The document describes a method for classifying vehicles based on their audio signals using quadratic discriminant analysis. Feature vectors containing short-time energy, average zero-crossing rate, and pitch frequency are extracted from periodic segments of vehicle audio signals. The method achieves better classification accuracy by only considering feature vectors with high energy, as these correspond better to the vehicle sounds and exclude low energy background noise regions. Simulation results show the proposed method of separating high energy vectors based on thresholds of average energy and zero-crossing rate improves classification performance compared to considering all vectors.
Jamming Detection based on Doppler Shift Estimation in Vehicular Communicatio...IJCNCJournal
Since Doppler shift is one of the most important parameters in wireless propagation, the evaluation of the Doppler shift at the base station (BTS) in vehicular communications improves BTS in many aspects such as channel varying rate, jamming detection, and handover operations. Therefore, in this study, we propose a novel method at a base station based on the received user signal to estimate the channel Doppler shift seen by BTS. Utilizing the inherent information existed in common receivers, a level crossing rate (LCR) based Doppler shift estimation algorithm is developed without any excessive hardware. Moreover, a jamming detection algorithm is improved based on the proposed Doppler shift estimation scheme. The performance of the proposed scheme is evaluated in a Terrestrial Trunked Radio (TETRA) network, and comprehensive experimental results have shown superior performance in a wide range of velocities, signal to noise ratios and jammers.
AN MINIMUM RECONFIGURATION PROBABILITY ROUTING ALGORITHM FOR RWA IN ALL-OPTIC...sipij
In this paper, we present a detailed study of Minimum Reconfiguration Probability Routing (MRPR) algorithm, and its performance evaluation in comparison with Adaptive unconstrained routing (AUR) and Least Loaded routing (LLR) algorithms. We have minimized the effects of failures on link and router failure in the network under changing load conditions, we assess the probability of service and number of light path failures due to link or route failure on Wavelength Interchange(WI) network. The computation complexity is reduced by using Kalman Filter(KF) techniques. The minimum reconfiguration probability
routing (MRPR) algorithm selects most reliable routes and assign wavelengths to connections in a manner that utilizes the light path(LP) established efficiently considering all possible requests.
Enhancing Performance for Orthogonal Frequency Division Multiplexing in Wirel...IRJET Journal
The document discusses enhancing the performance of Orthogonal Frequency Division Multiplexing (OFDM) in wireless systems. It proposes using a technique called Selective Level Mapping (SLM) to reduce the peak-to-average power ratio (PAPR) of OFDM signals. PAPR reduction is important for OFDM systems to improve power amplifier efficiency. The document describes a "Class-III SLM scheme" that can generate multiple alternative OFDM signal sequences using only one inverse fast Fourier transform, helping to reduce complexity. It proposes a selection method for rotation values that can achieve optimal PAPR reduction while balancing the load across components. Simulation results show the proposed method achieves better PAPR reduction performance than conventional methods
Performance analysis of adaptive filter channel estimated MIMO OFDM communica...IJECEIAES
Advanced Communication Systems are wideband systems to support multiple applications such as audio, video and data so and so forth. These systems require high spectral efficiency and data rates. In addition, they should provide multipath fading and inter-symbol interference (ISI) free transmission. Multiple input multiple output orthogonal frequency division multiplexing (MIMO OFDM) meets these requirements Hence, MIMOOFDM is the most preferable technique for long term evaluation advanced (LTEA). The primary objective of this paper is to control bit error rate (BER) by proper channel coding, pilot carriers, adaptive filter channel estimation schemes and space time coding (STC). A combination of any of these schemes results in better BER performance over individual schemes. System performance is analyzed for various digital modulation schemes. In this paper, adaptive filter channel estimated MIMO OFDM system is proposed by integrating channel coding, adaptive filter channel estimation, digital modulation and space time coding. From the simulation results, channel estimated 2×2 MIMO OFDM system shows superior performance over individual schemes.
Routing in All-Optical Networks Using Recursive State Space Techniquesipij
In this papr, we have minimized the effects of failures on network performace, by using suitable Routing
and Wavelenghth Assignment(RWA) method without disturbing other performance criteria such as blocking
probability(BP) and network management(NM). The computation complexity is reduced by using Kalaman
Filter(KF) techniques. The minimum reconfiguration probability routing (MRPR) algorithm must be
able to select most reliable routes and assign wavelengths to connections in a manner that utilizes the light
path(LP) established efficiently considering all possible requests.
Analyzing the performance of the dynamicIJCNCJournal
In this paper, we are focused to analyse the performance of the two dimensional dynamic
Position Location and Tracking (PL&T) of mobile nodes. The architecture of the dynamic PL&T
is developed based on determining the potential zone of the target node (s) and then tracking
using the triangulation. We assume that the nodes are mobile and have one omnidirectional
antenna per node. The network architecture under consideration is cluster based Mobile Ad Hoc
Network (MANET) where at an instance of time, three nodes are used as reference nodes to track
target node(s) using triangulation method. The novel approach in this PL&T tracking method is
the “a priori” identification of the zone of the target node(s) within a circle with a reasonable
radios, and then placing the three reference nodes for the zone such that a good geometry is
created between the reference nodes and the target nodes to improve the accuracy of
triangulation method. The geometry of the reference nodes’ triangle is closer to equilateral
triangle and all potential target nodes are inside the circle. We establish the fact that when the
target node is moving linearly, the predictive method of zone finding is sufficient to track the
target node accurately. However, when the target node changes the direction, the predictive
method of zone finding will fail and we need to place the three references outside the zone such
that proper geometry with no one angle is less than 30 degrees is maintained to get accurate
PL&T location of the target node at each instance of time. The new zone is always formed for
each instance of time prior to triangulation.
In this paper, we demonstrate the accuracy of integrated zone finding and triangulation for
detecting the PL&T location the node at each instance of time within 1.5 foot accuracy. It should
be noted that as the target node is tracked continuously by applying the integrated zone finding
and triangulation algorithm at different instances of time, one foot accuracy can no longer be
maintained. Periodically, the good PL&T data on each node has to be established by
reinitializing the PL&T locations of the nodes including those that are used as reference nodes.
In this paper, the performance of the dynamic PL&T system is derived using Additive White
Gaussian Noise (AWGN) channel; and using AWGN plus Multi-path fading channel. The impact
of multipath fading on tracking accuracy is analysed using Rician Fading channel for MANET
applications outdoors. Our real time simulations show the PL&T tracking accuracy for the
mobile target nodes in both cases to be within 1.5 foot accuracy.
This document presents a new hybrid method for estimating the topology of digital subscriber lines using a combination of Correlation Time Domain Reflectometry (CTDR) and Frequency Domain Reflectometry (FDR). The method obtains an approximate loop topology estimate from initial CTDR measurements, then uses an optimization algorithm based on FDR to predict a more accurate topology. It compares measured FDR data to simulated FDR data for the approximate topology to define an objective function, which is minimized using an optimization method to estimate the accurate loop topology without requiring prior network knowledge. Tests on typical loops showed good prediction capability of the proposed hybrid CTDR/FDR method.
An Improved of Multiple Harmonic Sources Identification in Distribution Syste...IAES-IJPEDS
This paper introduces an improved of multiple harmonic sources
identification that been produced by inverter loads in power system using
time-frequency distribution (TFD) analysis which is spectrogram.
The spectrogram is a very applicable method to represent signals in
time-frequency representation (TFR) and the main advantages
of spectrogram are the accuracy, speed of the algorithm and use low memory
size such that it can be computed rapidly. The identification of multiple
harmonic sources is based on the significant relationship of spectral
impedances which are the fundamental impedance (Z1) and harmonic
impedance (Zh) that extracted from TFR. To verify the accuracy of the
proposed method, MATLAB simulations carried out several unique cases
with different harmonic producing loads on IEEE 4-bus test feeder cases. It is
proven that the proposed method is superior with 100% correct identification
of multiple harmonic sources. It is envisioned that the method is very
accurate, fast and cost efficient to localize harmonic sources in distribution
system.
