Resource-constrained Wireless Research Articles

AdaptiveMesh: Adaptive Federate Learning for Resource-Constrained Wireless Environments

AdaptiveMesh: Adaptive Federate Learning for Resource-Constrained Wireless Environments

TSRP: A Novel Trust-Based and Energy-Aware Secure Routing Protocol for Resource-Constrained Wireless Sensor Networks

TSRP: A Novel Trust-Based and Energy-Aware Secure Routing Protocol for Resource-Constrained Wireless Sensor Networks

Theoretical and Practical Bounds on the Initial Value of Clock Skew Compensation Algorithm Immune to Floating-Point Precision Loss for Resource-Constrained Wireless Sensor Nodes

ABSTRACT We revisit our prior work on clock skew compensation immune to floating-point precision loss and provide practical as well as theoretical bounds on the initial value of the skew-compensated clock based on a systematic analysis of the errors of floating-point operations; by practical bounds, we mean the actual values of the theoretical bounds calculated at resource-constrained computing platforms like WSN nodes equipped with 32-bit single-precision floating-point format. Numerical examples demonstrate that the proposed practical bounds on single-precision floating-point format do not violate the theoretical bounds and thereby can guarantee the correctness of the clock skew compensation on resource-constrained computing platforms.

Federated Split Learning for Edge Intelligence in Resource-Constrained Wireless Networks

Federated Split Learning for Edge Intelligence in Resource-Constrained Wireless Networks

Agent Selection Framework for Federated Learning in Resource-Constrained Wireless Networks

Agent Selection Framework for Federated Learning in Resource-Constrained Wireless Networks

Federated Learning Over-the-Air by Retransmissions

Motivated by the increasing computational capabilities of wireless devices, as well as unprecedented levels of user- and device-generated data, new distributed machine learning (ML) methods have emerged. In the wireless community, Federated Learning (FL) is of particular interest due to its communication efficiency and its ability to deal with the problem of non-IID data. FL training can be accelerated by a wireless communication method called Over-the-Air Computation (AirComp) which harnesses the interference of simultaneous uplink transmissions to efficiently aggregate model updates. However, since AirComp utilizes analog communication, it introduces inevitable estimation errors. In this paper, we study the impact of such estimation errors on the convergence of FL and propose retransmissions as a method to improve FL accuracy over resource-constrained wireless networks. First, we derive the optimal AirComp power control scheme with retransmissions over static channels. Then, we investigate the performance of Over-the-Air FL with retransmissions and find two upper bounds on the FL loss function. Numerical results demonstrate that the power control scheme offers significant reductions in mean squared error. Additionally, we provide simulation results on MNIST classification with a deep neural network that reveals significant improvements in classification accuracy for low-SNR scenarios.

DISTRIBUTED ARTIFICIAL INTELLIGENCE MODELS IN PERSPECTIVE COMMUNICATION NETWORKS

6G technology will create an intelligent, highly scalable, dynamic and programmable wireless network capable of serving a variety of heterogeneous wireless devices. Various 6G modules and devices will generate colossal amounts of distributed data, so post-NGN (New Generation Networks) will need to implement a number of machine learning methods that will solve significantly complicated network problems. To overcome these problems, distributed learning methods can be used, allowing devices to train models jointly, without exchanging raw data, which reduces communication costs, delays, and increases data privacy level as well. Distributed machine learning models will play an important role in 6G networks, since they have a number of advantages over centralized methods, however, the implementation of distributed algorithms in resource-constrained wireless environments can be challenging. It is important to take into account the wireless environment uncertainty associated with various disturbing factors and limited wireless (transmission power, radio frequency spectrum) and hardware resources (computing power). Consequently, it is important to choose the suitable machine learning algorithm based on the wireless environment characteristics and the resource requirements of the learning process. The article reviews the application of distributed artificial intelligence models in new generation communication networks for resource management and data processing purposes. The general algorithms and approaches of distributed machine learning, applications, methods and models are described. The article analyzes the ways in which distributed artificial intelligence models can solve various problems in communication networks, including optimizing resource use and ensuring high performance and availability of network services.

GoMORE: Global Model Reuse for Resource-Constrained Wireless Federated Learning

Due to the dynamics of wireless channels and limited wireless resources (i.e., spectrum), deploying federated learning (FL) over wireless networks is challenged by frequent FL parameter transmission errors and incomplete FL model aggregation from devices. In order to overcome these challenges, we propose a Global MOdel REuse strategy (GoMORE) that reuses global FL models received at previous FL training iterations to replace erroneous local models caused by imperfect wireless transmissions. We analytically prove that the proposed GoMORE is strictly superior over the existing strategy, especially at low signal-to-noise ratios (SNRs). Based on the derived expression of weight divergence, we optimize the number of devices that participate in the model aggregation to maximize the FL performance with limited communication resources. Numerical results verify that the proposed GoMORE successfully approaches the performance upper bound by an ideal transmission. It also mitigates the negative impacts of non-independent and non-identically distributed (non-IID) data while achieving over 5 dB reduction in energy consumption.

