Utilising multiple disjoint paths in multiple channels can improve network performance by enabling a node to reroute data along discovered paths seamlessly when link failure is detected. However, depending on a stale/invalid route to recover from a broken link could increase the delay to recover from the broken link and degrade the network performance. In this paper, we propose a new MAC protocol (RIVC-MB) to enhance communication reliability in the multipath multichannel routing protocol. The reliability of transmitting/re-routing the data packet in multipath multichannel routing protocol is improved by providing early route invalidity detection and early switchover. Waiting time to access the medium is also improved, when a node is attempting to access a busy medium, by re-routing the data packet via the alternative route. The RIVC-MB protocol uses the cross-layer interaction between MAC and routing protocols to achieve these goals. The proposed protocol is implemented and extensively evaluated using the NS-2 simulator. Simulation results show that the new proposed protocol reduces the end-to-end delay, and reduces both the number of route error control packets and the number of dropped data packets in the routing layer. It also reduces the collision rate in the MAC layer in a dense network.

The route recovery algorithm is a crucial part of an ad hoc routing protocol. Designing an efficient and fast route recovery mechanism scheme without incurring extra overheads or delays to repair the broken link is a desirable goal for any routing protocol. The route recovery algorithm in reactive routing protocols like AODV occurs only after a link failure is detected. However, we propose a new route discovery mechanism called Route Migration over Multiple link failure in Multi-Channel (RMMMC), which combines multiple paths and channels to provide a disjointed alternative route. When link failure is detected, the RMMMC reroutes the data packet through the alternative route in alternative channels that have been previously discovered. Furthermore, the RMMMC is resilient to multiple link failures. Unlike other route recovery approaches, nodes in RMMMC are considered to be equipped with a single transceiver. The RMMMC scheme is implemented and extensively evaluated by using an ns-2 simulator. Simulation results show that the proposed scheme achieves a better packet delivery ratio, throughput and reduces the routing overhead and end-to-end delay in a dense network.

MANET routing protocols are designed based on the assumption that all nodes cooperate without maliciously disrupting the operation of the routing protocol. AODV is a reactive MANET routing protocol that is vulnerable to a dramatic collapse of network performance in the presence of blackhole attack. The paper introduces a new concept of Self-Protocol Trustiness (SPT) in which detecting a malicious intruder is accomplished by complying with the normal protocol behavior and lures the malicious node to give an implicit avowal of its malicious behavior. We present a Blackhole Resisting Mechanism (BRM) to resist such attacks that can be incorporated into any reactive routing protocol. It does not require expensive cryptography or authentication mechanisms, but relies on locally applied timers and thresholds to classify nodes as malicious. No modifications to the packet formats are needed, so the overhead is a small amount of calculation at nodes, and no extra communication. Using NS2 simulation, we compare the performance of networks using AODV under blackhole attacks with and without our mechanism to SAODV, showing that it significantly reduces the effect of a blackhole attack.

This paper describes a study of different approximation techniques used to predict the response times of database transactions represented as patterns of resource consumption and modelled with non-product-form queueing networks. The techniques are applied to a range of examples. The experiments show that none of the approximation techniques has a consistent advantage over the others for all cases considered. On the other hand, a simple heuristic rule is formulated which provides an acceptable approximation to the average transaction response time for the entire range of examples. The rule specifies a procedure for labelling each queue in a queueing network as either an M/M/1 or an M/G/1 resource. The resulting network can then be solved to obtain the mean response time of individual transactions.

Packet radio networks that employ several parallel multiple-access channels are considered. An architecture for such a network dictates the selection of channels for packet transmission. We propose and analyze two multichannel architectures. In the first, each node employs a single radio and is assigned a channel on which it listens when it does not transmit. To transmit a packet, the node tunes its radio to the channel of the intended receiver for that transmission only. The second architecture requires each radio to use a single channel for both transmission and reception, but provides some of the nodes with more than one radio each, allowing them to serve as bridges between channels. Within these architectures, one can further select the amount of routing information held by each node and the channel-access protocol, both of which greatly affect the network performance. To ascertain the effects of the various parameters, we calculate the throughput in both architectures. The channel-access protocols we consider are slotted ALOHA and CSMA with and without capture. We also evaluate the effect of increasing the amount of routing information held by the nodes.

Models for local area networks of the slotted ring style of architecture are developed and evaluated. The hardware protocol is modeled using a BCMP network. The Basic Block protocol of the Cambridge ring is modeled using an approximate solution method of the fixed-point type. A limited comparison between the Cambridge Ring and another ring architecture—the token ring—is carried out.

The tradeoffs between efficiency and reliability in M / M / N queueing systems subject to breakdowns are studied numerically. The dependence of the optimal number of servers on the system parameters is investigated under two different sets of assumptions about the pattern of breakdowns and repairs.