Jamming Detection based on Doppler Shift Estimation in Vehicular Communicatio...IJCNCJournal
Since Doppler shift is one of the most important parameters in wireless propagation, the evaluation of the Doppler shift at the base station (BTS) in vehicular communications improves BTS in many aspects such as channel varying rate, jamming detection, and handover operations. Therefore, in this study, we propose a novel method at a base station based on the received user signal to estimate the channel Doppler shift seen by BTS. Utilizing the inherent information existed in common receivers, a level crossing rate (LCR) based Doppler shift estimation algorithm is developed without any excessive hardware. Moreover, a jamming detection algorithm is improved based on the proposed Doppler shift estimation scheme. The performance of the proposed scheme is evaluated in a Terrestrial Trunked Radio (TETRA) network, and comprehensive experimental results have shown superior performance in a wide range of velocities, signal to noise ratios and jammers.
AN MINIMUM RECONFIGURATION PROBABILITY ROUTING ALGORITHM FOR RWA IN ALL-OPTIC...sipij
In this paper, we present a detailed study of Minimum Reconfiguration Probability Routing (MRPR) algorithm, and its performance evaluation in comparison with Adaptive unconstrained routing (AUR) and Least Loaded routing (LLR) algorithms. We have minimized the effects of failures on link and router failure in the network under changing load conditions, we assess the probability of service and number of light path failures due to link or route failure on Wavelength Interchange(WI) network. The computation complexity is reduced by using Kalman Filter(KF) techniques. The minimum reconfiguration probability
routing (MRPR) algorithm selects most reliable routes and assign wavelengths to connections in a manner that utilizes the light path(LP) established efficiently considering all possible requests.
Enhancing Performance for Orthogonal Frequency Division Multiplexing in Wirel...IRJET Journal
The document discusses enhancing the performance of Orthogonal Frequency Division Multiplexing (OFDM) in wireless systems. It proposes using a technique called Selective Level Mapping (SLM) to reduce the peak-to-average power ratio (PAPR) of OFDM signals. PAPR reduction is important for OFDM systems to improve power amplifier efficiency. The document describes a "Class-III SLM scheme" that can generate multiple alternative OFDM signal sequences using only one inverse fast Fourier transform, helping to reduce complexity. It proposes a selection method for rotation values that can achieve optimal PAPR reduction while balancing the load across components. Simulation results show the proposed method achieves better PAPR reduction performance than conventional methods
Performance analysis of adaptive filter channel estimated MIMO OFDM communica...IJECEIAES
Advanced Communication Systems are wideband systems to support multiple applications such as audio, video and data so and so forth. These systems require high spectral efficiency and data rates. In addition, they should provide multipath fading and inter-symbol interference (ISI) free transmission. Multiple input multiple output orthogonal frequency division multiplexing (MIMO OFDM) meets these requirements Hence, MIMOOFDM is the most preferable technique for long term evaluation advanced (LTEA). The primary objective of this paper is to control bit error rate (BER) by proper channel coding, pilot carriers, adaptive filter channel estimation schemes and space time coding (STC). A combination of any of these schemes results in better BER performance over individual schemes. System performance is analyzed for various digital modulation schemes. In this paper, adaptive filter channel estimated MIMO OFDM system is proposed by integrating channel coding, adaptive filter channel estimation, digital modulation and space time coding. From the simulation results, channel estimated 2×2 MIMO OFDM system shows superior performance over individual schemes.
Routing in All-Optical Networks Using Recursive State Space Techniquesipij
In this papr, we have minimized the effects of failures on network performace, by using suitable Routing
and Wavelenghth Assignment(RWA) method without disturbing other performance criteria such as blocking
probability(BP) and network management(NM). The computation complexity is reduced by using Kalaman
Filter(KF) techniques. The minimum reconfiguration probability routing (MRPR) algorithm must be
able to select most reliable routes and assign wavelengths to connections in a manner that utilizes the light
path(LP) established efficiently considering all possible requests.
Analyzing the performance of the dynamicIJCNCJournal
In this paper, we are focused to analyse the performance of the two dimensional dynamic
Position Location and Tracking (PL&T) of mobile nodes. The architecture of the dynamic PL&T
is developed based on determining the potential zone of the target node (s) and then tracking
using the triangulation. We assume that the nodes are mobile and have one omnidirectional
antenna per node. The network architecture under consideration is cluster based Mobile Ad Hoc
Network (MANET) where at an instance of time, three nodes are used as reference nodes to track
target node(s) using triangulation method. The novel approach in this PL&T tracking method is
the “a priori” identification of the zone of the target node(s) within a circle with a reasonable
radios, and then placing the three reference nodes for the zone such that a good geometry is
created between the reference nodes and the target nodes to improve the accuracy of
triangulation method. The geometry of the reference nodes’ triangle is closer to equilateral
triangle and all potential target nodes are inside the circle. We establish the fact that when the
target node is moving linearly, the predictive method of zone finding is sufficient to track the
target node accurately. However, when the target node changes the direction, the predictive
method of zone finding will fail and we need to place the three references outside the zone such
that proper geometry with no one angle is less than 30 degrees is maintained to get accurate
PL&T location of the target node at each instance of time. The new zone is always formed for
each instance of time prior to triangulation.
In this paper, we demonstrate the accuracy of integrated zone finding and triangulation for
detecting the PL&T location the node at each instance of time within 1.5 foot accuracy. It should
be noted that as the target node is tracked continuously by applying the integrated zone finding
and triangulation algorithm at different instances of time, one foot accuracy can no longer be
maintained. Periodically, the good PL&T data on each node has to be established by
reinitializing the PL&T locations of the nodes including those that are used as reference nodes.
In this paper, the performance of the dynamic PL&T system is derived using Additive White
Gaussian Noise (AWGN) channel; and using AWGN plus Multi-path fading channel. The impact
of multipath fading on tracking accuracy is analysed using Rician Fading channel for MANET
applications outdoors. Our real time simulations show the PL&T tracking accuracy for the
mobile target nodes in both cases to be within 1.5 foot accuracy.
This document presents a new hybrid method for estimating the topology of digital subscriber lines using a combination of Correlation Time Domain Reflectometry (CTDR) and Frequency Domain Reflectometry (FDR). The method obtains an approximate loop topology estimate from initial CTDR measurements, then uses an optimization algorithm based on FDR to predict a more accurate topology. It compares measured FDR data to simulated FDR data for the approximate topology to define an objective function, which is minimized using an optimization method to estimate the accurate loop topology without requiring prior network knowledge. Tests on typical loops showed good prediction capability of the proposed hybrid CTDR/FDR method.
An Improved of Multiple Harmonic Sources Identification in Distribution Syste...IAES-IJPEDS
This paper introduces an improved of multiple harmonic sources
identification that been produced by inverter loads in power system using
time-frequency distribution (TFD) analysis which is spectrogram.
The spectrogram is a very applicable method to represent signals in
time-frequency representation (TFR) and the main advantages
of spectrogram are the accuracy, speed of the algorithm and use low memory
size such that it can be computed rapidly. The identification of multiple
harmonic sources is based on the significant relationship of spectral
impedances which are the fundamental impedance (Z1) and harmonic
impedance (Zh) that extracted from TFR. To verify the accuracy of the
proposed method, MATLAB simulations carried out several unique cases
with different harmonic producing loads on IEEE 4-bus test feeder cases. It is
proven that the proposed method is superior with 100% correct identification
of multiple harmonic sources. It is envisioned that the method is very
accurate, fast and cost efficient to localize harmonic sources in distribution
system.
DORA: Server Based VANETs and its ApplicationsIRJET Journal
This document discusses a server-based vehicle communication system called DORA that aims to efficiently allocate network resources for vehicles uploading files to roadside access points. It formulates the problem as a finite-horizon sequential decision process and proposes algorithms like dynamic optimal random access and joint DORA to compute optimal transmission policies for vehicles at single and multiple access points. The performance is evaluated using simulations and shows efficiency over existing solutions. Key aspects covered are the system model, traffic and channel models, distributed medium access control, problem formulation and proposed optimization algorithms.
Abstract In optical circuit switching the high values of blocking probability is resolved by dynamic wavelength routing algorithms with wavelength conversion. The aim of this paper is to study these algorithms. Then the algorithm is selected which gives good results with and without wavelength conversion. The selected algorithm is then checked for other parameters of networking namely throughput, packet delivery ratio, and delay. A comparative study is then carried out for increasing traffic. We try to prove that these algorithms satisfy the criteria of QoS parameters by this comparative study. The results of simulation show that the parameters follow the trend of blocking probability of the selected algorithm. Keywords: optical burst switching, throughput, packet delivery ratio, delay.
