Bruno P. Santos

eXtend collection tree protocol

In this work, we propose eXtend Collection Tree Protocol (XCTP), a routing protocol that is an extension of the Collection Tree Protocol (CTP). CTP is the de-facto standard collection routing protocol for Wireless Sensor Network (WSN). CTP creates a routing tree to transfer data from one or more sensors to a root (sink) node. But, CTP does not create the reverse path between the root node and sensor nodes. This reverse path is important, for example, for feedback commands or acknowledgment packets. XCTP enables communication in both ways: root to node and node to root. XCTP accomplishes this task by exploring the CTP control plane packets. XCTP requires low storage states and very low additional overhead in packets. With the reverse path, it is possible to implement reliable transport layer protocols for Wireless Sensor Network (WSN). Thus, we designed Transport Automatic Piggyback Protocol (TAP2), a transport protocol with Automatic Repeat-reQuest (ARQ) error-control on top of XCTP. We implemented these protocols on TinyOS and evaluated on TOSSIM. We compared XCTP with CTP, Routing Protocol for low-power and lossy networks (RPL), and Ad hoc On Demand Distance Vector (AODV) protocols. We conducted scalability and stress tests, evaluating them with different loads and number of nodes. Our results shows that XCTP is more reliable than CTP, delivering 100% of the packets. XCTP sends fewer control packets than RPL. XCTP is faster to recovery from network failures and also stores fewer states than AODV, thus being efficient and agile.

Matrix: Multihop Address Allocation and Dynamic Any-to-Any Routing for 6LoWPAN

Standard routing protocols for IPv6 over Low power Wireless Personal Area Networks (6LoWPAN) are mainly designed for data collection applications and work by establishing a tree-based network topology, which enables packets to be sent upwards, from the leaves to the root, adapting to dynamics of low-power communication links. The routing tables in such unidirectional networks are very simple and small since each node just needs to maintain the address of its parent in the tree, providing the best-quality route at every moment. In this work, we propose Matrix, a platform-independent routing protocol that utilizes the existing tree structure of the network to enable reliable and efficient any-to-any data traffic. Matrix uses hierarchical IPv6 address assignment in order to optimize routing table size, while preserving bidirectional routing. Moreover, it uses a local broadcast mechanism to forward messages to the right subtree when persistent node or link failures occur. We implemented Matrix on TinyOS and evaluated its performance both analytically and through simulations on TOSSIM. Our results show that the proposed protocol is superior to available protocols for 6LoWPAN, when it comes to any-to-any data communication, in terms of reliability, message efficiency, and memory footprint.

Towards intra-vehicular sensor data fusion

Urban mobility aspects have become a challenge with the constant growth of global population. In the same time, more data has become available, which allows new information technologies to improve the mobility systems, especially the transportation system. Thus, a low cost strategy to handle these issues, rises as a new concept named ITS - Intelligent Transportation Systems. These systems depend on various data types and sources, and aggregating it is an important task, which can be accomplished by performing heterogeneous data fusion. In this work, we conducted an exploratory analysis over real vehicular data to show for each listed data issues (i.e imperfection, correlation, inconsistencies, among others) which of them have been found in our data set. Indeed, we found out several issues in the data implying that they must be treated before fusion process. As future extensions of this work, we will apply heterogeneous data fusion techniques to enhance, for example vehicular mobility traces by adding contextual information such as traffic conditions and driver behavior.

Mobile Matrix: A Multihop Address Allocation and Any-to-Any Routing in Mobile 6LoWPAN

In this work, we present Mobile Matrix, a routing protocol for 6LoWPAN that uses hierarchical IPv6 address allocation to perform any-to-any routing and mobility management without changing a nodes IPv6 address. In this way, device mobility is transparent to the application level. The protocol has low memory footprint, adjustable control message overhead and achieves optimal routing path distortion. Moreover, it does not rely on any particular hardware for mobility detection, such as an accelerometer. Instead, it provides a passive mechanism to detect that a device has moved. We present analytic proofs for the computational complexity and efficiency of Mobile Matrix, as well as an evaluation of the protocol through simulations. Finally, we propose a new mobility model, to which we refer as cyclical random waypoint mobility model, that we use to simulate mobility scenarios, where communication is carried out in environments with limited mobility, such as 6LoWPANs deployed in office buildings, university campuses, concert halls or sports stadiums. Results show that μMatrix deliveries 3x times more packets than RPL for top-down traffic over high mobility scenario.

