TCP was designed to provide stream-oriented communication service for bulk data transfer applications (e.g., FTP and Email). With four-decade development, Internet applications have undergone significant changes, which now involve highly dynamic traffic pattern and message-oriented communication paradigm. However, the impact of this substantial evolution on congestion control (CC) has not been fully studied. Most of the network transports today still make the long-held assumption about application traffic, i.e., a byte stream with an unlimited data arrival rate.

Modern mobile wireless networks have been demonstrated to be high-speed but lossy, while mobile applications have more strict requirements including reliability, goodput guarantee, bandwidth efficiency, and computation efficiency. Such a complicated combination of requirements and conditions in networks pushes the pressure to transport layer protocol design. We analyze and argue that few of existing network transport layer solutions are able to handle all these requirements. We design and implement Sphinx to satisfy the four requirements in high-speed and lossy networks. Sphinx has (1) a proactive coding-based method named semi-random LT codes for loss recovery, which estimates packet loss rate and adjusts the redundancy level accordingly, (2) a reactive retransmission method named Instantaneous Compensation Mechanism (ICM) for loss retransmission, which compensates the lost packets once actual loss exceeds the estimation, and (3) a parallel coding architecture, which leverages multi-core, shared memory and kernel-bypass DPDK. Prototype and evaluation show that Sphinx outperforms TCP and other coding solutions significantly in microbenchmarks across all four requirements, and improves the performance of applications such as video streaming and block data transfer.

There is growing interest in new physical-layer transmission methods based on known-interference cancellation (KIC). These KIC-based methods share the common idea that the interference can be cancelled when the bit-sequence of it is known, which can improve the efficiency of wireless data communications. Existing work on KIC mainly focuses on single-hop or two-hop networks, with physical-layer network coding (PNC) and full-duplex (FD) communications as typical examples. This paper extends the idea of KIC to multi-hop networks, and proposes a bidirectional end-to-end KIC (BE2E-KIC) transmission method for the scenario where two nodes intend to exchange packets through multiple intermediate nodes. With BE2E-KIC, the involved nodes can simultaneously transmit and receive on the same channel. We first discuss the procedure of BE2E-KIC and provide a theoretical analysis on its feasibility and effectiveness. Then, we propose a medium access control (MAC) scheme that supports BE2E-KIC, which schedules packet transmissions in more realistic cases with the presence of packet-loss. Simulation results illustrate that BE2E-KIC can improve the network throughput and reduce the end-to-end delay compared with other existing transmission methods.

Physical-layer network coding (PNC) is an effective strategy for increasing the throughput of wireless networks. In the current literatures, PNC without rate and power adaptation is mainly focused. Realizing that the transmission efficiency can be improved through rate and power adaptation in wireless networks, this paper focuses on developing a rate and power adaptation scheme for PNC. Through formulating how the data rate and transmission power affect the bit error rate (BER) of involved links in PNC, we observe that with a given data rate, the transmission power has to satisfy some constraints. Using these power constraints, we obtain a candidate set of optimal transmission power. By traversing the candidate set and the data rates supported by nodes, a rate and power adaptation scheme is developed. To test its performance, we apply the proposed scheme into an existing PNC-supported MAC protocol. Simulation results demonstrate that the proposed scheme can improve the throughput and delay performance in various scenarios.