On the performance of heterogeneous MANETs

Recent decades have witnessed the birth of major applications of wireless communication technology, further supported by the increasing capabilities of portable devices, low cost and ubiquitous presence. Despite radio technology diversity, a great deal of existing research focuses on a single and isolated wireless technology at a time, where homogeneous elements are identified by IP addresses. This work presents a heterogeneous technology routing (HTR) Framework, targeted towards scenarios where the heterogeneity of devices and networking technologies is present. Our contribution is many fold. It consists of a framework, which encompasses a process for bootstrapping networks, a routing protocol capable of dealing with multiple network interfaces, and a tuning with multipath extensions. We evaluate the performance of the bootstrap, routing and multipath mechanisms by way of simulation and an actual testbed implementation. The multipath evaluation simulates HTR networks with WiMAX, 3GPP LTE and Wi-Fi support. Results show that our proposal can effectively improve the data delivery ratio for ad-hoc networks and that it reduces the end-to-end delay without major impact on network energy consumption. As part of HTR tuning, we investigate next the impacts of tuning the HELLO refresh interval timer on route convergence and its subsequent energy consumption reduction during this phase. We also compare our tuned HTR with the widely used optimized link state routing protocol. Results show that varying the HELLO refresh interval can improve the convergence time and reduce the energy consumption without major impact on network behavior. Our proposal also includes a new distributed address allocation algorithm, namely, the dynamic node configuration protocol (DNCP). This paper conducts a comparative analysis between the Prime, Prophet and the DNCP schemes using static and dynamic topologies in terms of network setup time, energy consumption and control message overhead. Results show that the DNCP had a lower battery power consumption and less control message overhead while it slightly suffers with regard to setup.

     

Approaches to resource reservation for migrating real-time sessions in future mobile wireless networks

This paper presents two novel frameworks for session admission control and resource reservation in the context of next generation mobile and cellular networks. We also devised a special scheme that avoids per-user reservation signaling overhead in order to meet scalability requirements needed for next generation multi-access networks. The first proposal, Distributed Call Admission Control with Aggregate Resource Reservation (VR), uses mobility prediction based on mobile positioning system location information and takes into account the expected bandwidth to be used by calls handing off to and from neighboring cells within a configurable estimation time window. In conjunction, a novel concept called virtual reservation has been devised to prevent per-user reservation. Our second proposal, Local Call Admission Control and Time Series-based Resource Reservation, takes into account the expected bandwidth to be used by calls handed off from neighboring cells based only on local information stored into the current cell a user is seeking admission to. To this end, we suggest the use of two time series-based models for predicting handoff load: the Trigg and Leach (TL), which is an adaptive exponential smoothing technique, and Autoregressive Integrated Moving Average (ARIMA) that uses the Box and Jenkins methodology. It is worth to emphasize that the use of bandwidth prediction based on ARIMA technique still exist for wireless networks. The novelty of our approach is to build an adaptive framework based on ARIMA technique that takes into account the measured handoff dropping probability in order to tuning the prediction time window size so increasing the prediction accuracy. The proposed schemes are compared through simulations with the fixed guard channel (GC) and other optimized dynamic reservation-based proposals present in the literature. The results show that our schemes outperform many others and that the simpler local proposal based on TL can grant nearly similar levels of handoff dropping probability as compared to those from more the complex distributed approach.     

The murine immune response to the male-specific antigen mouse testicular cytochrome c

