Edge Inference Research Articles

Self-aware collaborative edge inference with embedded devices for IIoT

Edge inference and other compute-intensive industrial Internet of Things (IIoT) applications suffer from a bad quality of experience due to the limited and heterogeneous computing and communication resources of embedded devices. To tackle these issues, we propose a model partitioning-based self-aware collaborative edge inference framework. Specifically, the device can adaptively adjust the local model inference scheme by sensing the available computing and communication resources of surrounding devices. When the inference latency requirement cannot be met by local computation, the model should be partitioned for collaborative computation on other devices to improve the inference efficiency. Furthermore, for two typical IIoT scenarios, i.e., bursting and stacking tasks, the latency-aware and throughput-aware collaborative inference algorithms are designed, respectively. Via jointly optimizing the partition layer and collaborative device selection, the optimal inference efficiency, characterized by minimum inference latency and maximum inference throughput, can be obtained. Finally, the performance of our proposal is validated through extensive simulations and tests conducted on 10 Raspberry Pi 4Bs using popular models. Specifically, in the case of two collaborative devices, our platform reaches up to 92.59% latency reduction for bursting tasks and 16.19× throughput growth for stacking tasks. In addition, the divergence between simulations and tests ranges from 1.64% to 9.56% for bursting tasks and from 3.24% to 11.24% for stacking tasks, which indicates that the theoretical performance analyses are solid. For the general case where the data privacy is not considered and the number of collaborative devices is optimally determined, up to 14.76× throughput speed up and 84.04% latency reduction can be obtained.

Coal-rock interface real-time recognition based on the improved YOLO detection and bilateral segmentation network

To improve the accuracy and efficiency of coal-rock interface recognition, this study proposes a model built on the real-time detection algorithm, you only look once (YOLO), and the lightweight bilateral segmentation network. Simultaneously, the regional similarity transformation function and dragonfly algorithm are introduced to enhance the quality of coal-rock images. The comparison with three other models demonstrates the superior edge inference performance of the proposed model, achieving a mean Average Precision (mAP) of 90.2 at the Intersection over Union (IoU) threshold of 0.50 (mAP50) and 81.4 across a range of IoU thresholds from 0.50 to 0.95 (mAP[50,95]). Furthermore, to maintain high accuracy and real-time recognition capabilities, the proposed model is optimized using the open visual inference and neural network optimization toolkit, resulting in a 144.97% increase in the mean frames per second. Experimental results on four actual coal faces confirm the efficacy of the proposed model, showing a better balance between accuracy and efficiency in coal-rock image recognition, which supports further advancements in coal mining intelligence.

Integrated Sensing and Communication for Edge Inference With End-to-End Multi-View Fusion

Integrated Sensing and Communication for Edge Inference With End-to-End Multi-View Fusion

Unveiling hidden connections in omics data via pyPARAGON: an integrative hybrid approach for disease network construction.

Network inference or reconstruction algorithms play an integral role in successfully analyzing and identifying causal relationships between omics hits for detecting dysregulated and altered signaling components in various contexts, encompassing disease states and drug perturbations. However, accurate representation of signaling networks and identification of context-specific interactions within sparse omics datasets in complex interactomes pose significant challenges in integrative approaches. To address these challenges, we present pyPARAGON (PAgeRAnk-flux on Graphlet-guided network for multi-Omic data integratioN), a novel tool that combines network propagation with graphlets. pyPARAGON enhances accuracy and minimizes the inclusion of nonspecific interactions in signaling networks by utilizing network rather than relying on pairwise connections among proteins. Through comprehensive evaluations on benchmark signaling pathways, we demonstrate that pyPARAGON outperforms state-of-the-art approaches in node propagation and edge inference. Furthermore, pyPARAGON exhibits promising performance in discovering cancer driver networks. Notably, we demonstrate its utility in network-based stratification of patient tumors by integrating phosphoproteomic data from 105 breast cancer tumors with the interactome and demonstrating tumor-specific signaling pathways. Overall, pyPARAGON is a novel tool for analyzing and integrating multi-omic data in the context of signaling networks. pyPARAGON is available at https://github.com/netlab-ku/pyPARAGON.

