| dc.description.abstract |
With the rapid growth of multimedia applications and the Internet, streaming video over the Internet is becoming more and more attractive to users especially live video streaming. Many application-layer multicast pro-tocols have been proposed recently for this demand and each method has di erent advantages. However, these current methods do not solve e ec-tively peer to peer (P2P) live video streaming's problems. These problems are the randomly churn of nodes and long bu ering time since the variance between the arrival times of data of di erent sub-streams is large. They
also do not address the problem of free-riding nodes. This thesis presents our low-delay push-pull based application layer mul-ticast for live video streaming on P2P networks. The main goal of ourwork is to optimize content delivery on P2P networks to guarantee the time
constraints of live video streaming. We achieve this goal by constructing multiple balanced sub-trees for pushing data and optimizing pulling connec-
tions between nodes in di erent sub-trees to reduce the time gap between arrival times of pushing data and pulling data. As a result, our mechanism
can reduce bu ering time at each node. Our mechanism also includes atit-for-tat method to promote node contribution. In order to evaluate our
method's performance, we also build a simulation program in SMPL sched-uler. We also compare our method's performance with other methods. The simulation result demonstrated the e ectiveness of our method. |
vi |