Towards better performance: phase congruency based face recognitionTELKOMNIKA JOURNAL
Phase congruency is an edge detector and measurement of the significant feature in the image. It is a robust method against contrast and illumination variation. In this paper, two novel techniques are introduced for developing alow-cost human identification system based on face recognition. Firstly, the valuable phase congruency features, the gradient-edges and their associate dangles are utilized separately for classifying 130 subjects taken from three face databases with the motivation of eliminating the feature extraction phase. By doing this, the complexity can be significantly reduced. Secondly, the training process is modified when a new technique, called averaging-vectors is developed to accelerate the training process and minimizes the matching time to the lowest value. However, for more comparison and accurate evaluation,three competitive classifiers: Euclidean distance (ED),cosine distance (CD), and Manhattan distance (MD) are considered in this work. The system performance is very competitive and acceptable, where the experimental results show promising recognition rates with a reasonable matching time.
This document describes a proposed two-stage traffic light system using fuzzy logic to minimize vehicle delay at intersections. The system has two modules: a traffic urgency decision module that selects the next phase to turn green based on traffic urgency, and an extension time decision module that determines how long to extend the green light phase based on vehicle numbers. Software was developed in MATLAB to simulate this system at an isolated intersection and evaluate its performance using average vehicle delay. The document reviews other related works applying fuzzy logic to traffic light control and adaptive signal systems.
This document provides an overview and critical analysis of pilot-based and blind channel estimation techniques for OFDM systems. It first describes the basic OFDM system model and how it is used for pilot-based channel estimation, where pilot signals are inserted into subcarriers periodically to estimate the channel. It then discusses blind channel estimation techniques that do not require transmitting a training sequence. The document analyzes various channel estimation techniques like least squares, minimum mean square error, constant modulus algorithm, and linear precoding. It aims to aid the development of new blind channel estimation methods by comparing existing techniques.
This paper proposes algorithms for dynamic travel time prediction to provide reliable real-time
travel time information using probe travel time data collected by a dedicated short range communication (DSRC)
system. The travel time predictions were performed using arrival-time-based travel time; subsequently, the
accuracy of these predictions was evaluated using the concurrent departure-time-based travel time data, which
were also collected by the DSRC system. The prediction methodologies proposed in this research include the
Kalman filter and a newly developed algorithm that uses weighting factors according to probe sample size. An
evaluation of the performance of the two algorithms showed their errors ranged from 5 to 7%, thereby showing
satisfactory results. Considering the fact that the Kalman filter requires historical travel time for prediction, the
similarity between the historical and current data is core factor for reliable travel time prediction. On the other
hand, the newly developed algorithm does not need historical data, thereby the benefit could be enhanced
especially when historical travel time data analogous to current ones are not easily available.
Smart Antenna is a device with signal processing
capability combining multiple antenna elements to optimize its
radiation and reception patterns as per designed specifications.
Smart antennas basically comprise of two functionalities, i.e.,
Direction of Arrival and Beamforming. This paper explains the
estimation of Direction of Arrival using MLM method and a
novel approach called MUltiple Signal Classification which takes
advantage of orthogonal property and performs subspace
computation. With a comparative study of both the algorithms,
we shall prove the advantages of MUltiple Signal Classification
over the MLM method with the aid of MATLAB
This document compares the performance of three mobile ad hoc network (MANET) routing protocols: AODV, FSR, and IERP. It uses the QualNet network simulator to evaluate these protocols based on various metrics like throughput, average jitter, average end-to-end delay, and packet delivery ratio. The protocols are evaluated under different node speeds on a grid topology network with 90 nodes over an area of 1500x1500 meters. Simulation results show that AODV generally performs best in terms of throughput and packet delivery ratio across varying node speeds, while FSR performs worst for these metrics. IERP shows the worst performance for average end-to-end delay and average jitter as node speed increases.
- The document presents a new maximum likelihood technique for estimating the parameters of a moving target in noncoherent MIMO radar systems.
- It models target motion within a processing interval as a polynomial of general order, where the first three coefficients correspond to initial location, velocity, and acceleration for a second-order polynomial.
- The technique develops an ML estimator that requires multidimensional search over the unknown polynomial coefficients to estimate the target motion parameters. It shows the ML problem can be interpreted as a nonlinear least squares problem.
Traffic Light Signal Parameters Optimization Using Modification of Multielement...IJECEIAES
The document describes a modification to the Multielement Genetic Algorithm (MEGA) called H-MEGA to optimize traffic light signal parameters and reduce traffic congestion. H-MEGA adds a hash table to store the best populations found so far to guide the recombination process. Testing showed H-MEGA improved vehicle throughput over the original MEGA and Particle Swarm Optimization methods by 10-21% on a road network in Kumamoto City, Japan.
A Path Planning Technique For Autonomous Mobile Robot Using Free-Configuratio...CSCJournals
This paper presents the implementation of a novel technique for sensor based path planning of autonomous mobile robots. The proposed method is based on finding free-configuration eigen spaces (FCE) in the robot actuation area. Using the FCE technique to find optimal paths for autonomous mobile robots, the underlying hypothesis is that in the low-dimensional manifolds of laser scanning data, there lies an eigenvector which corresponds to the free-configuration space of the higher order geometric representation of the environment. The vectorial combination of all these eigenvectors at discrete time scan frames manifests a trajectory, whose sum can be treated as a robot path or trajectory. The proposed algorithm was tested on two different test bed data, real data obtained from Navlab SLAMMOT and data obtained from the real-time robotics simulation program Player/Stage. Performance analysis of FCE technique was done with existing four path planning algorithms under certain working parameters, namely computation time needed to find a solution, the distance travelled and the amount of turning required by the autonomous mobile robot. This study will enable readers to identify the suitability of path planning algorithm under the working parameters, which needed to be optimized. All the techniques were tested in the real-time robotic software Player/Stage. Further analysis was done using MATLAB mathematical computation software.
1) The document discusses algorithms for finding optimal bus routes between locations, including Dijkstra's algorithm and improvements made to address its limitations.
2) It analyzes shortest path algorithms based on graph theory, least transfers, and station matrices. An improved Dijkstra's algorithm is proposed to find shortest paths between any two nodes.
3) The results show the improved algorithm can determine the shortest distance and transfer routes between any four bus stations, demonstrating its accuracy and feasibility for route planning applications.
IRJET- Survey Paper on Human Following RobotIRJET Journal
The document summarizes research on developing an autonomous human following robot. It discusses using triangulation of radio signals from a tag worn by a human to calculate the tag's location using multiple antennas on the robot. The robot would use triangulation and received signal strength to determine the tag's position and direction to follow the human. It reviews several localization algorithms and navigation techniques used in other projects. The proposed method is to use triangulation of signals from three antennas to accurately calculate the tag's position and allow the robot to autonomously follow or be remotely controlled via Bluetooth.
The railway capacity optimization problem deals with the maximization of the number of trains running on
a given network per unit time. In this study, we frame this problem as a typical asymmetrical Travelling
Salesman Problem (ATSP), with the ATSP nodes representing the train arrival /departure events and the
ATSP total cost representing the total time-interval of the schedule. The application problem is then
optimized using the standard Ant Colony Optimization (ACO) algorithm. The simulation experiments
validate the formulation of the railway capacity problem as an ATSP and the ACO algorithm produces
optimal solutions superior to those produced by the domain experts.
IRJET-Study of Performance analysis of Wind Tunnel Simulation of Pollutant Di...IRJET Journal
This document reviews high throughput polar encoders. It discusses how polar codes achieve channel capacity and are becoming popular error correcting codes. It then summarizes the methodology used in the paper, which proposes a pipelined FPGA architecture for polar code encoding to reduce hardware complexity for long codes. This involves a partially parallel approach compared to fully parallel architectures. The conclusions discuss providing this pipelined architecture to optimize hardware usage.
Artificial Intelligence in Robot Path Planningiosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
This document proposes a hybrid optimization algorithm using ant colony optimization and particle swarm optimization to solve the multiobjective multicast routing problem in wireless sensor networks. The goal is to optimize two objectives simultaneously - end-to-end delay and total transmitted power. ACO and PSO are combined to find Pareto-optimal solutions efficiently. Simulation results show the algorithm can find near-optimal solutions for minimizing delay and power consumption when routing data from a source to multiple destinations in wireless sensor networks.
