach"  , a design approach based on autonomic components and cross-layer monitoring and control is presented. This approach optimizes the performance of the WiOptiMo system, which supplies internet work roaming by handling mobility at the application layer.
The parameters are identified in Section (188.8.131.52). The approach is classified according to the target layers of these parameters:
- Transport to application
TCP provides packet loss and throughput information to application layer. The application layer uses this input to adapt its sending rate. In  , the results show that UDP Lite fails to improve the media quality substantially because of the nature of errors induced by the 802.11b channel. Therefore, FEC protection is required at the application layer, regardless of the underlying transport layer protocol, for the delivery of high bit rate multimedia.
The upward cross-layer approach can also include multiple-layer interactions, such as that presented in "An analysis of Wireless TCP Network"  , which integrated information among the physical, link/MAC, and transport layers. To maximize the TCP throughput, the frame size needs to be adjusted as a function of the BER and the link/MAC layer retransmission protocol. Thus, to achieve the best throughput, an upward cross-layer approach is used to feed the TCP layer with the BER measured at the physical layer and the retransmission protocol used in link/MAC layer, thus incorporating "physical to transport" and "link/MAC to transport" interactions in the solution.
5.1.3 Hybrid Approach
This approach combines both downward and upward approaches, resulting in a multi-layer hybrid solution. For example, "Tramcar"  proposed a cross-layer solution for managing vertical handovers based on user-defined preferences from the application layer. These preferences concern cost of service, power consumption, security, network conditions, and network performance. This solution uses Stream Control Transmission Protocol at transport layer with mobility extensions to accomplish handovers through IP multi-homing. The network layer is adapted based on some of the parameters preferred by the user.
In "A new QoS parameter adaptation scheme"  , a cross-layer design is proposed to reduce losses in video transmission over 1xEV DO Revision 0 reverse link when mobile terminals travel at vehicular speed. The variation pair values of RSSI in the link/MAC layer are passed to the network layer affecting PLR. In addition, video resolution and frame rates in the application layer are adjusted based on a decision-distribution process. This process results in abrupt changes in the PLR in the network layer. In "End-to-End Reference QoS Architecture"  , an end-to-end reference QoS architecture for WLAN is presented. The architecture contains QoS components and mechanisms from different layers. The proposed solution involves the link/MAC, network, transport, middleware, and application layers [Figure 4]. This solution is divided into control plane and data plane modules. The control plane comprises QoS components and mechanisms that are responsible for QoS control and management. The architecture comprises QoS components from the link/MAC layer to the middleware layer. For QoS mechanisms, at the link/MAC layer, a network monitor keeps track of wireless channel conditions, such as the perceived bandwidth. The network scheduler at the network layer schedules packets from applications according to their QoS requirements. At the middleware layer, a packet classifier marks packet importance for registered QoS applications, enabling the network scheduler to perform appropriate scheduling. This architecture can help QoS designers in understanding the importance and the complexity of various QoS components and in selecting such QoS components appropriately.
In  , the Dynamic Multi-Attribute Cross-Layer Design (DMA-CLD) framework is proposed. This framework allows interactions between the network layer and upper, as well as lower layers of the OSI model. The network layer ranks routes based on inter-layer feedback, as well as on information gathered from intermediate nodes. The Interface MAC passes MAC information, including the number of one-hop neighbors and the contention index, to the network layer. Moreover, the Interface Physical (IP) passes Information from the physical to the network layer to calculate route outages. The application layer constructs the attributes of the comparison matrix based on the application requirements and network conditions. This matrix is passed on to the network layer via the Interface Application, allowing for the integration of the proper inter-layer feedback into DMA-CLD.
5.1.4 MAC-centric Approach
Initially, the mechanisms of the cross-layer were limited to the interactions between the physical and data link layers , . Subsequently, several studies emerged, which suggested that interactions with the upper layers, such as the application layer  , result in a MAC-centric approach. The MAC-centric approach obtains information from the application layer, such as traffic load and QoS requirements, and determines the current channel conditions from the physical layer to adapt to MAC parameters. This approach solves a number of problems, such as on-demand dynamic channel selection in Code Division Multiple Access (CDMA)-based ad hoc networks, by jointly considering the MAC layer scheduling and the physical layer request detector  .
Another example is a "MAC level content-aware ARQ scheme"  , which is a MAC-centric cross-layer approach for H.264 video streaming over HSDPA wireless links. This approach is based on the minimal interactions between the radio link control layer and the application layer (i.e., H.264 encoder), giving packets different importance values according to their content semantics [Figure 5]. In  , a TS-DCF (Distributed Coordination Function) channel access protocol has been devised for real-time video applications. This protocol is a MAC-centric approach that passes timestamp information from the video application layer to the link/ MAC layer. The proposed simulation limited the physical rate in the physical layer to 6 Mbps. This model achieves lower cumulative distributions of video frame delay compared with the legacy of DCF protocol.
In  , a time-based adaptive retry (TAR) mechanism for video streaming over WLANs is proposed. TAR improves the retransmission mechanism for video packets in DCF and adaptively decides whether to discard or resend a packet based on a retransmission deadline (D) attached to the packet. The retransmission deadline is assigned by the application layer according to the applications' specific requirements for the transmitted media data. This deadline is also used by the MAC layer to decide when to stop the retransmission process for each packet. An optimal packet size is defined by physical layer parameters, such as Signal to Interference and Noise Ratio, BER, symbol rate, and constellation size of maximal throughput.