Cryptographic Algorithms with Data Shorter than the Encryption Key, Based on LZW and Huffman Coding.

Modern, commonly used cryptosystems based on encryption keys require that the length of the stream of encrypted data is approximately the length of the key or longer. In practice, this approach unnecessarily complicates strong encryption of very short messages commonly used for example in ultra-low-power and resource-constrained wireless network sensor nodes based on microcontrollers (MCUs). In such cases, the data payload can be as short as a few bits of data while the typical length of the key is several hundred bits or more. The article proposes an idea of employing a complex of two algorithms, initially applied for data compression, acting as a standard-length encryption key algorithm to increase the transmission security of very short data sequences, even as short as one or a few bytes. In this article, we present and evaluate an approach that uses LZW and Huffman coding to achieve data transmission obfuscation and a basic level of security.

A novel self-adaptive multi-strategy artificial bee colony algorithm for coverage optimization in wireless sensor networks

A novel self-adaptive multi-strategy artificial bee colony algorithm for coverage optimization in wireless sensor networks

Collision-free output-feedback super-twisting control of robotic surface vehicles for coordinated path following with experiments

Collision-free output-feedback super-twisting control of robotic surface vehicles for coordinated path following with experiments

DRL-Driven Dynamic Resource Allocation for Task-Oriented Semantic Communication

Semantic communication has been regarded as a promising technology to serve upcoming intelligent applications. However, few studies have addressed the problem of resource allocation in semantic communication networks. Most resource allocation mechanisms act fairly to all original data, ignoring the meaning behind the transmitted bits. In this paper, a dynamic resource allocation scheme for the task-oriented semantic communication network (TOSCN) based on deep reinforcement learning (DRL) is proposed, which allows data with richer semantic information to preferentially occupy limited communication resources. This paper aims to design a deep deterministic policy gradient (DDPG) agent at the micro base station to maximize the long-term transmission efficiency of tasks. Firstly, the relationship between semantic information and task performance is investigated. Subsequently, a novel wireless resource allocation model for TOSCN is proposed by taking the image classification task as an example. Then, a joint optimization problem of the semantic compression ratio, transmit power, and bandwidth of each user is formulated. The agent is trained in an interactive learning environment to obtain a decent trade-off between the amount of data delivered to the receiver and the accuracy of intelligent tasks. Simulation results demonstrate that the proposed scheme achieves significant advantages in relieving communication pressure and improving task performance in resource-constrained wireless networks.

Progressive Feature Transmission for Split Classification at the Wireless Edge

We consider the scenario of inference at the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">wire-less edge</i> , in which devices are connected to an edge server and ask the server to carry out remote classification, that is, classify data samples available at edge devices. This requires the edge devices to upload high-dimensional features of samples over resource-constrained wireless channels, which creates a communication bottleneck. The conventional feature pruning solution would require the device to have access to the inference model, which is not available in the current split inference scenario. To address this issue, we propose the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">progressive feature transmission</i> (ProgressFTX) protocol, which minimizes the overhead by progressively transmitting features until a target confidence level is reached. A control policy is proposed to accelerate inference, comprising two key operations: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">importance-aware feature selection</i> at the server and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">transmission-termination control</i> . For the former, it is shown that selecting the most important features, characterized by the largest discriminant gains of the corresponding feature dimensions, achieves a sub-optimal performance. For the latter, the proposed policy is shown to exhibit a threshold structure. Specifically, the transmission is stopped when the incremental uncertainty reduction by further feature transmission is outweighed by its communication cost. The indices of the selected features and transmission decision are fed back to the device in each slot. The control policy is first derived for the tractable case of linear classification, and then extended to the more complex case of classification using a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">convolutional neural network</i> . Both Gaussian and fading channels are considered. Experimental results are obtained for both a statistical data model and a real dataset. It is shown that ProgressFTX can substantially reduce the communication latency compared to conventional feature pruning and random feature transmission strategies.

Distributed and Efficient Slot Assignment–Alignment Protocol for Resource-Constrained Wireless IoT Devices

Distributed and Efficient Slot Assignment–Alignment Protocol for Resource-Constrained Wireless IoT Devices

Reinforcement learning based flow and energy management in resource-constrained wireless networks

Reinforcement learning based flow and energy management in resource-constrained wireless networks

Energy – Efficient routing and scheduling using clustering in geographic routing protocol

In the resource-constrained wireless sensor network (WSN) geographic routing has been considered as an attractive method where it exploits the location data instead of global topology to transmit the data. The geographic routing protocol faces the routing issues when it is used by a heterogeneous device and utilizes high energy during the propagation of data. The lifespan of the sensor network depends on the efficiency of energy and capacity of the battery. Hence, successful data transmission, enrichment of battery capacity and energy utilization is necessary for WSN. To attain this requirements an effective change is made in the data transmission environment and network topology. In this paper proposed a dynamic cluster based duty cycle scheduling is initiated for the data transmission. The cluster-based scheduling and routing in geographic routing protocol (CSRGR) utilize the clustering mechanism which in turn reduces the consumption of energy and maximizes the throughput. The objective function of the proposed approach provides a scheduling and routing strategy. The demonstration of simulation results shows the effective cluster size balancing with data transmission range dynamically. The proposed algorithm is compared with the existing approach and from the results, the energy consumption is minimum for diverse scenarios.