An enhanced available bandwidth estimation technique for an end to end networ...redpel dot com
The document describes an enhanced technique for estimating available bandwidth (ABW) of an end-to-end network path. It proposes a unique probing scheme using a probing train structure with a high-density region to detect the turning point more accurately. It also includes a rate adjustment algorithm and a modified excursion detection algorithm to estimate ABW more accurately and less intrusively. Experimental results using an Android device over a 4G/LTE network and a testbed show the technique outperforms other state-of-the-art methods in terms of accuracy, intrusiveness, and convergence time.
This Spanish lesson focuses on vocabulary words related to items found in a house. The objectives are for students to identify, spell, and pronounce Spanish words for furniture and other household items with 90% accuracy. Students will practice filling in blanks, matching words to definitions, creating a vocabulary book with a partner, and completing sentences using the new vocabulary words. The lesson materials include worksheets, a PowerPoint, stapler, and scissors, and is designed to take a full class period.
The document discusses creating a virtual heritage site based on Seth's fictional Canadian town of Dominion City from his comics. It analyzes how Seth represents the town through multiple iterations over time in the comics versus his cardboard models that lack narrative. The proposed website would allow users to manipulate and tag objects from Dominion City across different media, encouraging new connections and narratives rather than a linear tour. This emphasizes personal engagement with the ephemera of the site's past over rigidly preserving its physical architecture.
DORA: Server Based VANETs and its ApplicationsIRJET Journal
This document discusses a server-based vehicle communication system called DORA that aims to efficiently allocate network resources for vehicles uploading files to roadside access points. It formulates the problem as a finite-horizon sequential decision process and proposes algorithms like dynamic optimal random access and joint DORA to compute optimal transmission policies for vehicles at single and multiple access points. The performance is evaluated using simulations and shows efficiency over existing solutions. Key aspects covered are the system model, traffic and channel models, distributed medium access control, problem formulation and proposed optimization algorithms.
Abstract In optical circuit switching the high values of blocking probability is resolved by dynamic wavelength routing algorithms with wavelength conversion. The aim of this paper is to study these algorithms. Then the algorithm is selected which gives good results with and without wavelength conversion. The selected algorithm is then checked for other parameters of networking namely throughput, packet delivery ratio, and delay. A comparative study is then carried out for increasing traffic. We try to prove that these algorithms satisfy the criteria of QoS parameters by this comparative study. The results of simulation show that the parameters follow the trend of blocking probability of the selected algorithm. Keywords: optical burst switching, throughput, packet delivery ratio, delay.
Towards better performance: phase congruency based face recognitionTELKOMNIKA JOURNAL
Phase congruency is an edge detector and measurement of the significant feature in the image. It is a robust method against contrast and illumination variation. In this paper, two novel techniques are introduced for developing alow-cost human identification system based on face recognition. Firstly, the valuable phase congruency features, the gradient-edges and their associate dangles are utilized separately for classifying 130 subjects taken from three face databases with the motivation of eliminating the feature extraction phase. By doing this, the complexity can be significantly reduced. Secondly, the training process is modified when a new technique, called averaging-vectors is developed to accelerate the training process and minimizes the matching time to the lowest value. However, for more comparison and accurate evaluation,three competitive classifiers: Euclidean distance (ED),cosine distance (CD), and Manhattan distance (MD) are considered in this work. The system performance is very competitive and acceptable, where the experimental results show promising recognition rates with a reasonable matching time.
This document describes a proposed two-stage traffic light system using fuzzy logic to minimize vehicle delay at intersections. The system has two modules: a traffic urgency decision module that selects the next phase to turn green based on traffic urgency, and an extension time decision module that determines how long to extend the green light phase based on vehicle numbers. Software was developed in MATLAB to simulate this system at an isolated intersection and evaluate its performance using average vehicle delay. The document reviews other related works applying fuzzy logic to traffic light control and adaptive signal systems.
This document provides an overview and critical analysis of pilot-based and blind channel estimation techniques for OFDM systems. It first describes the basic OFDM system model and how it is used for pilot-based channel estimation, where pilot signals are inserted into subcarriers periodically to estimate the channel. It then discusses blind channel estimation techniques that do not require transmitting a training sequence. The document analyzes various channel estimation techniques like least squares, minimum mean square error, constant modulus algorithm, and linear precoding. It aims to aid the development of new blind channel estimation methods by comparing existing techniques.
This paper proposes algorithms for dynamic travel time prediction to provide reliable real-time
travel time information using probe travel time data collected by a dedicated short range communication (DSRC)
system. The travel time predictions were performed using arrival-time-based travel time; subsequently, the
accuracy of these predictions was evaluated using the concurrent departure-time-based travel time data, which
were also collected by the DSRC system. The prediction methodologies proposed in this research include the
Kalman filter and a newly developed algorithm that uses weighting factors according to probe sample size. An
evaluation of the performance of the two algorithms showed their errors ranged from 5 to 7%, thereby showing
satisfactory results. Considering the fact that the Kalman filter requires historical travel time for prediction, the
similarity between the historical and current data is core factor for reliable travel time prediction. On the other
hand, the newly developed algorithm does not need historical data, thereby the benefit could be enhanced
especially when historical travel time data analogous to current ones are not easily available.
Smart Antenna is a device with signal processing
capability combining multiple antenna elements to optimize its
radiation and reception patterns as per designed specifications.
Smart antennas basically comprise of two functionalities, i.e.,
Direction of Arrival and Beamforming. This paper explains the
estimation of Direction of Arrival using MLM method and a
novel approach called MUltiple Signal Classification which takes
advantage of orthogonal property and performs subspace
computation. With a comparative study of both the algorithms,
we shall prove the advantages of MUltiple Signal Classification
over the MLM method with the aid of MATLAB
This document compares the performance of three mobile ad hoc network (MANET) routing protocols: AODV, FSR, and IERP. It uses the QualNet network simulator to evaluate these protocols based on various metrics like throughput, average jitter, average end-to-end delay, and packet delivery ratio. The protocols are evaluated under different node speeds on a grid topology network with 90 nodes over an area of 1500x1500 meters. Simulation results show that AODV generally performs best in terms of throughput and packet delivery ratio across varying node speeds, while FSR performs worst for these metrics. IERP shows the worst performance for average end-to-end delay and average jitter as node speed increases.
- The document presents a new maximum likelihood technique for estimating the parameters of a moving target in noncoherent MIMO radar systems.
- It models target motion within a processing interval as a polynomial of general order, where the first three coefficients correspond to initial location, velocity, and acceleration for a second-order polynomial.
- The technique develops an ML estimator that requires multidimensional search over the unknown polynomial coefficients to estimate the target motion parameters. It shows the ML problem can be interpreted as a nonlinear least squares problem.
Traffic Light Signal Parameters Optimization Using Modification of Multielement...IJECEIAES
The document describes a modification to the Multielement Genetic Algorithm (MEGA) called H-MEGA to optimize traffic light signal parameters and reduce traffic congestion. H-MEGA adds a hash table to store the best populations found so far to guide the recombination process. Testing showed H-MEGA improved vehicle throughput over the original MEGA and Particle Swarm Optimization methods by 10-21% on a road network in Kumamoto City, Japan.
A Path Planning Technique For Autonomous Mobile Robot Using Free-Configuratio...CSCJournals
This paper presents the implementation of a novel technique for sensor based path planning of autonomous mobile robots. The proposed method is based on finding free-configuration eigen spaces (FCE) in the robot actuation area. Using the FCE technique to find optimal paths for autonomous mobile robots, the underlying hypothesis is that in the low-dimensional manifolds of laser scanning data, there lies an eigenvector which corresponds to the free-configuration space of the higher order geometric representation of the environment. The vectorial combination of all these eigenvectors at discrete time scan frames manifests a trajectory, whose sum can be treated as a robot path or trajectory. The proposed algorithm was tested on two different test bed data, real data obtained from Navlab SLAMMOT and data obtained from the real-time robotics simulation program Player/Stage. Performance analysis of FCE technique was done with existing four path planning algorithms under certain working parameters, namely computation time needed to find a solution, the distance travelled and the amount of turning required by the autonomous mobile robot. This study will enable readers to identify the suitability of path planning algorithm under the working parameters, which needed to be optimized. All the techniques were tested in the real-time robotic software Player/Stage. Further analysis was done using MATLAB mathematical computation software.
1) The document discusses algorithms for finding optimal bus routes between locations, including Dijkstra's algorithm and improvements made to address its limitations.
2) It analyzes shortest path algorithms based on graph theory, least transfers, and station matrices. An improved Dijkstra's algorithm is proposed to find shortest paths between any two nodes.