5.1.5 Joint Adaptation Approach
Because of its complexity, only a few studies adopted this approach. In this approach, the adaptation strategies or schemes at different layers are jointly designed to optimize the overall performance, including the determination of the best procedure to obtain optimal strategy for various modulation and channel coding schemes existing in the physical layer, different packetization, ARQ, scheduling, admission control, FEC mechanisms at the link/MAC layer, and adaptation of video compression parameters at the application layer , . The authors in "Integrative cross-layer optimization"  formulated the cross-layer design problem as an optimization with the objective of selecting a joint strategy across multiple OSI layers that minimizes the video distortion at the application layer. They assume a number of adaptation and protection strategies available at the physical, link/MAC, and application layers. For instance, the strategies at physical layer may represent various modulation and channel-coding schemes for a particular WLAN standard. The strategies at link/MAC layer correspond to different packetization, ARQ, scheduling, admission control, and FEC mechanisms. The strategies at application layer may include adaptation of video compression parameters and packetization. Furthermore, each strategy can take a variety of possible values. The joint cross-layer strategy is assumed to complement all these joint design strategies at different layers (Sn). The cross-layer optimization problem seeks to determine the optimal solution from the array of possible solutions. Hence, (i) definition of one or more objective functions and (ii) an algorithm to find the desired optimal solution are needed. Examples of possible objective functions include distortion D (Sn) and delay Delay (Sn) of the received video, rate R(Sn), and complexity/power C (Sn) as a function of the joint strategy Sn, n = 1,.., N solution. To simplify this cross-layer optimization problem, joint optimized Min D (Sn), Min Delay (Sn), Min R (Sn), and Min C (Sn) are considered [Figure 6]. "Cross-layer wireless multimedia transmission: challenges, principle, and new paradigms"  introduced this approach to reduce the computational complexity and the joint Physical-MAC-Application optimization technique to find the best solutions for video-streaming applications.
|Figure 6: Joint design strategy cross‑layer optimization based on real‑time classification and off‑line MOO .|
Click here to view
5.1.6 Summary of Direction Flow Information-based Approaches
From these five approaches of cross-layer design, determining the advantages and disadvantages helps in the selection of the most appropriate and best design. For example, most existing systems use the downward approach. Previously, the upward approach was rarely used, especially for video transmission, because this approach is not optimal for multimedia transmission. The upward approach yields a large delay in information transfer, thereby resulting in throughput reduction. However, a large number of studies are currently attempting to mitigate the issues on this approach , .
In the hybrid approach, a number of studies attempted to validate the benefits of both downward and upward approaches ,,, . The MAC-centric approach faces a problem because of the inability of the link/MAC layer to perform adaptive source channel coding trade-offs, given the time-varying channel conditions and the multimedia requirements  . In the joint adaptation approach, complexity remains a major problem because this approach strives to use all possible strategies and parameters to choose the best strategy for optimizing QoS. To solve this problem, several learning and classification techniques were introduced  . The advantages and disadvantages of these approaches can help determine the appropriate approach. The best solution also depends on the application requirements, protocols, and algorithms used at the various layers, as well as their levels of complexity.
Initial research on cross-layer design started by mixing link/MAC layer interaction with the physical layer , and then extending such a design to the application layer  . Hence, research literature referred to such cross-layer schemes as the MAC-centric approach. Nonetheless, this approach can be classified as a hybrid approach using the proposed taxonomy. The joint adaptation approach could also be considered as hybrid approach for involving multi-layer downward or multi-layer upward information flows. However, the joint adaptation approach is considered as distinct approach because it focuses on finding the best solution by jointly optimizing the overall performance, determining the optimal strategy for selecting parameters from different layers such as the physical, link/MAC, and application layers, to obtain the best solution.
5.2 Cross-layer Design based on Information
This cross-layer design is based on the information exchanged among layers. This information depends on the availability of a function for cross-layer interaction and system architecture. The classification is divided into four categories, namely, channel state information (CSI), QoS-related parameters, resource information, and application-based data  .
5.2.1 Channel State Information
This category includes channel response, mobile speed, location information, condition numbers, signal strength, and interference levels, among others. An "interference-aware cross-layer design"  is presented to increase the throughput of the wireless mesh network. Multiple-access interference is a major limiting factor in wireless communication. Hence, enhancing multi-hop routing and scheduling with minimal interference to existing nodes is achieved through the present work. In  , a cross-layer scheduler design for the uplink of multi-user SIMO systems is introduced. The system performance rapidly improves quickly because of spatial multiplexing with perfect CSI. In "T-AS"  , a cross-layer transmit antenna selection (T-AS) approach for multiple-input, multiple-output, spatial multiplexing (MIMO-SM) systems employing a decision feedback detector (DFD) over Ricean flat-fading channels is investigated. DFD cancels interference and improves the detection of the transmitted packets. The result shows that the cross-layer T-AS delivers higher throughput gains compared with the capacity-based T-AS. Also, in  , a novel cross-layer scheduling scheme for a single-cell OFDMA wireless system with partial CSI at transmitter (CSIT) is given. This design determines optimal subcarrier and power allocation policies based on partial CSIT and individual user's QoS requirements to attain a more realistic resolution that maximizes the system's throughput and satisfies heterogeneous delay requirements with significantly low-power consumption.
5.2.2 Quality of Service-related Parameters
QoS parameters, such as throughput, delay, BER, and Packet Error Rate (PER) measurements, are used in "Unified cross-layer error-control"  , where a design of error-control with cross-layer interaction over a wireless LAN and an efficient low overhead ARQ are proposed, as illustrated in [Figure 7]. This framework combines the cross-layer error protection technique and the error correction code in the link/MAC layer, the erasure code in the application layer, and ARQ in the link/MAC and application layers. This combination reduces the overhead ARQ adaptation. "LLE-TCP" also shows performance enhancement of TCP over wireless links  . This scheme is conducted through the cross-layer optimization of the ARQ at different layers to avoid transmission acknowledgement (ACKs) over wireless links by performing local generation of ACKs at the sender node or at the BS. In Reference  , a study on PLR, an analytical expression was derived, and a cross-layer solution using error-locating code and FEC in the header/control packet was introduced to withdraw the effect of BER on network performance. In  , a cross-layer enhancement packet scheduling algorithm is presented to schedule uplink multimedia traffic in time-division multi-code CDMA networks, optimizing the cross-layer QoS parameter and the PLR and accounting for both the physical layer PER and the link-layer Packet Drop Rate (PDR). In  , a new adaptive distributive video-coding control algorithm is introduced. This algorithm was adapted to available bit rate and PLR using the cross-layer approach. When packet loss occurs in case the encoder transmits data available to the decoder, the decoder receives the amount Ravail (1-PLR), and the encoder decides the QP according to Ravail (1-PLR). The amount of data to be sent is changed as the QP changes.
