Abstract — Campus and enterprise wireless networks are increasingly characterized by ubiquitous coverage and rising traffic demands. Efficiently assigning channels to access points (APs) in these networks can significantly affect the performance and capacity of the WLANs. The state-of-the-art approaches assign channels statically, without considering prevailing traffic demands. In this paper, we show that the quality of a channel assignment can be improved significantly by incorporating observed traffic demands at APs and clients into the assignment process. We refer to this as traffic-aware channel assignment. We conduct extensive trace-driven and synthetic simulations and identify deployment scenarios where traffic-awareness is likely to be of great help, and scenarios where the benefit is minimal. We address key practical issues in using traffic-awareness, including measuring an interference graph, handling non-binary interference, collecting traffic demands, and predicting future demands based on historical information. We present an implementation of our assignment scheme for a 25-node WLAN testbed. Our testbed experiments show that traffic-aware assignment offers superior network performance under a wide range of real network configurations. On the whole, our approach is simple yet effective. It can be incorporated into existing WLANs with little modification to existing wireless nodes and infrastructure. I.

Active Networking adds programmability to the elements of the network, most aggressively by using programmable packets, or capsules. ANTS [1, 2] and PLANet [3, 4] are the most mature examples of capsule-based systems, both having been publicly available for several years. This paper presents our experience with these systems and the lessons they hold for the future of capsule-based Active Networking. The paper focuses on four key issues: flexibility, performance, security, and usability. We consider how ANTS and PLANet address these issues, noting that despite substantial surface differences, both systems identify similar key problems and use closely related solutions. Based on our experience with these systems we conclude that capsule-based systems can achieve useful levels of flexibility, performance, and usability. Many aspects of security can also be adequately addressed, but some important problems related to denial of service remain as open problems. Keywords: Activ...

Accurate cross-layer associations play an essential role in today’s network management tasks such as backbone planning, maintenance, and failure diagnosis. Current techniques for manually maintaining these associations are complex, tedious, and error-prone. One possible approach is to widen the interfaces between layers to support auto discovery. We argue instead that it is less useful to export additional data between layers than to import information into a separate, logically centralized management database. The specification of an interface to this database enables independent evolution of individual layers, side-stepping the challenges inherent in wide layer interfaces. Furthermore, management tools can leverage the network-wide cross-layer visibility provided by such a database to deliver enhanced services that depend on physical- or link-layer diversity. 1

Effective network troubleshooting is critical for maintaining efficient and reliable network operation. Troubleshooting is especially challenging in multihop wireless networks because the behavior of such networks depends on complicated interactions between many unpredictable factors such as RF noise, signal propagation, node interference, and traffic flows. In this paper we propose a new direction for research on fault diagnosis in wireless networks. Specifically, we present a diagnostic system that employs trace-driven simulations to detect faults and perform root cause analysis. We apply this approach to diagnose performance problems caused by packet dropping, link congestion, external noise, and MAC misbehavior. In a 25 node multihop wireless network, we are able to diagnose over 10 simultaneous faults of multiple types with more than 80 % coverage. Our framework is general enough for a wide variety of wireless and wired networks.

We present a new scheme for active, or programmable, packets based upon a new packet language, SNAP (Safe Networking with Active Packets). SNAP's semantics permit us to prove that all SNAP programs are safe with respect to network resource usage and evaluation isolation. Furthermore, we describe an implementation of a SNAP interpreter, snapd, which achieves high performance for standard networking tasks. This work represents the first active packet system that is demonstrated to be both safe and efficient.

Even though CMIP has achieved the level of an international standard for communication between management entities, very few management agents have been written to support CMIP. For the most part, the management agents being built today support SNMP, and a large number of management applications are being written to use this protocol. This situation adds to the inertia slowing the introduction of CMIP agents, since the existing SNMP-based applications cannot manage CMIP-capable devices. The absence of CMIP-based applications as an alternative leaves network managers with little choice but to stay with SNMP in the search for management interoperability. In this paper we present an architecture for building management agents that can support both SNMP and CMIP protocols. In particular, we analyze the naming schemes used by each of the two protocols for identifying information in a MIB, and show how we can use multiple resource name registration tables to support both protoc...

The increasing size of networks makes their management more and more difficult. While large networks are already structured hierarchically in two or three layers, network management has not yet moved from flat to hierarchical structures. The resulting problems are discussed in this article. In search for solutions, the deployment of midlevel management tools seems to be a way out of the crisis. By reducing the number of management values with rather low information content and calculating high order values, micromanagement tries to reduce the flood of information. This paper presents our solution, namely a prototype SubManager, which is a mid-level tool used to stucture network management hierarchically. Some examples are given to demonstrate its use. Our approach has the following advantages: 1) Interfaces to both managers and agents conform to SNMPv1 and SNMPv2. 2) The SubManager can easily be integrated into existing network management systems. 3) Compared with other systems alrea...

Effective network troubleshooting is critical for maintaining efficient and reliable network operation. Troubleshooting is especially challenging in multi-hop wireless networks because the behavior of such networks depends on complicated interactions between many factors such as RF noise, signal propagation, node interference, and traffic flows. In this paper we propose a new direction for research on fault diagnosis in wireless mesh networks. Specifically, we present a diagnostic system that employs trace-driven simulations to detect faults and perform root cause analysis. We apply this approach to diagnose performance problems caused by packet dropping, link congestion, external noise, and MAC misbehavior. In a 25 node mesh network, we are able to diagnose over 10 simultaneous faults of multiple types with more than 80 % coverage. Categories and Subject Descriptors

Business and economic considerations are driving the extensive use of service differentiation in Virtual Private Networks (VPNs) operated for business enterprises today. The resulting Class of Service (CoS) designs embed complex policy decisions based on the described priorities of various applications, extent of bandwidth availability, and cost considerations. These inherently complex high-level policies are realized through low-level router configurations. The configuration process is tedious and error-prone given the highly intertwined nature of CoS configuration, the multiple router configurations over which the policies are instantiated, and the complex access control lists (ACLs) involved. Our contributions include (i) a formal approach to modeling CoS policies from router configuration files in a precise manner; (ii) a practical and computationally efficient tool that can determine the CoS treatments received by an arbitrary set of flows across multiple routers; and (iii) a validation of our approach in enabling applications such as troubleshooting, auditing, and visualization of network-wide CoS design, using router configuration data from a cross-section of 150 diverse enterprise VPNs. To our knowledge, this is the first effort aimed at modeling and analyzing CoS configurations.

The trend towards value-added Internet services causes network providers to deploy new network based quality-of-service and security services. Today, however, the customer has only limited means of controlling the service delivery. For example the network security guaranteed by virtual private network providers cannot be checked with a traditional static approach. This paper presents a novel approach for controlling new IP services using mobile code, and motivates the approach with two examples of new IP services proposed by the Internet engineering task force (IETF). 1 Introduction The rapid growth of the transport capacity of the Internet and the global trend towards liberalisation of the telecommunication market forces the Internet service providers (ISP) to look for new revenues beyond pure connectivity offerings. Therefore, ISPs that control their own network try to introduce new Internet services including quality features such as premium transport or traffic privacy. Si...