Please login first

List of accepted submissions

Show results per page
Find papers
  • Open access
  • 39 Reads
Engineering design for mobile brain imaging helmet – AM-PET

It has been proposed that a portable unit which can be worn on a patient's head to provide a lightweight, low cost unit can conduct AM-PET (Ambulatory Micro-dose Positron Emission Tomography) scans in diverse locations. This is also conducive to the testing of patients while performing selected activities that stimulate certain brain regions such as walking, playing a game, clapping, and other such tasks; this unit is specifically intended for those with Alzheimer's, cancer, drug addiction, or any kind of brain injury to learn more about the condition for findings of a cure and treatment options. The need for this type of device has led to the design of a portable head held unit that currently uses twelve PET module sensors in specific locations around the head. Current PET scanners need a patient to remain still for several minutes, whereas this unit you can see defined parts of the brain working. This manuscript is focused on the engineering design aspects of the Am-PET project. The goal of this work is to further design this portable head unit to alleviate weight of the current head unit, improve comfort for the patient, and diversify its application amongst research activities. Specific gains include the design and manufacture of a system that reduces the relative movement of head and helmet significantly and does not restrict the user's head motion. The continuation of this work includes design review of the current model, design improvements based on the identified project requirements, and development of a working prototype. The success of this work will be measured by comparing the developed prototype to the original device and the newly developed metrics.

  • Open access
  • 16 Reads
Statistical Evaluation of the Positioning Error in Sequential Localization Techniques for Sensor Networks

In wireless sensor networks (WSN) localization of the nodes is relevant, especially for the task of identification of events that occur in the environment being monitored. Thus, positioning of the sensors is essential to satisfy such task. In WSN, sensors use techniques for self-localization based on some reference or anchor nodes (AN) that know their own position in advance. These ANs are fusion centers or nodes with more processing power. Assuming that the number of ANs given in the network is N, we carry out the localization algorithm to position sensors sequentially using those N ANs. Now, when a sensor has been localized, it becomes a new AN, and now, other sensors will use N+1 ANs, this is repeated until all the sensors in the network have been localized. In this sequential localization algorithm, the positioning error (difference between true and estimated position) increases as the sensor to be located is farther away from the group of original ANs in the network. This error becomes critical when propagation issues such as mutlipath propagation and shadowing in indoor environments are considered. In this paper, we characterize statistically positioning error in WSN for one and two-dimensional indoor environments when sensors are deployed randomly with different distributions. We also evaluate the performance of the localization algorithm and determine correcting factors based on the statistical characterization to minimze positioning error. We present results from simulations and measurements in an indoor environment.

  • Open access
  • 56 Reads
Techniques to compensate propagation impairments for greater accuracy in localization for sensors in indoor environments.

Position location estimation in sensor networks is a valuable supplement since it supports the deployment of location-based services. Sensor networks have changing conditions in the environment due to propagation issues, noise and placement of sensors, which represent challenges that position location algorithms must deal with. Accuracy of the location estimation technique is relevant since it allows minimizing positioning error. In indoor environments, propagation issues such as multipath signals, affect adversely the precision of the positioning algorithm. Also, the use of parameters such as time of arrival has a trade-off between the small distances that the signals traverse and the precision of the hardware used to capture such measurements. In this paper, we use received signal strength indicator (RSSI) to estimate the coordinates of individual sensors in an area of study. The RSSI parameter is measured and processed by a set of reference nodes installed in the area. We show that performance of the location estimation algorithm needs additional techniques to obtain improved accuracy rate. We develop additional techniques based on the use of polynomial interpolation and spline functions to balance propagation issues. These techniques help us to implement correcting factors that are used in the propagation model to compensate the RSSI measurements. We use these techniques to show how the positioning error is reduced in the area of study with simulations and measurements using sensors.

  • Open access
  • 35 Reads
Change detection of Lakes in Pokhara, Nepal using Landsat Data

Pokhara, city of lakes, is second largest and most beautiful tourist place in Nepal. Out of seven lakes, the large three: Phewa, Begnas and Rupa are famous for tourist attraction, whereas the rest are small and less known. Lakes are not only have economic value, but are also ecological and environmental resources. But, these lakes are facing challenges due to climatic and anthropogenic activities. As these changes are slow and takes long time, the damage unnoticed to take measures. Hence, long historic data provided such as of remote sensors are concrete evidence of change, which help us understand the cause and prevent further change. Landsat series provide the continuous data with high temporal resolution freely to the scientific community. For such data, many simple and low cost index methods has been developed to identify water bodies. In this study, we use these indices to detect the change of lakes in Pokhara city using Landsat data of 25 years gap. A model is developed in ArcGIS by differencing the water bodies derived form index methods and difference were calculated for positive and negative change. The result can be helpful in reclaiming and restoration of lake area, preserve and maintain the wetland ecosystem in the city.