|Figure 7: Overall structure of the unified error‑control framework with cross‑layer interactions .|
Click here to view
5.2.3 Resources Information
Multi-user reception capabilities, number and type of antenna, battery depletion level, and so on, are used in this approach. In "Cross-layer design for CDMA resource allocation"  , a cross-layer design is presented to support real-time video over time-varying CDMA channels, where the link layer resource allocation benefits from information at the application and physical layers -[Figure 8]. In "Exploiting Macrodiversity in Dense Multihop Networks and Relay Channels"  , the battery-aware physical layer declares a battery status to adapt the operation mode. Optimal transmitting power is important for reducing power. This condition is achieved by properly exploiting the macro diversity inherent in dense ad hoc networks. The gain can be achieved by a decentralized and cross-layer design. In "cross-layer design framework for multi-user multi-antenna systems"  , an analytical cross-layer design framework is established to exploit the spatial multiplexing gain, as well as the multi-user selection diversity gain. Moreover, the delay constraints of the delay sensitive users are maintained. The results show delay performance and high-throughput gain.
|Figure 8: The cross‑layer information for IP‑based CDMA resource allocation .|
Click here to view
5.2.4 Application-based Data
Application-related data, such as data traffic information, knowledge of data rate (constant or variable), data fragmentation, packet sizes, information on queue sizes, and so on, are used in this approach. The framework in "A generic, cross-layer optimization"  considers medium access collisions, routing decisions, and wireless channel effects to determine the optimal packet size. Four objective functions are used to investigate various performance factors, resulting in the determination of the optimal packet size. These functions are throughput, energy consumption per bit, latency, and packet error rate. "A new QoS parameter adaptation scheme"  is also an example using format video resolution and frame rate to generate data rate, which reverts to decision distribution that affects the quality of video sequences. In  , a cross-layer approach to packet scheduling in Mobile WiMAX is presented. WiMAX uses AMC. Hence, the wireless link should be considered in the scheduling. The algorithm is suitable for implementation at the BS. This algorithm is based on a number of parameters, such as packet queue and packet size.
5.2.5 Summary of Information-based Approaches
This classification involves multiple layers. For example, "CSI," which includes channel response, mobile speed, location information, condition numbers, signal strength, and interference level, involve the physical and link/MAC layers. "QoS-related parameters" involves the physical, network, and transport layers. "Resources information" involves the physical, link/MAC, and application layers. Finally, "Application-based data" involves the transport and application layers. The information-based approaches can be considered as a type of hybrid direction approach, section (5.1.3), except for the fact that the focus is on the information itself, instead of the manner by which the parameters are transferred.
| 6. Cross-layer Usage|| |
6.1 Cross-layer Solution on Pre-stored Video Streaming
Pre-stored video streaming refers to on-demand requests for compressed audio/video files that are stored in servers. The communication is unicast, and on-demand. It is divided into four approaches according to the method of downloading the files from a server :
Generally, pre-stored video-streaming application faces the main issues of throughput and packet loss in terms of QoS. The cross-layer method was employed to solve the issues mentioned above. A cross-layer framework that supports the video-on-demand service in multi-hop WiMax mesh networks is introduced. First, a joint solution of admission control and channel scheduling for video streams is proposed. Then, an efficient and lightweight multicast routing technique is also proposed to minimize the bandwidth cost of joining a multicast tree. Furthermore, the patching technique is adopted in the application layer to improve the capacity of the video server. The efficient cooperation of these techniques that are proposed in different layers improves the overall quality of the video-on-demand service  . In  , an upward cross-layer design consisting of an intra-layer optimization of operating modes and an inter-layer optimization of operating points was presented, as mentioned in upward section. The client requests the video streaming of a file that is stored in the streaming server. The result shows that QoS is maximizing. In "a Cross-Layer QoS Framework, CLQM"  , downward cross-layer QoS mapping framework for supporting per-class service differentiation in MANETs is introduced. Flows are grouped into four classes based on throughput and delay sensitivity of the video-streaming applications. One of these classes is for video-on-demand. QoS classes are achieved by:
- Using the web server: the client sends a request message and the server responds to the request message by sending the compressed file to the browser, which uses an application, such as a media player, to play the file. The user needs to wait for few seconds before the file can be played using contemporary data rates.
- Using the web server with metafile: the media player is directly connected to the web server. The web server contains two files for each stored file, including the audio/video file and the metafile, which holds information about the audio/video file. The client sends a request message and the server responds with the metafile information, which is passed on to the media player. The media player uses the URL in the metafile to access the audio/video file and the server responds.
- Using the media server: this approach solves the problem of the stated second approach, which runs the http over TCP. The TCP retransmits a lost or corrupted segment that affects the concept of the streaming. In this approach, UDP is used instead of TCP for streaming by using another server (media server) in the previous approach.
- Using the media server and Real Time Streaming Protocol (RTSP): RTSP is a control protocol designed to add more functionalities to the streaming process. These functionalities include setup connection, pause message that temporarily stops streaming, play message that resumes streaming, and teardown message that breaks the connection.
