Wireless network compression: Code design and trade offs

Traces derived from the real-world mobile network traffic show significant inter-client redundancy among packets. This has inspired new solutions to reduce the amount of redundancy present in the data in order to manage the explosive traffic. In this paper, we propose a novel approach to leverage this redundancy by employing a novel network compression technique using wireless helper nodes. A Helper node provides the side information, that it obtained via overhearing, to mobile clients and assists the wireless gateway to more effectively compress each new packet when serving every new client. We investigate the potential benefits of wireless network compression from an information-theoretic point of view. We also describe a coding mechanism which adapts the two-part coding scheme to wireless network compression. We optimize this coding scheme to achieve minimum cost communication in the network. We also characterize the trade-off between the number of bits sent by the wireless gateway and the number of bits sent by the helper to a client.