3) The results show the improved algorithm can determine the shortest distance and transfer routes between any four bus stations, demonstrating its accuracy and feasibility for route planning applications.
IRJET- Survey Paper on Human Following RobotIRJET Journal
The document summarizes research on developing an autonomous human following robot. It discusses using triangulation of radio signals from a tag worn by a human to calculate the tag's location using multiple antennas on the robot. The robot would use triangulation and received signal strength to determine the tag's position and direction to follow the human. It reviews several localization algorithms and navigation techniques used in other projects. The proposed method is to use triangulation of signals from three antennas to accurately calculate the tag's position and allow the robot to autonomously follow or be remotely controlled via Bluetooth.
The railway capacity optimization problem deals with the maximization of the number of trains running on
a given network per unit time. In this study, we frame this problem as a typical asymmetrical Travelling
Salesman Problem (ATSP), with the ATSP nodes representing the train arrival /departure events and the
ATSP total cost representing the total time-interval of the schedule. The application problem is then
optimized using the standard Ant Colony Optimization (ACO) algorithm. The simulation experiments
validate the formulation of the railway capacity problem as an ATSP and the ACO algorithm produces
optimal solutions superior to those produced by the domain experts.
IRJET-Study of Performance analysis of Wind Tunnel Simulation of Pollutant Di...IRJET Journal
This document reviews high throughput polar encoders. It discusses how polar codes achieve channel capacity and are becoming popular error correcting codes. It then summarizes the methodology used in the paper, which proposes a pipelined FPGA architecture for polar code encoding to reduce hardware complexity for long codes. This involves a partially parallel approach compared to fully parallel architectures. The conclusions discuss providing this pipelined architecture to optimize hardware usage.
Artificial Intelligence in Robot Path Planningiosrjce
IOSR Journal of Computer Engineering (IOSR-JCE) is a double blind peer reviewed International Journal that provides rapid publication (within a month) of articles in all areas of computer engineering and its applications. The journal welcomes publications of high quality papers on theoretical developments and practical applications in computer technology. Original research papers, state-of-the-art reviews, and high quality technical notes are invited for publications.
This document proposes a hybrid optimization algorithm using ant colony optimization and particle swarm optimization to solve the multiobjective multicast routing problem in wireless sensor networks. The goal is to optimize two objectives simultaneously - end-to-end delay and total transmitted power. ACO and PSO are combined to find Pareto-optimal solutions efficiently. Simulation results show the algorithm can find near-optimal solutions for minimizing delay and power consumption when routing data from a source to multiple destinations in wireless sensor networks.
An enhanced available bandwidth estimation technique for an end to end networ...redpel dot com
The document describes an enhanced technique for estimating available bandwidth (ABW) of an end-to-end network path. It proposes a unique probing scheme using a probing train structure with a high-density region to detect the turning point more accurately. It also includes a rate adjustment algorithm and a modified excursion detection algorithm to estimate ABW more accurately and less intrusively. Experimental results using an Android device over a 4G/LTE network and a testbed show the technique outperforms other state-of-the-art methods in terms of accuracy, intrusiveness, and convergence time.
This Spanish lesson focuses on vocabulary words related to items found in a house. The objectives are for students to identify, spell, and pronounce Spanish words for furniture and other household items with 90% accuracy. Students will practice filling in blanks, matching words to definitions, creating a vocabulary book with a partner, and completing sentences using the new vocabulary words. The lesson materials include worksheets, a PowerPoint, stapler, and scissors, and is designed to take a full class period.
The document discusses creating a virtual heritage site based on Seth's fictional Canadian town of Dominion City from his comics. It analyzes how Seth represents the town through multiple iterations over time in the comics versus his cardboard models that lack narrative. The proposed website would allow users to manipulate and tag objects from Dominion City across different media, encouraging new connections and narratives rather than a linear tour. This emphasizes personal engagement with the ephemera of the site's past over rigidly preserving its physical architecture.
This document provides an overview and introduction to waste-to-energy technology. It discusses the historic development of waste-to-energy, outlines the presentation topics which include the environmental performance and impacts of waste-to-energy facilities as well as responses to common myths. The document also notes that waste-to-energy has evolved over 100 years, there are now almost 1000 well-functioning plants worldwide, and waste-to-energy supports waste reduction and recycling while efficiently generating energy and reducing landfill volumes.
TREATMENT BY ALTERNATIVE METHODS OF REGRESSION GAS CHROMATOGRAPHIC RETENTION ...ijsc
The study treated two closer alternative methods of which the principal characteristic: a non-parametric
method (the least absolute deviation (LAD)) and a traditional method of diagnosis OLS.This was applied to
model, separately, the indices of retention of the same whole of 35 pyrazines (27 pyrazines with 8 other
pyrazines in the same unit) eluted to the columns OV-101 and Carbowax-20M, by using theoretical
molecular descriptors calculated using the software DRAGON. The detection of influential observations for
non-parametric method (LAD) is a problem which has been extensively studied and offers alternative
dicapproaches whose main feature is the robustness.here is presented and compared with the standard
least squares regression .The comparison between methods LAD and OLS is based on the equation of the
hyperplane, in order to confirm the robustness thus to detect by the meaningless statements and the points
of lever and validated results in the state approached by the tests statistics: Test of Anderson-Darling,
shapiro-wilk, Agostino, Jarque-Bera, graphic test (histogram of frequency) and the confidence interval
thanks to the concept of robustness to check if the distribution of the errors is really approximate.
Rapid increases in information technology also changed the existing markets and transformed them into emarkets
(e-commerce) from physical markets. Equally with the e-commerce evolution, enterprises have to
recover a safer approach for implementing E-commerce and maintaining its logical security. SOA is one of
the best techniques to fulfill these requirements. SOA holds the vantage of being easy to use, flexible, and
recyclable. With the advantages, SOA is also endowed with ease for message tampering and unauthorized
access. This causes the security technology implementation of E-commerce very difficult at other
engineering sciences. This paper discusses the importance of using SOA in E-commerce and identifies the
flaws in the existing security analysis of E-commerce platforms. On the foundation of identifying defects,
this editorial also suggested an implementation design of the logical security framework for SOA supported
E-commerce system.
You can earn money by completing online surveys for survey companies. Some of the top survey sites recommended in the document include Surveysavvy, Brandinstitute, Mysurvey, Opinionoutpost, Acop, Mindfieldonline, Esearch, Ithinkinc, Beginsurvey, and Surveyspot. Completing surveys on these sites can result in cash payments depending on the survey instructions and level of completion.
MARKOV CHAIN AND ADAPTIVE PARAMETER SELECTION ON PARTICLE SWARM OPTIMIZERijsc
Particle Swarm Optimizer (PSO) is such a complex stochastic process so that analysis on the stochastic
behavior of the PSO is not easy. The choosing of parameters plays an important role since it is critical in
the performance of PSO. As far as our investigation is concerned, most of the relevant researches are
based on computer simulations and few of them are based on theoretical approach. In this paper,
theoretical approach is used to investigate the behavior of PSO. Firstly, a state of PSO is defined in this
paper, which contains all the information needed for the future evolution. Then the memory-less property of
the state defined in this paper is investigated and proved. Secondly, by using the concept of the state and
suitably dividing the whole process of PSO into countable number of stages (levels), a stationary Markov
chain is established. Finally, according to the property of a stationary Markov chain, an adaptive method
for parameter selection is proposed.
Minggu 1 media pembelajaran pendahuluan 2013 pbiArie Susanti
Dokumen tersebut memberikan informasi tentang teknologi pembelajaran pada Universitas Ahmad Dahlan tahun 2013. Terdapat sarana komunikasi seperti forum dan Facebook untuk konsultasi, sharing informasi, dan pengumuman. Metode pembelajaran menggunakan PAKEM yaitu pembelajaran aktif, kreatif, efektif, dan menyenangkan. Mahasiswa membuat portofolio pekerjaan mereka dan dosen berperan merencanakan pembelajaran serta menilai mahasiswa.
Multisectoral Approaches for Improving Nutrition: Lessons from Global Experie...POSHAN-IFPRI
This document discusses multisectoral approaches for improving nutrition based on global experiences. It outlines that nutrition has multiple underlying and basic causes related to food, health, care, education and the economic structure. An effective approach requires all key actors to agree on the problem, causes and solutions; roles and responsibilities; and work to implement solutions. Global experiences demonstrate the importance of high-level political prioritization of nutrition, creating shared understanding among actors, establishing accountability, and strong leadership at all levels.
