Juliano F. Kazienko

Secure Secret Key Distribution and Storage in Wireless Sensor Networks

Security in Wireless Sensor Networks (WSNs) is commonly bootstrapped by cryptographic key distribution among nodes. In this scenario, one of the greatest research challenges is to ensure secure keys storage by sensors due to their constrained resources and major exposure to tampering attempts. The main contribution of this work is to propose and analyze a mechanism for secure symmetric key distribution solving the stored key exposure problem. Additionally, a tampering attempt alarm for sensors is designed in order to isolate compromised nodes or keys from the network. Hence, this approach aims to increase the systems security against the tampering and compromising of sensor nodes.

On the performance of a secure storage mechanism for key distribution architectures in wireless sensor networks

Security in wireless sensor networks (WSNs) demands efficient key management schemes. In particular, one of the main research challenges is to ensure secure key storage by sensors due to their constrained resources and high exposure to tampering attempts. To address this issue, we have proposed SENSORLock, a secure key storage mechanism which can be applied to different key distribution architectures for WSNs. In this work, we evaluate SENSORLock through different study cases considering three key distribution architectures: TinySec, SPINS, and NCD. Our goal is to demonstrate its feasibility in large-scale sensor networks. Simulation results indicate that our mechanism introduces an average overhead of 1.9% in terms of end to end delay and provides a similar estimated power consumption compared to the aforementioned architectures. Hence, we argue that it is feasible to use SENSORLock (i) in large-scale sensor networks and (ii) under different key distribution schemes.

SENSORLock: a lightweight key management scheme for wireless sensor networks†

Security in wireless sensor networks demands an efficient key management scheme. As sensors typically operate unattended, it becomes quite important to ensure security to cryptographic keys stored in their memories. In this scenario, the development of lightweight encryption mechanisms is a challenge because of sensor-constrained resources. In this work, we present a mechanism tailored to sensor networks called SENSORLock applying it to a specific case. Our main contribution is to propose, analyze, and demonstrate the feasibility of SENSORLock for secure symmetric key distribution solving the stored key exposure problem. Analytical results demonstrate that this approach increases the systems security against the tampering of sensor nodes. Additionally, the mechanism is evaluated using simulation and practical experiments, using the TinyOS platform. Simulation results reveal that this scheme introduces very low processing overhead, in the order of nanoseconds, and an estimated power consumption quite similar to existing approaches. Besides, practical experiments indicate that the scheme can be deployed by off-the-shelf sensors, such as MicaZ and TelosB. Copyright

A Novel iOS m-Health Application to Assist the Hospital-Acquired Pneumonia Diagnosis and Treatment

For the treatment of Hospital-Acquired Pneumonia (HAP) is necessary accuracy in diagnosis, according to the guidelines of a hospital. Moreover, the diagnosis and treatment of HAP should be performed immediately. To assist in this process, this work presents the PneumoH, which consists in a mobile application for supporting to physicians in the HAPs diagnosis and treatment. In order to evaluate PneumoH, an acceptance test was accomplished in a high circulation public hospital, relying on the participation of an infectious diseases expert. Our preliminary evaluation of the PneumoH indicates that it is useful for clinical practice.

A secure architecture based on ubiquitous computing for medical records retrieval

Electronic health record systems are important tools employed for accessing and maintaining patient data, such as the history of hospitalizations and medical exams. Nowadays, physicians, nurses and hospital staff need fast and secure access to medical records avoiding bureaucracy for patient information retrieval and imprecision in patients data maintenance. Ubiquitous and pervasive computing can contribute to overcoming such challenges; however the device impersonation problem should be carefully addressed in this scenario. To deal with such problem, this paper presents a secure architecture based on ubiquitous and pervasive computing for medical records retrieval and maintenance. Such architecture relies on Near Field Communication (NFC) for message exchange between smartphones and tags. An authentication mechanism is presented and validated to ensure device authentication. Analytical results reveal that such mechanism is efficient in providing mutual authentication. Additionally, another important security properties are reached, as confidentiality, message anti-replay and device anti-tracking. Finally, as proof of concept, we present a medication delivery study case based on a developed prototype.

MobiCAP: A mobile application prototype for management of community-acquired pneumonia

This demonstration aims at presenting the MobiCAP prototype. MobiCAP is a mobile application system for management of Community-Acquired Pneumonia (CAP). The system supports experienced end beginner physicians to deal with the CAPs diagnosis process problem. Thus, MobiCAP enables physicians to handle a high amount of variables and take a quick and precise diagnosis based on rules described by the internal guidelines previously established by hospital physicians. MobiCAP was designed and developed in conjunction with specialist physicians from the Hospital Nossa Senhora da Conceicao (HNSC)-placed in Porto Alegre city-and based on HNSC internal guidelines for diagnosis and management of the CAP. The system was written in the Objective-C language using a development approach called Model View Controller (MVC). The application was implemented to run on the iOS platform, more specifically under iPhone type devices. As programming environment, it was used the Xcode 5.0.2 for the overall development of the application. In summary, three graphical interfaces correspond to the core functionality of the application, namely: (a) a graphical user interface for the risk stratification of pneumonia; (b) a graphical interface for the algorithm treatment; and (c) a graphical interface to the treatment according to etiology.