A quality-of-service (QoS) routing protocol is developed for mobile ad hoc networks. It can establish QoS routes with reserved bandwidth on a per flow basis in a network employing TDMA. An efficient algorithm for calculating the end-to-end bandwidth on a path is developed and used together with the route discovery mechanism of AODV to setup QoS routes. In our simulations the QoS routing protocol produces higher throughput and lower delay than its best-effort counterpart.

We propose a framework for single-source, satellite-based multicast dissemination of bulk files. The framework trades off between reception delay and bandwidth usage and coexists with terrestrial background network traffic; specifically TCP traffic utilizing a short term congestion control mechanism. The framework consists of two major components: (1) a multicast rate scheduling mechanism that uses long-term, end-to-end multicast packet survival statistics in order to deal with the bandwidth-delay trade-off issue, and (2) a fair queueing algorithm that regulates the points where multicast traffic from the satellite meets terrestrial background traffic. We show through simulation the performance of this framework under a number of scenarios.

In this paper we propose a multicast routing protocol for mobile ad hoc networks (MANETs). The protocol-termed differential destination mmulticast (DDM)-differs from common approaches proposed for MANET multicast routing in two ways. Firstly, instead of distributing membership control throughout the network, DDM concentrates this authority at the data sources (i.e. senders) thereby giving sources knowledge of group membership. Secondly, differentially-encoded, variable-length destination headers are inserted in data packets which are used in combination with unicast routing tables to forward multicast packets towards multicast receivers. Instead of requiring that multicast forwarding state to be stored in all participating nodes, this approach also provides the option of stateless multicasting. Each node independently has the choice of caching forwarding state or having its upstream neighbor to insert this state into self-routed data packets, or some combination thereof. The protocol is best suited for use with small multicast groups operating in dynamic networks of any size.

This paper presents a distributed channel probing scheme for wireless networks. By transmitting a probing signal in a channel and measuring the signal-to-interference ratio (SIR), a wireless node can estimate channel admissibility and predict its required transmission power without fully powering up. The channel probing scheme can be used as part of a distributed channel allocation algorithm, and simulations have shown that it outperforms other schemes.

We study a long-term, end-to-end rate scheduling problem of multicasting a large bulk file over a hybrid satellite-terrestrial network to heterogeneous receivers. With selected receivers periodically reporting their long-term packet survival probabilities at each of the sender's transmission rates, we create a system which is both capable of trading off bandwidth consumption and receiver latency through a control parameter and proactively maintaining a certain degree of reliable delivery. To increase bandwidth efficiency, we assume forward error correction (FEC) in the form of the Reed Solomon erasure (RSE) coding. As an exact solution is computationally costly, we develop a rate scheduling algorithm based on a Gaussian approximation and study its performance by simulations.

We present a reliable broadcast protocol designed for use in ad hoc networks. The protocol ensures reliable, in-order, duplicate-free delivery of objects to a node's one-hop neighbor set. The flexibility and efficiency of the protocol hinges on the fact that the responsibility for reliable delivery rests solely at the receiver. Furthermore, when used on top of an 802.11-like MAC layer, we show that characteristics of an RTS/CTS mechanism can be used to further improve protocol performance.

This paper summarizes work underway at the University of Maryland on the design of a protocol kernel for mobile ad hoc networks. The kernel's architecture is modular, and follows a composition-based approach to protocol kernel construction. In this approach, interacting policies are logically, vertically closely-coupled, but physically loosely-coupled. Thus, various components of the kernel may be changed more easily. The architecture supports the notion of routing through an IP-layer routing "fabric", a construction which is a logical union of multiple, multi-hop topologies-each network using a different link layer technology. The architecture explicitly separates the policy functions of data routing and forwarding-an approach that more clearly associates a policy with its function, requires independent specification of both policies and results in a more flexible and configurable kernel.

A new single channel, time division multiple access (TDMA) based broadcast, scheduling protocol, termed the five-phase reservation protocol (FPRP), is presented for mobile ad hoc networks. The protocol jointly and simultaneously performs the tasks of channel access and node broadcast scheduling. The protocol allows nodes to make reservations within TDMA broadcast schedules. It employs a contention-based mechanism with which nodes compete with each other to acquire TDMA slots. The FPRP is free of the "hidden terminal" problem, and is designed such that reservations can be made quickly and efficiently with negligible probability of conflict. It is fully-distributed and parallel (a reservation is made through a localized conversation between nodes in a 2-hop neighborhood), and is thus arbitrarily scalable. A "multihop ALOHA" policy is developed to support the FPRP. This policy uses a multihop, pseudo-Bayesian algorithm to calculate contention probabilities and enable faster convergence of the reservation procedure. The performance of the protocol is studied via simulation, and the node coloring process is seen to be as effective as an existing centralized approach. Some future work and applications are also discussed.

