In the contemporary landscape of large language models (LLMs) development, it is crucial to address the challenges of deploying these models on hardware-constrained consumer electronic devices (CEDs), especially within complex dynamic task scheduling in multi-cloud environments (MCE). We propose a novel methodology leveraging a lightweight LLM to enhance task scheduling decisions in MCE.Our approach involves creating a task scheduling expert database informed by optimization objectives to fine-tune the lightweight LLM. This enables the model to generate a schedulable candidate set of tasks based on the current state of tasks and operational conditions within CEDs across MCE, optimizing scheduling decisions and enhancing overall efficiency. Simulations using both synthetic and real-world datasets demonstrate that our method outperforms three other algorithms in cost minimization, makespan reduction, and energy consumption. In summary, our methodology empowers CEDs to optimize the utilization of multi-cloud resources and harness the capabilities of lightweight LLMs to effectively minimize makespan, operational costs, and energy consumption during the task scheduling process, thereby facilitating efficient task scheduling.

Amid the ever-accelerated development of wireless communication technology, we have become increasingly demanding for location-based service; thus, passive indoor positioning has gained widespread attention. Channel State Information (CSI), as it can provide more detailed and fine-grained information, has been followed by researchers. Existing indoor positioning methods, however, are vulnerable to the environment and thus fail to fully reflect all the position features, due to limited accuracy of the fingerprint. As a solution, a CSI-based passive indoor positioning method was proposed, Wavelet Domain Denoising (WDD) was adopted to deal with the collected CSI amplitude, and the CSI phase information was unwound and transformed linearly in the offline phase. The post-processed amplitude and phase were taken as fingerprint data to build a fingerprint database, correlating with reference point position information. Results of experimental data analyzed under two different environments show that the present method boasts lower positioning error and higher stability than similar methods and can offer decimeter-level positioning accuracy.

In this paper, we study a tracking service vehicular edge computing (VEC) network that provides computation offloading service for Intelligent vehicles, where computational tasks with different urgency and dependency are required to be completed efficiently within strict time constraints. We consider the actual scenario where the environmental parameters fluctuate randomly and their distributions are unknown, thus, a longterm scheduling policy optimization problem needs to be solved. For this motivation, we first define a queueing criterion to sort the subtasks into a scheduling queue, and then model a specific Markov decision process (MDP) according to the scheduling queue. Furthermore, we propose our vehicular task scheduling policy optimizing (VTSPO) algorithm based on the most advanced policy-based deep reinforcement learning (DRL). The experimental results compared with known value-based DRL algorithms verify the advantages of the proposed VTSPO algorithm.

The device-free channel state information indoor fingerprint localization method may lead to phase offset errors, strong fingerprint noise and low sampling classification accuracy. In light of these characteristics, this article presents an indoor localization algorithm that is based on phase difference processing and principal component analysis. First, during the offline phase, this algorithm calculates phase differences to correct for random phase shifts and random time shifts in communication links. Second, the principal component analysis method is used to reduce the dimensionality of the denoised data and establish a robust fingerprint database. During the online phase, the algorithm trains a back-propagation neural network using the fingerprint data and determines the modelled mapping relationship between the fingerprint data and the physical localization after carrying out the phase difference correction and the principal component analysis–based dimensionality reduction. The experiments show that compared with existing fingerprint location methods, this algorithm has the advantages of significant denoising effectiveness and high localization accuracy.

Data stream classification is widely used in Internet of Things (IoT) scenarios such as health monitoring, anomaly detection and online diagnosis. Due to the continuous data stream changing dynamically over time, it is impossible to classify all the data simultaneously. Moreover, labeling each sample in practical data stream applications is time-and resource-consuming. The realistic situation is that only a few instances in a data stream are labeled. Therefore, classifying data streams with limited labels has become challenging in IoT scenarios. In this paper, we propose an incremental dynamic weighted semi-supervised method for classifying IoT data streams. Considering the dynamics and continuity in data streams, we use a chunk-based approach to learn the features in the data stream and assign weights to the classifier dynamically. Moreover, we deploy incremental learning methods to continuously learn from the sampled labeled data stream to update the classifier model, which can take advantage of newly incoming labeled data to improve learning performance. Experimental evaluations on seven IoT datasets show that the proposed method outperforms semi-supervised methods in accuracy, precision, and geometric mean (Gmean) by 10% and 5% over supervised methods, respectively.