Structured segment rescaling with Gaussian processes for parameter efficient ConvNets

We introduce a novel mechanism for structured pruning on ConvNet blocks and channels. Our mechanism, Structured Segment Rescaling (SSR) down-samples a ConvNet’s dimensions using depth and width modifiers that respectively remove whole blocks and channels. SSR is a systematic approach for constructing ConvNets that can replace arbitrary design heuristics. The SSR modifiers rescale logical partitions (segments) of a ConvNet with grouped layers. Different modifiers on segments yield many different architectures with unique rescales for their blocks. This diversity of architectures is then systemically explored using a Gaussian Process (GP) that optimizes for modifiers that maintain accuracy and reduce parameters. We analyze SSR in the context of resource constrained environments using ResNets trained on the CIFAR datasets. An initial set of depth and width modifiers explore extreme rescales of ResNet segments, where we find up to 70% parameter reduction. The GP then generalizes on these initial rescales by being trained on them and then predicts the accuracy of other rescaled ConvNet given their segment modifiers. SSR produces over 105 ConvNets that can be trained selectively based on their GP predicted accuracy. The GP enabled SSR pushes compression to over 80% with minimal accuracy impact. While both depth and width modifiers can reduce parameters, we show reducing blocks is better for reducing latency with up to 80% faster ConvNets. Using our mechanism, we can efficiently customize ConvNets using their parameter-accuracy trade-offs. SSR only requires 101 GPU hours and modest engineering to yield efficient new ConvNets that can facilitate edge inference.

Holistic Energy Awareness and Robustness for Intelligent Drones

Drones represent a significant technological shift at the convergence of on-demand cyber-physical systems and edge intelligence. However, realizing their full potential necessitates managing the limited energy resources carefully. Prior work looks at factors such as battery characteristics, intelligent edge sensing considerations, planning, and robustness in isolation. But a global view of energy awareness that considers these factors and looks at various tradeoffs is essential. To this end, we present results from our detailed empirical study of battery charge-discharge characteristics and the impact of altitude and lighting on edge inference accuracy. Our energy models, derived from these observations, predict energy usage while performing various manoeuvres with an error of 5.6%, a 2.5X improvement over the state-of-the-art. Furthermore, we propose a holistic energy-aware multi-drone scheduling system that decreases the energy consumed by 21.14% and the mission times by 46.91% over state-of-the-art baselines. To achieve system robustness in the event of link or drone failure, we observe trends in Packet Delivery Ratio to propose a methodology to establish reliable communication between nodes. We release an open-source implementation of our system. Finally, we tie all of these pieces together using a people-counting case study.

Gradient-descent hardware-aware training and deployment for mixed-signal neuromorphic processors

Mixed-signal neuromorphic processors provide extremely low-power operation for edge inference workloads, taking advantage of sparse asynchronous computation within spiking neural networks (SNNs). However, deploying robust applications to these devices is complicated by limited controllability over analog hardware parameters, as well as unintended parameter and dynamical variations of analog circuits due to fabrication non-idealities. Here we demonstrate a novel methodology for offline training and deployment of SNNs to the mixed-signal neuromorphic processor DYNAP-SE2. Our methodology applies gradient-based training to a differentiable simulation of the mixed-signal device, coupled with an unsupervised weight quantization method to optimize the network’s parameters. Parameter noise injection during training provides robustness to the effects of quantization and device mismatch, making the method a promising candidate for real-world applications under hardware constraints and non-idealities. This work extends Rockpool, an open-source deep-learning library for SNNs, with support for accurate simulation of mixed-signal SNN dynamics. Our approach simplifies the development and deployment process for the neuromorphic community, making mixed-signal neuromorphic processors more accessible to researchers and developers.

6G Goal-Oriented Communications: How to Coexist with Legacy Systems?

6G will connect heterogeneous intelligent agents to make them natively operate complex cooperative tasks. When connecting intelligence, two main research questions arise to identify how artificial intelligence and machine learning models behave depending on (i) their input data quality, affected by errors induced by interference and additive noise during wireless communication; (ii) their contextual effectiveness and resilience to interpret and exploit the meaning behind the data. Both questions are within the realm of semantic and goal-oriented communications. With this paper, we investigate how to effectively share communication spectrum resources between a legacy communication system (i.e., data-oriented) and a new goal-oriented edge intelligence one. Specifically, we address the scenario of an enhanced Mobile Broadband (eMBB) service, i.e., a user uploading a video stream to a radio access point, interfering with an edge inference system, in which a user uploads images to a Mobile Edge Host that runs a classification task. Our objective is to achieve, through cooperation, the highest eMBB service data rate, subject to a targeted goal effectiveness of the edge inference service, namely the probability of confident inference on time. We first formalize a general definition of a goal in the context of wireless communications. This includes the goal effectiveness, (i.e., the goal achievability rate, or the probability of achieving the goal), as well as goal cost (i.e., the network resource consumption needed to achieve the goal with target effectiveness). We argue and show, through numerical evaluations, that communication reliability and goal effectiveness are not straightforwardly linked. Then, after a performance evaluation aiming to clarify the difference between communication performance and goal effectiveness, a long-term optimization problem is formulated and solved via Lyapunov stochastic network optimization tools to guarantee the desired target performance. Finally, our numerical results assess the advantages of the proposed optimization and the superiority of the goal-oriented strategy against baseline 5G-compliant legacy approaches, under both stationary and non-stationary communication (and computation) environments.

