Bayesian Nonparametric Modeling for Predicting Dynamic Dependencies in Multiple Object Tracking

The paper considers the problem of tracking an unknown and time-varying number of unlabeled moving objects using multiple unordered measurements with unknown association to the objects. The proposed tracking approach integrates Bayesian nonparametric modeling with Markov chain Monte Carlo methods to estimate the parameters of each object when present in the tracking scene. In particular, we adopt the dependent Dirichlet process (DDP) to learn the multiple object state prior by exploiting inherent dynamic dependencies in the state transition using the dynamic clustering property of the DDP. Using the DDP to draw the mixing measures, Dirichlet process mixtures are used to learn and assign each measurement to its associated object identity. The Bayesian posterior to estimate the target trajectories is efficiently implemented using a Gibbs sampler inference scheme. A second tracking approach is proposed that replaces the DDP with the dependent Pitman–Yor process in order to allow for a higher flexibility in clustering. The improved tracking performance of the new approaches is demonstrated by comparison to the generalized labeled multi-Bernoulli filter.

Characterizing several properties of high-dimensional random Apollonian networks

In this article, we investigate several properties of high-dimensional random Apollonian networks, including two types of degree profiles, the small-world effect (clustering property), sparsity and three distance-based metrics. The characterizations of the degree profiles are based on several rigorous mathematical and probabilistic methods, such as a two-dimensional mathematical induction, analytic combinatorics and Pólya urns, etc. The small-world property is uncovered by a well-developed measure—local clustering coefficient and the sparsity is assessed by a proposed Gini index. Finally, we look into three distance-based properties; they are total depth, diameter and Wiener index.

Data Exchange in Cluster Structure for Longevity of IoT

In the Internet of Things (IoT), the scope of wireless sensor nodes is extended to things deployed in a pervasive world. For various IoT service applications, things can gather and share their information with each other through self-decision-making. Therefore, we cannot apply the existing information aggregation methods of wireless sensor networks to the IoT environment, which aim to transmit the collected data to only a sink node or a central server. Moreover, since the existing methods involve all the sensor nodes in the process of data exchange, they can cause an increase in the network traffic, delay of data transmission, and amount of energy consumed by things. In this paper, we propose a clustering-property-based data exchange method for efficient energy consumption in IoT networks. First, the proposed method assigns properties to each thing according to the characteristics of the obtained data. Second, it constructs a cluster network considering the location of things and their energy consumption. Finally, the things in a cluster communicate with other things in a different cluster based on their properties. In the experiment, the proposed method exhibits a better performance than the existing method. Owing to the energy-saving effect, we demonstrate that the proposed method results in a more reliable network and improves the longevity of IoT networks.

Exponential Decay of Truncated Correlations for the Ising Model in any Dimension for all but the Critical Temperature

The truncated two-point function of the ferromagnetic Ising model on $${\mathbb {Z}}^d$$Zd ($$d\ge 3$$d≥3) in its pure phases is proven to decay exponentially fast throughout the ordered regime ($$\beta >\beta _c$$β>βc and $$h=0$$h=0). Together with the previously known results, this implies that the exponential clustering property holds throughout the model’s phase diagram except for the critical point: $$(\beta ,h) = (\beta _c,0)$$(β,h)=(βc,0).

Clustering of extremely red objects in the AKARI NEP-deep field

We study the clustering property of extremely red objects (EROs) using Canada–France–Hawaii Telescope (CFHT) surveys with 0.55 deg2 in the AKARI north ecliptic pole (NEP) deep field. EROs are selected by the color criterion of r′ − Ks > 3.66, which is equivalent to (R − Ks)Vega > 5. We conducted the clustering analysis for two magnitude-limited cases, Ks < 20.3 (N = 363) and Ks < 20.9 (N = 727), using two-point angular correlation represented by a single power-law function. By fixing a power-law (with 0.8), the correlation lengths of EROs with Ks < 20.3 and Ks < 20.9 are 9.10 ± 1.86 and 7.81 ± 1.21 h−1 Mpc, respectively. We find that bias factors of EROs with Ks < 20.3 and Ks < 20.9 are 3.19 ± 0.59 and 2.83 ± 0.40, respectively, revealing that EROs reside in dark matter halos heavier than $\sim 10^{13}\, M_{\odot }$. To investigate possible descendants of EROs with Ks < 20.9, we calculate how the bias for dark matter halos that host EROs evolves by accounting for mass growth of halos along the redshift. We find that halos hosting EROs evolve into halos hosting local massive galaxies with 2–$7\, L^{*}$. It suggests that passive EROs with Ks < 20.9 are likely to be progenitors of massive galaxies in the present universe. The comparison between passive EROs (pEROs) and star-forming EROs (sEROs) classified by near-infrared colors shows that pEROs seem to be connected with more massive local galaxies. By fitting spectral energy distributions (SEDs), we estimate active galactic nucleus (AGN) contribution for 68 sEROs which are selected in mid-IR bands. AGN contributions to the IR luminosity are less than $10\%$ except for six sEROs. At least in the IR-selected sEROs, the contribution of AGN seems to be not significant.