A receiver-centric transport protocol for mobile hosts with heterogeneous wireless interfaces

Abstract

Numerous transport protocols have been proposed in related work for use by mobile hosts over wireless environments. A common theme among the design of such protocols is that they specifically address the distinct characteristics of the last-hop wireless link, such as random wireless errors, round-trip time variations, blackouts, handoffs, etc. In this paper, we argue that due to the defining role played by the wireless link on a connection's performance, locating the intelligence of a transport protocol at the mobile host that is adjacent to the wireless link can result in distinct performance advantages. To this end, we present a receiver-centric transport protocol called RCP (Reception Control Protocol) that is a TCP clone in its general behavior, but allows for better congestion control, loss recovery, and power management mechanisms compared to sender-centric approaches. More importantly, in the context of recent trends where mobile hosts are increasingly being equipped with multiple interfaces providing access to heterogeneous wireless networks, we show that a receiver-centric protocol such as RCP can enable a powerful and comprehensive transport layer solution for such multi-homed hosts. Specifically, we describe how RCP can be used to provide: (i) a scalable solution to support interface specific congestion control for a single active connection; (ii) seamless server migration capability during handoffs; and (iii) effective bandwidth aggregation when receiving data through multiple interfaces, either from one server, or from multiple replicated servers. We use both packet level simulations, and real Internet experiments to evaluate the proposed protocol.

On cloud-centric network architecture for multi-dimensional mobility

“…• Provisioning: While existing mobile-centric approach (e.g., [7]) or distributed approach (e.g., home agent of Mobile IP [8]) might have an advantage accessing mobile and network conditions, they are physically distant from the control points of services in the cloud. These control points are where quality provisioning takes place.…”

On Transport Layer Support for Peer-to-Peer Networks

“…R 2 CP was originally proposed in [16] for mobile hosts with heterogeneous (such as 3G and WiFi) wireless interfaces. In the following, we provide a synopsis of the R 2 CP protocol, and show that R 2 CP encompasses the desired transport protocol design, allowing it to be considered for peerto-peer networks.…”

“…The RCP receiver drives the progression of the connection, while the RCP sender merely responds to the instructions sent by the receiver. It is shown in [16] that RCP is indeed TCPfriendly.…”

“…Since R 2 CP provides in-sequence data delivery to the application, minimizing out-of-order arrivals can also minimize head-of-line blocking at the receive socket buffer (head-of-line blocking occurs when R 2 CP is unable to bind more data to any RCP pipe for requests due to the buffer being filled up). It has been shown in [16] that such a packet scheduling algorithm allows R 2 CP to effectively aggregate the bandwidths available along individual paths with bandwidth/latency mismatches and fluctuations.…”

“…While we refer interested readers to [16] for a more detailed presentation of the R 2 CP protocol and its performance, it is clear from the above discussion that R 2 CP follows the design elements outlined in Section 3, allowing it to address the unique characteristics of peer-to-peer networks such as peer heterogeneity and peer transience. Our ongoing work includes developing R 2 CP to support load balancing, peer selection, and reliable replication mechanisms for use with peer-topeer networks.…”