Learning low-rank latent mesoscale structures in networks

Researchers in many fields use networks to represent interactions between entities in complex systems. To study the large-scale behavior of complex systems, it is useful to examine mesoscale structures in networks as building blocks that influence such behavior. In this paper, we present an approach to describe low-rank mesoscale structures in networks. We find that many real-world networks possess a small set of latent motifs that effectively approximate most subgraphs at a fixed mesoscale. Such low-rank mesoscale structures allow one to reconstruct networks by approximating subgraphs of a network using combinations of latent motifs. Employing subgraph sampling and nonnegative matrix factorization enables the discovery of these latent motifs. The ability to encode and reconstruct networks using a small set of latent motifs has many applications in network analysis, including network comparison, network denoising, and edge inference.

ULEEN: A Novel Architecture for Ultra-low-energy Edge Neural Networks

‘‘Extreme edge” 1 devices, such as smart sensors, are a uniquely challenging environment for the deployment of machine learning. The tiny energy budgets of these devices lie beyond what is feasible for conventional deep neural networks, particularly in high-throughput scenarios, requiring us to rethink how we approach edge inference. In this work, we propose ULEEN, a model and FPGA-based accelerator architecture based on weightless neural networks (WNNs). WNNs eliminate energy-intensive arithmetic operations, instead using table lookups to perform computation, which makes them theoretically well-suited for edge inference. However, WNNs have historically suffered from poor accuracy and excessive memory usage. ULEEN incorporates algorithmic improvements and a novel training strategy inspired by binary neural networks (BNNs) to make significant strides in addressing these issues. We compare ULEEN against BNNs in software and hardware using the four MLPerf Tiny datasets and MNIST. Our FPGA implementations of ULEEN accomplish classification at 4.0–14.3 million inferences per second, improving area-normalized throughput by an average of 3.6× and steady-state energy efficiency by an average of 7.1× compared to the FPGA-based Xilinx FINN BNN inference platform. While ULEEN is not a universally applicable machine learning model, we demonstrate that it can be an excellent choice for certain applications in energy- and latency-critical edge environments.

TinyML Olive Fruit Variety Classification by Means of Convolutional Neural Networks on IoT Edge Devices

Machine learning (ML) within the edge internet of things (IoT) is instrumental in making significant shifts in various industrial domains, including smart farming. To increase the efficiency of farming operations and ensure ML accessibility for both small and large-scale farming, the need for a low-cost ML-enabled framework is more pressing. In this paper, we present an end-to-end solution that utilizes tiny ML (TinyML) for the low-cost adoption of ML in classification tasks with a focus on the post-harvest process of olive fruits. We performed dataset collection to build a dataset that consists of several varieties of olive fruits, with the aim of automating the classification and sorting of these fruits. We employed simple image segmentation techniques by means of morphological segmentation to create a dataset that consists of more than 16,500 individually labeled fruits. Then, a convolutional neural network (CNN) was trained on this dataset to classify the quality and category of the fruits, thereby enhancing the efficiency of the olive post-harvesting process. The goal of this study is to show the feasibility of compressing ML models into low-cost edge devices with computationally constrained settings for tasks like olive fruit classification. The trained CNN was efficiently compressed to fit into a low-cost edge controller, maintaining a small model size suitable for edge computing. The performance of this CNN model on the edge device, focusing on metrics like inference time and memory requirements, demonstrated its feasibility with an accuracy of classification of more than 97.0% and minimal edge inference delays ranging from 6 to 55 inferences per second. In summary, the results of this study present a framework that is feasible and efficient for compressing CNN models on edge devices, which can be utilized and expanded in many agricultural applications and also show the practical insights for implementing the used CNN architectures into edge IoT devices and show the trade-offs for employing them using TinyML.