We present a relative performance comparison of the temporally-ordered routing algorithm (TORA) with an ideal link state (ILS) routing algorithm. The performance metrics evaluated include bandwidth efficiency for both control and data, as well as end-to-end message packet delay and throughput. The routing algorithms are compared in the context of a dynamic, multihop, wireless network employing broadcast transmissions. The network parameters varied include network size, average rate of topological changes and average network connectivity. While the average network connectivity was found not to be a significant factor, the relative performance of TORA and ILS was found to be critically dependent on the network size, and the average rate of topological changes. The results further indicate that for a given available bandwidth-as either the size of network increases or the rate of network topological change increases, the performance of TORA eventually exceeds that of ILS.

We present a new distributed routing protocol for mobile, multihop, wireless networks. The protocol is one of a family of protocols which we term "link reversal" algorithms. The protocol's reaction is structured as a temporally-ordered sequence of diffusing computations; each computation consisting of a sequence of directed link reversals. The protocol is highly adaptive, efficient and scalable; being best-suited for use in large, dense, mobile networks. In these networks, the protocol's reaction to link failures typically involves only a localized "single pass" of the distributed algorithm. This capability is unique among protocols which are stable in the face of network partitions, and results in the protocol's high degree of adaptivity. This desirable behavior is achieved through the novel use of a "physical or logical clock" to establish the "temporal order" of topological change events which is used to structure (or order) the algorithm's reaction to topological changes. We refer to the protocol as the temporally-ordered routing algorithm (TORA).

We propose a protocol "imbedding" technique whereby a secondary, slotted Aloha protocol is imbedded within the idle slots of an existing primary, slotted-time, multiple access scheme via "idle sensing", a form of carrier sense multiple access. We consider two applications: mixed-priority data transmission in multistation networks and integrated voice/data operation in cellular systems. For mixed-priority data transmission, we propose a two-level, imbedded, slotted Aloha protocol which consists of a primary and secondary level protocol. The secondary's operation is imbedded within the idle slots of the primary. Higher priority traffic is given to the primary protocol and lower priority traffic is given to the secondary. Since the performance of the primary slotted Aloha protocol (which operates independently of the secondary) is well known, we focus on the performance of the secondary protocol and analyze its throughput and delay. The protocol's operation is studied in both single cell and multicell settings where, in the latter, intercell interference is modeled as feedback errors. We present a novel approximation method for analyzing the secondary protocol's performance based on the notion that the non-empty slots of the primary protocol can also be viewed as the "feedback error interference" at the secondary protocol. This reduces the order of the underlying Markov chain from two dimensions to one with almost no loss of accuracy. Using these same methods, we also study the usage of this imbedding technique for data transmission in a voice-only, cellular TDMA system.

Considers a cellular system architecture based on Multi-Receiver Multiple Access (MRMA). The upper and lower bounds on the spectrum efficiency of this architecture are calculated and compared with the efficiency of existing cellular systems. In some cases, a MRMA-based system is shown to have a higher potential capacity, nearly double that of existing systems, but at the cost of higher complexity. Previous work has shown these systems to be well suited for data traffic and work is underway to determine the suitability of such systems for voice and integrated voice/data traffic.>

An architecture for incorporating mobile data users into an existing MAP/TOP (Manufacturing Automation Protocol/Technical and Office Protocol) system is presented. The scheme attaches a small number of radio base stations to the TOP network through which the mobiles may access the system by executing a capture-based, multi-receiver extension of the traditional nonadaptive slotted Aloha protocol. A two-component capture model motivated by the factory channel is developed, and the multireceiver system's throughput performance is analyzed under two transmission models: base location independent and spatially correlated. It is seen that there is significant throughput degradation when base coverage areas overlap, indicating that this overlap should be minimized during system design. >

The authors propose an architecture for incorporating mobile data users into an existing MAP/TOP system. The scheme involves attaching a small number of radio base stations to the TOP network through which the mobiles may access the system. The system's central feature is a new distributed, multi-receiver, multiple access protocol used by the mobiles to access the bases. The protocol permits users to gracefully switch between single and multi-base operation while maintaining satisfactory performance characteristics. They utilize a multi-receiver multiple access model derived from the traditional single-receiver model to measure the new protocol's performance via simulation.>