Optimization and Visual Analytics Research Group, Faculty of Computing and Informatics, Universiti Malaysia Sabah, Labuan International Campus, Labuan, Malaysia
Auto-generated profile, based on publicly available information
Say Leng Goh
Optimization and Visual Analytics Research Group, Faculty of Computing and Informatics, Universiti Malaysia Sabah, Labuan International Campus, Labuan, Malaysia
A comprehensive review is presented on the latest approaches to solutions focusing on multi-hop and mesh LoRa networks through the evaluation of simulations and real-world experiments, based on papers published between 2015 and 2023. The approaches are systematically classified into four (4) aspects: energy awareness, concurrent access and duty cycle regulations, routing protocols, and security. The first aspect encompasses parameter selection, distance, and clustering. The second aspect focuses on network segregation and time synchronisation. The third aspect covers proactive, reactive, and hybrid routing protocols. The aspect of security includes privacy and Denial-of-Service (DoS). Findings show a consensus that multi-hop networks and adherence to radio duty cycle are able to lower average power consumption. Concurrent access method can improve network performance, but more research is needed due to different conclusions. Reactive protocols have a slight advantage over proactive protocols in terms of node deployment scalability and lower power consumption due to less frequent routing table updates. Hybrid protocols are only beneficial in specific topology deployments. Security is still a major concern for LoRa networks, particularly on attacks such as Man-In-The-Middle (MITM), spoofing, wormholes, flooding, packet replay, and information disclosure. The upcoming trend of implementing machine learning algorithms to multi-hop and mesh LoRa networks opens up a wide range of possibilities, ranging from airspace efficiency, efficient route selection, and improving data throughput. A detailed discussion of the aspects’ issues and recommended industry applications that will benefit from multi-hop and mesh LoRa networks are also provided.
1 August 2017 in European Journal of Operational Research
This paper presents an overview of recent advances for the Nurse Rostering Problem (NRP) based on methodological papers published between 2012 to 2021. It provides a comprehensive review of the latest solution methodologies, particularly computational intelligence (CI) approaches, utilized in benchmark and real-world nurse rostering. The methodologies are systematically categorised (Heuristics, Meta-heuristics, Hyper-heuristics, Mathematical Optimisation, Matheuristics and Hybrid Approaches). The NRP benchmark repositories and the respective state-of-the-art methods are also presented. A distinctive feature of this survey is its focus on the emerging trends in terms of solution methodologies and benchmark datasets. Meta-heuristics are the most popular choices in addressing NRP. Matheuristics, one of most popular methodologies in addressing the NRP, has been an emerging trend in recent years (2018 onwards). The INRC-I dataset is the most popular benchmark currently in use by researchers to test their algorithms. An in-depth discussion on the challenges and research opportunities is provided. The summary and analysis of the recently published NRP methodological papers in this survey is valuable for the CI and Operational Research (OR) communities especially early career researchers seeking to find gaps and identify emerging trends in this fast-developing, important research area.
3 June 2018 in Journal of the Operational Research Society
In this paper, we utilise a two-stage approach for addressing the post enrolment course timetabling (PE-CTT) problem. We attempt to find a feasible solution in the first stage. The solution is further improved in terms of soft constraint violations in the second stage. We present an enhanced variant of the Simulated Annealing with Reheating (SAR) algorithm, which we term Simulated Annealing with Improved Reheating and Learning (SAIRL). We propose a reinforcement learning-based methodology to obtain a suitable neighbourhood structure for the search to operate effectively. We incorporate the average cost changes into the reheating temperature function. The proposed enhancements are tested on three widely studied benchmark data-sets. Our algorithm eliminates the need for tuning parameters in conventional SA as well as neighbourhood structure composition in SAR. The results are highly competitive with SAR and other state of the art methods. In addition, SAIRL is scalable when the runtime is extended. The algorithm achieves new best results for 6 instances and new mean results for 14 instances.
23 June 2023 in Intelligent Systems with Applications