Longitudinal human behavior modeling has received increasing attention over the years due to its widespread applications to patient monitoring, dietary and lifestyle recommendations, and just-in-time intervention for at-risk individuals (e.g., problematic drug users and struggling students), to name a few. Using in-the-moment health data collected via ubiquitous devices (e.g., smartphones and smartwatches), this multidisciplinary field focuses on developing predictive models for certain health or well-being outcomes (e.g., depression and stress) in the short future given the time series of individual behaviors (e.g., resting heart rate, sleep quality, and current feelings). Yet, most existing models on these data, which we refer to as ubiquitous health data, do not achieve adequate accuracy. The latest works that yielded promising results have yet to consider realistic aspects of ubiquitous health data (e.g., containing features of different types and high rate of missing values) and the consumption of various resources (e.g., computing power, time, and cost). Given these two shortcomings, it is dubious whether these studies could translate to realistic settings. In this paper, we propose MuHBoost, a multi-label boosting method for addressing these shortcomings, by leveraging advanced methods in large language model (LLM) prompting and multi-label classification (MLC) to jointly predict multiple health or well-being outcomes. Because LLMs can hallucinate when tasked with answering multiple questions simultaneously, we also develop two variants of MuHBoost that alleviate this issue and thereby enhance its predictive performance. We conduct extensive experiments to evaluate MuHBoost and its variants on 13 health and well-being prediction tasks defined from four realistic ubiquitous health datasets. Our results show that our three developed methods outperform all considered baselines across three standard MLC metrics, demonstrating their effectiveness while ensuring resource efficiency.

As wireless devices continue to proliferate, spectrum management is essential to a healthy and functioning digital ecosystem. Here we present an evolutionary analysis of how interbrand relationships can be expected to evolve in the cognitive radio domain over long time scales. We find that a range of trajectories are possible, and that the eventual outcomes depend on a variety of system parameters including the number of users and transmission band switching costs. Starting from previous bio-socially inspired fair spectrum sharing protocols, we put forward an extended model of secondary user etiquette that allows for a range of inter-group dynamics to arise in the natural course of competition over and co-use of spectrum resources. We show that as populations grow, increases in transmission switching costs lead to evolutionary pressures toward increasing antagonism between brands, and that in such scenarios devices tend to segregate by brand across bands. Understanding the drivers behind emerging inter-brand dynamics from an evolutionary perspective is an important input to the long term view of the successful application of distributed spectrum access and cognitive radio.

Wireless communication is an increasingly ubiquitous and important resource substrate of the digital ecosystem. In the face of the rapid growth in the population of Internet of Things (IoT), however, uncoordinated access to limited resources of radio spectrum is likely to lead to mass starvation. Here we put forward a new bio-social paradigm for cognitive radio, extending previous models in which the secondary users of spectrum alternate stochastically between foraging and consuming behaviors. In this paper, we ask and resolve two questions: (1) What costs and benefits does social deference to the group yield for each of the individuals therein? and (2) Can a notion of individual “hunger” form the basis of a distributed social deference scheme that is free of group coordination costs? Through a series of simulation experiments grounded in a well-specified formal model, we show that social deference improves both the fairness and the reliability of spectrum resource allocation, and moreover, that the concept of individual “hunger” can be used to implement social deference with minimal group coordination overhead. The results have consequences both in suggesting potential improvements for distributed spectrum access, and in understanding the evolutionary pressures on the behaviors of individual devices within emerging digital IoT societies.

Emerging standards in Cognitive Radio seek to alleviate the problem of spectrum scarcity by leveraging spectrum under-utilization through dynamic spectrum access (DSA). Unlicensed users (“secondary”) may opportunistically make use of channels during times when the licensed (“primary”) user is absent. The “open access” paradigm proposed by FCC mandates the study of primary-secondary dynamics, as well as secondary-secondary dynamics to better understand the implications of uncoordinated competition for limited number of resources between secondary users. In this paper, we consider long term etiquette between secondary users engaged in spectrum co-use. In order to maximize throughput, by avoiding over-crowded channels, we propose a novel approach wherein secondary users have the ability to “sense” very approximate contention levels at a single spectrum band at each point in time. System performance under the contention-sensing paradigm is compared to two schemes: first, a previously defined non-cooperative game model; second, a new model which allows for spectrum foraging, but without providing secondary users the ability to sense contention in a band. Through simulation experiments, we show that the contention-sensing paradigm results in significantly better co-use of spectrum resources. Additionally, our experiments demonstrate that the proposed contention-sensing paradigm continues to outperform when the ratio of secondary users to spectrum bands is increased. Thus, the new schemes described here may be of critical importance in a future where we expect the number of secondary users to grow exponentially and thus require a scalable mechanism for opportunistic scavenging of unused spectrum.

