arXiv subjects

L. Srikar Muppirisetty

Publications and source records attributed to L. Srikar Muppirisetty.

4 recordsLinked to original sources

arXiv2020

FuSSI-Net: Fusion of Spatio-temporal Skeletons for Intention Prediction Network

Pedestrian intention recognition is very important to develop robust and safe autonomous driving (AD) and advanced driver assistance systems (ADAS) functionalities for urban driving. In this work, we develop an end-to-end pedestrian intention framework that performs well on day- and night- time scenarios. Our framework relies on objection detection bounding boxes combined with skeletal features of human pose. We study early, late, and combined (early and late) fusion mechanisms to exploit the skeletal features and reduce false positives as well to improve the intention prediction performance. The early fusion mechanism results in AP of 0.89 and precision/recall of 0.79/0.89 for pedestrian intention classification. Furthermore, we propose three new metrics to properly evaluate the pedestrian intention systems. Under these new evaluation metrics for the intention prediction, the proposed end-to-end network offers accurate pedestrian intention up to half a second ahead of the actual risky maneuver.

cs.CV↗

arXiv2018

Proactive Resource Allocation with Predictable Channel Statistics

The behavior of users in relatively predictable, both in terms of the data they request and the wireless channels they observe. In this paper, we consider the statistics of such predictable patterns of the demand and channel jointly across multiple users, and develop a novel predictive resource allocation method. This method is shown to provide performance benefits over a reactive approach, which ignores these patterns and instead aims to satisfy the instantaneous demands, irrespective of cost to the system. In particular, we show that our proposed method is able to attain a novel fundamental bound on the achievable cost, as the service window grows. Through numerical evaluation, we gain insights into how different uncertainty sources affect the decisions and the cost.

cs.IT↗

arXiv2018

Location-Aided Pilot Contamination Avoidance for Massive MIMO Systems

Pilot contamination, defined as the interference during the channel estimation process due to reusing the same pilot sequences in neighboring cells, can severely degrade the performance of massive multiple-input multiple-output systems. In this paper, we propose a location-based approach to mitigating the pilot contamination problem for uplink multiple-input multiple-output systems. Our approach makes use of the approximate locations of mobile devices to provide good estimates of the channel statistics between the mobile devices and their corresponding base stations. Specifically, we aim at avoiding pilot contamination even when the number of base station antennas is not very large, and when multiple users from different cells, or even in the same cell, are assigned the same pilot sequence. First, we characterize a desired angular region of the target user at the serving base station based on the number of base station antennas and the location of the target user, and make the observation that in this region the interference is close to zero due to the spatial separability. Second, based on this observation, we propose pilot coordination methods for multi-user multi-cell scenarios to avoid pilot contamination. The numerical results indicate that the proposed pilot contamination avoidance schemes enhance the quality of the channel estimation and thereby improve the per-cell sum rate offered by target base stations.

cs.IT↗

arXiv2015

Spatial Wireless Channel Prediction under Location Uncertainty

Spatial wireless channel prediction is important for future wireless networks, and in particular for proactive resource allocation at different layers of the protocol stack. Various sources of uncertainty must be accounted for during modeling and to provide robust predictions. We investigate two channel prediction frameworks, classical Gaussian processes (cGP) and uncertain Gaussian processes (uGP), and analyze the impact of location uncertainty during learning/training and prediction/testing, for scenarios where measurements uncertainty are dominated by large-scale fading. We observe that cGP generally fails both in terms of learning the channel parameters and in predicting the channel in the presence of location uncertainties.\textcolor{blue}{{} }In contrast, uGP explicitly considers the location uncertainty. Using simulated data, we show that uGP is able to learn and predict the wireless channel.

cs.IT↗