TY  - CONF
N2  - Compressive sampling (CS) has emerged as a promising approach for data compression in Internet-of-Things (IoT) applications, offering an alternative to conventional compression techniques. Existing literature shows a wide adoption of CS across diverse applications. Despite many works applied CS, many of these works did not explicitly mention the process of selecting an appropriate CS reconstruction technique. This work shows that different reconstruction technique will achieve different reconstruction accuracy and performance. We opined that many studies overlook the importance of comparative analysis when choosing reconstruction technique, potentially impacting the reconstruction accuracy and performance of IoT applications. In this work, various reconstruction techniques are tested, including Basis Pursuit (BP), Least Absolute Shrinkage and Selection Operator (LASSO), Matching Pursuit (MP), Orthogonal Matching Pursuit (OMP), Stagewise Orthogonal Matching Pursuit (StOMP), Compressive Sampling Matching Pursuit (CoSaMP) and Iterative Hard Thresholding (IHT), across different signal types to show their performance differences and identify the appropriate technique for each signal type. The experimental results show that different reconstruction techniques yield varying outcomes in terms of reconstruction accuracy and processing time, highlighting the importance of considering both reconstruction accuracy and processing time when choosing a reconstruction technique. © 2024 IEEE.
KW  - Data compression; Signal sampling; Compressive sampling; Internet-of-thing; Matching pursuit; Performance; Processing time; Reconstruction accuracy; Reconstruction techniques; Signal reconstruction technique; Signals reconstruction; Signal reconstruction
AV  - none
SN  - 9798331528553
ID  - scholars20420
A1  - Chew, Kexin
A1  - Ooi, Boonyaik Yaik
A1  - Kh'ng, Xin Yi
A1  - Beh, W. L.
SP  - 75 
PB  - Institute of Electrical and Electronics Engineers Inc.
N1  - Cited by: 0
EP  -  80
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-85209648938&doi=10.1109%2FAiDAS63860.2024.10730014&partnerID=40&md5=14030a6f25c0e09155e498b6ab756251
Y1  - 2024///
TI  - Adaptive Reconstruction Methods Selection Framework for Compressive Sampling
ER  -