Male and female A/J mice were examined for their ability to elicit T lymphocyte and antibody (Ab) responses to the male-specific Ag, mouse testicular cytochrome c (Mt cyt). T lymphocytes from both male and female mice primed in vivo responded to the Ag in in vitro proliferation assays, and the dose-response curves were statistically indistinguishable. In addition, similar levels of Ab to Mt cyt were observed in immunized male and female mice. The B cells producing the Ab had switched isotypes to IgG1 and IgG2a, indicating that the self-reactive T helper (Th) cells in male mice were functional. Thus, male mice do not appear to be immunologically tolerant to Mt cyt, at least at the Th and B lymphocyte levels. No evidence for disease was found in male mice primed with Mt cyt. Major histocompatibility complex (MHC) class II-positive antigen-presenting cells are present in the testes and these were shown in vitro to process and present Mt cyt to a T cell hybridoma specific for the synthetic peptide Mt cyt 93-104. However, the hybridoma was not activated in the absence of exogenous Mt cyt 93-104 or Mt cyt, indicating that endogenous Mt cyt is not normally processed in sufficient quantity to effectively load MHC class II molecules with this particular Mt cyt-derived peptide. Notwithstanding any immunologic privilege of the testes, the lack of tolerance to Mt cyt and its failure to elicit an autoimmune disease could extend from the low levels of processed Mt cyt Ag available for T cell recognition. The T cell response elicited by Mt cyt contrasts the lack of response to mouse somatic cytochrome c which differs from Mt cyt at 13 amino acid residues and is expressed in most tissues and at higher levels.     

SPIDER: An availability-aware framework for the service function chain placement in distributed scenarios

The network function virtualization (NFV) paradigm replaces hardware-dependent network functions by virtual network functions (VNFs) that can be deployed in commodity hardware, including legacy servers. Consequently, the use of NFV is expected to reduce operating and capital expenses, as well as improve service deployment operation and management flexibility. For many use cases, the VNFs must be visited and invoked following a specific order of execution in order to compose a complete network service, named service function chain (SFC). Nonetheless, despite the benefits from NFV and SFC virtualization technologies, their introduction must not harm network performance and service availability. On the one hand, redundancy is seen by network service planners as a mechanism well established to combat availability issues. At same time, there is a goal to optimize resource utilization in order to reduce operational expenditure. In this article, we share our experience in the design use of a framework, named SPIDER, focused on SFC placement that considers the network infrastructure condition and the required SFC availability to define the placement strategy. The SPIDER monitors the status of infrastructure nodes and links and defines which servers the VNFs should be placed on and the number of redundant replicas needed. We present a proof-of-concept of SPIDER using Kubernetes to launch the VNFs as containers. We also use Kubernetes to forward the traffic between the VNFs, composing the service chain. We perform experiments to evaluate the runtime of SPIDER and the SFC delay under different network conditions.     

Gripper Design for Radio Base Station Autonomous Maintenance System

For an autonomous system to perform maintenance tasks in a networking device or a radio base station (RBS), it has to deal with a series of technological challenges ranging from identifying hardware-related problems to manipulating connectors. This paper describes the development of a robot maintenance system dedicated to detect and resolve faulty links caused by unplugged or poorly connected cables. Although the maintenance system relies on four subsystems, we significantly focus on our low-cost and efficient custom gripper solution developed to handle RJ45 Ethernet connectors. To examine our gripper, we conducted three experiments. First, a qualitative questionnaire was submitted to 30 users in the case of the teleoperated scenario of the gripper attached to a robotic arm. Similarly, we also tested the automatic operation mode. The results showed that our system is reliable and delivers a highly efficient maintenance tool in both teleoperated and autonomous operation modes. The practical experiment containing the removal or unplugging of connectors demonstrated our gripper’s ability to adequately handle these, whereas the feedback from the questionnaire pointed to a positive user experience. The automatic test assessed the gripper’s robustness against the continuous operation.

     

A methodology to assess the availability of next-generation data centers

The Journal of Supercomputing - Cloud data center providers benefit from software-defined infrastructure once it promotes flexibility, automation, and scalability. The new paradigm of...     

Nearest Neighbor Searches on the GPU

We introduce a GPU grid-based data structure for massively parallel nearest neighbor searches for dynamic point clouds. The implementation provides real-time performance and it is executed on GPU, both grid construction and nearest neighbors (approximate or exact) searches. This minimizes the memory transfer between device and system memories, improving overall performance. The proposed algorithm may be used across different applications with static and dynamic scenarios. Moreover, our data structure supports three-dimensional point clouds and given its dynamic nature, the user can change the data structure’s parameters at runtime. The same applies to the number of neighbors to be found. Performance comparisons were made against previous works, endorsing the benefits of our solution. Finally, we were able to develop a real-time Point-Based Rendering application for validation of the data structure. Its drawbacks and data distribution’s impact on performance were analysed and some directions for further investigation are given.