Mikko Nieminen

Multi-sensor Logical Decision Making in the Single Location Surveillance Point System

As demand for surveillance of physical locations increases, automated decision making software can help maintain the rising costs of human monitoring. The variety in different types of sensors is also growing, and making use of their consolidated data can improve the decision making process. Using the constructive research method, we aim to define a design of a surveillance systems decision making component that utilizes data fusion from multiple types of sensors. As a solution we present the Logical Decision Making Server (LDMS), used in the Single Location Surveillance Point (SLSP), a system designed for monitoring an indoors location. The decision making capabilities in the LDMS are based on user-configurable security rules, which allow security personnel to define threats based on current and recent event reports from any or all of the environment’s sensors. The LDMS has been successfully developed and integrated into an SLSP implementation.

Representing user definable rules for decision making in The Single Location Surveillance Point

The Single Location Surveillance Point (SLSP) is a multi-sensor surveillance system that supports human security personnel in monitoring a physical indoors location. The systems Logical Decision Making Server (LDMS) component features rule-based automated logical deduction capabilities used for detecting threats and reacting to them. We have constructed a representation format for user definable surveillance rules based on the Extensible Markup Language (XML), with the aim of providing a solution that is both human-readable and expressive enough to support the features of the SLSP system. The syntax is capable of representing complex logical rules consisting of multiple conditions and reactive actions. The design features a vocabulary for rule definition, specific to the capabilities of the SLSP system. A prototype rule set has been successfully developed as a part of an SLSP prototype implementation.

Sensor Data Collection of the Single Location Surveillance Point System

The single location surveillance point (SLSP) is a distributed multi-sensor surveillance software system. It contains an arbitrary amount of sensors that collect readings from a single location, which is the surveillance point. The SLSP system contains the following realized sensors: a fingerprint sensor, a video camera, an audio sensor, and a network analyzing monitor. The sensors are located in an indoor region. Each sensor automatically collects information from its environment. Each sensor automatically routes its crude sensor data to a session server, which handles the connections among the components. The session server conveys the crude sensor data to the logical decision making service. The logical decision making server (LDMS) automatically derives the situation at the surveillance point based on the received sensor data. The intention is to deduct the situation which is transpiring in the surveyed area based on the received crude data from the sensors. By deriving the situation of a surveyed area, the surveillance personnel may utilize refined information cogent to occurring events of the surveyed area. This branch of the SLSP intends to facilitate the collection of data from a surveillance point and decrement the amount of superfluous information and rendered to the surveillance personnel, by acquiring automatically sensor data and providing automatically derived information to the surveillance personnels end-device. The operability of the constructed prototype indicates that this endeavor is attained. The research is based on the constructive method of the related publications and technologies and the results are derived by the implemented branch of the SLSP system.

Detecting abnormal activities from multi-sensor surveillance systems

The main purpose of a surveillance system is to monitor valuable assets, such as office buildings and homes, and report any occurring security incidents. Sensor malfunctions or abnormal usages of the system are possible scenarios in a real life and a surveillance system with hundreds of sensors is creating a vast amount of data which is impossible to handle manually. This renders the fixing of these potential faults slow and expensive. This paper proposes a system which can analyse data received from a surveillance system. The proposed system will report abnormal activities, such as malfunctioning or dead sensors, abnormal usage, and abnormal events created by the surveillance system. The experimental evaluation is performed by using six cases describing different types of abnormal activity. The experiments indicate that the proposed system can effectively pinpoint faulty sensors and other abnormal activities. This will ease the task of the maintenance personnel to locate and fix possible problem in the surveillance system.

Active Service Quality Management in ASEMA system

One of the purposes of the Active Service Environment Management system (ASEMA) is to provide the best possible multimedia service experience to end users. This is realized by dynamically adapting to changes in service environment¿s configuration and used network infrastructure¿s quality of service (QoS) changes. The research problem of the paper is to find a solution for proactive approach to manage multimedia service performance in a dynamic service environment. The research is based on the constructive method by analyzing the related publications and experimentation with the selected technologies by implementing the solutions part of the ASEMA prototype. The prototype has profile negotiation functionality that dynamically updates ASEMA profiles according to dynamic changes in an end user¿s service environment. The quality of service management functionality in ASEMA monitor¿s actively underlying network¿s QoS level. It adjusts quality of video stream according to the changes in QoS levels of the used network infrastructure.

