School of Electrical Engineering and Technology (SEET)

Permanent URI for this communityhttp://197.211.34.35:4000/handle/123456789/24

School of Electrical Engineering and Technology (SEET)

Browse

Subject: search.filters.subject.Monitoring

Search Results

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Sensed Outlier Detection for Water Monitoring Data and a Comparative Analysis of Quantization Error Using Kohonen Self-Organizing Maps
    (IEEE, 2018) Dogo, E. M.; Nwulu, N. I.; Twala, B. and Aigbavboa, C. O.
    Measurement values obtained from sensors deployed in the field are sometimes prone to deviation from known patterns of the sensed data which is referred to as outlier or anomalous readings. The reasons for this outlier may include noise, faulty sensor errors, environmental events and cyber-attack on the sensor network, resulting in faulty and missing data that greatly affects quality of the raw data and its subsequent analysis. This paper employs the Self-Organizing Maps (SOM) algorithm to visualise and interpret clusters of sensed data obtained from fresh water monitoring sites, with patterns of similar expressions in a graphical form. With the aim of detecting potential anomalous sensed data, so that they could be investigated and possibly removed to guarantee the quality of the overall dataset. Furthermore, a comparative study of the effects of four different well known neighborhood functions (gaussian, bubble, triangle and mexican hat) with varying neighborhood radius (σ) and learning rate (η) values on Quantization Error (QE) metric was conducted. From the experiment conducted a 3.45% potentially anomalous sensed data were discovered from the entire dataset, in addition, our initial finding suggests a very insignificant variation of the QE based on our dataset and the experiments conducted.
  • No Thumbnail Available
    Item
    Development of a Wireless Sensor Network Based Water Quality Monitoring and Notification System
    (2019) Sithole, M. P. P.; Nwulu, N. I.; and Dogo, E. M.
    In this paper, we present a water quality monitoring and notification system. It is also integrated with a consumer alert system on the safety of the water in accordance with the WHO water quality standard. The consumer alert system is made up of a buzzer for notification, a red-Light Emitting Diode (LED) and green-LED as an indicator for unsafe water and safe water respectively. Five sources of contaminants in water namely, soil, chlorine, vinegar, salt, washing powder and their combination were used in this to validate the performance of the system. Wireless communication between the measuring subsystem and the analysis and notification subsystem was established for mobility using radio frequency modules. The error in measurements and the consumption risk per water parameter were calculated on MS Excel as part of the analysis and presented in this document. This paper also presents an added functionality by using a Light Dependent Resistor (LDR) for turbidity measurement and LEDs in the notification subsystem. The notification system accommodates people with disabilities as the buzzer can be heard by those who can't see, and the LEDs can be seen by those who can't hear. The developed system was compared for functionality and performance using the quality of the results measured in comparison with the expected results.
  • No Thumbnail Available
    Item
    Design and Implementation of a Wireless Patient Health Monitoring System
    (IEEE, 2019) O. Manzombi, O.; Dogo, E. M.; Nwulu, N. I.
    This paper presents the design and implementation of an IoT wireless patient's health monitoring system. The system can be used to continuously monitor the body temperature and pulse rate of a patient located in a hospital room using biomedical sensors. The temperature and pulse rate values are taken from the sensors and processed by an Arduino Uno. Furthermore, they are sent wirelessly via RF communication using a 433 MHz transmitter and receiver kit. The readings are encoded and sent to the receiver where they are decoded and displayed on an LCD screen. Finally, the temperature and pulse rate values are also displayed and stored online using an Arduino Ethernet Shield 2 for future analysis.