Browsing by Author "Semira Mohammed"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Real-time Feature Extraction in a Distributed Acoustic Sensor Based on Phase Demodulation With Fast Hilbert Transform
    (Addis Ababa University, 2024-03) Semira Mohammed; Yonas Seifu (PhD); Bisrat Derebssa (PhD)
    Phase-sensitive optical Time Domain Reflectometry (Φ-OTDR) is the most common implementation of a Distributed Acoustic Sensor (DAS) system. It employs the observation of speckles resulting from Rayleigh Back-scattering from coherent pulses in an optical fiber[1]. Since they are sensitive to local disturbances altering the intensity and phase of light, perturbations induced by events cause changes in the speckle pattern whose precise measurement provides information on the amplitude and frequency of vibrations distributed along the fiber. Demodulation of the local phase change is key to the precise measurement of events since it is more linearly related to the strain applied to the fiber. One of the key issues in distributed sensing is that phase demodulation schemes usually require additional post-processing algorithm runs for each spatial location, which introduces delays, and hence reductions in dynamic sensing capability when scaled along the whole sensing distance. In this research, we analyze the impact of the post-processing in different phase demodulation techniques employing Phase-Generated Carrier (PGC) on the bandwidth of distributed feature extraction in a typical DAS system by quantifying the total computation time needed for a benchmark, 10-km sensing range at meterscale and sub-meter-scale spatial resolutions. We then design, implement, and analyze a signal processing scheme for phase extraction in Φ -OTDR enabling real-time dynamic measurements based on a Fast Hilbert Transform (FHT). Particular focus is given to the choice of this demodulation scheme for optimizing the bandwidth of distributed feature extraction as it enables the use of parallel processing of adjacent blocks in such a way that the overall throughput of spatially resolved concurrent demodulation allows dynamic vibration sensing at speeds relevant to most distributed monitoring applications. Our analysis shows that that on average 3 orders magnitude reductions in computation times are achieved when employing the Fast Hilbert transform for demodulation compared to the commonly used PGC-arctan algorithm, while there is a three-fold reduction compared to PG-DCM and PG-DMS algorithms.
  • No Thumbnail Available
    Item
    Vision to Auditory Substitution for an Artificial Agent
    (Addis Ababa University, 2025-01) Semira Mohammed; Menore Tekeba (PhD)
    Sensory substitution technology converts raw visual input into auditory soundscapes, allowing individuals to “see” with sound. However, mastering this skill requires significant cognitive adaptation, extensive training, and practical application in realistic, everyday scenarios. Experiments with humans have shown the potential for auditory substitution of vision, but these efforts are limited by high costs, ethical concerns, and the risk of unintended side effects, such as impaired auditory skills. To address these challenges, this study develops a Vision-to-Auditory Sensory Substitution system for artificial agents. By simulating sensory substitution in a controlled reinforcement learning (RL) framework, this approach eliminates the need for human experimentation while retaining the ability to explore learning dynamics and decision-making behaviors. Using the Proximal Policy Optimization (PPO) algorithm, agents were trained in two OpenAI Gym environments—CarRacing-v2 and LunarLander-v2—to compare the performance of vision-based and auditory-based agents. The results demonstrate that auditory agents, despite inherent challenges in interpreting sound-encoded visual inputs, achieved means rewards of 427.91 in CarRacing-v2 environments and 259.85 in the LunarLander-v2 environment over 100 episodes. These findings highlight the potential of sensory substitution systems in enabling artificial agents to act effectively using auditory cues. This research contributes to advancing assistive technologies while addressing the limitations and risks of human-based sensory substitution experiments.