In aerial base station (ABS) placement studies, leveraging information on topographic features for air-to-ground (A2G) channel analysis has been considered a promising approach. Recently, this approach has been studied in several works. However, these studies have predominantly focused on simplistic 3-D cuboid representations of topographic features, which do not adequately capture the complexity of real-world environments, thereby impeding accurate A2G channel analysis. In this article, we propose a more advanced strategy by generalizing the shapes and arrangements of topographic features to enable their realistic and accurate representations. Utilizing these generalized topographic models, we study the problem of finding the ABS location for coverage maximization. To solve this problem, we identify so-called line-of-sight (LoS) and non-LoS (NLoS) zones formed by these generalized topographic features through geometric analysis and obtain straightforward derivations of the coverage area for any given ABS location. Building upon this foundation, we develop a new ABS placement strategy, termed the polygonal feature-aware ABS placement algorithm (FA-poly). Our simulation results demonstrate that the proposed FA-poly significantly outperforms existing baseline methods in terms of the coverage area, as it accurately identifies channel conditions by leveraging information on generalized topographic features and obtains the coverage area based on them.

Optimizing deep neural networks (DNNs) running on resource-constrained devices, such as energy harvesting sensor devices, poses unique challenges due to the limited memory and varying energy conditions. Existing efforts have shown that deploying a multi-exit network mitigates the problem by allowing tradeoffs between accuracy and computational complexity. However, previous works did not fully consider two essential requirements: optimized neural architecture and optimized inference policy. In this paper, we present HarvNet, which comprises two complementary techniques for generating and operating a multi-exit network for energy harvesting devices. First, we provide a neural architecture search scheme, HarvNAS, which configures the best multi-exit architecture while meeting memory and energy constraints. Second, HarvSched learns and constructs the best progressive inference policy with different energy constraints by considering runtime factors, such as the harvesting status and the energy storage level. We implemented HarvNAS using the TensorFlow framework and then implemented and evaluated HarvSched on an MSP430-based sensor device. The evaluation showed that HarvNAS generated a model with up to 2.6%p higher accuracy while saving up to 70% of memory compared to the existing technique, and HarvSched enabled zero-downtime operation of the generated model.

The live streaming of omnidirectional video (ODV) on mobile devices demands considerable network resources; thus, current mobile networks are incapable of providing users with high-quality ODV equivalent to conventional flat videos. We observe that mobile devices, in fact, underutilize graphics processing units (GPUs) while processing ODVs; hence, we envisage an opportunity exists in exploiting video super-resolution (VSR) for improved ODV quality. However, the device-specific discrepancy in GPU capability and dynamic behavior of GPU frequency in mobile devices create a challenge in providing VSR-enhanced ODV streaming. In this paper, we propose OmniLive, an on-device VSR system for mobile ODV live streaming. OmniLive addresses the dynamicity of GPU capability with an anytime inference-based VSR technique called Omni SR. For Omni SR, we design a VSR deep neural network (DNN) model with multiple exits and an inference scheduler that decides on the exit of the model at runtime. OmniLive also solves the performance heterogeneity of mobile GPUs using the Omni neural architecture search (NAS) scheme. Omni NAS finds an appropriate DNN model for each mobile device with Omni SR-specific neural architecture search techniques. We implemented OmniLive as a fully functioning system encompassing a streaming server and Android application. The experiment results show that our anytime VSR model provides four times upscaled videos while saving up to 57.15% of inference time compared with the previous super-resolution model showing the lowest inference time on mobile devices. Moreover, OmniLive can maintain 30 frames per second while fully utilizing GPUs on various mobile devices.