The Internet today continues to be vulnerable to distributed denial of service (DDoS) attacks. We consider the design of a scalable agent-based system for collecting information about the structure and dynamics of DDoS attacks. Our system requires placement of agents on inter-autonomous system (AS) links in the Internet. The agents implement a self-organizing and totally decentralized mechanism capable of reconstructing topological information about the spatial and temporal structure of attacks. The system is effective at recovering DDoS attack structure, even at moderate levels of deployment. In this paper, we demonstrate how careful placement of agents within the system can improve the system's effectiveness and provide better tradeoffs between system parameters and the quality of structural information the system generates. We introduced two agent placement algorithms for our agent-based DDoS system. The first attempts to maximize the percentage of attack flows detected, while the second tries to maximize the extent to which we are able to trace back detected flows to their sources. We show, somewhat surprisingly, these two objectives are concomitant. Placement of agents in a manner which optimizes in the first criterion tends also to optimize with respect to the second criterion, and vice versa. Both placement schemes show a marked improvement over a system in which agents are placed randomly, and thus provide a concrete design process by which to instrument a DDoS flow reconstruction system that is effective at recovering attack structure in large networks at moderate levels of deployment.

One of the ways that malware infects is by exploiting weaknesses in computer systems, often through conditions in software. When this happens, software and operating system vendors must repair these vulnerabilities by patching their software. However, vendors can release patches but cannot force users to apply them. Malware attempts to proliferate without regard to the state of the infected system; it is only once that the malware infection is stopped that we can truly say that systems are patched to eliminate that exploit. By examining appearance and disappearance of malware types, as determined through dynamic analysis of malware samples, classified by behavioral profiles correlated with a timeline of discovery dates, we can determine a more real-world average time for effective patch times, as opposed to the time it takes for a vendor to release a patch for a discovered vulnerability.

The practical limitations and challenges of radio frequency (RF) based communication networks have become increasingly apparent over the past decade, leading researchers to seek new hybrid communication approaches. One promising strategy that has been the subject of considerable interest is the augmentation of RF technology by Free Space Optics (FSO), using the strength of each communication technology to overcome the limitations of the other. In this article, we introduce a new scheme for controlling the topology in hybrid Radio-Frequency/Free Space Optics (RF/FSO) wireless mesh networks. Our scheme is based on adaptive adjustments to both transmission power (of RF and FSO transmitters) and the optical beam-width (of FSO transmitters) at individual nodes, with the objective of meeting specified Quality of Service (QoS) requirements, specifically end-to-end delay and throughput. We show how one can effectively encode the instantaneous objectives and constraints of the system as an instance of Integer Linear Programming (ILP). We demonstrate that the technique of Lagrangian Relaxation (LR), augmented with iterative repair heuristics, can be used to determine good (albeit sub-optimal) solutions for the ILP problem, making the approach feasible for mid-sized networks. We make the proposed scheme viable for large-scale networks in terms of number of nodes, number of transceivers, and number of source-destination pairs by solving the ILP problem using a Particle Swarm Optimization (PSO) implementation.

