Recently, artificial intelligence (AI) based on IoT sensors has been widely used, which has increased the risk of attacks targeting AI. Adversarial examples are among the most serious types of attacks in which the attacker designs inputs that can cause the machine learning system to generate incorrect outputs. Considering the architecture using multiple sensor devices, hacking even a few sensors can create a significant risk; an attacker can attack the machine learning model through the hacked sensors. Some studies demonstrated the possibility of adversarial examples on the deep neural network (DNN) model based on IoT sensors, but it was assumed that an attacker must access all features. The impact of hacking only a few sensors has not been discussed thus far. Therefore, in this study, we discuss the possibility of attacks on DNN models by hacking only a small number of sensors. In this scenario, the attacker first hacks few sensors in the system, obtains the values of the hacked sensors, and changes them to manipulate the system, but the attacker cannot obtain and change the values of the other sensors. We perform experiments using the human activity recognition model with three sensor devices attached to the chest, wrist, and ankle of a user, and demonstrate that attacks are possible by hacking a small number of sensors.

Meningkatkan Keterampilan Berpikir, Me-Manage Keuangan Pada Masyarakat Pesisir Pulau Setokok Kota Batam

A neural network is a form of artificial intelligence that has the ability to learn, grow, and adapt in a dynamic environment. Neural network began since 1890 because a great American psychologist named William James created the book "Principles of Psycology". James was the first one publish a number of facts related to the structure and function of the brain. The history of neural network development is divided into 4 epochs, the Camelot era, the Depression, the Renaissance, and the Neoconnectiosm era. Neural networks used today are not 100 percent accurate. However, neural networks are still used because of better performance than alternative computing models. The use of neural network consists of pattern recognition, signal analysis, robotics, and expert systems. For risk analysis of the neural network, it is first performed using hazards and operability studies (HAZOPS). Determining the neural network requirements in a good way will help in determining its contribution to system hazards and validating the control or mitigation of any hazards. After completion of the first stage at HAZOPS and the second stage determines the requirements, the next stage is designing. Neural network underwent repeated design-train-test development. At the design stage, the hazard analysis should consider the design aspects of the development, which include neural network architecture, size, intended use, and so on. It will be continued at the implementation stage, test phase, installation and inspection phase, operation phase, and ends at the maintenance stage.

Ensuring worker safety in open-pit mining remains challenging due to dynamic terrain, heavy machinery, and limited visibility from dust and varying illumination. Real-time detection of small human figures under these conditions is vital to prevent fatal accidents. This study evaluates three advanced versions of the YOLO (You Only Look Once) object detection framework-YOLOv10, YOLOv11, and YOLOv12-for detecting small-sized humans in open-pit coal mining environments. A proprietary dataset from an active mining site in East Kalimantan, Indonesia, was utilized to fine-tune and benchmark fifteen YOLO variants (nano, small, medium, large, extra-large). Performance was systematically evaluated using precision, recall, F1-score, and confidence threshold optimization. Results indicate YOLOv10x achieves the highest detection accuracy (F1score of 0.92). However, YOLOv12s and YOLOv11s exhibit comparable accuracy but significantly faster inference speeds (latencies below 16 milliseconds), making them suitable for edge computing platforms such as NVIDIA Jetson. An analysis of trade-offs between accuracy, inference latency, and optimal confidence thresholds provides practical recommendations for real-time deployment in safety-critical mining operations. This research establishes a robust, efficient, and deployable detection framework for small human figures, addressing existing gaps in mining safety literature and enhancing intelligent collision avoidance systems for high-risk industrial environments.

