Øivind Kure

A survey of key management in ad hoc networks

The wireless and dynamic nature of mobile ad hoc networks (MANETs) leaves them more vulnerable to security attacks than their wired counterparts. The nodes act both as routers and as communication end points. This makes the network layer more prone to security attacks. A main challenge is to judge whether or not a routing message originates from a trustworthy node. The solution thus far is cryptographically signed messages. The general assumption is that nodes in possession of a valid secret key can be trusted. Consequently, a secure and efficient key-management scheme is crucial. Keys are also required for protection of application data. However, the focus here is on network-layer management information. Whereas keymanagement schemes for the upper layers can assume an already running network service, schemes for the protection of the network layer cannot. Keys are a prerequisite to bootstrap a protected network service. This article surveys the state of the art within key management for ad hoc networks, and analyzes their applicability for network-layer security. The analysis puts some emphasis on their applicability in scenarios such as emergency and rescue operations, as this work was initiated by a study of security in MANETs for emergency and rescue operations.

Quality of Service in mobile ad hoc networks: a survey

To support multimedia applications, it is desirable that an ad hoc network has a provision of Quality of Service (QoS). However, the provision of QoS in a mobile ad hoc network is a challenging task. In this paper, we present a review of the current research related to the provision of QoS in an ad hoc environment. We examine issues and challenges involved in providing QoS in an ad hoc network. We discuss methods of QoS provisioning at different levels including those at the levels of routing, Medium Access Control (MAC), and cross layer. Also, we discuss schemes for admission control and scheduling that are proposed in the literature for the provision of QoS. We compare salient features of various solutions and approaches and point out directions for future work.

Constrained-based Multiple Sink Placement for Wireless Sensor Networks

A Wireless Sensor Network (WSN) consists of a large number of small, low-cost and low-power wireless sensing nodes. WSNs can gather information about the environment automatically and unattended and are suitable for many applications.The typical characteristic of WSNs is that they are energy and bandwidth constrained. Hence, routing protocols and algorithms for WSN must aim to conserve these two scarce resources. WSNs are also highly application-specific. This mean, firstly, that there is a tight bound between the application layer and the different protocol layers. Secondly, there are some WSN target applications that require certain protocol functionality that is not mandatory for other WSNs. In other words, both the general challenges and the specific application challenges must be addressed.This thesis aims to address routing in WSNs both from a general and an application specific perspective. Among the general energy-and bandwidth related topics the work in this thesis focuses on aggregation and routing-efficiency. Among the application-related topics the work focuses on localization and interoperability.The main contributions are:• A method for letting the routing protocol contribute in node localization.• A method for increasing the energy and bandwidth utilization with passive clustering.• A method for increasing the energy and bandwidth utilization using multiple sinks.• A data-aggregation scheme for WSNs that interoperates with external networks via a standardized interface.• A hybrid routing mechanism that are able to operate in high-interference scenarios.• Lessons learned from a real-world test campaign of a surveillance WSN.

The use of service level agreements in tactical military coalition force networks

Tactical military coalition force IP networks will have to offer quality of service. Service level agreements and service level specifications are important elements of the QoS architecture in civilian networks. However, SLAs/SLSs in military coalition networks should not be applied in the same fashion as in a commercial network. Our contribution is to identify the useful role SLAs/SLSs can play in network engineering and QoS management of tactical coalition force networks. The SLS definition needs to be more detailed than for a commercial SLS, and its monitoring must be performed on a finer timescale than in comparable commercial networks. The security architecture of military coalition networks may also restrict the monitoring and SLA management. Finally, we sketch a measurement-based approach showing how SLSs can be used in tactical military coalition force networks to support both call admission control and network engineering. Parts of this analysis include a discussion of SLSs and the proposed NATO standards for tactical communications network.

Gateway load balancing in future tactical networks

In future tactical networks, gateway nodes will have an important role in connecting different military communications platforms together to form a consolidated network. To increase capacity for upstream/downstream traffic as well as resiliency, more than one gateway should be deployed. In this context, performing gateway load balancing is vital in order to take full advantage of the resources available and thereby improve the performance. Previous work has shown that a number of factors such as the level of asymmetry, offered load, and gateway location may influence the performance of load balancing. However these parameters alone cannot explain why the performance of load balancing is high for certain topologies while it is very poor for others. Obviously, the specific layout of a topology also plays a crucial role on the efficiency of load balancing. We question what are the differences between topologies where load balancing is efficient from the topologies where it is inefficient? The work in this paper thus aims to find the answer to this question, and to explore the nature of performing load balancing in wireless multi-hop networks. Through the knowledge acquired we propose a Radio load based Load Balancing scheme (RLLB). Simulations of many randomly generated topologies show that the performance of RLLB is promising.