  • Open access
  • 43 Reads
A low-cost environmental monitoring system: how to prevent systematic errors in the design phase through the combined use of Additive Manufacturing and thermographic techniques

The nEMoS (nano Environmental Monitoring System ) device is an all-in-one, low-cost, web-connected and 3D-printed device aimed at assessing the Indoor Environmental Quality (IEQ) of buildings. It is built using some low-cost sensors connected to an Arduino microcontroller board. The device is assembled in a small size case and the integrated air temperature and relative humidity sensor and the globe thermometer could be affected by thermal effect due to overheating of some nearby components. A thermographic analysis was made to rule out this possibility. The paper shows how the pervasive technique of Additive Manufacturing can be combined with the more traditional thermographic technique to redesign the case and to verify the accuracy of the optimized system in order to prevent instrumental systematic errors in terms of difference between experimental and the actual values of air temperature, relative humidity and radiant temperature. 

  • Open access
  • 30 Reads
Characterization of radio propagation channel in Urban Vehicle to Infrastructure environments to support WSNs

Vehicular ad hoc Networks (VANET’s) enable vehicles to communicate with each other (V2V) as well with roadside infrastructure units (V2I) and, nowadays the focus of the VANET’s shifts from research topics to preparing deployment. Various research projects V2V and V2I, where the dynamic nature of vehicular traffic varies on both small and large scale, assume a full-fledge functional wireless communication channel as a prerequisite. Although there is a significant research effort in channel modelling, those works are mainly focused in V2V environments and not much work has been done for V2I.Empirical methods have been widely used to channel characterization for these type of environments. They are very rapid, but lacks of accuracy. On the other hand, deterministic methods are known in the literature to present very good agreement with the channel propagation phenomena in real world.  

This work aims to evaluate some important characteristics of a V2I wireless channel link in a road intersection, using a deterministic simulation model based on an in-house 3D Ray-Launching algorithm, validating its findings with real measurements obtained through the field-test using Wireless Sensor Networks (WSNs). Results of Path Loss, Power Delay Profile, Delay Spread, Doppler Shift and Doppler Spread will be presented. This results are very important in the deployment of a radio-planning V2I environment. Moreover, we will investigate a statistical analysis that let us the comparison with defined statistical models for large and small scale fading.  Finally, we will make some recommendations which will take into account the scope of the simulated scenario and its findings when is compared with the real measurements framework.

  • Open access
  • 56 Reads
Data-Driven Representation of Soft Deformable Objects From Force-Torque Sensor Data and 3D Vision Measurements

The realistic representation of deformations is still an active area of research, especially for soft objects whose behavior cannot be simply described in terms of elasticity parameters. Most of existing techniques assume that the elasticity parameters describing the object behavior are known a priori based on assumptions on the object material, such as its isotropy or homogeneity, or values for these parameters are chosen by manual tuning until the results seem plausible. This is a subjective process and cannot be employed where accuracy is expected. This paper proposes a data-driven neural-network-based model for capturing implicitly deformations of a soft object, without requiring any knowledge on the object material. Visual data, in form of 3D point clouds gathered by a Kinect sensor, is collected over an object while forces are exerted by means of the probing tip of a force-torque sensor. A novel approach advantageously combining distance-based clustering, stratified sampling and neural gas-tuned mesh simplification is then proposed to describe the particularities of the deformation. The compact representation of the object is denser in the region of the deformation (an average of 97% perceptual similarity with the collected data), while still preserving the object overall shape (71% similarity over the entire surface) and only using on average 30% of the number of vertices in the mesh.

  • Open access
  • 113 Reads
Sensor-based Smart Oven System to Enhance Cooking Safety

Sensor-based Smart Oven system to enhance cooking safety

Kitchen is the second place where the majority of domestic accidents occur, and in particular oven presents the most principal source of fire accidents in residence. Therefore, enabling kitchen safety is a major factor particularly, for ageing people independent living. The paper presents our sensor-based smart oven system that targets enhancing safety of ageing people while cooking. The system is based on our insightful study on cooking risks analysis and assessment that enables to determine the pertinent parameters to be monitored around oven in order to identify and measure risk situations while cooking. We also introduce in this paper a solution for sensors positioning and system integration in a real-world cooking environment. Furthermore, we present the results of sensors testing in real-world configurations and mainly focus on the results of our experimental study that leads us to select the appropriate sensors that constitute the basic building block of our smart oven system. The system is composed of sensor nodes to monitor events around oven, then the sensory data is transmitted to a computing unit. The system proactively reacts to hazards in order to prevent cooking associated risks.

Results. We identified three major risks during cooking that are: fire, burn or intoxication by gas/smoke. Following are a summary of the studied parameters: 1) Fire: we observed the Volatile Organic Compound (VOC) and Alcohol gases’ concentration in the cooking smoke. 2) Burn: for both, burn risk by splash of a hot liquid and by contacting hot objects, we observed the relative humidity, utensil temperature, burner temperature, and presence of object on burner. 3) Intoxication by gas/smoke: we observed the concentration of CO gas in the cooking smoke.

Smart-oven system overview. Our sensor-based smart oven system allows sensing contextual cooking activities and offering appropriate context-aware interventions. Determining a risk situation and the corresponding interventions are adaptable to user needs.