Recopilación de algunos artículos, comentarios, etc. de periodistas y personas conocedoras del proceso que llevó a España a aprobar su Constitución de 1978
Working multisectorally to improve maternal and child nutrition in India: Odi...POSHAN-IFPRI
The document summarizes Odisha's strategies for improving maternal and child nutrition through multisectoral collaboration. Key strategies include strengthening delivery systems like ICDS, collaborating across sectors like health and agriculture, decentralizing nutrition programs through self-help groups, and targeting vulnerable groups in high burden districts through district-specific planning. Mechanisms for convergence include nutrition councils, joint monitoring committees, and engaging communities through mothers' committees and growth monitoring. The impact of these efforts is seen in improved indicators for infant and young child feeding practices, immunization coverage, and reduced malnutrition according to survey data.
Working multisectorally to improve maternal and child nutrition in India: The...POSHAN-IFPRI
This document discusses malnutrition in India and proposes multi-sectoral solutions. Some key points:
- India has high levels of malnutrition, with over 40% of underweight children globally. Malnutrition has multiple causes including poverty, lack of access to water/sanitation, and lack of nutrition awareness.
- Malnutrition affects all ages and is intergenerational - with undernourished mothers more likely to have low birth weight babies who become undernourished children.
- A multi-sectoral approach is needed that addresses the various physical, socioeconomic, governance and behavioral causes. Key sectors include women and child development, health, food, agriculture, education and rural development.
- Proposed essential
Multisectoral Approach to Nutrition: India, Presentation by Suneetha KadiyalaPOSHAN-IFPRI
India has taken several steps towards a multisectoral approach to nutrition over the past few decades:
1) The 1993 National Nutrition Policy called for inter-ministerial coordination and a National Plan of Action was adopted in 1995.
2) A National Nutrition Mission was established in 2003 to implement the nutrition policies and plans.
3) In 2010, a multistakeholder retreat recommended a multisectoral program for maternal and child malnutrition prevention in high burden districts.
4) Several states have also established nutrition missions to coordinate multisectoral efforts, though operational details are still lacking in many cases.
The document discusses managed security services and the benefits they can provide. It notes that threats are increasing as technology makes fraud easier to commit. It then discusses how managed security services can help by analyzing security data in real time, detecting threats, and freeing up companies to focus on their core business instead of security administration. However, it cautions that these services require careful planning, resources, and don't replace the need for internal security processes. The document provides advice on selecting a managed security provider based on needs.
Fault detection of motorcycles using the Slopes of the estimated pseudospectr...ijcsa
This document describes a methodology for detecting faults in motorcycles using the slopes of estimated pseudospectra of the sounds produced. The methodology involves recording motorcycle sounds, segmenting the sounds, estimating the pseudospectrum using MUSIC algorithm, extracting features from the pseudospectrum slopes, and classifying the sounds as healthy or faulty using an artificial neural network classifier. Experiments showed the method achieved an average accuracy of 85% in fault detection. The study could be extended to localize specific faults in motorcycle subsystems.
Describe The Main Functions Of Each Layer In The Osi Model...Amanda Brady
Tone injection is a technique used to increase the constellation size of a signal constellation. It works by mapping each point in the original constellation to multiple equivalent points in an expanded constellation. This allows for embedding additional information by substituting points, improving spectral efficiency. However, it also increases implementation complexity and may degrade performance due to increased decision regions. Tone injection is useful for applications requiring high data rates within bandwidth constraints.
EXPLORING SOUND SIGNATURE FOR VEHICLE DETECTION AND CLASSIFICATION USING ANNijsc
This paper attempts to explore the possibility of using sound signatures for vehicle detection and
classification purposes. Sound emitted by vehicles are captured for a two lane undivided road carrying
moderate traffic. Simultaneous arrival of different types vehicles, overtaking at the study location, sound of
horns, random but identifiable back ground noises, continuous high energy noises on the back ground are
the different challenges encountered in the data collection. Different features were explored out of which
smoothed log energy was found to be useful for automatic vehicle detection by locating peaks. Melfrequency
ceptral coefficients extracted from fixed regions around the detected peaks along with the
manual vehicle labels are utilised to train an Artificial Neural Network (ANN). The classifier for four
broad classes heavy, medium, light and horns was trained. The ANN classifier developed was able to
predict categories well.
EXPLORING SOUND SIGNATURE FOR VEHICLE DETECTION AND CLASSIFICATION USING ANNijsc
This document summarizes a study that explored using sound signatures to detect and classify vehicles. The researchers collected audio recordings of vehicle sounds on a road carrying moderate traffic. They analyzed the recordings and identified smoothed log energy as useful for automatic vehicle detection by locating peaks. Mel-frequency cepstral coefficients extracted from regions around detected peaks, along with manual labels, were used to train an artificial neural network classifier for four broad vehicle classes. The trained ANN was able to predict vehicle categories with an accuracy of around 67% based on testing with unlabeled vehicle sound data. The study demonstrated the potential of using sound signatures for vehicle detection and classification.
Exploring Sound Signature for Vehicle Detection and Classification Using ANN ijsc
This paper attempts to explore the possibility of using sound signatures for vehicle detection and classification purposes. Sound emitted by vehicles are captured for a two lane undivided road carrying moderate traffic. Simultaneous arrival of different types vehicles, overtaking at the study location, sound of horns, random but identifiable back ground noises, continuous high energy noises on the back ground are the different challenges encountered in the data collection. Different features were explored out of which smoothed log energy was found to be useful for automatic vehicle detection by locating peaks. Melfrequency ceptral coefficients extracted from fixed regions around the detected peaks along with the manual vehicle labels are utilised to train an Artificial Neural Network (ANN). The classifier for four broad classes heavy, medium, light and horns was trained. The ANN classifier developed was able to predict categories well.
IRJET- Emotion recognition using Speech Signal: A ReviewIRJET Journal
This document provides a review of speech emotion recognition techniques. It discusses how speech emotion recognition systems work, including common features extracted from speech like MFCCs and LPC coefficients. Classification techniques used in these systems are also examined, such as DTW, ANN, GMM, and K-NN. The document concludes that speech emotion recognition could be useful for applications requiring natural human-computer interaction, like car systems that monitor driver emotion or educational tutorials that adapt based on student emotion.
A coustic pseudo spectrum based fault Localization in motorcyclesijcsa
This document presents a methodology for fault localization in motorcycles using acoustic pseudospectrum analysis. Sound samples are collected from healthy and faulty motorcycles, and segmented. Pseudospectra are estimated from the segments using MUSIC algorithm. Chaincodes are constructed by tracing pseudospectrum gradients. Eigenvectors of the chaincode matrices are used as features for an artificial neural network classifier. The methodology achieves 88% classification accuracy in identifying six common faults: mis-set valves, faulty crank, cylinder problems, muffler leakage, silencer leakage, and timing chain issues.
Acoustic Scene Classification by using Combination of MODWPT and Spectral Fea...ijtsrd
Acoustic Scene Classification ASC is classified audio signals to imply about the context of the recorded environment. Audio scene includes a mixture of background sound and a variety of sound events. In this paper, we present the combination of maximal overlap wavelet packet transform MODWPT level 5 and six sets of time domain and frequency domain features are energy entropy, short time energy, spectral roll off, spectral centroid, spectral flux and zero crossing rate over statistic values average and standard deviation. We used DCASE Challenge 2016 dataset to show the properties of machine learning classifiers. There are several classifiers to address the ASC task. We compare the properties of different classifiers K nearest neighbors KNN , Support Vector Machine SVM , and Ensembles Bagged Trees by using combining wavelet and spectral features. The best of classification methodology and feature extraction are essential for ASC task. In this system, we extract at level 5, MODWPT energy 32, relative energy 32 and statistic values 6 from the audio signal and then extracted feature is applied in different classifiers. Mie Mie Oo | Lwin Lwin Oo "Acoustic Scene Classification by using Combination of MODWPT and Spectral Features" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-5 , August 2019, URL: https://www.ijtsrd.com/papers/ijtsrd27992.pdfPaper URL: https://www.ijtsrd.com/computer-science/multimedia/27992/acoustic-scene-classification-by-using-combination-of-modwpt-and-spectral-features/mie-mie-oo
Fault Diagnostics of Rolling Bearing based on Improve Time and Frequency Doma...ijsrd.com
The neural network based approaches a feed forward neural network trained with Back Propagation technique was used for automatic diagnosis of defects in bearings. Vibration time domain signals were collected from a normal bearing and defective bearings under various speed conditions. The signals were processed to obtain various statistical parameters, which are good indicators of bearing condition, then best features are selected from graphical method and these inputs were used to train the neural network and the output represented the bearing states. The trained neural networks were used for the recognition of bearing states. The results showed that the trained neural networks were able to distinguish a normal bearing from defective bearings with 83.33 % reliability. Moreover, the network was able to classify the bearings into different states with success rates better than those achieved with the best among the state-of-the-art techniques.