6.2 Cross-layer Solution on Real-time Streams
- A class-based scheduler in the network layer
- Assigning non-overlapping contention window ranges a cross different QoS classes
- Dynamic adaptation contention window ranges based on changing network loads and achieved QoS in per-class basis
Real-time streaming refers to the broadcast of live streaming, such as live broadcast, online gaming, and video conferencing, through networks in real-time. The communication can be unicast or multicast and live. Real-time streaming has the following characteristics:
Delay, jitter, packet loss, and throughput are considered as the main issues in real-time streams application. In the "Reliable multimedia transmission over non-reliable networks: advanced source/channel coding techniques"  project, downward cross-layer optimization strategies based on the IEEE 802.11e standard is proposed, enabling robust video transmission using FEC channel codes at the transport layer, as well as multiple description coding architecture. The first strategy is for packet loss. The first strategy divides the characteristics of the sequence to be coded and selects the most appropriate coding mode according to the channel conditions obtained through a cross-layer signaling protocol. The second strategy depends on a parametric model of the estimated distortion during coding operations. In the proposed study, a video RTP server transmits the video sequence  . A delay in a multimedia wireless network is addressed using the downward cross-layer approach shown in  , where a cross-layer error control mechanism for low-delay and robust video transmission in wired/wireless networks is introduced. The wireless multimedia streaming system in this study consists of a media source that uses a real-time encoder, and the source packets are packetized using RTP. The proposed model is developed analytically and implemented through ARQ at both the application and link layers. The proposed model determines the optimal protection level, given the delay and channel rate constraints. Overall, PLR is effectively minimized, increasing the throughput, and the video quality is substantially improved. Reference  uses the same mechanism adopted in  with the addition of FEC in the application layer. A joint adaptation cross-layer optimization algorithm is presented in  to provide an efficient solution for real-time video transmission. This algorithm aims to maximize the decoded video quality of delay-constrained streaming in a multi-hop wireless mesh network that supports QoS and optimizes the following different control parameters at each node of the multi-hop network across the protocol layers: (1) optimal modulation at the physical layer; (2) optimal retry limit at the link/MAC layer; (3) optimal path to the receiver in the remaining part of the mesh network; and (4) the application layer optimized packet scheduling, given a predetermined topology and time reservation per link using the concepts of IEEE 802.11e HCCA.
- Time relationship:
Real-time data on packet-switched network requires the preservation of the time relationship among the packets of a session. Suppose real-time video data contain three packets, each given 10 seconds to transfer to the destination and has a 1 second delay for each packet. Then, the first, second, and third packets start at the 1 st , 11 th , and 21 st second, respectively. If the packets arrive at different delays, then a jitter issue would occur.
The timestamp is a solution for jitter. Each packet contains a timestamp that shows the time the packet was produced relative to the previous packet. Then, the receiver adds this time to the time at which it starts the playback so that it knows when each packet is to be played.
- Playback buffer:
The packets should be timestamped and the arrival time should be separated from the playback time to prevent jitter and this can be done by using a playback buffer to store data until they are played back.
The timestamp alone cannot inform the receiver of the lost packet, and thus, the sequence number for each packet is needed to solve this case.
6.3 Cross-layer Solution for Congestion Loss on High-definition Three-dimensional Television
3DTV signals consist in standard video streams, captured simultaneously by several cameras placed around the recorded scene, and geometry signals that associate to each video frame some information about the volumes of the captured elements. Congestion collapse is the main issue that occurs in 3DTV, which leads to packet loss, because of large volumes of multi-view video streaming. 3DTV is presented as independent subsection from video-streaming applications in this paper because of distinguishes characteristics of 3DTV such as depth-controlled object insertion and scene geometry information (e.g., per-pixel depth).
3D video can bring 3D and realistic perceptual experiences to viewers by simultaneously projecting different views to the left and right eyes of the users  .
The internet protocol has proven to be very flexible in streaming video over a wide range of networks, including 2.5G and 3G mobile networks. Considering these advancements, the transport of 3DTV signals over 4G IP packet networks and IP packet networks appears to be a natural choice  . A 3DTV streaming architecture can be classified as (1) server unicasting to a single client; (2) server multicasting to several clients; (3) P2P unicasting, where each peer forward packets to another peer; and (4) P2P multicasting, where each peer forward packets to several other peers  .
Recently, HD videos have offered a wide field of view, and 3DTV has supported a natural viewing experience in true dimensions. One of the primary functionalities of 3D natural videos is the generation of multi-view images, enabling the consumers to experience more realistic visual scenes  .
Multi-view video-streaming protocols can be implemented on top of RTP/UDP/IP, which is the current state of the art, or RTP/DCCP/IP, which is the next-generation protocol  . At present, RTP/UDP is the most widely used transport protocol for multimedia. However, this protocol can result in congestion collapse when large volumes of multi-view video are delivered because it does not contain any congestion control mechanism  . The datagram congestion control protocol (DCCP)  is designed to replace UDP and TCP for media delivery, running directly over the IP to provide congestion control without reliability. Unlike UDP, DCCP provides reliable handshakes for connection setup/teardown, reliable negotiation of options, and accommodates a choice of modular congestion control mechanisms. Hence, future 3DTV is expected to employ the DCCP protocol over IP services because of its effective video rate adaptation, matching the TCP Friendly Rate Control rate.
The "extension of DCCP over a wireless network"  focuses on the problems of video streaming over DCCP in a wireless network. The study proposes an upward cross-layer solution, in which the wireless packet loss information at the MAC layer is utilized by DCCP to distinguish congestion losses from wireless losses. The tests performed with the modified DCCP confirm that using cross-layer loss information prevents unnecessary decreasing rates and results in better video-streaming experiences.
| 7. Interaction between Layers|| |
This section summarizes the interaction between layers based on the parameters that each layer can provide to the other layers to improve QoS [Table 1].
|Table 1: Interaction between layers by provided parameters of layers to the other layers to show the effects|
Click here to view
[Table 1] shows the effects of the parameters of the layers when they are provided to other layers. For example, when the BER passes from the physical layer to the network layer, the latter acquires the capability to shift to another interface.
| 8. Comparison and Evaluation|| |
[Table 2] illustrates the comparison between the cross-layer design approaches based on the direction of information. The cost factor explains that cost is minimized in the downward, hybrid, and MAC-centric approaches, whereas the cost increases to maximize the QoS parameters  such as Joint adaptation. In the downward cross-layer design, the QoS parameters (i.e., packet loss, delay, throughput) are proven efficient, although not optimally, in multimedia transmission. The applications in this design explain their expectations of the required network behavior. The lower layers can use this information to optimize lower layer parameters. The upward cross-layer design is not optimal for multimedia transmission because of delays, packet loss incurred, and throughput reduction. The QoS parameters are also efficient for the MAC-centric cross-layer design, where the MAC layer uses the parameters from the application and physical layers to provide an efficient solution. The hybrid approach is similar to the MAC-centric approach in terms of the efficiency of QoS parameters. The joint cross-layer design proves to be the best in this classification, achieving more efficiency in the QoS parameters because this design is determined jointly by all OSI layers. Moreover, the joint cross-layer design extracts all possible strategies and their parameters, generating best quality performance. However, the complexity of this design is high  . In contrast, the complexity of the downward and upward designs is less  and is intermediate in the hybrid and MAC-centric approaches.
|Table 2: Comparison between cross-layer design approaches based on the direction of information|
Click here to view
Based on the information in the cross-layer design, the best solution does not depend on the best approach, but on the functionality of the cross-layer design and the system concept.