Abstract In this article, we describe how to construct physical computer network topologies, which can support the establishment of overlays that reduce or increase the distances between nodes. Reducing pairwise distances (i.e., compression) implies that the overlay enjoys significantly lower inter‐node latencies compared to the ambient physical network; such an overlay can be used to implement a “high‐performance mode” for disaster situations in which network responsiveness is of critical importance. On the other hand, increasing pairwise distances (i.e., expansion) implies that the overlay exhibits significantly higher internode latencies compared to the ambient physical network; such an overlay can be used to implement a brief “dilated state” in networks that have been infected by a malicious worm, where slowing down the infection spread allows greater time for antidote generation. We show that it is possible to design physical networks which support overlays whose logical link bandwidth is equal to the physical link bandwidth while providing arbitrarily high compression or expansion. We also show that it is possible to “grow” such networks over time in a scalable way, that is to say, it is possible to retain the compression/expansion properties while augmenting the network with new nodes, by making relatively small adjustments to the physical and overlay network structure. © 2010 Wiley Periodicals, Inc. NETWORKS, Vol. 58(1), 36–42 2011

We consider the problem of control traffic overhead in MANETs with long-lived connections, operating under a reactive routing protocol (e.g. AODV). In such settings, control traffic overhead origins can be traced principally to connection link failures, which trigger expensive global route discoveries. In this paper, we introduce a route maintenance scheme developed with the objective of reducing global route discoveries in such settings. The proposed scheme decrease the expected number of route discovery attempts by taking preemptive action to counteract impending link disconnections due to node movement. The proposed scheme was implemented as an extension of AODV in ns2, and compared with the standard AODV under different network regimes. Through the analysis of data derived from extensive simulations, we demonstrate that the proposed scheme significantly decreases overall control traffic while maintaining comparable packet delivery rates, at the cost of only very minor degradation in path optimality.

In reactive routing protocols, active routes for multihop connections retain their topological structure in spite of node movement over time. Unfortunately, node movements may make the connection route sub-optimal in terms of hop length, thereby resulting in unnecessarily high end-to-end delays, energy consumption and channel contention. In AODV, for example, a connection route is recomputed only if one of its constituent links suffers catastrophic failure, at which point global route discovery attempts repair, and after which the topological structure of the connection again returns to near-optimality.

We describe a general-purpose distributed system capable of traceback of malicious flow trajectories in the wide area despite possible source IP spoofing. Our system requires the placement of agents on a subset of the inter-autonomous system (AS) links of the Internet. Agents are instrumented with a uniform notion of attack criterion. Deployed, these agents implement a self-organizing, decentralized mechanism that is capable of reconstructing topological and temporal information about malicious flows. For example, when the attack criterion is taken to be based on excessive TCP connection establishment traffic to a destination, the system becomes a traceback service for distributed denial of service (DDoS) attacks. As another special case, when the attack criterion is taken to be based on malicious payload signature match as defined by an intrusion detection system (IDS), the agents provide a service for tracing malware propagation pathways. The main contribution of this paper, is to demonstrate that the proposed system is effective at recovering malicious flow structure even at moderate levels of deployment in large networks, including within the present Internet topology.

In this paper we describe an effective new technique by which to guarantee cooperativeness in Hybrid Radio-Frequency/Free Space Optics (RF/FSO) networks. Our approach is based on a novel Bayesian game-theoretic model, and uses a pricing scheme in which each destination node pays some amount of virtual money to the source node in order to acquire a reliable connection. We describe both single-stage and multi-stage solutions for the game in terms of its Nash and Perfect Bayesian Equilibriums. Pure strategies are found when the required conditions are met; otherwise the game is played as a mixed-strategy. Our numerical results quantify the inherent tradeoffs involved in changing the game's parameters vis-a-vis the equilibrium player strategies and game's outcomes.

Modern mobile ad-hoc networks (MANETs) frequently consist of nodes which exhibit a wide range of autonomy needs. This is particularly true in the settings where MANETs are most compelling, i.e. battlefield, response & rescue, and contexts requiring rapid deployment of mobile users. The time-critical nature of the underlying circumstances frequently requires deployment of both manned and unmanned nodes, and a coordination structure which provides prioritized tasking to them. Unlike consumer MANETs, these settings bring with them a common group purpose, making inter-node cooperation plausible. In this paper, we focus on how cooperation can improve MANET communications. We begin by taxonomizing all prior approaches and noting that no existing approach adequately captures networks where nodes exhibit a wide range of autonomy with respect to their mobility. To this end we present a new cooperative mobility model, developing a cost- benefit framework which enables us to explore the impact of cooperation in MANETs where nodes are, to varying extents, willing to move for the common good. In the second half of the paper, we describe the design of CoopSim, a platform for conducting simulation experiments to evaluate the impact of parameter, policy and algorithm choices on any system based on the proposed model. Finally, we present a small but illustrative case study and use the experimental evidence derived from it to give an initial evaluation of the merits of the proposed model and the efficacy of the CoopSim software.