Rolling Element Bearing Condition Monitoring using Filtered Acoustic Emission IJECEIAES
This document summarizes research on using filtered acoustic emission signals to monitor the condition of rolling element bearings. The researchers collected acoustic emission data from both healthy and defective bearings. They applied three active noise cancellation techniques (LMS, EMD, wavelet) to filter the noisy acoustic signals and compared their performance based on SNR and MSE, finding that EMD provided the best filtering. Time, frequency, and time-frequency analyses were then used to analyze the filtered signals and diagnose bearing faults. The analyses clearly showed differences between healthy and defective bearings and could detect different types of defects. The research demonstrates that acoustic emission monitoring combined with noise filtering is effective for rolling element bearing condition monitoring and fault diagnosis.
Multi-Target Classification Using Acoustic Signatures in Wireless Sensor Netw...CSCJournals
This document surveys techniques for classifying vehicles based on acoustic signals detected by wireless sensor networks. It discusses three main areas: feature extraction, classification techniques, and information fusion. For feature extraction, it describes techniques in the time, frequency and time-frequency domains like FFT, wavelet transforms, MFCC. It evaluates classifiers like ANN, SVM, HMM using features from spectrum analysis and wavelet packet transform. Finally, it discusses information fusion techniques like majority voting, Bayesian methods to improve classification performance.
Signal Processing and Soft Computing Techniques for Single and Multiple Power...idescitation
This paper reviews various techniques used for detection and classification of power quality events. It divides the techniques into those for single events versus combined events. For single events, techniques like wavelet transforms, statistical estimators, and intelligent methods are discussed. For combined events, papers addressing harmonic disturbances combined with others are summarized. The paper also includes a table providing a comparative analysis of several references based on aspects like classification accuracy, noise tolerance, and computation time. It concludes that the field is growing and future work could address techniques for large data and detection of both single and combined disturbances simultaneously.
Automatic recognition of speech using computers is a challenging issue. This paper describes a technique that uses Auto associative Neural Network (AANN) to recognized speech based on features using Sonogram. Modeling techniques such as AANN were used to model each individual word which is trained to the system. Each isolated word Segment using Voice Activity Detection (VAD) from the test sentence is matched against these models for finding the semantic representation of the test input speech. Experimental results of AANN shows good performance in recognized rate.
IRJET- Performance Evaluation of DOA Estimation using MUSIC and Beamformi...IRJET Journal
This document presents a simulation study comparing the MUSIC algorithm and LMS adaptive beamforming algorithm for direction of arrival (DOA) estimation and beamforming in a smart antenna system. The MUSIC algorithm uses eigendecomposition to estimate the DOA of multiple signals and finds the position location of the desired user. The LMS algorithm then adapts the beam pattern by adjusting weights to maximize gain towards the desired user while nulling interference from other directions. The simulation results show sharp peaks in the MUSIC spectrum to accurately locate the desired user and deep nulls in the LMS beam pattern to suppress interference.
This document summarizes a study that uses signal processing and optimization techniques to detect faults in roller bearings. Specifically, it applies minimum entropy deconvolution (MED) and the Teager-Kaiser energy operator (TKEO) to enhance the discrimination of defect-induced signals in bearing vibration data. It also uses empirical mode decomposition (EMD) to decompose vibration signals into intrinsic mode functions (IMFs), and a genetic algorithm to optimize the weights of IMFs to further improve fault detection sensitivity as measured by kurtosis values. Experimental results on a test bearing show that the EMD-GA approach successfully improved the resolution of detectability between a roller bearing with defect and an intact bearing system.
Microstrip circular patch array antenna for electronic toll collectioneSAT Journals
Abstract Electronic toll collection will reduce the wastage of time at toll gates and collects the money in fast manner. At toll collection stations on the highway, the automatic toll collection will collect the money using one sensing antenna and the signal processing unit with the help of computer interfacing. The technology will reduce the man power usage, time and cost with customer friendly environment. The present paper deals with the design and analysis of electronic toll collection antenna operating at 5.8 GHz with moderate gain and bandwidth. An array of 6X6 elements patch antenna is used in this design, which increased the gain considerably for the desired operation. Keywords: Parallel Feeding, Electronic Toll Collection (ETC), Microstrip Patch Array.
Microstrip circular patch array antenna for electronic toll collectioneSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Channel Estimation in MIMO OFDM Systems with Tapped Delay Line ModelIJCNCJournal
The continuous increase in the user demands fornew-generation communication systems, is making the wireless channel more complex and challenging for estimation, developing a simulation model for the channel,and evaluating the performance of different MIMO systems. In this work, a simulation model for multipath fading channels in wireless communication is performed. The model includes a selection of typical Tapped-Delay-Line channel models that can be implemented to reproduce the effects of representative channel distortion and interference. Based on the simulation results, the proposed method exhibits accurate channel estimation performance for frequency-selective fading channels. The proposed work employed LS, MMSE, and ML methods for channel estimation, using 16 and 32 pilots and fixed pilot locations in each frame. Results are obtained for 4x4, 8x8, 16x16, 16x8, and 16x4 MIMO systems and tapped delay line systems.
Channel Estimation in MIMO OFDM Systems with Tapped Delay Line ModelIJCNCJournal
The continuous increase in the user demands fornew-generation communication systems, is making the wireless channel more complex and challenging for estimation, developing a simulation model for the channel,and evaluating the performance of different MIMO systems. In this work, a simulation model for multipath fading channels in wireless communication is performed. The model includes a selection of typical Tapped-Delay-Line channel models that can be implemented to reproduce the effects of representative channel distortion and interference. Based on the simulation results, the proposed method exhibits accurate channel estimation performance for frequency-selective fading channels. The proposed work employed LS, MMSE, and ML methods for channel estimation, using 16 and 32 pilots and fixed pilot locations in each frame. Results are obtained for 4x4, 8x8, 16x16, 16x8, and 16x4 MIMO systems and tapped delay line systems.
This document summarizes a research paper on using time-domain signal cross-correlation for spectrum sensing in cognitive radio systems applied to vehicular ad-hoc networks (VANETs). It aims to address spectrum scarcity issues in VANETs by allowing vehicles to opportunistically access TV white space spectrum when licensed spectrum is unavailable. The time-domain symbol cross-correlation technique is analyzed for spectrum sensing performance over Rayleigh fading channels. Analytical expressions for average miss detection probability are derived and simulation results show the probability of miss detection decreases with increasing SNR and number of secondary users. The time-domain symbol cross-correlation method provides good spectrum sensing performance at low SNRs for cognitive radio in VANETs.
Similar to Classification of vehicles based on audio signals (20)
Introduction- e - waste – definition - sources of e-waste– hazardous substances in e-waste - effects of e-waste on environment and human health- need for e-waste management– e-waste handling rules - waste minimization techniques for managing e-waste – recycling of e-waste - disposal treatment methods of e- waste – mechanism of extraction of precious metal from leaching solution-global Scenario of E-waste – E-waste in India- case studies.
6th International Conference on Machine Learning & Applications (CMLA 2024)ClaraZara1
6th International Conference on Machine Learning & Applications (CMLA 2024) will provide an excellent international forum for sharing knowledge and results in theory, methodology and applications of on Machine Learning & Applications.
Advanced control scheme of doubly fed induction generator for wind turbine us...IJECEIAES
This paper describes a speed control device for generating electrical energy on an electricity network based on the doubly fed induction generator (DFIG) used for wind power conversion systems. At first, a double-fed induction generator model was constructed. A control law is formulated to govern the flow of energy between the stator of a DFIG and the energy network using three types of controllers: proportional integral (PI), sliding mode controller (SMC) and second order sliding mode controller (SOSMC). Their different results in terms of power reference tracking, reaction to unexpected speed fluctuations, sensitivity to perturbations, and resilience against machine parameter alterations are compared. MATLAB/Simulink was used to conduct the simulations for the preceding study. Multiple simulations have shown very satisfying results, and the investigations demonstrate the efficacy and power-enhancing capabilities of the suggested control system.