Robust architecture for distributed intelligence in an IP-based mobile wide-area surveillance system

Modern IP-based wide-area surveillance systems often build on networks of multi-modal, intelligent and mobile sensor units. Detection of complex events is performed on intelligent sensors and fusing input in the sensor units or centralized control room components. The domain of surveillance and public safety creates requirement for robustness and fault-tolerance. This article will present an automated intelligence architecture for mobile surveillance, which provides capabilities for combining on-board event detection in sensor units, centralized decision making on the server side, and automated exploitation of mobile surveillance unit positioning data. This architecture must be very reliable to provide services in the face of challenges such as natural disasters and fire, potentially damaging the infrastructure of the surveillance system. To increase its reliability and robustness, we study the introduction of a self-healing system into the architecture and examine the combined system’s operation in three case studies.

Integration of Root Cause Analysis into a Model-Based Testing Process of a Mobile Switching Server

Model-based testing of software has proved effective for automation of testing and efficient error discovery, by utilizing modeled behavior of the system under test and automated test case generation. One of the important challenges in model-based testing is locating the fundamental sources of encountered errors. A root cause analysis solution should be able to find the causes of errors among different components in a model-based testing process, while automating the analysis to eliminate daunting data-intensive manual work. We present a design for a Root Cause Analyzer (RCA) component, aimed at automated test analysis utilizing the outputs generated in a model-based testing process, and producing human and machine readable analysis reports. A prototype RCA implementation is integrated into a tool chain used for offline functional testing of a Mobile Switching Server (MSS) in cellular networks. The RCA prototype is shown to discover causes of errors encountered during testing and pinpointing them in several components of the testing environment. The RCA design also demonstrates potential for integration into other model-based testing processes.

Secure communication platform for distributed city-wide surveillance systems

Managing a modern IP-based city-wide surveillance system requires that sensors and actuators in the system are visible to multiple computation and co-ordination nodes deployed in the surveillance network. As current and future surveillance systems can often consist of multiple independently managed local-area networks utilizing network address translation, it is not guaranteed that the nodes can communicate with each other. One approach for overcoming this challenge is to utilize a dedicated proxy visible to all parties to mediate all communication between the nodes. The mediated data must be secure from end-to-end particularly in the context of surveillance systems due to the confidentiality of the data. In this paper a secure communication platform called dROS is presented. The platform enables secure end-to-end communication between nodes deployed in multiple networks.

Resilient and adaptive public-key infrastructure for distributed city-wide surveillance systems

Third generation city-wide distributed surveillance systems are built upon networks of sensors, actuators and computation nodes. In such large systems, different groups of nodes may belong to separate parties who want to control access to the resources provided by their nodes. The access control mechanism must verify the identities of nodes and check their authorization for accessing resources. The system must also support dynamic group reconfiguration mandated by factors such as changes in organizational structure or confidentiality rules and recover gracefully from security breaches. This paper presents a mechanism based on Public Key Infrastructure (PKI) and certificate hierarchies for fulfilling these requirements in a secure communication platform supporting both synchronous and asynchronous operations in a distributed surveillance system. Dynamic reconfiguration of groups and expulsion of compromised parts of the system is implemented by utilizing certificate revocation.

Unobtrusive stress detection on the basis of smartphone usage data

Stress has become an important health problem, but existing stress detectors are inconvenient in long-term real-life use because users either have to wear dedicated devices or expend notable interaction efforts in system adaptation to specifics of each person. Adaptation is necessary because individuals significantly differ in their perception of stress and stress responses, but typical adaptation employs supervised learning methods and hence requires fairly large sets of labelled data (i.e. information on whether each reporting period was stressful or not) from every user. To address these problems, we propose a novel unsupervised stress detector, based on using a smartphone as the only device and using discrete hidden Markov models (HMM) with maximum posterior marginal (MPM) decisions for analysis of phone data. Our detector requires neither additional hardware nor data labelling and hence is truly unobtrusive and suitable for lifelong use. Its accuracy was evaluated using two real-life datasets: in the first case, adaptation was based on very short (a few days) phone interaction histories of each individual, and in the second case—on longer histories. In these tests, the proposed HMM-MPM achieved 59 and 70% accuracies, respectively, which is comparable with results of fully supervised methods, reported by other works.