InferLoc: Hypothesis-Based Joint Edge Inference and Localization in Sparse Sensor Networks

Ranging-based localization is a fundamental problem in the Internet of Things and unmanned aerial vehicle networks. However, the nodes’ limited-ranging scope and users’ broad coverage purpose inevitably cause network sparsity or subnetwork sparsity. The performances of existing localization algorithms are extremely unsatisfactory in sparse networks. A crucial way to deal with the sparsity is to exploit the hidden knowledge provided by the unmeasured edges, which inspires this work to propose a hypothesis-based Joint Edge Inference and Localization algorithm called InferLoc . InferLoc mines the Unmeasured but Inferable Edges (UIEs). Each UIE is an unmeasured edge, but it is restricted through other edges in the network to be inside a rigid component, so it has only a limited number of possible lengths. We propose an efficient method to detect UIEs and geometric approaches to infer possible lengths for UIEs in 2D and 3D networks. The inferred possible lengths of UIEs are then treated as multiple hypotheses to determine the node locations and the lengths of UIEs simultaneously through a joint graph optimization process. In the joint graph optimization model, to make the 0/1 decision variables for hypotheses selection differentiable, differentiable functions are proposed to relax the 0/1 selections, and rounding is applied to select the final length after the optimization converges. We also prove the condition when a UIE can contribute to sparse localization. Extensive experiments show remarkably better accuracy and efficiency performances of InferLoc than the state-of-the-art network localization algorithms. In particular, it reduces the localization errors by more than 90% and speeds up the convergence time more than 100 times than that of the widely used G2O-based methods in sparse networks.

A Comprehensive Review and a Taxonomy of Edge Machine Learning: Requirements, Paradigms, and Techniques

The union of Edge Computing (EC) and Artificial Intelligence (AI) has brought forward the Edge AI concept to provide intelligent solutions close to the end-user environment, for privacy preservation, low latency to real-time performance, and resource optimization. Machine Learning (ML), as the most advanced branch of AI in the past few years, has shown encouraging results and applications in the edge environment. Nevertheless, edge-powered ML solutions are more complex to realize due to the joint constraints from both edge computing and AI domains, and the corresponding solutions are expected to be efficient and adapted in technologies such as data processing, model compression, distributed inference, and advanced learning paradigms for Edge ML requirements. Despite the fact that a great deal of the attention garnered by Edge ML is gained in both the academic and industrial communities, we noticed the lack of a complete survey on existing Edge ML technologies to provide a common understanding of this concept. To tackle this, this paper aims at providing a comprehensive taxonomy and a systematic review of Edge ML techniques, focusing on the soft computing aspects of existing paradigms and techniques. We start by identifying the Edge ML requirements driven by the joint constraints. We then extensively survey more than twenty paradigms and techniques along with their representative work, covering two main parts: edge inference, and edge learning. In particular, we analyze how each technique fits into Edge ML by meeting a subset of the identified requirements. We also summarize Edge ML frameworks and open issues to shed light on future directions for Edge ML.

Model Inversion Attacks Against Graph Neural Networks

Many data mining tasks rely on graphs to model relational structures among individuals (nodes). Since relational data are often sensitive, there is an urgent need to evaluate the privacy risks in graph data. One famous privacy attack against data analysis models is the model inversion attack, which aims to infer sensitive data in the training dataset and leads to great privacy concerns. Despite its success in grid-like domains, directly applying model inversion attacks on non-grid domains such as graph leads to poor attack performance. This is mainly due to the failure to consider the unique properties of graphs. To bridge this gap, we conduct a systematic study on model inversion attacks against Graph Neural Networks (GNNs), one of the state-of-the-art graph analysis tools in this paper. Firstly, in the white-box setting where the attacker has full access to the target GNN model, we present GraphMI to infer the private training graph data. Specifically in GraphMI, a projected gradient module is proposed to tackle the discreteness of graph edges and preserve the sparsity and smoothness of graph features; a graph auto-encoder module is used to efficiently exploit graph topology, node attributes, and target model parameters for edge inference; a random sampling module can finally sample discrete edges. Furthermore, in the hard-label black-box setting where the attacker can only query the GNN API and receive the classification results, we propose two methods based on gradient estimation and reinforcement learning (RL-GraphMI). With the proposed methods, we study the connection between model inversion risk and edge influence and show that edges with greater influence are more likely to be recovered. Extensive experiments over several public datasets demonstrate the effectiveness of our methods. We also evaluate our attacks under two defenses: one is the well-designed differential private training, and the other is graph preprocessing. Our experimental results show that such defenses are not sufficiently effective and call for more advanced defenses against privacy attacks.