Wireless edge device intelligent task offloading in mobile edge computing using hyper-heuristics

To overcome with the computation limitation of resource-constrained wireless IoT edge devices, providing an efficient task computation offloading and resource allocation in distributed mobile edge computing environment is consider as a challenging and promising solution. Hyper-heuristic in recent times is gaining popularity due to its general applicability of same solution to solve different types of problems. Hyper-heuristic is generally a heuristic method or framework which iteratively evaluates and chooses the best low-level heuristic, to solve different types of problems. In this paper, we try to solve wireless device task offloading in mobile edge computing, which is a non-convex and NP-Hard problem by using a proposed novel Hyper-Heuristic Framework using Stochastic Heuristic Selection (HHFSHS) using Contextual Multi-Armed Bandit (CMAB) with Epsilon-Decreasing strategy, considering two key Quality of Service (QoS) objectives computation time and energy consumption. These multiobjective criteria are modeled as single-objective optimization problem with the goal to minimize latency and energy consumption of wireless devices without losing the pareto optimality. Finally, evaluate its performance by comparing with other individual meta-heuristic algorithms.

Analysis of distributed ledger technologies for industrial manufacturing

In recent years, industrial manufacturing has undergone massive technological changes that embrace digitalization and automation towards the vision of intelligent manufacturing plants. With the aim of maximizing efficiency and profitability in production, an important goal is to enable flexible manufacturing, both, for the customer (desiring more individualized products) and for the manufacturer (to adjust to market demands). Manufacturing-as-a-service can support this through manufacturing plants that are used by different tenants who utilize the machines in the plant, which are offered by different providers. To enable such pay-per-use business models, Distributed Ledger Technology (DLT) is a viable option to establish decentralized trust and traceability. Thus, in this paper, we study potential DLT technologies for efficient and intelligent integration of DLT-based solutions in manufacturing environments. We propose a general framework to adapt DLT in manufacturing, and then we introduce the use case of shared manufacturing, which we utilize to study the communication and computation efficiency of selected DLTs in resource-constrained wireless IoT networks.

Situation-Aware Orchestration of Resource Allocation and Task Scheduling for Collaborative Rendering in IoT Visualization

Three dimensional rendering enabled IoT visualization provides an immersive operation view across large physical environments by contextually aggregating and visualizing numerous data streams from various systems. The massive resource demand to support real-time and high-quality rendering services can be fulfilled by collaborative rendering among resource-constrained wireless devices. To deliver reliable performance, one main challenge is to achieve reliable and sustainable collaboration in a dynamic IoT system with heterogeneous resource capacity and changing user intent. To overcome such issues, we propose a situation-aware orchestration mechanism of resource allocation and task scheduling. The proposed technique achieves objective-driven exploration of collaboration opportunity among heterogeneous resource by three steps: recognizing dynamic condition of resource and task, including resource reliability and computational demand; understanding the mutual impact of resource condition and task performance in the aspect of energy consumption and latency; precise alignment of resource capacity and task demands via a redundant task scheduling scheme. The proposed task scheduling problem is formulated as an optimization model with the objective of maximizing collaboration utility. A genetic algorithm (GA) with adaptive mating-distance is designed to tackle the NP-hard problem, which improves the optimal solution in simulation by approximately 25% and 30% compared to conventional GA and Greedy algorithm, respectively.

Optimal Model Placement and Online Model Splitting for Device-Edge Co-Inference

Device-edge co-inference opens up new possibilities for resource-constrained wireless devices (WDs) to execute deep neural network (DNN)-based applications with heavy computation workloads. In particular, the WD executes the first few layers of the DNN and sends the intermediate features to the edge server that processes the remaining layers of the DNN. By adapting the model splitting decision, there exists a tradeoff between local computation cost and communication overhead. In practice, the DNN model is re-trained and updated periodically at the edge server. Once the DNN parameters are regenerated, part of the updated model must be placed at the WD to facilitate on-device inference. In this paper, we study the joint optimization of the model placement and online model splitting decisions to minimize the energy-and-time cost of device-edge co-inference in presence of wireless channel fading. The problem is challenging because the model placement and model splitting decisions are strongly coupled, while involving two different time scales. We first tackle online model splitting by formulating an optimal stopping problem, where the finite horizon of the problem is determined by the model placement decision. In addition to deriving the optimal model splitting rule based on backward induction, we further investigate a simple one-stage look-ahead rule, for which we are able to obtain analytical expressions of the model splitting decision. The analysis is useful for us to efficiently optimize the model placement decision in a larger time scale. In particular, we obtain a closed-form model placement solution for the fully-connected multilayer perceptron with equal neurons. Simulation results validate the superior performance of the joint optimal model placement and splitting with various DNN structures.