In mobile wireless networks, livestreaming in high user density areas presents two typical challenges: the wireless bandwidth is depleted and the number of users is limited. In this study, a media distribution model utilizing peer to peer communications, Active Control in an Intelligent and Distributed Environment, is proposed for bandwidth efficient livestreaming. The basic idea is to group users with identical livestream interest in a cluster of n peers. Instead of sending n copies of a livestream package, only one copy is sent to the cluster. A package is divided into n blocks. Each user receives one block from the base station and the remaining n-1 blocks from the other peers. Two optimization problems are addressed. The first problem is minimizing the bandwidth needed to guarantee a continuous live media play on all peers. A solution is proposed to find the optimal block sizes such that the wireless bandwidth is minimized. The second problem is maximizing the number of peers admitted to a cluster, given a fixed wireless bandwidth. This problem is NP-complete and a greedy strategy is proposed to calculate a feasible solution for peer selection. The proposed model improves the bandwidth efficiency and allows more users to be served.

翻译：在移动无线网络中，高用户密度区域的直播流面临两大典型挑战：无线带宽耗尽与用户数量受限。本研究提出一种利用对等通信的媒体分发模型——智能分布式环境中的主动控制（Active Control in an Intelligent and Distributed Environment, ACIDE），用于实现带宽高效的直播流。其基本思想是将具有相同直播兴趣的用户分组至包含n个对等节点的集群中。不同于向集群发送n份直播流数据包，仅发送一份副本至该集群。数据包被划分为n个区块，每个用户从基站接收一个区块，其余n-1个区块从其他对等节点获取。本文解决两个优化问题：第一，在保证所有对等节点连续播放直播媒体的前提下最小化所需带宽，提出求解最优区块大小以最小化无线带宽的方案；第二，在固定无线带宽条件下最大化集群可接纳的对等节点数量——该问题属于NP完全问题，提出一种贪心策略以计算可行的对等节点选择方案。所提模型提高了带宽效率，并能够服务更多用户。