Predictive worker safety assessment through on-site correspondence using multi-layer fuzzy logic in outdoor construction environments

Construction sites remain highly perilous work environments globally, exposing employees to numerous hazards that can result in severe injuries or fatalities. To resolve this several solutions based on quantitative approaches have been developed. However the wide adoption of preexisting solutions is hindered by lack of accuracy. To this aim the development of an efficient fuzzy inference system has become a de-facto necessity. In this paper, we propose an edge inference framework based on multi-layered fuzzy logic for safety of construction workers. The proposed system employs an edge computing-based framework where IoT devices collect, store, and manage data to offer safety services. Multi-layer fuzzy logic is applied to infer the worker safety index based on rules that consist of construction environment factors. The multi-layer fuzzy logic is fed with weather, building and worker data collected from IoT nodes as inputs. The safety risk assessment process involves analyzing various factors. Weather information, such as temperature, humidity, and rainfall data, is considered to assess the risk to safety. The condition of the building is evaluated by analyzing load, strain, and inclination data. Additionally, the safety risk to workers is analyzed by taking into account their heart rate and location information. The initial layer's outputs are utilized as inputs for the subsequent layer, where an integrated safety index is inferred. Ultimately, the safety index is generated as the final outcome. The system's results are conveyed through warnings and an error measurement on a safety scale ranging from 1 to 10. Furthermore, web service is developed to allow the construction management to check the worker safety condition of the construction site in real-time, while also monitoring the operational status of the IoT devices, allowing for the early detection of sensor malfunction and the subsequent guarantee of worker safety. Extensive evaluations conducted to test the performance of the developed framework verify its efficiency to provide improved risk assessment, real-time monitoring, and proactive safety actions, encouraging a safer and more productive work environment.

Robust Information Bottleneck for Task-Oriented Communication With Digital Modulation

Task-oriented communications, mostly using learning-based joint source-channel coding (JSCC), aim to design a communication-efficient edge inference system by transmitting task-relevant information to the receiver. However, only transmitting task-relevant information without introducing any redundancy may cause robustness issues in learning due to the channel variations, and the JSCC which directly maps the source data into continuous channel input symbols poses compatibility issues on existing digital communication systems. In this paper, we address these two issues by first investigating the inherent tradeoff between the informativeness of the encoded representations and the robustness to information distortion in the received representations, and then propose a task-oriented communication scheme with digital modulation, named discrete task-oriented JSCC (DT-JSCC), where the transmitter encodes the features into a discrete representation and transmits it to the receiver with the digital modulation scheme. In the DT-JSCC scheme, we develop a robust encoding framework, named robust information bottleneck (RIB), to improve the communication robustness to the channel variations, and derive a tractable variational upper bound of the RIB objective function using the variational approximation to overcome the computational intractability of mutual information. The experimental results demonstrate that the proposed DT-JSCC achieves better inference performance than the baseline methods with low communication latency, and exhibits robustness to channel variations due to the applied RIB framework.

Energy-Efficient Approximate Edge Inference Systems

The rapid proliferation of the Internet of Things and the dramatic resurgence of artificial intelligence based application workloads have led to immense interest in performing inference on energy-constrained edge devices. Approximate computing (a design paradigm that trades off a small degradation in application quality for disproportionate energy savings) is a promising technique to enable energy-efficient inference at the edge. This article introduces the concept of an approximate edge inference system ( AxIS ) and proposes a systematic methodology to perform joint approximations between different subsystems in a deep neural network (DNN)-based edge inference system, leading to significant energy benefits compared to approximating individual subsystems in isolation. We use a smart camera system that executes various DNN-based image classification and object detection applications to illustrate how the sensor, memory, compute, and communication subsystems can all be approximated synergistically. We demonstrate our proposed methodology using two variants of a smart camera system: (a) Cam Edge , where the DNN is executed locally on the edge device, and (b) Cam Cloud , where the edge device sends the captured image to a remote cloud server that executes the DNN. We have prototyped such an approximate inference system using an Intel Stratix IV GX-based Terasic TR4-230 FPGA development board. Experimental results obtained using six large DNNs and four compact DNNs running image classification applications demonstrate significant energy savings (≈ 1.6× -4.7× for large DNNs and ≈ 1.5× -3.6× for small DNNs), for minimal (<1%) loss in application-level quality. Furthermore, results using four object detection DNNs exhibit energy savings of ≈ 1.5× -5.2× for similar quality loss. Compared to approximating a single subsystem in isolation, AxIS achieves 1.05× -3.25× gains in energy savings for image classification and 1.35× -4.2× gains for object detection on average, for minimal (<1%) application-level quality loss.

