Santosh Nannuru

Radio-Frequency Tomography for Passive Indoor Multitarget Tracking

Radio-frequency (RF) tomography is the method of tracking targets using received signal-strength (RSS) measurements for RF transmissions between multiple sensor nodes. When the targets are near the line-of-sight path between two nodes, they are more likely to cause substantial attenuation or amplification of the RF signal. In this paper, we develop a measurement model for multitarget tracking using RF tomography in indoor environments and apply it successfully for tracking up to three targets. We compare several multitarget tracking algorithms and examine performance in the two scenarios when the number of targets is 1) known and constant, and 2) unknown and time varying. We demonstrate successful tracking for experimental data collected from sensor networks deployed in three different indoor environments posing different tracking challenges. For the fixed number of targets, the best algorithm achieves a root-mean-squared error tracking accuracy of approximately 0.3 m for a single target, 0.7 m for two targets and 0.8 m for three targets. Tracking using our proposed model is more accurate than tracking using previously proposed observation models; more importantly, the model does not require the same degree of training.

Multi-target device-free tracking using radio frequency tomography

This paper examines device-free tracking of multiple targets based on radio-frequency received signal strength (RSS) measurements recorded by a sensor network. We report experimental results that validate, for relatively uncluttered outdoor environments, a recently proposed model in which targets cause additive attenuation. Using this model, we examine the performance of three multi-target tracking algorithms using a experimental sensor network testbed consisting of 24 nodes that conducts surveillance of an outdoor area of size 50m2. The experiments are restricted to the case of a fixed number of targets (up to four). For four targets, all algorithms are able to track with average error less than 1m (as measured using the second-order OMAT metric); for two targets the error is close to 0.2m.

Computationally-Tractable Approximate PHD and CPHD Filters for Superpositional Sensors

In this paper we derive computationally-tractable approximations of the Probability Hypothesis Density (PHD) and Cardinalized Probability Hypothesis Density (CPHD) filters for superpositional sensors with Gaussian noise. We present implementations of the filters based on auxiliary particle filter approximations. As an example, we present simulation experiments that involve tracking multiple targets using acoustic amplitude sensors and a radio-frequency tomography sensor system. Our simulation study indicates that the CPHD filter provides promising tracking accuracy with reasonable computational requirements.

Multisnapshot Sparse Bayesian Learning for DOA

The directions of arrival (DOA) of plane waves are estimated from multisnapshot sensor array data using sparse Bayesian learning (SBL). The prior for the source amplitudes is assumed independent zero-mean complex Gaussian distributed with hyperparameters, the unknown variances (i.e., the source powers). For a complex Gaussian likelihood with hyperparameter, the unknown noise variance, the corresponding Gaussian posterior distribution is derived. The hyperparameters are automatically selected by maximizing the evidence and promoting sparse DOA estimates. The SBL scheme for DOA estimation is discussed and evaluated competitively against LASSO (ℓ1-regularization), conventional beamforming, and MUSIC.

General solution and approximate implementation of the multisensor multitarget CPHD filter

Random finite set (RFS) based filters such as the cardinalized probability hypothesis density (CPHD) filter have been successfully applied to the problem of single sensor multitarget tracking. Various multisensor extensions of these filters have been proposed in the literature, but exact update equations for the multisensor CPHD filter have not been identified. In this paper, we provide the update equations and propose an approximate implementation. The exact implementation of the multisensor CPHD filter is infeasible even for very simple scenarios. We develop an algorithm that greedily searches for the most likely groups of measurement subsets. This enables a computationally tractable implementation. Numerical simulations are performed to compare the proposed filter implementation with other random finite set based filters.

Particle filter implementation of the multi-Bernoulli filter for superpositional sensors

The multi-Bernoulli filter is a promising method for computationally efficient and accurate multi-target tracking. Computationally tractable approximations of the multi-Bernoulli filter equations for superpositional sensors were recently derived. In this paper we present a particle filter implementation of these approximate update filter equations. We describe how the filter could be employed to address the radio-frequency tomographic tracking task and conduct a simulation study to compare performance with the probability hypothesis density (PHD) and cardinalized probability hypothesis density (CPHD) filters.

Multisensor CPHD filter

The single-sensor probability hypothesis density (PHD) and cardinalized probability hypothesis density (CPHD) filters have been developed in the literature using the random finite set framework. The existing multisensor extensions of these filters have limitations such as sensor-order dependence, numerical instability, or high computational requirements. In this paper, we derive update equations for the multisensor CPHD filter. The multisensor PHD filter is derived as a special case. Exact implementation of the multisensor CPHD involves sums over all partitions of the measurements from different sensors and is thus intractable. We propose a computationally tractable approximation that combines a greedy measurement partitioning algorithm with the Gaussian mixture representation of the PHD. Our greedy approximation method allows the user to control the trade-off between computational overhead and approximation accuracy.

Hybrid multi-Bernoulli and CPHD filters for superpositional sensors

In this paper we present an approximate multi-Bernoulli filter and an approximate hybrid multi-Bernoulli cardinalized probability hypothesis density filter for superpositional sensors. The approximate-filter equations are derived by assuming that the predicted and posterior multitarget states have the same form and propagating the probability hypothesis density function for each independent component of the multitarget state. We examine the performance of the filters in a simulated acoustic sensor network and a radio frequency tomography application.

Multi-frequency sparse Bayesian learning for robust matched field processing

The multi-snapshot, multi-frequency sparse Bayesian learning (SBL) processor is derived and its performance compared to the Bartlett, minimum variance distortionless response, and white noise constraint processors for the matched field processing application. The two-source model and data scenario of interest includes realistic mismatch implemented in the form of array tilt and data snapshots not exactly corresponding to the range-depth grid of the replica vectors. Results demonstrate that SBL behaves similar to an adaptive processor when localizing a weaker source in the presence of a stronger source, is robust to mismatch, and exhibits improved localization performance when compared to the other processors. Unlike the basis or matching pursuit methods, SBL automatically determines sparsity and its solution can be interpreted as an ambiguity surface. Because of its computational efficiency and performance, SBL is practical for applications requiring adaptive and robust processing.

Hybrid multi-Bernoulli CPHD filter for superpositional sensors

We propose, for the super-positional sensor scenario, a hybrid between the multi-Bernoulli filter and the cardinalized probability hypothesis density (CPHD) filter. We use a multi-Bernoulli random finite set (RFS) to model existing targets and we use an independent and identically distributed cluster (IIDC) RFS to model newborn targets and targets with low probability of existence. Our main contributions are providing the update equations of the hybrid filter and identifying computationally tractable approximations. We achieve this by defining conditional probability hypothesis densities (PHDs), where the conditioning is on one of the targets having a specified state. The filter performs an approximate Bayes update of the conditional PHDs. In parallel, we perform a cardinality update of the IIDC RFS component in order to estimate the number of newborn targets. We provide an auxiliary particle filter based implementation of the proposed filter and compare it with CPHD and multi-Bernoulli filters in a simulated multitarget tracking application