Mission-oriented MANETs are characterized by implicit common group objectives which make inter-node cooperation both logical and feasible. We propose new techniques to leverage two optimizations for cognitive radio networks that are specific to such contexts: opportunistic channel selection and cooperative mobility. We present a new formal model for MANETs consisting of cognitive radio capable nodes that are willing to be moved (at a cost). We develop an effective decentralized algorithm for mobility planning, and powerful new Altering and fuzzy based techniques for both channel estimation and channel selection. Our experiments are compelling and demonstrate that the communications infrastructure-specifically, connection bit error rates-can be significantly improved by leveraging our proposed techniques. In addition, we find that these cooperative/opportunistic optimization spaces do not trade-off significantly with one another, and thus can be used simultaneously to build superior hybrid schemes. Our results have significant applications in high-performance mission-oriented MANETs, such as battlefield communications and domestic response & rescue missions.

Rapid-deployment mobile ad-hoc networks (MANETs) are frequently characterized by common overarching mission objectives which make it reasonable to expect some degree of cooperativeness on the part of their constituent nodes. In this article we demonstrate new strategies to improve MANET communications, based on inter-node cooperation with respect to node mobility. We present our model for cooperative mobility, and use this cost-benefit framework to explore the impact of cooperation in MANETs where nodes are - to varying extents - willing to be moved for the common good. We develop a mixed-integer linear programming (MILP) formulation of the model, accurately capturing its objectives and constraints. The MILP model is evaluated through simulations and found to be very effective, albeit for small networks. To make the proposed technique scale to large networks we develop a new technique for converting a large global MILP into a sequence of smaller local MILP optimizations, and demonstrate that the resulting approach is scalable and succeeds at efficiently moving cooperative nodes in a manner which optimizes connection bit error rates.

Memory scanning is a common technique used by malicious programs to read and modify the memory of other programs. Guarding programs against such exploits requires memory encryption, which is presently achievable either by (i) re-writing software to make it encrypt sensitive memory contents, or (ii) employing hardware-based solutions. These approaches are complicated, costly, and present their own vulnerabilities. In this paper, we describe new secure software technology that enables users to transparently add memory encryption to their existing software, without requiring users to invest in costly encryption hardware or requiring programmers to undertake complicated software redesign/redeployment. The memory encryption and transparent aegis library (METAL) functions as a shim library, allowing legacy applications to transparently enjoy an assurance of memory confidentiality and integrity. The proposed solution is tunable in terms of trade-offs between security and computational overhead. We describe the design of the library and evaluate its benefits and performance trade-offs.

Rapid-deployment mobile ad-hoc networks (MANETs) are frequently characterized by common over-arching mission objectives which predicate a cooperativeness on the part of constituent nodes. In this article we present a new strategy to improve MANET communications based on node cooperation with respect to mobility. We present our model for cooperative mobility, and use this cost-benefit framework to explore the impact of cooperation in MANETs where nodes are-to varying extents--willing to be moved for the common good. We develop an effective centralized algorithm for mobility planning based on multigrid techniques. Our simulation results are compelling and demonstrate that the communication infrastructure-specifically, connection bit error rate-can be significantly improved by leveraging this proposed scheme.

In this paper, we develop a Cooperative Mobility Modelthat captures new salient features of collaborative and mission-oriented MANETs. We propose new techniques to leverage two optimizations for cognitive radio networks that are specific to such contexts: cooperative mobility and opportunistic channel selection. We present a new formal model for MANETs consisting of cognitive radio capable nodes that are willing to be moved (at a cost). We develop an effective decentralized algorithm for mobility planning, and powerful newfiltering and fuzzy based techniques for both channel estimation and channel selection. Our experiments are compelling and demonstrate that the communications infrastructure-specifically, connectionbit error rates-can be significantly improved by leveraging our proposed techniques. In addition, we find that these cooperative/opportunistic optimization spaces do not trade-off significantly with one another, and thus can be used simultaneously to build superior hybrid schemes. Our results have significant applications in high-performance mission-oriented MANETs, such as battlefield communications and domestic response rescue missions.