A 2-Transistor-2-Capacitor Ferroelectric Edge Compute-in-Memory Scheme With Disturb-Free Inference and High Endurance

This paper proposes C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> FeRAM, a 2T2C/cell ferroelectric compute-in-memory (CiM) scheme for energy-efficient and high-reliability edge inference and transfer learning. With certain area overhead, C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> FeRAM achieves the following highlights: (i) compared with FeFET/FeMFET, it achieves disturb-free CiM and much higher write endurance (equal to FeRAM), leading to >100x inference time with <1% accuracy drop for VGG8 in CIFAR-10 dataset, along with the enhanced endurance for weight updates, e.g., CiM-based transfer learning; (ii) compared with 1T1C FeRAM inference cache, the achieved disturb-free feature and CiM capability in C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> FeRAM lead to improvements of 4x energy, 200x speed, and 3.2e5x life cycles. Such benefits highlight an intriguing solution for future intelligent edge AI.

Understanding Node Localizability in Barycentric Linear Localization

The barycentric linear localization (BLL) methods provide a lightweight, distributed way to calculate locations for resource-limited IoT devices. A crucial requirement for BLL is that the nodes participating in the iterative location propagation are localizable. Otherwise, the unlocalizable nodes will continuously pose error information in the location propagation process, making even the theoretically localizable nodes converge to the wrong locations. However, the research on node localizability in BLL is much lacked, greatly limiting the application scope of BLL. In specific, BLL node localizability is detected on a generated graph <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal{G^A}$</tex-math> </inline-formula> . For any node, its neighbors appear in <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal{G^A}$</tex-math> </inline-formula> only when the neighbors can form triangle(s), so that <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal{G^A}$</tex-math> </inline-formula> is much sparser than the original <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal G$</tex-math> </inline-formula> . Thus, the node localizability condition in BLL is harder to be satisfied than that in traditional localization methods. Moreover, the distributed algorithm to detect BLL localizable nodes is still open. This paper thoroughly investigates the node localizability conditions and distributed localizable node detection algorithms in BLL. At first, an efficient and fully distributed Negative Edge Inference (NEI) algorithm is proposed for each node to infer implicit edges in its neighborhood. NEI strengthens the distance graph by revealing more distance constraints so that enables more neighboring triangles. Then a new sufficient condition, i.e., the recursive three disjoint path condition (Recursive-3DP) on the strengthened distance graph is proposed to identify BLL localizable nodes much more accurately. Secondly, a distributed Path Extension and Pruning (PEP) algorithm is proposed for distributed localizable node detection. PEP is proved to detect all the theoretically Recursive-3DP nodes in the strengthened distance graph. A Fast-PEP algorithm is further proposed, which misses very limited Recursive-3DP nodes while bringing significant improvement in efficiency. PEP and Fast-PEP guarantee to identify BLL localizable nodes in <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2H$</tex-math> </inline-formula> rounds, where <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$H$</tex-math> </inline-formula> is the maximum hop number of the node disjoint paths. Finally, by using NEI and PEP (Fast-PEP), a localizability-aware BLL (LABEL) method is proposed, which correctly identifies localizable nodes and guarantees their correct location convergence. Extensive analysis and experiments show the advantages in localizability and location accuracy of the proposed schemes over the state-of-the-art methods.

Simulation of a Fully Digital Computing-in-Memory for Non-Volatile Memory for Artificial Intelligence Edge Applications.

In recent years, digital computing in memory (CIM) has been an efficient and high-performance solution in artificial intelligence (AI) edge inference. Nevertheless, digital CIM based on non-volatile memory (NVM) is less discussed for the sophisticated intrinsic physical and electrical behavior of non-volatile devices. In this paper, we propose a fully digital non-volatile CIM (DNV-CIM) macro with compressed coding look-up table (LUT) multiplier (CCLUTM) using the 40 nm technology, which is highly compatible with the standard commodity NOR Flash memory. We also provide a continuous accumulation scheme for machine learning applications. When applied to a modified ResNet18 network trained under the CIFAR-10 dataset, the simulations indicate that the proposed CCLUTM-based DNV-CIM can achieve a peak energy efficiency of 75.18 TOPS/W with 4-bit multiplication and accumulation (MAC) operations.