Jonas Gustafsson

Integration of Wireless Sensor and Actuator Nodes With IT Infrastructure Using Service-Oriented Architecture

A large number of potential applications for Wireless Sensor and Actuator Networks (WSAN) have yet to be embraced by industry despite high interest amongst academic researchers. This is due to various factors such as unpredictable costs related to development, deployment and maintenance of WSAN, especially when integration with existing IT infrastructure and legacy systems is needed. Service-Oriented Architecture (SOA) is seen as a promising technique to bridge the gap between sensor nodes and enterprise applications such as factory monitoring, control, and tracking systems where sensor data is used. To date, research efforts have focused on middleware software systems located in gateway devices that implement standard service technology, such as Devices Profile for Web Services (DPWS), for interacting with the sensor network. This paper takes a different approach-deploying interoperable Simple Object Access Protocol (SOAP)-based web services directly on the nodes and not using gateways. This strategy provides for easy integration with legacy IT systems and supports heterogeneity at the lowest level. Twofold analysis of the related overhead, which is the main challenge of this solution, is performed; Quantification of resource consumption as well as techniques to mitigate it are presented, along with latency measurements showing the impact of different parts of the system on system performance. A proof-of-concept application using Mulle-a resource-constrained sensor platform-is also presented.

Thermodynamic Simulation of a Detached House with District Heating Subcentral

A physical thermodynamic model of a detached house connected to a low-tempered district heating network is presented. The model is created in Mathworks Simulink@ with a pedagogic approach in mind, e.g. masked subsystems divided in to physical components. The house model is easily modified to any detached house. Provision is also made to make it scalable to multi-family houses. The district heating substation modeled is a parallel coupled plate heat exchanger, which is the most common substation in smaller buildings such as villas. The purpose of creating the model was to provide a platform for test and evaluation of new control methods for district heating system based on wireless sensor networks. Initial validation of the model is presented.

Independence and interdependence of systems in district heating

In this paper, a fully operational wireless sensor and actuator network is presented. The network has the ability to control a district heating substation to ensure indoors comfort while minimizing energy waste by maximizing heat extraction from the distribution network. Introduced here is the foundation for a systems of systems approach within a district heating application, where several substations cooperate with the heat production plant. Presented are also the first steps to a service oriented architecture (SOA) where sensor nodes in the district heating network can cooperate with other nodes and systems, for example a ventilation control network.

Application of Service Oriented Architecture for Sensors and Actuators in District Heating Substations

Hardwired sensor installations using proprietary protocols found in todays district heating substations limit the potential usability of the sensors in and around the substations. If sensor resources can be shared and re-used in a variety of applications, the cost of sensors and installation can be reduced, and their functionality and operability can be increased. In this paper, we present a new concept of district heating substation control and monitoring, where a service oriented architecture (SOA) is deployed in a wireless sensor network (WSN), which is integrated with the substation. IP-networking is exclusively used from sensor to server; hence, no middleware is needed for Internet integration. Further, by enabling thousands of sensors with SOA capabilities, a System of Systems approach can be applied. The results of this paper show that it is possible to utilize SOA solutions with heavily resource-constrained embedded devices in contexts where the real-time constrains are limited, such as in a district heating substation.

Energy harvesting technologies for wireless sensors in rotating environments

Using sensors to measure parameters of interest in rotating environments and communicating the measurements in real-time over wireless links, requires a reliable power source. In this paper, we have investigated the possibility to generate electric power locally by evaluating six different energy-harvesting technologies. The applicability of the technology is evaluated by several parameters that are important to the functionality in an industrial environment. All technologies are individually presented and evaluated, a concluding table is also summarizing the technologies strengths and weaknesses. To support the technology evaluation on a more theoretical level, simulations has been performed to strengthen our claims. Among the evaluated and simulated technologies, we found that the variable reluctance-based harvesting technology is the strongest candidate for further technology development for the considered use-case.

