580 California St., Suite 400
San Francisco, CA, 94104
Key research themes
1. How can multipath video transmission optimize reliability and quality over wireless ad hoc and heterogeneous networks?
This theme investigates the use of multipath transmission strategies to enhance robustness, bandwidth utilization, and overall video quality in dynamic wireless environments such as ad hoc and heterogeneous networks. The research addresses challenges like frequent link failures, channel variability, and congestion by employing multipath routing, caching, and concurrent multipath transfer techniques to improve real-time video streaming QoS and QoE.
Key finding: Proposes a multipath transport architecture leveraging two disjoint paths and video proxy nodes for buffering and local error management, demonstrating enhanced video stream stability and error resilience over highly dynamic... Read more
Key finding: Develops an analytical framework for video streaming over wireless heterogeneous networks utilizing Concurrent Multipath Transfer (CMT) that increases throughput, reliability, and effective bandwidth aggregation. The... Read more
Key finding: Identifies error resilience and path diversity as key enablers for robust video streaming in Mobile Ad Hoc Networks (MANETs) and Vehicular Ad Hoc Networks (VANETs). Demonstrates that simplified multiple description coding... Read more
Key finding: Introduces a convex optimization framework for rate allocation in multipath wireless ad hoc networks to minimize the aggregate video distortion across multiple competing video sources. The framework models trade-offs between... Read more
Key finding: Presents a cross-layer design for Unequal Error Protection (UEP) of scalable video bitstreams over wireless channels leveraging multipath and multi-layer error control. The approach jointly optimizes Forward Error Correction... Read more
2. What video coding and error control mechanisms best support low-latency and energy-efficient video transmission over wireless sensor and embedded systems?
This theme covers how video compression schemes and error resilience techniques can be tailored for resource-constrained wireless multimedia sensor networks (WMSNs) and embedded devices, focusing on low energy consumption, low complexity encoding, and real-time constraints. Research examines distributed video coding, scalable codecs, and cross-layer designs aimed at balancing computational load, latency, and robustness to achieve efficient wireless video capture and streaming.
Key finding: Proposes an enhanced Pixel-Domain Wyner-Ziv (PDWZ) distributed video coding scheme with slice structure and adaptive rate selection to reduce system delay and computational complexity for wireless multimedia sensor networks.... Read more
Key finding: Develops a PDA-based wireless real-time video communication system using MPEG-4 compression optimized through multithreading, buffer design, and wireless transmission techniques. The system achieves real-time... Read more
Key finding: Introduces the SNP/VQR real-time video codec exploiting forward unilateral prediction and vector quantization to achieve spatial and temporal scalability. Incorporates bandwidth adaptation integrated with error concealment... Read more
3. How do video streaming parameters and network adaptations impact video quality and coverage in wireless mesh and sensor networks?
This theme studies the interaction between video compression parameters (e.g., GOP size), network performance metrics (bandwidth, delay, packet loss), and coverage in wireless mesh and video-based wireless sensor networks. Investigations assess how video encoding choices and network routing affect perceived video quality and sensor coverage, informing adaptive encoding and protocol design for efficient video broadcasting.
Key finding: Evaluates the impact of varying GOP sizes on high-resolution MPEG video quality transmitted over multihop wireless mesh networks. Experimental results demonstrate that optimal GOP size depends on network conditions, with... Read more
Key finding: Analyzes limitations of traditional coverage and routing algorithms when applied to video-based wireless sensor networks, highlighting that the unique field-of-view and data correlation characteristics of cameras require... Read more
Key finding: Designs and implements a client-server video broadcasting system with control over wireless network parameters such as packet loss rate, bandwidth, and delay. The system optimizes video transmission quality by dynamically... Read more