Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

http://www.csun.edu/~andrzej/COMP529-S05/papers/Mesh%20Networks%20Survey.pdf

Wireless mesh networks: a survey

We present Tor, a circuit-based low-latency anonymous communication service. This second-generation Onion Routing system addresses limitations in the original design by adding perfect forward secrecy, congestion control, directory servers, integrity checking, configurable exit policies, and a practical design for location-hidden services via rendezvous points. Tor works on the real-world Internet, requires no special privileges or kernel modifications, requires little synchronization or coordination between nodes, and provides a reasonable tradeoff between anonymity, usability, and efficiency. We briefly describe our experiences with an international network of more than 30 nodes. We close with a list of open problems in anonymous communication.

/pdf/tor-the-second-generation-onion-router-2qfrg0ppqb.pdf

Tor: the second-generation onion router

We consider routing security in wireless sensor networks. Many sensor network routing protocols have been proposed, but none of them have been designed with security as a goal. We propose security goals for routing in sensor networks, show how attacks against ad-hoc and peer-to-peer networks can be adapted into powerful attacks against sensor networks, introduce two classes of novel attacks against sensor networks sinkholes and HELLO floods, and analyze the security of all the major sensor network routing protocols. We describe crippling attacks against all of them and suggest countermeasures and design considerations. This is the first such analysis of secure routing in sensor networks.

/pdf/secure-routing-in-wireless-sensor-networks-attacks-and-4630unozk9.pdf

Secure routing in wireless sensor networks: attacks and countermeasures

Advances in sensing and tracking technology enable location-based applications but they also create significant privacy risks. Anonymity can provide a high degree of privacy, save service users from dealing with service providers’ privacy policies, and reduce the service providers’ requirements for safeguarding private information. However, guaranteeing anonymous usage of location-based services requires that the precise location information transmitted by a user cannot be easily used to re-identify the subject. This paper presents a middleware architecture and algorithms that can be used by a centralized location broker service. The adaptive algorithms adjust the resolution of location information along spatial or temporal dimensions to meet specified anonymity constraints based on the entities who may be using location services within a given area. Using a model based on automotive traffic counts and cartographic material, we estimate the realistically expected spatial resolution for different anonymity constraints. The median resolution generated by our algorithms is 125 meters. Thus, anonymous location-based requests for urban areas would have the same accuracy currently needed for E-911 services; this would provide sufficient resolution for wayfinding, automated bus routing services and similar location-dependent services.

/pdf/anonymous-usage-of-location-based-services-through-spatial-4o8wvs27a8.pdf

Anonymous Usage of Location-Based Services Through Spatial and Temporal Cloaking

Disruption-tolerant networks (DTNs) attempt to route network messages via intermittently connected nodes. Routing in such environments is difficult because peers have little information about the state of the partitioned network and transfer opportunities between peers are of limited duration. In this paper, we propose MaxProp, a protocol for effective routing of DTN messages. MaxProp is based on prioritizing both the schedule of packets transmitted to other peers and the schedule of packets to be dropped. These priorities are based on the path likelihoods to peers according to historical data and also on several complementary mechanisms, including acknowledgments, a head-start for new packets, and lists of previous intermediaries. Our evaluations show that MaxProp performs better than protocols that have access to an oracle that knows the schedule of meetings between peers. Our evaluations are based on 60 days of traces from a real DTN network we have deployed on 30 buses. Our network, called UMassDieselNet, serves a large geographic area between five colleges. We also evaluate MaxProp on simulated topologies and show it performs well in a wide variety of DTN environments.

/pdf/maxprop-routing-for-vehicle-based-disruption-tolerant-1qv0k3f197.pdf

MaxProp: Routing for Vehicle-Based Disruption-Tolerant Networks

Most recent ad hoc network research has focused on providing routing services without considering security. We detail security threats against ad hoc routing protocols, specifically examining AODV and DSR. In light of these threats, we identify three different environments with distinct security requirements. We propose a solution to one, the managed-open scenario where no network infrastructure is pre-deployed, but a small amount of prior security coordination is expected. Our protocol, authenticated routing for ad hoc networks (ARAN), is based on certificates and successfully defeats all identified attacks.

/pdf/a-secure-routing-protocol-for-ad-hoc-networks-2ifoapah6c.pdf

A secure routing protocol for ad hoc networks

Many DTN routing protocols use a variety of mechanisms, including discovering the meeting probabilities among nodes, packet replication, and network coding. The primary focus of these mechanisms is to increase the likelihood of finding a path with limited information, so these approaches have only an incidental effect on such routing metrics as maximum or average delivery latency. In this paper, we present RAPID, an intentional DTN routing protocol that can optimize a specific routing metric such as worst-case delivery latency or the fraction of packets that are delivered within a deadline. The key insight is to treat DTN routing as a resource allocation problem that translates the routing metric into per-packet utilities which determine how packets should be replicated in the system.We evaluate RAPID rigorously through a prototype of RAPID deployed over a vehicular DTN testbed of 40 buses and simulations based on real traces. To our knowledge, this is the first paper to report on a routing protocol deployed on a real DTN at this scale. Our results suggest that RAPID significantly outperforms existing routing protocols for several metrics. We also show empirically that for small loads RAPID is within 10% of the optimal performance.

/pdf/dtn-routing-as-a-resource-allocation-problem-22qq5k3jz5.pdf

DTN routing as a resource allocation problem

IP multicast offers the scalable point-to-multipoint delivery necessary for using group communication applications on the Internet. However, the IP multicast service has seen slow commercial deployment by ISPs and carriers. The original service model was designed without a clear understanding of commercial requirements or a robust implementation strategy. The very limited number of applications and the complexity of the architectural design-which we believe is a consequence of the open service model-have deterred widespread deployment as well. We examine the issues that have limited the commercial deployment of IP multicast from the viewpoint of carriers. We analyze where the model fails and what it does not offer, and we discuss requirements for successful deployment of multicast services.

Deployment issues for the IP multicast service and architecture

Underwater sensor networks are attracting increasing interest from researchers in terrestrial radio-based sensor networks. There are important physical, technological, and economic differences between terrestrial and underwater sensor networks. Previous surveys have provided thorough background material in underwater communications, and an introduction to underwater networks. This has included detail on the physical characteristics of the channel [1], on underwater acoustic communications [2, 3, 4], and surveys of underwater acoustic networks [5, 6, 7, 8]. In this survey, we highlight a number of important practical issues that are not emphasized in the recent surveys of underwater networks, with an intended audience of researchers who are moving from radio-based terrestrial networks into underwater networks. We focus on issues relevant to medium access control (MAC) protocols, which are an area of continuing work both in terrestrial sensor networks and especially in underwater networks. Underwater networks are often characterized by more expensive equipment, higher mobility, sparser deployments, and different energy regimes when compared with terrestrial sensor networks. We discuss the role of these factors in the different set of challenges that face underwater networks. We identify several of these points in the outline below, and we expand upon them in later sections. In Section 2, we provide a classification scheme for underwater networks. Link-layer range, node density, and geographic coverage of nodes are key factors in determining the type of network deployed. The key differentiating factor for underwater net-

/pdf/a-survey-of-practical-issues-in-underwater-networks-3t1v0coh2p.pdf

Brian Neil Levine

Papers

MaxProp: Routing for Vehicle-Based Disruption-Tolerant Networks

A secure routing protocol for ad hoc networks

DTN routing as a resource allocation problem

Deployment issues for the IP multicast service and architecture

A survey of practical issues in underwater networks