We present the results of simulation experiments that compare end-to-end error management (used in controlled access MAC protocols) against hop-by-hop error management (used in random access MAC protocols). Our experiments are novel in that we restrict both MAC schemes to identical power budgets and power distribution strategies. By making such a normalized comparison, we observe that end-to-end schemes are more effective than hop-by-hop schemes at reducing connection BER. We are also able to quantify the sensitivity of this relative advantage to various environmental parameters, including power budget size, geographic distance, the number of hops, and power distribution scheme.

We develop a new dynamic scheme which continuously redistributes a fixed power budget among the wireless nodes participating in a multi-hop wireless connection, with the objective of minimizing the end-to-end wireless connection bit error rate (BER). We compare the efficacy of our scheme with two static schemes: one that distributes power uniformly, and one that distributes it proportionally to the square of inter-hop distances. In our experiments we observed that the dynamic allocation scheme achieved superior performance, reducing BER by using its ability to distribute the power budget. We quantified the sensitivity of this performance improvement to various environmental parameters, including power budget size, geographic distance, and the number of hops.

In this paper we present new energy-efficient techniques to lower the packet-level error rates of application-layer connections in wireless ad-hoc networks. In our scheme, each application-layer connection is implemented at the physical level by an overlay network. Data packets submitted at the connection source are checksummed and replicated, flowing breadth-first across the overlay network towards the destination. The destination delivers the first error-free copy of each packet, in order, to the application layer, dropping packets that are corrupt or duplicate. Specifically in this paper, we consider overlays consisting of multiple parallel node-disjoint multi-hop paths. We compare this overlay scheme with the traditional scheme in which the source transmits to the destination along a single minimum-hop path. We show that even when the two schemes are constrained by identical power consumption bounds, an overlay scheme that uses multiple multi-hop paths provides significantly lower packet-level error rates in many common situations. We describe the relationship between packet error rate, the number of paths, and the lengths of each path, and show that the qualitative nature of the relationship changes significantly, depending on available power budget.

In this paper we present and evaluate a new technique to lower packet-level error rates of application layer connections in wireless ad-hoc networks. In our scheme, data packets submitted at a connection's source are checksummed and replicated, flowing breadth-first across an overlay network towards the destination. The destination delivers the first error-free copy of each packet, in order, to the application layer, dropping packets that are corrupt or duplicate. Specifically in this paper, we consider overlays consisting of multiple parallel multi-hop paths. We provide an algorithm which determines the optimal parameters of the overlay in terms of the number of paths, their lengths, and specific routes. We demonstrate experimentally that the proposed scheme significantly outperforms traditional routing and power allocation approaches in terms of bit error rate, even when the comparison is made under identical power consumption constraints.

Just a few years ago, the phrase "optical Internet" meant little more than a few WDM fibers whose wavelengths were serving as a link layer to interconnect routers via add/drop multiplexors. By comparison today, multiwavelength reconfigurable optical switches are becoming readily available, enabling for the first time, the development of large-scale high speed optical networks operating at rates in excess of 10 Gbit/s. At the core of this lightpath management problem for these switched optical networks is the challenge of developing effective mechanisms to coordinate the control planes of different optical switches. In this paper, we describe how the Multi-wavelength Optical Networking (MONET) Program is meeting this challenge today. Specifically, we describe the NRL MONET control plane and, though its implementation is not yet fully completed, provide some preliminary measurements and assessment of its effectiveness as a system for lightpath management.

The toolkit for routing in optical networks (TRON) is a freely available library developed to facilitate research experiments on OSPF-based routing protocols for optical networks. Currently, TRON supports the lightwave-OSPF routing protocol, which is our adaptation of the optical extensions to OSPF proposed in the Internet drafts of Kompella et al and Wang et al. TRON is implemented in C++ using the component architecture for simulating network objects (CASiNO). TRON software can be used in either simulation or emulation mode. In this paper, we describe lightwave-OSPF and the architecture of the TRON software.