Performance analysis of gateway load balancing in ad hoc networks with random topologies

In wireless multihop networks such as MANETs or Wireless Mesh Networks (WMN), an Internet gateway (IGW) is a node that provides Internet connectivity, linking the wireless network with the global Internet. Congestion around the IGW represents a potential bottleneck for all Internet traffic that has to pass through the IGW. To alleviate this problem, the common solution is to have multiple IGWs in the network. However in order to take advantage of the capacity provided by multiple gateways, the routing protocol utilized must efficiently load balance the traffic among available IGWs such that the network performance is optimized. In this context, it is questioned to which extent it is possible to enhance the performance by utilizing a load balancing metric instead for the traditional shortest path metric. Furthermore, what are the factors that may set an upper limit for the performance that can be achieved. The aim of our investigation is to seek the answers to these questions through extensive simulations of a large number of random topologies. While a number of other studies have reported potential benefits of load balancing with some specific network topologies, to the best of our knowledge none have conducted similar studies covering a larger number of random topologies.

Optimized group communication for tactical military networks

In tactical networks there is a need for group communication applications, such as position and information sharing (Situational Awareness data), and Push-to-Talk (PTT) voice communication. This paper focuses on group communication in tactical military ad hoc networks, where most of the nodes are interested receivers. In this case, an efficient flooding protocol will be the best solution for the group communication. Efficient flooding can be achieved with the Simplified Multicast Forwarding (SMF) framework. The performance of SMF depends on the chosen forwarding algorithm. Two plausible alternatives are S-MPR and NS-MPR. The former is the more bandwidth efficient, while the latter is more robust to mobility. This paper investigates the limitations of the forwarding algorithms and investigates measures to mend S-MPRs mobility problem. Further, the paper suggests combining S-MPR and NS-MPR using the radio load as metric. Finally, the PTT and Situational Awareness (SA) traffic types are evaluated when run simultaneously, and a preemptive switch to S-MPR is proposed for the SA traffic. Through employing the methods suggested in this paper, the performance for PTT and SA traffic forwarded using SMF in tactical military networks can be increased.

Rerouting Time and Queueing in Proactive Ad Hoc Networks

In a MANET network where nodes move frequently, the probability of connectivity loss between nodes might be high, and communication sessions may easily loose connectivity during transmission. The routing protocol is designed to find alternative paths in these situations. This rerouting takes time, and the latency is referred to as the rerouting time. This paper investigates the rerouting time of proactive routing protocols and shows that the rerouting time is considerably affected by queueing. Simulations and analysis are conducted to explore the problem. Finally, we propose a MAC-layer solution that reduces the rerouting time problems due to queueing. Simulations and analysis show that the solution is so effective that it eliminates the entire problem in many situations.

XTP over ATM

Abstract This paper discusses some of the issues involved in mapping the Express Transfer Protocol (XTP) over the ATM (Asynchronous Transfer Mode) network technology. Both ATM and XTP are designed to facilitate high bandwidth transfers. ATM offers virtual channels that can be utilized to carry XTP frames without any encapsulation. The added benefit is the ability to offer different Quality of Service with the XTP channels. The largest problem area is the mapping of XTP signalling onto the signalling used in ATM. There is a mismatch in mechanisms that will result in inefficiency in the implementations. Some of the mechanisms in XTP seem to be optimized for shared medium technologies like Ethernet and FDDI. In our opinion, this needs to be changed in future releases of the protocol.

Performance analysis of the AODV ad hoc routing protocol in a dual radio network

Existing ad hoc routing protocols supports multiple radio interfaces, but they are designed for radio interfaces with similar properties. In contrast, in a typical usage scenario the interfaces will have different properties. There will be a need for longer range resulting in a lower bandwidth, while some applications will have higher capacity requirements but for nodes in close proximity. A solution may be to use one interface with low bandwidth and long range for example in the UHF band, whilst the other interface has a higher bandwidth and shorter range in the VHF or SHF band. Such a solution will not be energy efficient, so it will be feasible only for vehicle-mounted radios like in tactical military or emergency service networks. With the shortest path routing, the long range interfaces tend to be selected. The traffic is therefore shifted towards the links with the lowest capacity. Firstly, we analyze and illustrate the problem for the typical reactive routing protocol, AODV. Secondly, we propose two techniques that take the underlying radio capacity into account when a new route is established. Compared to original AODV, the proposed algorithms provide higher network throughput, less routing overhead, less end-to-end delay and still provides connectivity. The applicability to this approach is one of practical interest to scenarios where nodes are typically unevenly distributed; a network where some nodes only are reachable over long range radios, while others are also reachable over short range high capacity radios.