Rajiv Punmiya

Mr. Rajiv Punmiya

The Catholic University of Korea
orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 January 2019 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
RPRajiv Punmiya
Rajiv Punmiya
SCSangho Choe
Sangho Choe
For the smart grid energy theft identification, this letter introduces a gradient boosting theft detector (GBTD) based on the three latest gradient boosting classifiers (GBCs): 1) extreme gradient boosting; 2) categorical boosting; and 3) light gradient boosting method. While most of existing machine learning (ML) algorithms just focus on fine tuning the hyperparameters of the classifiers, our ML algorithm, GBTD, focuses on the feature engineering-based preprocessing to improve detection performance as well as time-complexity. GBTD improves both detection rate and false positive rate (FPR) of those GBCs by generating stochastic features like standard deviation, mean, minimum, and maximum value of daily electricity usage. GBTD also reduces the classifier complexity with weighted featureimportance-based extraction techniques. Emphasis has been laid upon the practical application of the proposed ML for theft detection by minimizing FPR and reducing data storage space and improving time-complexity of the GBTD classifiers. Additionally, this letter proposes an updated version of the existing six theft cases to mimic real-world theft patterns and applies them to the dataset for numerical evaluation of the proposed algorithm.

9 September 2021 in IEEE Access
Verified
Journal Article
Open access
NSnavneet singh
navneet singh
SCSangho Choe
Sangho Choe
RPRajiv Punmiya
Rajiv Punmiya
In the era of the Internet of Things (IoT) and Industry 4.0, the indoor usage of smart devices is expected to increase, thereby making their location information more important. Based on various practical issues related to large delays, high design cost, and limited performance, conventional localization techniques are not practical for indoor IoT applications. In recent years, many researchers have proposed a wide range of machine learning (ML)-based indoor localization approaches using Wi-Fi received signal strength indicator (RSSI) fingerprints. This survey attempts to provide a summarized investigation of ML-based Wi-Fi RSSI fingerprinting schemes, including data preprocessing, data augmentation, ML prediction models for indoor localization, and postprocessing in ML, and compare their performance. Any ML-based study is heavily reliant on datasets. Therefore, we dedicate a significant portion of this survey to the discussion of dataset collection and open-source datasets. To provide good direction for future research, we discuss the current challenges and potential solutions related to ML-based indoor localization systems.

2 September 2022 in Sensors
Verified
Journal Article
Open access
PDF file
NSnavneet singh
navneet singh
SCSangho Choe
Sangho Choe
RPRajiv Punmiya
Rajiv Punmiya
NKNavneesh Kaur
Navneesh Kaur
Location-based indoor applications with high quality of services require a reliable, accurate, and low-cost position prediction for target device(s). The widespread availability of WiFi received signal strength indicator (RSSI) makes it a suitable candidate for indoor localization. However, traditional WiFi RSSI fingerprinting schemes perform poorly due to dynamic indoor mobile channel conditions including multipath fading, non-line-of-sight path loss, and so forth. Recently, machine learning (ML) or deep learning (DL)-based fingerprinting schemes are often used as an alternative, overcoming such issues. This paper presents an extreme gradient boosting-based ML indoor localization scheme, simply termed as XGBLoc, that accurately classifies (or detects) the positions of mobile devices in multi-floor multi-building indoor environments. XGBLoc not only effectively reduces the RSSI dataset dimensionality but trains itself using structured synthetic labels (also termed as relational labels), rather than conventional independent labels, that classify such complex and hierarchical indoor environments well. We numerically evaluate the proposed scheme on the publicly available datasets and prove its superiority over existing ML or DL-based schemes in terms of classification and regression performance.

4 January 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
RPRajiv Punmiya
Rajiv Punmiya
SCSangho Choe
Sangho Choe
In the near future, it is highly expected that smart grid (SG) utilities will replace existing fixed pricing with dynamic pricing, such as time-of-use real-time tariff (ToU). In ToU, the price of electricity varies throughout the whole day based on the respective utilities’ decisions. We classify the whole day into two periods with very high and low probabilities of theft activities, termed as the “theft window” and “non-theft window”, respectively. A “smart” malicious consumer can adjust his/her theft to mostly targeting the theft window, manipulate actual usage reporting to outsmart existing theft detectors, and achieve the goal of “paying reduced tariff”. Simulation results show that existing schemes do not detect well such window-based theft activities conversely exploiting ToU strategies. In this paper, we begin by introducing the core concept of window-based theft cases, which is defined at the basis of ToU pricing as well as consumption usage. A modified extreme gradient boosting (XGBoost) based machine learning (ML) technique called dynamic electricity theft detector (DETD) has been presented to detect a new type of theft cases.

1 June 2019
Verified
Conference Paper
Closed access
RPRajiv Punmiya
Rajiv Punmiya
OZOlga Zyabkina
Olga Zyabkina
SCSangho Choe
Sangho Choe
JMJan Meyer
Jan Meyer
Power quality (PQ) data analysis has become an issue of great interest in past few years. Due to numerous impacts, PQ can exhibit rare and significant deviations of a time series from its typical characteristic. Such unusual deviations are known as anomalies and the process of identifying such changes is referred to anomaly detection. In this paper, four proximity-based machine learning (ML) techniques are applied to original (nontransformed) PQ data for automatic anomaly detection. In contrast to standard classification tasks, these techniques are applied on unlabeled data, taking only the characteristics of the PQ data into account. The process of parameters tuning and their influence on methods performance are described. The methods performance has been evaluated on different time series, which represent the typical variation in PQ data. Finally, the ability to detect different types of anomalies is analyzed.