The current paper shows that further research is needed to provide QoS with low cost and with less complexity in video transmission over wireless networks, highlighting new approaches in cross-layer designs.
| 9. Acknowledgment|| |
The authors are thankful to Institute of Postgraduate Studies and Universiti Sains Malaysia for the fellowship awarded to Ali Bin Salem to support his research.
| References|| |
|1.||Z. Azri, T. Suzuki, and S. Tasaka, "Application-Level QoS and QoE Assessment of a Cross-Layer Packet Scheduling Scheme for Audio-Video Transmission over Error-Prone IEEE 802.11e HCCA Wireless LANs", Journal of IEICE Transaction communication, vol.E93-B, no.6, pp.1384-94, June 2010. |
|2.||P. Goudarzi and M.S. Moin, "Multi-source Video Transmission with Optimum Perceptual Quality over Wireless Ad Hoc Networks", Springer Berlin / Heidelberg; vol.5174, Aug. 2008, pp.61-71. |
|3.||A. Toufik, and D. Ismail, "Delivering Audiovisual Content with MPEG-21-Enabled Cross-Layer QoS Adaptation", Journal of Zhejiang University Science A, vol. 7, no.5, pp.784-93, 2006. |
|4.||S. Dobson, "Putting Meaning into the Network: Some Semantic Issues for the Design of Autonomic Communications Systems", Proceedings of the 1st IFIP Workshop on Autonomic Communications, pp.207-16, 2005. |
|5.||Z. J. Haas, "Design Methodologies for Adaptive and Multimedia Networks", IEEE Communications Magazine, vol.39, no.11, pp. 106-7, Nov. 2001. |
|6.||M. Abdur Razzaque, S. Dobson and P. Nixon, "Cross-Layer Architectures for Autonomic Communications", Journal of Network and Systems Management, vol.15, no.1, pp.13-27, Mar. 2007. |
|7.||T. Vijay and S. Iyer, "Cross Layer Design Optimizations in Wireless Protocol Stacks", Computer Communications (Elsevier) journal, vol. 27, no.8, pp.720-4, 2004. |
|8.||G. Wu, Y. Bai, J. Lai and A. Ogielski, "Interactions between TCP and RLP in Wireless Internet", Global Telecommunications Conference 1999 (GLOBECOM '99), Rio de Janeireo, pp.661-6, 1999. |
|9.||B.A. Forouzan, and S.C. Fegan, "Data Communication and networking" 4th ed, New York, NY, USA: McGrawHill; 2007. |
|10.||C. Wang, T. Lin, and P. Lin, "Cross-layer Analysis and Design for Multimedia Services in IEEE 802.11 WLANs", In Proceedings of the international Conference on Mobile Technology, Applications, and Systems, Taiwan, pp.1-6, Sept 2008. |
|11.||M. van der Schaar, and S. Shankar, "Cross-layer Wireless Multimedia Transmission: Challenges, Principles and New Paradigms", IEEE Wireless Communications Magazine, vol.12 no.4, pp.50-8, Nov. 2004. |
|12.||V. Srivastava, "Cross-Layer Design: A Survey and the Road Ahead", IEEE Communications Magazine, vol. 43, no.12, pp.112-9, Dec 2005. |
|13.||M. Gineste, N. Van Wambeke, E. Exposito, C. Chassot and L. Dairaine, "A Cross-Layer Approach to Enhance QoS for Multimedia Applications Over Satellite", Wireless Pers Commun, vol.50, no.3, pp.305-28, Aug. 2009. |
|14.||D Pradas, A Bouabdallah, J Lacan, M.A. Vazquez Castro and M. Bousquet, "Cross-layer Optimization of Unequal Protected Layered Video over Hierarchical Modulation", Global Telecommunications Conference (GLOBECOM 2009. IEEE), pp.1-6, Nov. 2009. |
|15.||Y. Lee, F. Xia, and G. Bernstein, "Problem Statement for Cross-Layer Optimization" [online], Available from World Wide Web: http:// tools.ietf.org/html/draft-lee-cross-layer-optimization-problem-00, July 2010. [Las Accessed on 12th December 2010]. |
|16.||A. Saul and G. Auer, "Analysis of Cross-Layer Optimization between Application and Link Layer", 4th International Symposium on Wireless Communication Systems, (ISWCS 2007)., Trondheim, pp. 838-843, Oct. 2007. |
|17.||J. Ren, X. Zhang and Y. Liu, "Services Driven Cross-Layer Routing Protocol for Mobile Ad Hoc Networks", in Proceedings of the 21st International Conference on Advanced Information Networking and Applications Workshops, IEEE Computer Society, pp.770-5, 2007. |
|18.||P. Sarolahti, S. Floyd and M. Kojo, "Transport-layer Considerations for Explicit Cross-layer Indications" [online], Available from World Wide Web: http://tools.ietf.org/html/ draft-sarolahti-tsvwg-crosslayer-01, March 2007. [Last Accessed on 12th December 2010]. |
|19.||T. Mahmoodi, V. Friderikos, O. Holland and A. Hamid Aghvami, "Cross-Layer Design to Improve Wireless TCP Performance with Link-Layer Adaptation", IEEE VTC07-Fall, Baltimore, MD, USA, pp.1504-8, Oct. 2007. |
|20.||T. Sanguankotchakorn, and K. Pudashine, "A cross-layer framework for efficient MPEG-4 video streaming over IEEE 802.11e in MANETs", 2011 8th International Conference on Electrical Engineering/ Electronics, Computer, Telecommunications and Information Technology (ECTI-CON), Khon Kaen, pp. 369-72, May 2011. |