Wireless infrastructure in a district heating substation

A new wireless sensor and actuator network architecture for a district heating substation is presented. The implementation allows for new control algorithms to save substancial amounts of energy and therefore reduce waste and indirectly CO2 emissions. The wireless network uses TCP/IP all the way to the sensor level, which can enable a standardized Service Oriented Architecture and the possibility to upgrade software of each node remotely over the Internet. New diagnostic services are also available to improve service and maintenance.

Piezoelectric energy harvesting modeled with SPICE

The generation of electricity from vibrations has attracted considerable attention over the past decade. For this purpose, a piezoelectric-type harvester tuned to resonate at a specific frequency is typically used. This paper proposes a SPICE model capable of modeling piezoelectric energy harvesters. The model is parametrized enabling simple introduction of relevant physical parameters. Underlying theory based on Euler-Bernoulli beam theory is presented and modeling simplifications are justified. The SPICE model is verified by comparison to examples for which analytical and experimental solutions has been published. The SPICE model is valid close to the analyzed mode center frequency and delivers results within 1% compared to experimental and analytical data. Furthermore, we also show that the harvester can be electrically tuned to match the excitation frequency. This makes it possible to maximize the power output for both linear and non-linear loads. The here presented model enables efficient analysis of harvester design and changes of harvesting conditions.

A passive Barkhausen noise sensor for low-power applications

This paper proposes a passive Barkhausen noise sensor design suitable for low power applications. The sensor uses a permanent magnet and the relative motion between itself and a measured specimen instead of the conventional method that uses a fixed sensor and an alternating magnetic field. Since this novel design is passive, the sensor is well suited for low power applications and could potentially be used in e.g. a condition monitoring system integrated into a rolling element bearing. Proof of concept testing has been performed showing that the proposed sensor produces similar results as conventional Barkhausen noise sensors when applied to specimens being cyclically loaded until failure in a rotating bending rig. The results imply that material fatigue detection using the Barkhausen noise can be performed with the proposed sensor at a fraction of the energy cost compared to a conventional sensor. This warrants future research into the development of the proposed sensor, its advantages, disadvantages, and functionality.

Interleaved Switch Harvesting on Inductor: Non-linear extraction, action and reaction

The circuit presented in this paper provides increased transfer efficiency for the electric charge generated by a piezoelectric element to the storage device, it also enables a simplified active control circuitry required for switched harvesting solutions as the control transistors are referenced to ground. This is achieved by inverting the voltage potential across the piezoelectric element and thereby maintaining a higher voltage while extracting energy. The increased potential allows vibration harvesters to operate over a broader frequency range. For weakly coupled piezoelectric harvesters, the increased potential allows the electrical power output to be increased. In this paper, the circuit is simulated in SPICE using existing models of discrete components in conjunction with a model of a piezoelectric vibration harvester. The power output and mechanical behavior of the harvester using the presented circuit are compared to two existing technologies: Synchronous Switch Harvesting on Inductor (SSHI) and a full-wave rectifying bridge. The comparison demonstrates that the novel Interleaved Switch Harvesting on Inductor (ISHI) circuit outperforms state-of-the-art active harvesting solutions, e.g., SSHI, as a result of the decreased energy losses from diodes. In addition, the ISHI circuit requires fewer components than does the SSHI circuit and solves the floating transistor source issue that arises in SSHI implementations. To achieve the maximum power conversion efficiency, impedance matching is required, which can be achieved by controlling the voltage of the storage device.

Building System of Systems with SOA Technology: A Smart House Use Case

The IMC-AESOP architecture has been used to implemente a smart house demonstration. Six different systems has been integrated with local (802.11, 802.15.4) and global (telecom) communication. The six systems integrated are: Car arrival detection system, Garage door opening system, House security system, External house lightning system, External electrical outlet system, House energy control system. The SOA technologies used are CoAP and EXI using SenML to encode the services. Engineering tools have been used to simulate the usage scenario and provide prediction of system behaviour.