Our smart oven system is composed of three main modules: A) Contextual data acquisition module. B) Reasoning engine module that determines a risk situation and its severity level based on fuzzy logic. C) Interventions module that is responsible to manage risk situations and to trigger the appropriate interventions.

  1. Contextual data acquisition module.

The system is based on a smart environment infrastructure, especially sensors and actuators:

  • Sensors. Installed around oven to perform context acquisition. They allow the system to infer the situation during cooking, or detect changes in the surrounding environment (e.g., smoke, burner temperature, utensil temperature, and presence of a utensil on burner).
  • Actuators. Distributed in the residence to ubiquitously alert user of cooking risk situations. They provide feedback through screens, speakers, or flashing lights, and control appliances in the kitchen (such as switch off oven power). The actuators provide a wide range of possibilities for human-machine interaction including appropriate intervention for each detected risk situation, and an adapted reaction according to user needs.Reasoning engine module.

B. Reasoning engine module.

The reasoning engine is based on our risk detection algorithms for each type of risk: fire, burn by splash, burn by contacting hot objects, and intoxication by CO gas. We present in the paper our threshold-based risk detection algorithms with three risk levels that correspond to two threshold values of the input parameters for each risk. The reasoning engine manages the detection of risk situations and determines their severity levels according to the contextual information around oven. Our reasoning engine is based on fuzzy logic approach that enables to efficiently implement an adaptable and flexible reasoning engine and to improve the compromise between generating false alarms, and an early accurate detection of a risk situation, as well as its severity level management, in presence of multiple input parameters.

We briefly mention our fire risk detection algorithm:

Fire-Risk-Level = 0

If (concentration of VOC or concentration of Alcohol between 200 and 250 ppm) then

Fire-Risk-Level = 1


If (concentration of VOC or concentration of Alcohol > 250 ppm) then

Fire-Risk-Level = 2

End if

C. Interventions engine module.

The intervention engine module triggers the appropriate interventions based on our intervention protocol presented in the paper. The interventions are triggered according to the type of the detected risk and its severity level . For example, if the level of the detected risk is low then, the intervention engine ubiquitously notifies the user about the current detected risk through screens, appropriate lights, and speakers in the rooms of the residence and switches off the oven power. If the level of the risk does not diminish during a predetermined amount of time a second level of interventions will be triggered by the intervention engine. A second level of interventions enables to notify the user, and/or family members and/or caregivers through mobile phone, and messages.

  • Open access
  • 28 Reads
Surface-enhanced Raman spectroscopy study of commercial fruit juices

Surface-enhanced Raman Spectroscopy (SERS) is a vibrational spectroscopy holding potentials for a rapid evaluation of quality and composition of food industry products without any need of sample preparation. In fact, SERS combines the advantages of the Raman effect such as the high specificity (ability to identify a given molecular species in the presence of many other chemicals) with the use of nanosized metallic materials enabling Raman signal enhancement. Among many nanomaterials, gold nanoparticles (GNPs) and their colloidal dispersions have attracted great interest for SERS applications due to their unique properties of small size, large surface area to volume ratio, high reactivity to the living cells, stability over high temperatures. In this frame, a low-cost home-made nanosized substrate has been designed and used for the investigation of commercial fruit juices. The substrate is based on home-made 50-nm sized GNPs. The use of the designed substrate has allowed us to observe the SERS spectra of commercial juices featuring a low-level of Raman signal with a commercial micro-Raman apparatus. Thanks to the use of a wavelet denoising procedure and background subtraction spectra with clear features have been obtained. Their quantitative analysis has enabled to evidence the presence of juice components of great importance for the quality evaluation of the products, such as fructose and pectin. The overall inspection of the results has confirmed the potentialities of SERS in food industry especially because of the use of home-made substrates well-suited to be employed for the eventual on-line product evaluation.

  • Open access
  • 57 Reads
Tactile profile classification using a multimodal MEMs-based sensing module

Robots are expected to perform complex dexterous operations in a variety of applications such as health and elder care, manufacturing, or high-risk environments. In this context, the most important task is to handle objects, the first step being the ability to recognize objects and their properties by touch. This paper concentrates on the issue of surface recognition by monitoring the interaction between a tactile probe in contact with a surface. A sliding motion is performed by a robot finger (i.e. kinematic chain composed of 3 motors) carrying the tactile probe on its end.  The probe comprises a 9-DOF MEMs MARG (Magnetic, Angular Rate, and Gravity) sensor and deep MEMs pressure (barometer) sensor, both embedded in a flexible compliant structure. The sensors are placed such that, when the tip is rubbed over a surface, the MARG unit vibrates and the deep pressure sensor captures the overall normal force exerted. The tactile probe collects data over seven synthetic shapes (profiles). The proposed method to distinguish them, in frequency and time domain, consists of applying multiscale principal components analysis prior to the classification with a multilayer neural network. The achieved classification accuracy of 85.1% demonstrates the usefulness of traditional MEMs as tactile sensors embedded into flexible substrates.

1 2 3 4 5