Ad-Hoc and Delay-Tolerant Networks (AHDTNs) can be very useful in environments where more traditional networking technologies fail. Determining the practical effectiveness of AHDTNs can however be challenging. We review design considerations and practical experience with a novel mechanism for monitoring and analyzing the performance of AHDTNs. This new mechanism, AllNet Trace, somewhat resembles a combination of ping and traceroute, and can be used for the same purposes as either ping, with or without IP record route, or traceroute. AllNet Trace can be used to obtain information about networks where packets may be forwarded to next-hop destinations more than once, or may be arbitrarily delayed. We describe the design, implementation, and performance of AllNet Trace within the context of the AllNet protocol, which is designed to securely deliver interpersonal data over both AHDTNs and the Internet.

In a distributed social network, no single system holds information about all the individuals in the network, and no single system is trusted by all the individuals in the network. It is nonetheless desirable to reliably compute the social distance among individuals. This must be done anonymously, without giving away any identifying information about individuals in the social network, and reliably, without allowing anyone to pretend to be socially closer to someone else than they actually are. The Social Network Connectivity Algorithm, or SoNCA, accomplishes these goals in a distributed manner. This paper describes both the high-level algorithm and a concrete design that is intended for future use with a network, AllNet, designed to provide secure interpersonal communication utilizing all available means, including Internet, cellular communications, ad-hoc networking and delay-tolerant networking.

Wireless sensor networks that are energy-constrained must transmit and receive data as efficiently as possible. If the transmission is delay tolerant, transferring blocks of accumulated data can be more efficient than transferring each sensed measurement as soon as it is available. This paper proposes a Block Transfer Protocol (BTP) designed for efficient and reliable transmission in wireless sensor networks. BTP reduces the time it takes to reliably transfer a block of packets compared to conventional link layer protocols, by piggybacking in data packets information about the transfer, minimizing the number of acknowledgements needed for reliable transmission, and reducing the need for timeouts, which can substantially slow down communication when transmission is unreliable. In addition, BTP improves reliability by handling false positive acknowledgements and by letting the receivers communicate to senders how many packets they are prepared to accept, providing a flow control mechanism to exert back-pressure on the senders. BTP has been evaluated on a real sensor node platform, as well as in simulation. BTP reduces the average time to transfer blocks of 60 packets, each 41 bytes long, by more than 20% over a perfect link with no packet loss, to 63% when 50% of the packets are lost. Furthermore, BTP ensures channel fairness and in our tests achieves a 100% delivery over multiple hops.

A new physical layer technique in Hong, Y-W abd Scaglione, A., (2003) for broadcasting in multi-hop wireless ad hoc networks makes broadcast a much more efficient building block than traditional flooding broadcast. This broadcasting technique requires that all nodes that receive a message retransmit it at the same time. Because the messages are identical, interference can be constructive rather than destructive, and the multiple transmitters act as an antenna array - an Opportunistic Large Array, or OLA -to transmit the message farther and ultimately with fewer hops than it could have been transmitted by conventional means. This paper presents several networking algorithms for utilizing such physical layer broadcast to solve problems that are common within distributed systems, including wireless ad hoc sensor networks, mesh networks, and mobile ad hoc networks. In such an environment, broadcasting may be more efficient than conventional algorithms and protocols that employ unicast or multicast communications, including routed communications. This paper uses broadcasting in wireless ad-hoc networks to addresses issues and requirements common to different applications of wireless ad hoc networks. Some of these requirements, such as broadcasting, multicasting, polling, geocasting, and synchronization, are straightforward given an efficient broadcasting mechanism. This paper considers some of the specific benefits available for these cases when employing broadcasting using OLAs. Other common requirements can also be satisfied by implementations using this building block. As well as considering the challenge of assigning unique identifiers, these include associative access to the data in the network as well as selecting a unique leader among all the nodes.

This paper describes the design of an implementation of the Transmission Control Protocol using an extension of the Standard ML (SML) language. SML supports higher-order functions, modularity, and type-safe module composition. We find that by using SML we can achieve good structure and good performance simultaneously. Good structure includes a modular decomposition of the protocol stack and of the TCP implementation, a control structure that imposes a total ordering on all events and processes them synchronously, and a test structure that allows component testing to catch problems before system integration. Strategies that help achieve good performance include using fast algorithms, using language constructs that make it easy to stage function evaluation, and language implementation features such as compacting garbage collection.