Research Excellences

  • Adaptive Instantaneous Traffic Signal Management through Cascade Object Detection
    Graduate Thesis, 2013

    sites/img/no-profile-pic.png

    A traffic signal duration management system aligned with interminably fluctuating traffic demand is proposed in this thesis work. Computer vision technique is used to judge traffic plea in specific road wing. Road traffic control is one of the prime issues of any busy town in the world. The situation is worst at the junction point of multiple roads. It is not always possible to predict upcoming traffic pressure intensity and direction with prior experience and information.

  • Fundamental and Technical Analysis by TPWS, Association Rules Mining and Text Mining
    Graduate Thesis, 2012

    sites/img/no-profile-pic.png

    The objective of this research is to classify or forecast the stock market from the general investor?s point of view. There are four parts in this research. In the first part we performed a survey on most of the well known data mining indicators, implemented the algorithms and calculated the accuracy by applying them on historical data.

  • Modified PolyHarmonic Broadcasting Scheme with Seamless Channel Transition
    Graduate Thesis, 2011

    sites/img/no-profile-pic.png

    A Modified PolyHarmonic Broadcasting Scheme with Seamless Channel Transition Due to continuous change in networking technology, video on demand (VOD) has gained enormous popularity. One of the VOD scheme is harmonic broadcasting which is concern about bandwidth. PolyHarmonic broadcasting (PHB) protocols provide an efficient way to distribute videos that are requested by clients and are simultaneously watched by many viewers.

  • SMERT: A Protocol for Multiple Event-to-Sink Reliability in WSN
    Graduate Thesis, 2010

    sites/img/no-profile-pic.png

    A wireless sensor network (WSN) consists of numerous distributed, self-sufficient micro sensor nodes to agreeably monitor physical or environmental circumstances, such as temperature, sound, vibration, pressure, moisture, traffic, motion, pollutant ts or habitat monitoring etc. Event based WSN rely on the combined effort of a number of micro sensor nodes. Reliable event detection at the sink is based on combined information provided by source nodes, not on any individual report.