Spectral-Efficient LoRa with Low Complexity Detection

In this paper, we propose a spectral-efficient LoRa (SE-LoRa) modulation scheme with a low complexity successive interference cancellation (SIC)-based detector. The proposed communication scheme significantly improves the spectral efficiency of LoRa modulation, while achieving an acceptable error performance compared to conventional LoRa modulation, especially in higher spreading factor (SF) settings. We derive the joint maximum likelihood (ML) detection rule for the SE-LoRa transmission scheme that turns out to be of high computational complexity. To overcome this issue, and by exploiting the frequency-domain characteristics of the dechirped SE-LoRa signal, we propose a low complexity SIC-based detector with a computation complexity at the order of conventional LoRa detection. By computer simulations, we show that the proposed SE-LoRa with low complexity SIC-based detector can improve the spectral efficiency of LoRa modulation up to $445.45\%$, $1011.11\%$, and $1071.88\%$ for SF values of $7$, $9$, and $11$, respectively, while maintaining the error performance within less than $3$ dB of conventional LoRa at symbol error rate (SER) of $10^{-3}$ in Rician channel conditions.

Resource-efficient medical image classification for edge devices

Medical image classification is a critical task in healthcare, enabling accurate and timely diagnosis. However, deploying deep learning models on resource-constrained edge devices presents significant challenges due to computational and memory limitations. This research investigates a resource-efficient approach to medical image classification by employing model quantization techniques. Quantization reduces the precision of model parameters and activations, significantly lowering computational overhead and memory requirements without sacrificing classification accuracy. The study focuses on the optimization of quantization-aware training (QAT) and post-training quantization (PTQ) methods tailored for edge devices, analyzing their impact on model performance across medical imaging datasets. Experimental results demonstrate that quantized models achieve substantial reductions in model size and inference latency, enabling real-time processing on edge hardware while maintaining clinically acceptable diagnostic accuracy. This work provides a practical pathway for deploying AI-driven medical diagnostics in remote and resource-limited settings, enhancing the accessibility and scalability of healthcare technologies.

Quantized and Interpretable Learning Scheme for Deep Neural Networks in Classification Task

Deep learning techniques have proven highly effective in image classification, but their deployment in resourceconstrained environments remains challenging due to high computational demands. Furthermore, their interpretability is of high importance which demands even more available resources. In this work, we introduce an approach that combines saliency-guided training with quantization techniques to create an interpretable and resource-efficient model without compromising accuracy. We utilize Parameterized Clipping Activation (PACT) to perform quantization-aware training, specifically targeting activations and weights to optimize precision while minimizing resource usage. Concurrently, saliency-guided training is employed to enhance interpretability by iteratively masking features with low gradient values, leading to more focused and meaningful saliency maps. This training procedure helps in mitigating noisy gradients and yields models that provide clearer, more interpretable insights into their decision-making processes. To evaluate the impact of our approach, we conduct experiments using famous Convolutional Neural Networks (CNN) architecture on the MNIST and CIFAR-10 benchmark datasets as two popular datasets. We compare the saliency maps generated by standard and quantized models to assess the influence of quantization on both interpretability and classification accuracy. Our results demonstrate that the combined use of saliency-guided training and PACT-based quantization not only maintains classification performance but also produces models that are significantly more efficient and interpretable, making them suitable for deployment in resource-limited settings.

Coordinated Deep Neural Networks: A Versatile Edge Offloading Algorithm

As artificial intelligence (AI) applications continue to expand, there is a growing need for deep neural network (DNN) models. Although DNN models deployed at the edge are promising to provide AI as a service with low latency, their cooperation is yet to be explored. In this paper, we consider the DNN service providers share their computing resources as well as their models' parameters and allow other DNNs to offload their computations without mirroring. We propose a novel algorithm called coordinated DNNs on edge (\textbf{CoDE}) that facilitates coordination among DNN services by creating multi-task DNNs out of individual models. CoDE aims to find the optimal path that results in the lowest possible cost, where the cost reflects the inference delay, model accuracy, and local computation workload. With CoDE, DNN models can make new paths for inference by using their own or other models' parameters. We then evaluate the performance of CoDE through numerical experiments. The results demonstrate a $75\%$ reduction in the local service computation workload while degrading the accuracy by only $2\%$ and having the same inference time in a balanced load condition. Under heavy load, CoDE can further decrease the inference time by $30\%$ while the accuracy is reduced by only $4\%$.

A Tutorial on Chirp Spread Spectrum for LoRaWAN: Basics and Key Advances

Chirps spread spectrum (CSS) modulation is the heart of long-range (LoRa) modulation used in the context of long-range wide area network (LoRaWAN) in internet of things (IoT) scenarios. Despite being a proprietary technology owned by Semtech Corp., LoRa modulation has drawn much attention from the research and industry communities in recent years. However, to the best of our knowledge, a comprehensive tutorial, investigating the CSS modulation in the LoRaWAN application, is missing in the literature. Therefore, in the first part of this paper, we provide a thorough analysis and tutorial of CSS modulation modified by LoRa specifications, discussing various aspects such as signal generation, detection, error performance, and spectral characteristics. Moreover, a summary of key recent advances in the context of CSS modulation applications in IoT networks is presented in the second part of this paper under four main categories of transceiver configuration and design, data rate improvement, interference modeling, and synchronization algorithms.

D2D-aided LoRaWAN LR-FHSS in Direct-to-Satellite IoT Networks

In this paper, we present a device-to-device (D2D) transmission scheme for aiding long-range frequency hopping spread spectrum (LR-FHSS) LoRaWAN protocol with application in direct-to-satellite IoT networks. We consider a practical ground-to-satellite fading model, i.e. shadowed-Rice channel, and derive the outage performance of the LR-FHSS network. With the help of network coding, D2D-aided LR-FHSS transmission scheme is proposed to improve the network capacity for which a closed-form outage probability expression is also derived. The obtained analytical expressions for both LR-FHSS and D2D-aided LR-FHSS outage probabilities are validated by computer simulations for different parts of the analysis capturing the effects of noise, fading, unslotted ALOHA-based time scheduling, the receiver's capture effect, IoT device distributions, and distance from node to satellite. The total outage probability for the D2D-aided LR-FHSS shows a considerable increase of 249.9% and 150.1% in network capacity at a typical outage of 10^-2 for DR6 and DR5, respectively, when compared to LR-FHSS. This is obtained at the cost of minimum of one and maximum of two additional transmissions per each IoT end device imposed by the D2D scheme in each time-slot.