P2P Live Streaming Towards Best Video Quality
read more
Citations
Avoiding Quality Bottlenecks in P2P Adaptive Streaming
Adaptive topology formation for peer-to-peer video streaming
Efficient utilization of peer's upload capacity in P2P networks using SVC
On topology construction in layered P2P live streaming networks
Daedalus: A media agnostic peer-to-peer architecture for IPTV distribution
References
A Measurement Study of a Large-Scale P2P IPTV System
Can internet video-on-demand be profitable?
Stochastic Fluid Theory for P2P Streaming Systems
Random walk based node sampling in self-organizing networks
Building heterogeneous peer-to-peer networks: protocol and analysis
Related Papers (5)
Supporting low-cost video-on-demand in heterogeneous peer-to-peer networks
Frequently Asked Questions (15)
Q2. How long does the playback rate adaptation take?
In addition, since the composition of peers changes slowly and smoothly, the playback rate adaptation is carried out over a longer time scale.
Q3. What is the salient feature of the link-level homogenous overlay?
A salient feature of the link-level homogenous overlay is that video flows along the path of connected connections do not encounter any bottleneck links, and therefore, peers can achieve the guaranteed data rates.
Q4. How is the random walk algorithm constructed?
Since the length of the random walks is related to the cost and accuracy of peer selection, the Markovian plane should be constructed to minimize the cost of random walks with the guaranteed accuracy.
Q5. How does the algorithm repair the graph?
To repair the graph, if the nodes’ degree is less than nmin, they rebuild a new link by randomly selecting a peer in the Markovian plane.
Q6. How can the authors adapt the proposed algorithm to the dynamic changes of the composition of peers?
the authors conclude that their proposed algorithm can effectively adapt to the dynamic changes of the composition of peers so as to fully utilize the varying available bandwidth resource towards the best video quality.
Q7. What is the abstract of the overlay network?
The authors abstract the overlay network as a directed graph G = {V,E} where V denotes the peer set, including the source node, and E denotes the link set.
Q8. How do the authors make the Markovian plane a self-loop?
the authors construct the Markovian plane to be an undirected graph with a self-loop connection at each node so that peers are possible to keep walkers without forwarding to others.
Q9. What is the effect of the curve on the download rate of peers?
The authors can see that the curve presents an impulse with a heavy tail, which means that in the formed overlay topology most participating peers have roughly identical capacity per out-degree value.
Q10. What is the effect of the out-degree evolution of the investigated node?
the investigated node has a node index of 1,000 and performs the same behaviors as other peers with the exception that it has an infinitely large life time and never departs from the network.
Q11. How can the authors calculate the download rate of peer?
The downloading rate of each peer can becomputed by summing up all its parent nodes’ capacity per out-degree values due to the feature of link-level homogeneity.
Q12. How does the random walk algorithm work?
After the random walk process converges, the walker will stay at a node according to the steady-state probability of the Markov chain and therefore finishes the selection.
Q13. how many peer i converges to the same bandwidth?
In this state, the downloading rate of each peer i converges asdi = δ×m, ∀i ∈V (5)Using their link-level homogeneity overlay, the authors show that in the formed network, all the peers converge to the identicalbandwidth per out-degree.
Q14. What is the effect of the new peer download rate?
As the capacity per out-degree values of the newly joined peers converge, the downloading rate of these nodes will also converge.
Q15. What is the effect of increasing the out-degree of the investigated node?
As the authors can see that with λ increasing the out-degree of the investigated node can adapt to the change of capacity much faster, and the simulation results match their analysis in Eqn (4).