|21.||Y. Liu and Q. Zhou, "State of the Art in Cross-Layer Design for Cognitive Radio Wireless Networks", 2009 International Symposium on Intelligent Ubiquitous Computing and Education, IEEE Computer Society, Washington, DC, USA, pp.366-9, 2009. |
|22.||N. Saxena, A. Roy and J. Shin, "Cross-Layer Algorithms for QoS Enhancement in Wireless Multimedia Sensor Networks", IEICE Transactions on Communications, vol. E91.B, no.8, pp.2716-9, Aug. 2008. |
|23.||Y.T. Mai, C.C. Yang and Y.H. Lin, "Design of Cross-Layer QoS Framework for the IEEE 802.16 PMP Networks", IEICE Transactions on Communications, vol.E91.B, no.5, pp.1360-9, May. 2008. |
|24.||P. Lettieri, M.B. Srivastava, "Adaptive Frame Length Control for Improving Wireless Link Throughput, Range and Energy Effciency", in: INFOCOM, vol. 2, pp.564-71, 1998. |
|25.||R. Ludwig, A. Konrad and A.D. Joseph, "Optimizing the End-to-End Performance of Reliable Flows over Wireless Links", Proceedings of the 5th annual ACM/IEEE International Conference on Mobile Computing and Networking (MobiCom '99), pp.113-9, 1999. |
|26.||I.F. Akyildiz and X. Wang, "Cross-Layer Design", "WirelessMesh Networks", John Wiley & Sons, Ltd, pp.215-44, May 2009. |
|27.||Y. Yang, Y. Chen and W. Yi, "Cross-Layer Forward Error Control for Reliable Transfer in Wireless Multimedia Sensor Networks", Consumer Communications and Networking Conference (CCNC), 7th IEEE, Las Vegas, NV, pp.1-2, Jan. 2010. |
|28.||S. Lee, and K. Chung, "Combining the Rate Adaptation and Quality Adaptation Schemes for Wireless Video Streaming", Journal of Visual Communication and Image Representation, vol.19, no.8, pp.508-19, 2008. |
|29.||L.U. Choi, M.T. Ivrla?c, E. Steinbach and J.A. Nossek, "Bottom-Up Approach to Cross-layer Design for Video Transmission over Wireless Channels", in Proc. IEEE Vehicular Technology Conference 2005-Spring (VTCS05), pp.3019-25, 2005. |
|30.||J.P. Ebert and A. Wolisz, "Combined Tuning of RF Power and Medium Access Control for WLANs", in Proc. Of 1999 IEEE International Workshop on Mobile Multimedia Communications (MoMuC99), San Diego, USA, Novemberpp.74-82, 1999. |
|31.||S. Vassilaras, "A Cross-Layer Optimized Adaptive Modulation and Coding Scheme for Transmission of Streaming Media over Wireless Links", Journal of Wireless Network, vol.16, no.4, pp.903-14, 2010. |
|32.||L. Iannone, R. Khalili, K. Salamat and S. Fdida, "Cross-Layer Routing in Wireless Mesh Networks", 1st International Symposium on Wireless Communication Systems, IEEE explore, pp.319-23, Sept. 2004. |
|33.||M. Belleschi, L. Balucanti, P. Soldati, M. Johansson, and A. Abrardo, "Fast power control for cross-layer optimal resource allocation in DS-CDMA wireless networks". In Proc. of the IEEE ICC International Conference on Communications, Dresden, Germany, June 2009. |
|34.||S. Khan, J. Brehmer, W. Kellerer, W. Utschick and E. Steinbach, "Application-Driven Cross-Layer Optimization for Coded OFDMA Systems", in Proc of the 9th International Symposium on Wireless Personal Multimedia Communications (WPMC 2006), San Diego, USA, September 2006. |
|35.||A. Sanmateu, F. Paint, L. Morand, S. Tessier, P. Foquart, and A. Sollund, et al, "Seamless Mobility across IP Networks using Mobile IP", vol. 40, no.1, pp.181-90, 2002. |
|36.||K. Gogo, R. Shibui, and F. Teraoka, "An L3-Driven Fast Handover Mechanism in IPv6 Mobility", In Proceedings of International Symposium on Applications and the Internet Workshops (SAINT2006), pp.10-3, Jan. 2006. |
|37.||Y.S. Baguda, N. Fisal, S.H. Syed, L.A. Latiff, S.K. Yusof, and R. Rashid, et al "Adaptive FEC Error Control Scheme for Wireless Video Transmission", The 12th International Conference on Advanced Communication Technology (ICACT), Phoenix Park, pp.565-9, Feb. 2010. |
|38.||S. Shetty, Y.Tang and W.Collani, "A Cross-Layer Packet Loss Identification Scheme to Improve TCP Veno Performance", International Journal of Computer Networks (IJCN) vol.1, no.1, pp. 36-45, Nov. 2009. |
|39.||G. Xylomenos, and G.C. Polyzos, "Quality of Service Support over Multi-Service Wireless Internet Links", Computer Networks journal, vol. 37, no.5, pp.601-15, 2001. |
|40.||R. Caceres and L. Iftode, "Improving the Performance of Reliable Transport Protocols in Mobile Computing Environments", IEEE Journal on Selected Areas in Communications, vol.13, no.5, pp.850-7, June 1995. |
|41.||H. Gharavi and K. Ban, "Cross-Layer Feedback Control for Video Communications via Mobile Ad-hoc Networks", Proceedings IEEE 2003 Fall Vehicular Technology Conference (VTC), pp.2941-5, Oct. 2003. |
|42.||G.A. Di Caro, S. Giordano, M. Kulig, D. Lenzarini, A. Puiatti and F. Schwitter, "A Cross-Layering and Autonomic Approach to Optimized Seamless Handover", Proceedings of the 3rd Annual Conference on Wireless on demand Network Systems and Services (WONS), Jan. 2006. |
