Pruet Boonma

BiSNET: A biologically-inspired middleware architecture for self-managing wireless sensor networks

This paper describes BiSNET (Biologically-inspired architecture for Sensor NETworks), a middleware architecture that addresses several key issues in multi-modal wireless sensor networks (MWSNs) such as autonomy, scalability, adaptability, self-healing and simplicity. Based on the observation that various biological systems have developed mechanisms to overcome these issues, BiSNET follows certain biological principles such as decentralization, food gathering/storage and natural selection to design MWSN applications. In BiSNET, each application consists of multiple software agents, which operate on the BiSNET middleware platform in individual sensor nodes, and each agent exploits certain biologically-inspired mechanisms such as energy exchange, pheromone emission, replication, migration and death. This is analogous to a bee colony (application) consisting of multiple bees (agents). This paper describes the biologically-inspired mechanisms in BiSNET, and evaluates their impacts on the autonomy, scalability, adaptability, self-healing and simplicity of MWSNs. Simulation results show that BiSNET allows sensor nodes (agents and platforms) to be scalable with respect to network size, autonomously adapt their sleep periods for power efficiency and responsiveness of data collection, adaptively aggregate data from different types of sensor nodes, and collectively self-heal (i.e., detect and eliminate) false positive sensor data. The BiSNET platform is implemented simple in its design and lightweight in its memory footprint.

MONSOON: A Coevolutionary Multiobjective Adaptation Framework for Dynamic Wireless Sensor Networks

Wireless sensor applications (WSNs) are often required to simultaneously satisfy conflicting operational objectives (e.g., latency and power consumption). Based on an observation that various biological systems have developed the mechanisms to overcome this issue, this paper proposes a biologically-inspired adaptation mechanism, called MONSOON. MONSOON is designed to support data collection applications, event detection applications and hybrid applications. Each application is implemented as a decentralized group of software agents, analogous to a bee colony (application) consisting of bees (agents). Agents collect sensor data and/or detect an event (a significant change in sensor reading) on individual nodes, and carry sensor data to base stations. They perform these data collection and event detection functionalities by sensing their surrounding environment conditions and adoptively invoking biologically- inspired behaviors such as pheromone emission, reproduction and migration. Each agent has its own behavior policy, as a gene, which defines how to invoke its behaviors. MONSOON allows agents to evolve their behavior policies (genes) and adapt their operations to given objectives. Simulation results show that MONSOON allows agents (WSN applications) to simultaneously satisfy conflicting objectives by adapting to dynamics of physical operational environments and network environments (e.g., sensor readings and node/link failures) through evolution.

Extensible and Precise Modeling for Wireless Sensor Networks

Developing applications for wireless sensor networks (WSN) is a complicated process because of the wide variety of WSN applications and low-level implementation details. Model-Driven Engineering offers an effective solution to WSN application developers by hiding the details of lower layers and raising the level of abstraction. However, balancing between abstraction level and unambiguity is challenging issue. This paper presents Baobab, a metamodeling framework for designing WSN applications and generating the corresponding code, to overcome the conflict between abstraction and reusability versus unambiguity. Baobab allows users to define functional and non-functional aspects of a system separately as software models, validate them and generate code automatically.

Middleware Support for Pluggable Non-Functional Properties in Wireless Sensor Networks

Wireless sensor networks (WSN) imposes stringent constraints on efficiency, memory footprint and power consumption. Since the need to satisfy these constraints often results in tightly coupled designs, WSN applications tend to be inflexible; it is hard to flexibly reuse, introduce, customize and replace various non-functional properties (e.g., data routing, concurrency, data aggregation and event filtering) for developing and maintaining WSN applications. In order to address this issue, this paper proposes the TinyDDS middleware, which decouples various non-functional properties from WSN applications and allows those applications to flexibly reuse and transparently configure non-functional properties according to their own requirements. Without breaking the generic architecture of TinyDDS, the proposed pluggable framework allows WSN applications to have fine-grained control over non-functional properties and specialize in their own requirements. Currently, TinyDDS supports two types of non-functional properties: application-level and middleware-level non-functional properties.

A Confidence-Based Dominance Operator in Evolutionary Algorithms for Noisy Multiobjective Optimization Problems

This paper describes a noise-aware dominance operator for evolutionary algorithms to solve the multiobjective optimization problems (MOPs) that contain noise in their objective functions. This operator takes objective value samples of given two individuals (or solution candidates), estimates the impacts of noise on the samples and determines whether it is confident enough to judge which one is superior/inferior between the two individuals. Since the proposed operator assumes no noise distributions a priori, it is well applicable to various MOPs whose objective functions follow unknown noise distributions. Experimental results show that it operates reliably in noisy MOPs and outperforms existing noise-aware dominance operators.