In this era, free access points are found available in various places. But this freedom comes with a price, and only a few users really understand the risk. In a recent survey, 70% of tablet owners and 53% of smartphone owners stated that they use public wifi hotspots. The biggest threat to public wifi security is how a hacker positions himself as a liaison between victims and Authorized Access Points. To do this the hacker creates an Unauthorized Access Point (Fake Access Point). We took a pentester/attacker POV in this artikel for educational purposes, so that users may know the stages of Fake Access Point attack based on Kali Linux, Fluxion. For the digital evidence analysis stage, we used the customized OSCAR (Obtain information, Strategies, Collect Evidence, Analyze and Report) methods, where attacking is the stage for preparation, determining which wifi Access Points is going to be the target of the attack, and carrying out attacks. While, analysis is the stage of analyzing the steps of attack and how to distinguish between AAP and UAP. The results of our research are that after determining the target, the pen tester/attacker will use aircrack-ng on Fluxion to get a handshake, create a fake web interface, then launch a deauth all attack, also known as DoS, to AAP so that the victim / client cannot connect with the AAP and switch to Fake Access Point. The fake web interface will then ask the victim to enter the password, where after the password is found, the pen tester/attacker can see it through Fluxion. As a precautionary measure, the difference between a Fake Access Point and an Authorized Access Point is found in the presence or absence of the padlock symbol (Android) or an exclamation point (Windows 10).Pada zaman ini, free access point telah tersedia di berbagai tempat. Namun, nyatanya kebebasan ini memiliki harga, dan hanya sedikit pengguna yang memahami benar risikonya. Ancaman terbesar terhadap kemanan wifi publik adalah bagaimana seorang hacker memposisikan dirinya sebagai penghubung antar korban dan Authorized Access Point. Untuk melakukan hal tersebut, hacker membuat Unauthorized Access Point (Fake Access Point). Dalam artikel ini pen tester/attacker diambil sudut pandang sebagai dengan tujuan edukasi, agar pengguna mengetahui tahapan serangan Fake Access Point dengan tool Fluxion berbasis OS Kali Linux. Tahapan analisis bukti digital menggunakan metode OSCAR (Obtain Information, Strategies, Collect Evidence, Analyze and Report) yang telah di kostumisasi, di mana attacking adalah tahapan untuk persiapan menentukan target wifi Access Point yang akan diserang serta menjalankan serangan. Analysis adalah tahapan menganalisa langkah penyerangan serta bagaimana cara membedakan Authorized Access Point dengan Unauthorized Access Point. Hasil penelitian yang dilakukan setelah menentukan target, pentester/attacker akan menggunakan Aircrack-ng pada Fluxion untuk mendapatkan handshake, membuat web interface palsu, kemudian melancarkan serangan Deauth all, dikenal DoS ke AAP, sehingga korban/client tidak dapat terkoneksi dan masuk ke Fake Access Point. Web interface palsu kemudian akan meminta korban untuk memasukkan password. Setelah password ditemukan, maka pen tester/attacker dapat melihatnya melalui Fluxion. Sebagai langkah pencegahan, perbedaan antara Fake Access Point dan yang Authorized Access Point ditemukan pada ada tidaknya simbol gembok (Android) atau tanda seru (Windows 10).

Object detection in open-pit mining environments presents a critical yet underexplored challenge: identifying personnel who appear as ultra-tiny targets in imagery captured from mining vehicles and surveillance systems. When persons occupy merely 9×9 pixels on average far below the 32×32 pixel threshold conventionally defining small objects standard detection architectures exhibit severe performance degradation. This study presents a systematic comparative evaluation of nine model configurations spanning three contemporary YOLO architectures (YOLO26, YOLOv12, and YOLOv11) across nano, small, and medium variants for ultra-tiny person detection in real mining imagery. Our experiments reveal a counterintuitive finding: nano variants consistently outperform their larger counterparts across all architectures, challenging prevailing assumptions about model scaling in deep learning. YOLOv11-Nano achieved the highest performance with 0.973 F1-score and 0.991 recall while maintaining 31 frames per second on NVIDIA Jetson Nano hardware comfortably exceeding real-time requirements for collision avoidance systems. The attention mechanisms in YOLOv12, despite theoretical promise for preserving spatial information, introduced computational overhead without accuracy benefits at ultra-tiny scales. These findings suggest that aggressive downsampling in deeper networks eliminates spatial information critical for ultra-tiny detection before meaningful feature encoding occurs. The results provide practical guidance 1 for deploying person detection systems in mining safety applications and contribute empirical evidence that conventional model scaling assumptions may not hold at extreme object scales.