|43.||S.A. Khayam, S. Karande, M. Krappel, and H. Radha, "Cross-Layer Protocol Design for Real-Time Multimedia Application over 802.11b Networks", Proceedings of the 2003 International Conference on Multimedia and Expo (ICME '03), pp.425-8, 2003. |
|44.||L.D.P. Mendes, and J.M.C. Brito,, "Some Analysis of a Cross-Layer Design for a Wireless TCP Network", Fifth International Conference on Wireless and Mobile Communications, pp.64-9, 2009. |
|45.||A. Hasswa, N. Nasser, and H. Hassanein, "Tramcar: A Context-Aware Cross-Layer Architecture for Next Generation Heterogeneous Wireless Networks", IEEE International Conference on Communications, 2006. ICC 06, 2006. |
|46.||R. Rivera-Rodr?guez, O.E. Olivares-Dom?nguez and A. Serrano Santoyo, "Cross-Layer Optimization of the Packet Loss Rate in Mobile Videoconferencing Applications", Journal of Applied Research and Technology, vol.8, no.1, pp.130-43, Apr. 2010. |
|47.||H. Nguyen, R. Rivas and K. Nahrstedt, "End-to-End Reference QoS Architecture for 802.11 WLAN", IEICE Transactions on Communications, vol.E93.B, no.6, pp. 1350-8, June 2010. |
|48.||A.M. Safwat, "A Novel Framework for Cross-Layer Design in Wireless Ad hoc and Sensor Networks", Globecom 2004 Workshops, IEEE Communications Society, pp.130-5, 2004. |
|49.||S. Khan, Y. Peng, E. Steinbach, M. Sgroi and W. Kellerer, "Application-Driven Cross-Layer Optimization for Video Streaming over Wireless Networks", IEEE Communications Magazine, vol.44, no.1, pp.122-30, Jan. 2006. |
|50.||A. Ortega, L. Tong, H. Zheng, and M. Zorzi, "Special Issue on - Cross-layer design for communications and signal processing systems - Editorial" , EURASIP Journal on Applied Signal Processing, vol. 2005, no. 2, pp. 113-5, 2005. |
|51.||S. Benayoune, N. Achir, K. Boussetta and K. Chen, "A MAC Centric Cross Layer Approach for H.264 Video Streaming over HSDPA", Journal of Communication, vol.4, no.9, pp.691-9, Oct. 2009. |
|52.||J.O. Kim, H. Tode and K. Murakami, "Channel Access Protocol for Video Packets in IEEE 802.11 Networks", International Conference on Wireless Networks, Communications and Mobile Computing, pp.498-503, June 2005. |
|53.||M.H. Lu, P. Steenkiste and T. Chen, "A Time-Based Adaptive Retry Strategy for Video Streaming in 802.11 WLANs", Journal of Wireless Communications and Mobile Computing, vol.7, no.2, pp. 187-203, Feb. 2007. |
|54.||M. Van Der Schaar, and M. Tekalp, "Integrated Multi-objective Cross-layer Optimization for Wireless Multimedia Transmission", IEEE Xplore, vol.4, pp.3543-6, 2005. |
|55.||C. Verikoukis, L.G. Alonso Z?rate, and T. Giamalis, "Cross-layer optimization for wireless systems: an European research key challenge", IEEE Communications Magazine, vol.47, no.7, pp. 2-3, 2005. |
|56.||H.Y. Wei, S. Ganguly, R. Izmailov, and Z.J. Haas, "Interference-Aware IEEE 802.16 WiMax Mesh Networks", in Proceedings of 61st IEEE Vehicular Technology Conference (VTC 2005 Spring), pp.3102-6, 2005. |
|57.||V.K.N. Lau, M. Jiang and Y. Liu, "Cross Layer Design of Uplink Multi-Antenna Wireless Systems with Outdated CSI", IEEE Transactions on Wireless Communications, vol.5, no.6, pp. 1250-3, June 2006. |
|58.||H.A. Abou Saleh and W. Hamouda, "Cross-Layer Criterion for MIMO Spatial Multiplexing Systems with Imperfect CSI", 69th Vehicular Technology Conference, 2009 (VTC Spring 2009) IEEE, Barcelona, pp.1-5, Apr. 2009. |
|59.||C. Zarakovitis, NI Qiang and D. Skordoulis, "Cross-Layer Design for Single-Cell OFDMA Systems with Heterogeneous QoS and Partial CSIT", Wireless Communications and Networking Conference, 2009 (WCNC 2009) IEEE, Budapest, pp.1-6, Apr. 2009. |
|60.||J.Y. Choi, and J. Shin, "Cross-layer Error-control with Low-overhead ARQ for H.264 Video Transmission over Wireless LANs", Computer Communications (Elsevier) journal, vol.30, no.7, pp.1476-86, 2007. |
|61.||D. Kliazovich, F. Granelli, and M. Gerla, "Performance Improvement in Wireless Networks Using Cross-layer ARQ", Computer Networks (Elsevier) journal, vol.51, no.15, pp.4396-411, 2007 |
|62.||A. Kimsas, H. Overby, S. Bjornstad. and V.L. Tuft, "A Cross Layer Study of Packet Loss in All-Optical Networks", Advanced International Conference Telecommunications and International Conference on Internet and Web Applications and Services, 2006 (AICT-ICIW '06), pp.65-71, Feb. 2006. |
|63.||C. Hui, H.C. Chan, V.C. Leung, and Z. Jie, "Cross-Layer Enhanced Uplink Packet Scheduling for Multimedia Traffic Over MC-CDMA Networks", Vehicular Technology, IEEE Transactions, vol.59, no.2, pp. 986-92, Feb. 2010. |
|64.||C.K. Kim, D.Y. Suh, and G.H. Park, "Cross Layer QoS Control for Distributed Video Coding", IEICE Transactions on Communications, vol.E93.B, no.9, pp. 2485-8, 2010. |
|65.||H. Jiang, W. Zhuang, and X. Shen, "Cross-Layer Design for Resource Allocation in 3G Wireless Networks and Beyond", IEEE Communications Magazine, vol.43, no.12, pp.120-6, 2005. |
