We consider the problem of online and real-time registration of partial point clouds obtained from an unseen real-world rigid object without knowing its 3D model. The point cloud is partial as it is obtained by a depth sensor capturing only the visible part of the object from a certain viewpoint. It introduces two main challenges: 1) two partial point clouds do not fully overlap and 2) keypoints tend to be less reliable when the visible part of the object does not have salient local structures. To address these issues, we propose DeepPRO, a keypoint-free and an end-to-end trainable deep neural network. Its core idea is inspired by how humans align two point clouds: we can imagine how two point clouds will look like after the registration based on their shape. To realize the idea, DeepPRO has inputs of two partial point clouds and directly predicts the point-wise location of the aligned point cloud. By preserving the ordering of points during the prediction, we enjoy dense correspondences between input and predicted point clouds when inferring rigid transform parameters. We conduct extensive experiments on the real-world Linemod and synthetic ModelNet40 datasets. In addition, we collect and evaluate on the PRO1k dataset, a large-scale version of Linemod meant to test generalization to real-world scans. Results show that DeepPRO achieves the best accuracy against thirteen strong baseline methods, e.g., 2.2mm ADD on the Linemod dataset, while running 50 fps on mobile devices.

In triangulation-based coverage, a group of three mobile sensor nodes (MSNs) position themselves to form an equilateral triangle. A larger area is covered with many such smaller triangles as three MSNs move from one triangle to another. Such a scheme has several applications in localization, 3D imaging and coordinated search operation. In this work, we introduce an efficient mobile traversal algorithm (MTA) that provides a triangulation-based coverage of a field that can be approximated as a rectangle. We analyze the MTA in terms of the MSNs' travel distance and time taken to complete the traversal process. The bounds derived are also useful in determining the amount of energy consumption by the MSNs.

A novel approach to geo-locate sensor-nodes in a Ricean fading environment is presented. For localization, this work considers autocorrelation characteristics of short term fading that are typically averaged out in traditional signal-strength based algorithms. Location attributes of the sensor nodes are determined by minimizing the squared error between the estimated autocorrelation function and its theoretical approximation. The theoretical approximation of the autocorrelation function is derived in closed form from Aulin's three-dimensional channel model. It is shown that despite noisy measurements and weak line-of-sight signal, the proposed method estimates coordinates of sensor nodes within an error of less than plusmn5% for signal-to-noise ratios higher than 20 dB.

This work presents a novel algorithm to improve energy conservation in sensor networks. The algorithm is based upon selective activation of sensor nodes using thresholding (SAT). The sensor continuously monitors the received signal and makes a binary decision to join the network if the average received signal power falls within the specified minimum and maximum threshold range. Thus, SAT divides all the receivers within the coverage range of the transmitter into sets of active and inactive nodes realizing a saving in power consumption proportional to the number of inactive nodes. The sensor life-time is enhanced by allowing the node to transmit at a power that is a constant fraction of the total residual energy. For two cases of linear and hexagonal networks considered in this work, it is shown that for each transmission, the fraction of inactive nodes may exceed by over 30% of the total number of nodes present within maximum one-hop distance of the transmitter. The cost associated with energy conservation through SAT is (a) increased sensor node density or (b) increased transmission power requirement.

Outage events occur when the instantaneous signal-to-noise ratio or bit error rate fall into an unacceptable operational regime. Outage performance analysis is traditionally carried out for diverse channel behaviors and transmission techniques, leaving the identification of unacceptable operational regime to user specified performance requirements. An alternative approach is taken in this work and an acceptable combination of system parameters are identified such that the user specified quality of service (QoS) constraints are satisfied. The QoS constraints are specified jointly in terms of outage probability, outage duration and probability of occurrence of errors during the outage event. This work presents inter-dependencies between the QoS constraints and modulation, diversity and signal parameters in closed form. The analysis assumes identically, independently distributed Nakagami interferers using maximal ratio diversity combining (MRC)

This work presents an analytical approach for determining the relative coordinates of a receiver with respect to the source node. The source is assumed to be location aware. Statistics of the received signal are used to estimate the orientation and distance of the receiver with respect to the source. The received signal power is used as a distance estimator and the orientation is determined using the inphase and quadrature components of the received signal. The proposed method is applicable only in presence of line-of-sight path.

Outage is an event of instantaneous signal-to-noise ratio or bit error rate falling in an unacceptable operational regime. Outage performance analysis is traditionally carried out for diverse channel behaviors and transmission techniques, leaving the identification of unacceptable operational regime to user specified performance requirements. The paper takes an alternative approach and an acceptable combination of system parameters are identified such that the user specified quality of service (QoS) constraints are satisfied. The QoS constraints are specified jointly in terms of outage probability, outage duration and probability of occurrence of errors during the outage event. The paper presents inter-dependencies between the QoS constraints and modulation, diversity and signal parameters in closed form. The analysis assumes identically, independently distributed Nakagami interferers using maximal ratio diversity combining (MRC).