Toward Interoperable Publish/Subscribe Communication between Wireless Sensor Networks and Access Networks

Traditional wireless sensor networks (WSNs) often do not consider interoperability between WSNs and access networks. To address the issue, this paper investigates interoperable publish/subscribe communication in WSNs. The proposed middleware, called TinyDDS, provides two types of interoperability, programming language interoperability and protocol interoperability, by customizing standard data types, data representation and session protocol. Evaluation results show that TinyDDS simplifies the development of publish/subscribe applications and it is implemented efficient in memory footprint and power consumption.

BiSNET: A Biologically-Inspired Architecture forWireless Sensor Networks

This paper describes a sensor network architecture, called BiSNET, which addresses several key issues in wireless sensor networks such as autonomy, adaptability, self-healing and simplicity. Based on the observation that various biological systems have developed mechanisms necessary to overcome these issues, BiSNET follows certain biological principles such as decentralization, food gathering/storage and natural selection to design sensor networks. This paper describes and evaluates the biologically-inspired mechanisms in BiSNET Simulation results show that BiSNET allows sensor nodes to autonomously adapt their duty cycles for power efficiency and responsiveness of data transmission, to collectively self-heal (i.e., detect and eliminate) false positives in their sensor readings, and to be lightweight

Moppet: A Model-Driven Performance Engineering Framework for Wireless Sensor Networks

This paper describes a model-driven performance engineering framework for applications embedded in individual nodes of wireless sensor networks (WSNs). The framework, called Moppet, is designed for developers, even non-programmers, to rapidly implement WSN applications, estimate their performance and feedback the estimated performance results for customizing their design/implementation. By leveraging the notion of feature modeling, Moppet allows developers to graphically and intuitively specify features (e.g., functionalities and configuration policies) required in their applications. It also validates a set of constraints among features and generates lightweight application code. Moreover, with event calculus and network calculus, Moppet estimates a WSN application’s performance without generating its code nor running it on simulators and real networks. It can approximate data yield, data transmission latency and network lifetime as well as each node’s power and bandwidth consumption. Evaluation results show that, in a large-scale WSN of 400 nodes, Moppet’s performance estimation is 46% more efficient than empirical performance measurement and its estimation error is lower than 10%. Moppet scales well to network size with respect to estimation efficiency and accuracy.

Leveraging Biologically-inspired Mobile Agents Supporting Composite Needs of Reliability and Timeliness in Sensor Applications

Wireless sensor networks are often used for event detection applications, which require a certain level of reliability and timeliness while minimizing energy consumption. Existing work considers reliability, timeliness and energy consumption largely in isolation. This paper proposes a solution to satisfy these conflicting requirements by using biologically-inspired mobile agents. The problem is formulated into an NP-hard problem, the vehicle routing problem, and decentralized and centralized heuristics are developed to govern agent behaviors. Simulation results show that proposed solution allows agents (i.e., sensor applications) to effectively balance the tradeoffs among reliability, timeliness and energy efficiency and outperform an existing similar mechanism.

Evolutionary constraint-based multiobjective adaptation for self-organizing wireless sensor networks

Wireless sensor applications (WSNs) are often required to simultaneously satisfy conflicting operational objectives (e.g., latency and power consumption). Based on an observation that various biological systems have developed the mechanisms to overcome this issue, this paper proposes a biologically-inspired adaptation mechanism, called MONSOON. With MONSOON, each application is designed as a decentralized group of software agents. This is analogous to a bee colony (application) consisting of bees (agents). Agents collect sensor data on individual nodes, and carry the data to base stations. They perform this data collection functionality by autonomously sensing their local and surrounding environment conditions and adaptively invoking biological behaviors such as pheromone emission, replication, reproduction and migration. Each agent has its own behavior policy, as a gene, which defines how to invoke its behaviors. MONSOON allows agents to evolve their behavior policies (i.e., genes) and simultaneously adapt to conflicting objectives. In addition to consider multiple objectives equally, MONSOON also allows agents to evolve in a constraint-based (or intentionally-biased) manner. A constraint is defined as an upper or lower bound for each objective. Simulation results show that MONSOON allows agents (WSN applications) to adapt to dynamics of the network (e.g., node/link failures) through evolution and simultaneously satisfy conflicting objectives in a self-organizing manner.