CGR: Centrality-based Green Routing for Low-Power and Lossy Networks

Abstract High throughput and energy are two important constraints of the Low-power and Lossy Network (L2N). We propose Centrality-based Green Routing for L2Ns (CGR) as a routing protocol that considers both centrality and energy to improve network performance and decrease power consumption. CGR is a collection routing protocol that combines the best features of both protocols Collection Tree Protocol (CTP) and Centrality Tree (CT). CGR uses centrality betweenness-based to choose intermediate nodes that can perform data fusion and employ Link Quality Estimation (LQE) to find routes of high throughput and delivery rate. The suitable combination of these techniques leads the protocol to improve the literature results in delivery rate, energy consumption, and time to deliver data. There is a trade-off by routing through central nodes and their power consumption. Thus we also propose a Policy-Aware algorithm to balance energy consumption and to increase the network lifetime.

Matrix: Multihop Address allocation and dynamic any-To-any Routing for 6LoWPAN

Abstract Standard routing protocols for IPv6 over Low power Wireless Personal Area Networks (6LoWPAN) are mainly designed for data collection applications and work by establishing a tree-based network topology, enables packets to be sent upwards, from the leaves to the root, adapting to dynamics of low-power communication links. In this work, we propose Matrix, a platform-independent routing protocol that utilizes the existing tree structure of the network to enable reliable and efficient any-to-any data traffic in 6LoWPAN. Matrix uses hierarchical IPv6 address assignment to optimize routing table size while preserving bidirectional routing. Moreover, it uses a local broadcast mechanism to forward messages to the right subtree when a persistent node or link failures occur. We implemented Matrix on TinyOS and evaluated its performance both analytically and through simulations on TOSSIM. Our results showed that the proposed protocol is superior to available protocols for 6LoWPAN when it comes to any-to-any data communication, concerning reliability, message efficiency, and memory footprint.

CRAL: A Centrality-Based and Energy Efficient Collection Protocol for Low Power and Lossy Networks

High throughput and energy are two important constraints of the Wireless Sensor Networks (WSNs). In this paper, we present the Centrality-based Routing Aware for L2Ns (CRAL)1-Low Drop and Fast Delivery to mitigate these problems. The routing protocols are centrality-based and employ link quality estimators Expected Transmission Count (ETX) and Expected Transmission Time (ETT) to find best paths in wireless links. The suitable combinations of these techniques lead the algorithms to improve the literature results in delivery rate, energy consumption, and time to delivery data packets. CRAL does this by building routing trees with high throughput and maintains low energy consumption. The simulation results show that CRAL is more reliable, efficient in energy consumption, robust, and favoring data fusion than Centrality Tree (CT) and Shortest Paths Tree (SPT).

Cultura organizacional e liderança

GESTAO E AVALIACAO DE POLITICAS SOCIAIS NO BRASIL. De Murilo Fahel e Jorge Alexandre Barbosa Neves (Orgs.). Belo Horizonte: PucMinas, 2007. 428 p. ISBN: 978-85-60778-17-1 Diante da pouca tradicao de estudos empiricamente orientados e de manejo de metodos quantitativos de analise no Brasil sobre o tema “gestao e avaliacao de politicas sociais”, este livro apresenta assuntos e praticas contemporâneas que busca atingir dois publicos interessados na esfera da administracao publica e gestao social: os tecnicos - profissionais e pesquisadores – e os academicos – universitarios e iniciantes no campo. Ambos os publicos encontrarao nesta obra um debate acerca de elementos importantes com forte interligacao com a realidade socio-politica do Brasil como: insercao social, transparencia, controle, territorio, participacao cidada, governanca local, controle, assistencia social, desigualdade, transferencia de recursos, capital social e descentralizacao. (Nota por Magnus Luiz Emmendoerfer – UFV). CULTURA ORGANIZACIONAL E LIDERANCA. De Edgar H. Schein. Sao Paulo: Editora Atlas, 2009. 441 p. ISBN 978-85-224-5497-6 Pioneiro do estudo em cultura organizacional, Edgar Schein, na traducao da 3a edicao americana de seu celebre livro, traz uma atualizacao de seu entendimento da cultura como conceito, tratando-a como uma abstracao, mas de suma importância para se entender o conflito intergrupal tambem em nivel organizacional. Aproximando-a de um tema caro aos estudos organizacionais, a lideranca, Schein mostra como os lideres criam a cultura na medida em que criam grupos, organizacoes; e como a cultura, uma vez criada define os criterios para a lideranca, determinado assim quem sera ou nao lider. Uma primorosa obra para aqueles voltados aos estudos organizacionais. (Nota por Bruno Santos – UFMG).