Scribd
Upload
67 views15 pages

Collaborative Resource Exchanges For Peer-to-Peer Video Streaming Over Wireless Mesh Networks

The document discusses collaborative resource exchanges for peer-to-peer video streaming over wireless mesh networks. It proposes partitioning scalable video streams into prioritized sub-flows that can be independently admitted to the network. Peers exchange information about wireless resources, channel conditions, and quality of service to determine which sub-flows to admit and through which paths in a way that maximizes the minimum quality experienced across all flows. Simulation results show the collaborative approach achieves near-optimal performance and outperforms a non-collaborative one.

Uploaded by

Koushik Kashyap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
67 views15 pages

Collaborative Resource Exchanges For Peer-to-Peer Video Streaming Over Wireless Mesh Networks

The document discusses collaborative resource exchanges for peer-to-peer video streaming over wireless mesh networks. It proposes partitioning scalable video streams into prioritized sub-flows that can be independently admitted to the network. Peers exchange information about wireless resources, channel conditions, and quality of service to determine which sub-flows to admit and through which paths in a way that maximizes the minimum quality experienced across all flows. Simulation results show the collaborative approach achieves near-optimal performance and outperforms a non-collaborative one.

Uploaded by

Koushik Kashyap
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 15

Collaborative Resource Exchanges for Peer-to-Peer

Video Streaming Over Wireless Mesh Networks


ECE-693
Koushik A M

Background:
Wireless multihop mesh networks has been applied in:
1. Video conferencing,
2. On-demand distributed services
3. Surveilance etc.

Many of the multimedia applications involve P2P transmission.


Link over P2P is dynamic so streaming should adapt to the variations.

Background:
Resource reservation for streaming.
Information exchange between peers.
Resource allocate on strategy impacts and impacted by other peers.
Information exchange:
Wireless resources (used, available)
Channel condition
Qos

VIDEO SUB-FLOWS AND QUALITY-RATE MODEL:


Partitioning Scalable Bitstream Into Sub-Flows:

Partition scalable video into several prioritized sub-flows.


Each subflow is viewed as a separate quality layer.
Each subflow is admitted to network independently.
Each aggregated flow, y is represented as set

Total number of subflows, N and each sub flow

and

with (1

VIDEO SUB-FLOWS AND QUALITY-RATE MODEL:


Quality-Rate Model : Utility for Collaborative Resource Exchange:
Incremental quality is provided by decoding of an individual sub-flow using:

Quality ,

received by aggregated flow

the sub flow.

is

depends on encoding parameters

VIDEO SUB-FLOWS AND QUALITY-RATE MODEL :

Quality-Rate Model : Utility for Collaborative Resource Exchange:


we have indicator function,

-1 : sub-flow admission decision hasnt made.


1 : sub-flow is admitted.
0 : sub-flow is rejected.

VIDEO SUB-FLOWS AND QUALITY-RATE MODEL :


:

Maximize the minimum quality experienced by any aggregated flow in the


system:

COLLABORATIVE RESOURCE EXCHANGES

Determince which sub-flow should be admitted and through which path


based on
Channel condition.
Bit-rate requirements of each subflow.

Goodput experienced by each sub-flow

is the maximum achievable rate and

is the packet error probability .

COLLABORATIVE RESOURCE EXCHANGES

Sub-flow admission control:


To maximize

Arrange subflows in the descending order, .

Admit subflows according to the order.

Admit subflows according to quality layer.


First admit quality-1 sub flows for all aggregated flows.

Within the same quality layer admit sublows in descending order according
to

COLLABORATIVE RESOURCE EXCHANGES

Collabarative Distributed Path Provisioning:


Source peers collaboratively determine the sorted list of subflows in decreasing order.
For each sub-flow, the corresponding source peer determines the available path.
If multiple path exists, the path with small congestion is selected.

Non-Collabarative Distributed Path Provisioning:


Source peer determine paths before other peers determine the path.

Cross layer design:

Application layer:
(Packet scheduling): Packets are ordered interms of increasing order of decoding
deadlines.

MAC layer:

Optimal Txn limit:

Physical layer:
Adaptive modulation according to variation in the channel conditions.

Observations from Simulation Results:

Monotonic relationship b/w


wheather we optimize

and quality layers, Results are much similar

The distributed collaborative alg have a performance similar to centralized


exhaustive search.
Collaborative approach outperforms the non-collaborative approach and have near
optimal performance.
With dynamic cross-layer optimization, achieved gain 2-5db over fixed retransmission
and modulation.
Addition of additional flows gracefully reduce the quality using collaborative
approach.

Possible improvements:

Consider channel SINR due to interference from neighbor peers.


Multicast routing.
Network coding

THANK
YOU

COLLABORATIVE RESOURCE EXCHANGES

Information exchange overheads:


o First message: sub-flow bit rates,

, priorities

o 2nd message: link and channel conditions (SINR).


o 3rd message: Path P_x selected for each subflow f_x.

You might also like