|66.||M.C. Valenti and N. Correal, "Exploiting Macrodiversity in Dense Multihop Networks and Relay Channels", Wireless Communications and Networking, 2003 (WCNC 2003), IEEE, New Orleans, LA, USA, pp. 1877-82, Mar. 2003. |
|67.||M. Jiang, D.S. Hui, V.K. Lau and W.H. Lam, "Cross-Layer Downlink Scheduling of Multi-user Multi-Antenna Systems with Heterogeneous Delay Constraints", IEEE International Symposium on Information Theory, Seattle, WA, pp.2894-8, July 2006. |
|68.||M.C. Vuran and I.F. Akyildiz, "Cross-layer Packet Size Optimization for Wireless Terrestrial, Underwater, and Underground Sensor Networks", publication in the IEEE INFOCOM 2008 proceedings, Phoenix-AZ, pp.226-30, 2008. |
|69.||D.S. Shuaibu, S.K. Syed-Yusof, and N. Fisal, "A Cross Layer Approach for Packet Scheduling at Downlink of WiMAX IEEE802.16e", European Journal of Scientific Research, vol.45, no.4, pp. 529-39, 2010. |
|70.||F. Xie, K.A. Hua , and N. Jiang, "A cross-layer framework for video-on-demand service in multi-hop WiMax mesh networks", Computer Communications, vol. 31, no. 8, pp. 1615-26, May. 2008. |
|71.||N. Sarma, and S. Nandi, "A Cross-layer QoS Mapping Framework for Mobile Ad Hoc Networks", IETE Technical Review, vol. 25, no. 6, pp. 346-58, 2008. |
|72.||S. Milani, and P. Zontone. "Cross-Layer Optimization Strategies for Video Transmission over IEEE 802.11E Networks", available from: http://www.gtti.it/GTTI08/papers/milani.pdf [online], [Last accessed on 2008]. |
|73.||A. Argyriou, "Cross-Layer Adaptive ARQ for Uplink Video Streaming in Tandem Wireless/Wireline Networks", IEEE 9th Workshop on Multimedia Signal Processing, 2007 (MMSP 2007), Crete, pp. 151-4, Oct. 2007. |
|74.||A. Argyriou, "Cross-Layer Error Control for Multimedia Streaming in Wireless/Wireline Packet Networks", IEEE Transactions on Multimedia, vol.10, no.6, pp. 1121-7, Oct. 2008. |
|75.||Y. Andreopoulos, N. Mastronarde and M. van der Schaar, "Cross-Layer Optimized Video Streaming over Wireless Multihop Mesh Networks", IEEE Journal on Selected Areas in Communications, vol.24, no.11, pp.2104-15, Nov. 2006. |
|76.||P.K. Tsung, L.F. Ding, W.Y. Chen, T.D. Chuang, Y.H. Chen, and P.H. Hisao, et al, "Video Encoder Design for High-Definition 3D Video Communication Systems", IEEE communication magazine, vol.48, no.4, pp.76-86, Apr. 2010. |
|77.||G.B. Akar, A.M. Tekalp, C. Fehn, and M. Reha Civanlar, "Transport Methods in 3DTVA Survey, IEEE Transactions on Circuits and Systems for Video Technology", vol.17, no.11, pp.1622-30, Nov. 2007. |
|78.||S.Y. Kim, S.B. Lee and Y.S. Ho, "Three-dimensional Natural Video System based on Layered Representation of Depth Maps", IEEE Trans. on Consumer Electronics, vol.52, no.3, pp.1035-42, June 2006. |
|79.||B. Gorkemli, M.O. Sunay, and A.M. Tekalp, "Video streaming over wireless DCCP", Image Processing, 2008. ICIP 2008. 15th IEEE International Conference, San Diego-CA, pp.2028-31, Oct. 2008. |
|80.||A.M. Baker, C.K. Ng, N.K. Noordin and S.Khatun, "PHY and MAC, Cross-Layer Optimization and Design", Proceedings of IEEE 2008 6th National Conference on Telecommunication Technologies and IEEE 2008 2nd Malaysia Conference on Photonics, pp.192-7, Aug. 2008. |
|81.||C.J. Hui, S. Khatun, and B.M. Ali, "Design of Cross Layer optimization of Space-Time Block Coded MIMO-OFDM System", 4th Student Conference on Research and Development (SCOReD 2006) on IEEE Xplore, pp.16-21, June 2006. |
|82.||W. Su and T.L. Lim, "Cross-Layer Design and Optimization for Wireless Sensor Networks", Proceedings of the Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD06), Las Vegas, pp.278-84, 2006. |
| Authors|| |
Ali Abdulqader Bin-Salem received B.S (Computer Science) degree from Al-Ahgaff University, Yemen, in 2006 and M.S (Computer Science) from Universiti Sains Malaysia (USM), Malaysia, in 2009. Currently, he is a PhD student at National Advanced IPv6 Center (NAv6), (USM). His current research interests include wireless LAN, multimedia QoS, video transmission over wireless, distributed system, P2P, and client-server architecture.
Tat-Chee Wan is an Associate Professor in the School of Computer Sciences, Universiti of Sains, Malaysia. He lectures in the area of Microprocessors, Computer Networks, and Operating Systems. He was formerly with Motorola Malaysia Sdn. Bhd. as a Senior R&D Engineer in Software Development for two-way radios. His current research interests include QoS mechanisms for Wireless networks, Satellite-based Internet, and Real-Time Systems. He is presently involved in the East Asia-wide AI3 [Ay-triple-Ei] (Asian Internet Interconnections Initiative) Project, to investigate the use of Unidirectional Links over Satellite for interactive multimedia communications.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2]