MambaLSTM: A Spatio-Temporal Framework for Enhanced Traffic Accident Risk Prediction

In traffic accident risk prediction, most studies overlook the extra noise that could be incorporated when fusing temporal features into spatial features, and some models struggle to capture global correlations among spatial regions. To address these challenges, we propose a novel traffic accident risk prediction framework named MambaLSTM. First, we develop a squeeze-and-excitation temporal feature fusion module to integrate temporal information without compromising spatio-temporal integrity. Second, we introduce a new patch embedding module for effectively capturing semantic relationships among spatially adjacent regions. Additionally, we introduce a Mamba block based on state-space models to model global spatial semantics in urban regions. Finally, we propose a MambaLSTM unit to efficiently capture long- and short-term temporal dependencies for identifying dynamic risk patterns. Extensive experiments on real-world datasets demonstrate the proposed model's superiority over state-of-the-art methods. The code is released at https://github.com/Zhenzovo/MambaLSTM.

Energy Efficiency Maximization in IRS-enabled Phase Cooperative PS-SWIPT based Self-sustainable IoT Network

Power splitting based simultaneous wireless information and power transfer (PS-SWIPT) appears to be a promising solution to support future self-sustainable Internet of Things (SS-IoT) networks. However, the performance of these networks is constrained by radio frequency signal strength and channel impairments. To address this challenge, intelligent reflecting surfaces (IRSs) are introduced in PS-SWIPT based SS-IoT networks to improve network efficiency by controlling signal reflections. In this article, an IRS-enabled phase cooperative framework is proposed to improve energy efficiency (EE) of the IoT network $({\mathtt {I}}^{net})$ using phase shifts of the user network $({\mathtt {U}^{net})}$, without constraining hardware resources at ${\mathtt {U}^{net}}$. We exploit transmit beamforming (BF) at access points (APs) and phase shifts optimization at the IRS end with phase effective cooperation between APs to enhance ${\mathtt {I}}^{net}$ EE performance. The maximization problem turns out to be NP-hard, so first, an alternating optimization (AO) is solved for the ${\mathtt {U}^{net}}$ using low computational complexity heuristic BF approaches, namely, transmit minimum-mean-square-error and zero-forcing BF, and phase optimization is performed using semidefinite relaxation (SDR) approach. To combat the computational complexity of AO, we also propose an alternative solution by exploiting heuristic BF schemes and an iterative algorithm, i.e., the element-wise block-coordinate descent method for phase shifts optimization. Next, EE maximization is solved for the ${\mathtt {I}^{net}}$ by optimizing the PS ratio and active BF vectors by exploiting optimal phase shifts of the ${\mathtt {U}}^{net}$. Simulation results confirm that employing IRS phase cooperation in PS-SWIPT based SS-IoT networks can significantly improve EE performance of ${\mathtt {I}^{net}}$ without constraining resources.

Lightweight Attribute Localizing Models for Pedestrian Attribute Recognition

Pedestrian Attribute Recognition (PAR) deals with the problem of identifying features in a pedestrian image. It has found interesting applications in person retrieval, suspect re-identification and soft biometrics. In the past few years, several Deep Neural Networks (DNNs) have been designed to solve the task; however, the developed DNNs predominantly suffer from over-parameterization and high computational complexity. These problems hinder them from being exploited in resource-constrained embedded devices with limited memory and computational capacity. By reducing a network's layers using effective compression techniques, such as tensor decomposition, neural network compression is an effective method to tackle these problems. We propose novel Lightweight Attribute Localizing Models (LWALM) for Pedestrian Attribute Recognition (PAR). LWALM is a compressed neural network obtained after effective layer-wise compression of the Attribute Localization Model (ALM) using the Canonical Polyadic Decomposition with Error Preserving Correction (CPD-EPC) algorithm.