Distortion-Aware Concurrent Multipath Transfer for Mobile Video Streaming in Heterogeneous Wireless Networks (1406.7054v1)

Abstract: The massive proliferation of wireless infrastructures with complementary characteristics prompts the bandwidth aggregation for Concurrent Multipath Transfer (CMT) over heterogeneous access networks. Stream Control Transmission Protocol (SCTP) is the standard transport-layer solution to enable CMT in multihomed communication environments. However, delivering high-quality streaming video with the existing CMT solutions still remains problematic due to the stringent QoS (Quality of Service) requirements and path asymmetry in heterogeneous wireless networks. In this paper, we advance the state of the art by introducing video distortion into the decision process of multipath data transfer. The proposed Distortion-Aware Concurrent Multipath Transfer (CMT-DA) solution includes three phases: 1) per-path status estimation and congestion control; 2) quality-optimal video flow rate allocation; 3) delay and loss controlled data retransmission. The term `flow rate allocation' indicates dynamically picking appropriate access networks and assigning the transmission rates. We analytically formulate the data distribution over multiple communication paths to minimize the end-to-end video distortion and derive the solution based on the utility maximization theory. The performance of the proposed CMT-DA is evaluated through extensive semi-physical emulations in Exata involving H.264 video streaming. Experimental results show that CMT-DA outperforms the reference schemes in terms of video PSNR (Peak Signal-to-Noise Ratio), goodput, and inter-packet delay.

• The paper proposes CMT-DA, a method that incorporates video distortion metrics into multipath data transfer to enhance streaming quality.
• It utilizes dynamic per-path congestion control and utility maximization theory to allocate flow rates effectively across heterogeneous networks.
• Empirical tests using Exata show CMT-DA achieves up to 11.3 dB improvement in PSNR and enhances goodput by 195 Kbps compared to existing schemes.

Distortion-Aware Concurrent Multipath Transfer for Mobile Video Streaming

The research presented in "Distortion-Aware Concurrent Multipath Transfer for Mobile Video Streaming in Heterogeneous Wireless Networks" addresses the ongoing challenge of high-quality video delivery over the myriad of wireless infrastructures with differing capabilities. The proliferation of mobile video services has exacerbated bandwidth demands, especially as video traffic increasingly dominates mobile data transmission.

This paper introduces an innovative approach to enhance mobile video streaming, termed Distortion-Aware Concurrent Multipath Transfer (CMT-DA). The research problem stems from inadequacies in existing CMT solutions, which are unable to meet stringent QoS prerequisites due to inherent asymmetry across heterogeneous networks. The proposed solution advances the field by incorporating video distortion metrics into the multipath data transfer process, seeking to minimize end-to-end video distortion through strategic data distribution.

Key Elements and Methodology

CMT-DA evaluates per-path status estimates and employs congestion control mechanisms tailored to each communication path. This strategy allows for dynamic flow rate allocation via utility maximization theory, a departure from typical throughput or delay-focused methods. The approach further incorporates controlled data retransmission, focusing on maintaining bandwidth efficiency while minimizing loss and delay.

Semiphysical emulations conducted within the Exata platform validate the effectiveness of CMT-DA, particularly in streaming H.264 video across networks with varied characteristics. The findings underscore CMT-DA’s superiority over reference schemes (such as CMT-QA, CMT-PF, and baseline CMT) regarding metrics like video PSNR, goodput, inter-packet delay, and effective loss rate. For example, CMT-DA achieved an average PSNR improvement of up to 11.3 dB and enhanced average goodput by as much as 195 Kbps compared to conventional methods.

Implications and Future Directions

The promising results reveal CMT-DA’s potential to fundamentally enhance mobile video streaming over heterogeneous wireless networks. The method stands to significantly impact both theoretical and practical domains of network optimization, suggesting a path towards more efficient and QoS-compliant streaming protocols. As mobile networks evolve, solutions based on distortion-awareness might support emerging services requiring higher video fidelity, such as augmented reality.

Looking forward, further research could explore the integration of CMT-DA with mobile cost minimization strategies, taking into account the economic constraints of data transmission in various network environments. Additionally, applying the concepts presented in this paper to environments of varying mobility, speed, and device constraints will likely elucidate additional benefits and scalability of the CMT-DA model.