Honoriu Valean

Smart home automation system using Wi-Fi low power devices

Home automation is gaining popularity nowadays. A smart home automation system is based on ensuring security and making user life easier. It contains a large number of sensors which can control or monitor objects distributed in three-dimensional space. The sensors can be specialized in measuring temperature, humidity, pressure, light, noise, dust air, and so on. In this paper, a solution to transform a normal house in a smart house while reducing the energy consumption is proposed. This can be realized with the help of wireless sensor networks and of the LabVIEW™ graphical programming environment, which use the NI LabVIEW™ Statechart Module for collecting data from sensors.

On multi-user web conference using WebRTC

The paper discusses the potential of the WebRTC in a multi-user video conference. All the signaling for multi-user video conferences is sustained by the XMPP server. Currently this technology is being deployed on Google Chrome, Opera and Firefox web-browsers. Without a standardized solution, service providers can implement various types of architectures. In this paper we propose a multi-video web conference solution that works in other browsers along with Firefox, Opera and Google Chrome. XMPP server is used as a signaling and transporting protocol.

Agent based system for home automation, monitoring and security

Collecting data from sensors, for various applications and scenarios, is becoming a norm. Each sensor is connected to an Access Point (AP), which forwards the data to a host computer. When an AP fails, it causes all sensors connected to it to be offline for a while. In this paper, an adaptive and fault-tolerant system that continues to function in presence of AP failures and sensor joins (churn) is proposed. As a case study, it is showed how the system can be used to automate a building. The APs have associated software agents that use Paxos protocol to allocate the sensors to APs under churn. Virtual redundancy is implemented via Paxos, thus when an AP fails, its role will be taken by another AP. Hence, it is achieved high reliability by using software agents with wireless sensors, and employing the Paxos protocol, without any component redundancy and with low implementation cost.

An Efficient Hybrid Soft Computing Approach to the Generalized Vehicle Routing Problem

The generalized vehicle routing problem (GVRP) is one of the challenging combinatorial optimization problems that finds a lot of practical applications. The GVRP is a natural extension of the classical vehicle routing problem (VRP) and it is an NP-hard optimization problem belonging to the class of generalized combinatorial optimization problems. The aim of this paper is to present a new approach to tackle this complex problem. Combining this approach with a genetic algorithm results an efficient hybrid soft computing technique for solving the generalized vehicle routing problem. The proposed algorithm is competitive with other heuristics published to date in both solution quality and computation time. The computational results for several benchmarks problems are reported and the results point out that our proposed algorithm is an appropriate method to explore the search space of this complex problem and leads to good solutions in a reasonable amount of time.

A novel intrusion detection method based on support vector machines

Security of computers and the networks that connect them is increasingly becoming of great significance. As an effect, building effective intrusion detection models with good accuracy and real-time performance are essential.

Learning classification rules with genetic algorithm

This paper aims to challenge the problem of finding accurate and relevant rules for the task of classification. The scope is to improve the accuracy, or at least to provide a comparable accuracy measure, for classification algorithms implemented so far. Because the task of classification must be as accurate as possible, the paper proposes a method based on genetic algorithms to enhance the speed and quality of classification. Thus, by using a genetic approach, there is a chance that the classification process will execute faster. A known fact is that genetic algorithms are well suited for the increase of performance.

Indoor localization based on Wi-Fi parameters influence

In indoor environments, received Wi-Fi signal strength is common used for human or objects localization. However, there are many factors which have an important influence on the signal propagation. In this paper, we analyze the parameters with the highest impact on the signal strength. We have determined the equations for distance calculation in which we have isolated the parameters with the highest effect on the signal. The experimental results show that, in a multipath propagation environment, the ratio between the receiver antenna gain and the system and environmental losses increases nonlinearly with the inverse level of the received signal. We have also tested a scenario in which the signal was measured in different directions in order to choose the proper method for indoor localization.

Agent based patient scheduling system

The continuous evolution of technology can be used to further improvements in our daily lives. Embedded smart technologies follow the same increasing trend of evolution. These technologies can greatly improve medical systems where, along with the services they provide, they can mimic intelligent behavior in hospitals. One such behavior allows doctors to be contacted in a short period of time in order to respond to the patients needs. This paper presents a system designed for this purpose, that uses embedded smart devices, tailored for hospitals, clinics, ambulatory services or treatment centers. The system drives to a patient scheduling scheme which can be used to store and distribute medical information according to its nature and importance. Emergencies are the main target of our system, due to the requirements of locating a relevant physician in a minimum period of time. In such cases, the proposed device can be used to locate a nearby doctor and notify him about the emergency within a few seconds.

Improving classification with Support Vector Machine for unbalanced data

A problem arises in data mining, when classifying unbalanced datasets using Support Vector Machines. Because of the uneven distribution and the soft margin of the classifier, the algorithm tries to improve the general accuracy of classifying a dataset, and in this process it might misclassify a lot of weakly represented classes, confusing their class instances as overshoot values that appear in the dataset, and thus ignoring them. This paper introduces the Enhancer, a new algorithm that improves the Cost-sensitive classification for Support Vector Machines, by multiplying in the training step the instances of the underrepresented classes. We have discovered that by oversampling the instances of the class of interest, we are helping the Support Vector Machine algorithm to overcome the soft margin. As an effect, it classifies better future instances of this class of interest.

Tracking and localization system using Android mobile phones

There are many situations which request a tracking and localization system. Some examples are car/persons tracking and localization in case of accidents or disasters. Due to the hardware and software characteristics of the smartphones, these devices are suitable to work as terminals for such a system. The goal of the paper is to present the design and implementation of a tracking and localization system based on Android phones, able to find persons in case of accident and give a set of necessary information for rescue. The system sends the GPS coordinates of the person, display the coordinates on a map and computes the shortest route to the accident site.