Harnessing WebAssembly for Real-time Stateless Streaming PipelinesChristina Lin
Traditionally, dealing with real-time data pipelines has involved significant overhead, even for straightforward tasks like data transformation or masking. However, in this talk, we’ll venture into the dynamic realm of WebAssembly (WASM) and discover how it can revolutionize the creation of stateless streaming pipelines within a Kafka (Redpanda) broker. These pipelines are adept at managing low-latency, high-data-volume scenarios.
Literature Review Basics and Understanding Reference Management.pptxDr Ramhari Poudyal
Three-day training on academic research focuses on analytical tools at United Technical College, supported by the University Grant Commission, Nepal. 24-26 May 2024
DEEP LEARNING FOR SMART GRID INTRUSION DETECTION: A HYBRID CNN-LSTM-BASED MODELgerogepatton
As digital technology becomes more deeply embedded in power systems, protecting the communication
networks of Smart Grids (SG) has emerged as a critical concern. Distributed Network Protocol 3 (DNP3)
represents a multi-tiered application layer protocol extensively utilized in Supervisory Control and Data
Acquisition (SCADA)-based smart grids to facilitate real-time data gathering and control functionalities.
Robust Intrusion Detection Systems (IDS) are necessary for early threat detection and mitigation because
of the interconnection of these networks, which makes them vulnerable to a variety of cyberattacks. To
solve this issue, this paper develops a hybrid Deep Learning (DL) model specifically designed for intrusion
detection in smart grids. The proposed approach is a combination of the Convolutional Neural Network
(CNN) and the Long-Short-Term Memory algorithms (LSTM). We employed a recent intrusion detection
dataset (DNP3), which focuses on unauthorized commands and Denial of Service (DoS) cyberattacks, to
train and test our model. The results of our experiments show that our CNN-LSTM method is much better
at finding smart grid intrusions than other deep learning algorithms used for classification. In addition,
our proposed approach improves accuracy, precision, recall, and F1 score, achieving a high detection
accuracy rate of 99.50%.
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsVictor Morales
K8sGPT is a tool that analyzes and diagnoses Kubernetes clusters. This presentation was used to share the requirements and dependencies to deploy K8sGPT in a local environment.
Understanding Inductive Bias in Machine LearningSUTEJAS
This presentation explores the concept of inductive bias in machine learning. It explains how algorithms come with built-in assumptions and preferences that guide the learning process. You'll learn about the different types of inductive bias and how they can impact the performance and generalizability of machine learning models.
The presentation also covers the positive and negative aspects of inductive bias, along with strategies for mitigating potential drawbacks. We'll explore examples of how bias manifests in algorithms like neural networks and decision trees.
By understanding inductive bias, you can gain valuable insights into how machine learning models work and make informed decisions when building and deploying them.
Low power architecture of logic gates using adiabatic techniquesnooriasukmaningtyas
The growing significance of portable systems to limit power consumption in ultra-large-scale-integration chips of very high density, has recently led to rapid and inventive progresses in low-power design. The most effective technique is adiabatic logic circuit design in energy-efficient hardware. This paper presents two adiabatic approaches for the design of low power circuits, modified positive feedback adiabatic logic (modified PFAL) and the other is direct current diode based positive feedback adiabatic logic (DC-DB PFAL). Logic gates are the preliminary components in any digital circuit design. By improving the performance of basic gates, one can improvise the whole system performance. In this paper proposed circuit design of the low power architecture of OR/NOR, AND/NAND, and XOR/XNOR gates are presented using the said approaches and their results are analyzed for powerdissipation, delay, power-delay-product and rise time and compared with the other adiabatic techniques along with the conventional complementary metal oxide semiconductor (CMOS) designs reported in the literature. It has been found that the designs with DC-DB PFAL technique outperform with the percentage improvement of 65% for NOR gate and 7% for NAND gate and 34% for XNOR gate over the modified PFAL techniques at 10 MHz respectively.
1. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
DOI: 10.5121/ijsc.2015.6105 53
Classification of Vehicles Based on Audio Signals
using Quadratic Discriminant Analysis and High
Energy Feature Vectors
A. D. Mayvana
, S. A. Beheshtib
, M. H. Masoomc
a
Department of Electrical Engineering, Iran University of Science and
Technology, Tehran, Iran
b
Department of Electrical Engineering, Iran University of Science and
Technology, Tehran, Iran.
c
Department of Mechanical Engineering, BabolNoshirvani University of
Technology, Babol, Iran.
ABSTRACT
The focusof this paper is on classification of different vehicles using sound emanated from the vehicles. In
this paper,quadratic discriminant analysis classifies audio signals of passing vehicles to bus, car, motor, and
truck categories based on features such as short time energy, average zero cross rate, and pitch frequency of
periodic segments of signals. Simulation results show that just by considering high energy feature vectors,
better classification accuracy can be achieved due to the correspondence of low energy regions with noises
of the background. To separate these elements, short time energy and average zero cross rate are used
simultaneously.In our method,we have used a few features which are easy to be calculated in time domain
and enable practical implementation of efficient classifier. Although, the computation complexity is low,
the classification accuracy is comparable with other classification methodsbased on long feature vectors
reported in literature for this problem.
KEYWORD
Classification accuracy; Periodic segments; Quadratic Discriminant Analysis; Separation criterion; Short
time analysis.
1. INTRODUCTION
Vehicle identification while it is in motion is a prerequisite for traffic and speed management,
classified vehicle count, traffic signal time optimization, gap/ headway measurement, and military
purposes. Moving vehicles affect the environment in different ways. Vehicle emits heats, sounds,
and magnetic field. There are many approaches investigated vehicle identification based on
different kinds of signals. Image processing techniques are used to classify vehicles under the real
time traffic management and for Intelligent Transportation Systems (ITS). Inductive loops based
2. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
54
systems are widely used for determination of vehicle counts [1]. The most promising approach
for vehicle identification is the one that is based on acoustic signals. Moving vehicles emit
characteristic sounds. These sounds are generated from moving parts, frictions, winds, emissions,
tires,etc. Assuming that similar vehicles which have the same working conditions generate the
same sounds; then these sounds can be used to classify vehicles [2]. Vehicle classification based
on sound signals have already attempted by researchers. Paulrajet al. (2013) [3] used
autoregressive modeling algorithm for the analysis to extract the features from the recorded
vehicle signals. Probabilistic neural network (PNN) models are developed to classify the vehicle
type and its distance. Nooralahiyan et al. (1997) [4] used a directional microphone connected to a
DAT (Digital Audio Tape) recorder. The digital signal was pre-processed by LPC (Linear
Predictive Coding) parameter conversion based on autocorrelation analysis. A Time Delay Neural
Network (TDNN) was chosen to classify individual travelling vehicles based ontheir speed
independent acoustic signature to four broad categories: buses or Lorries, small or large saloons,
various types of motorcycles, and light goods vehicles or vans. Ghiurcau and Rusu(2009) [5]
presented a vehicle sound classification system using time encoded signal processing and
recognition (TESPAR) method combined with the archetypes technique implemented on the
Matlab platform. The experimental results show the efficiency of the TESPAR method when
dealing with vehicle sounds. Effectiveness of features set is usually measured by how well they
represent the signal; however, features should also satisfy several conditions: their number should
be small (less than 15); also,to enable practical implementation they should be easy to calculate.
There is a number of feature extraction methods used in vehicle classification. Some of them
produce too many features for a single input vector like estimation of Power Spectrum Density
[6], [7], [8], and some are too complicated like Principal Component Analysis [9], [7].
Nevertheless, no feature extraction method fulfills all our requirements, and there is no systematic
way to select best feature according to our criterion. Harmonic line, Schur coefficients, and MEL
filters, were presented in [10] each relating to properties of vehicle audio. Methods were
compared in context of their separability and correct classification rate. Aljaafreh and Dong
(2010) [2] investigated two feature extraction methods for acoustic signals from moving vehicles.
The first one is based on spectrum distribution and the second one on wavelet packet transform.
They evaluated the performance of different classifiers such as K-nearest neighbor algorithm
(KNN) and support vector machine (SVM). It is found that for vehicle sound data, a discrete
spectrum based feature extraction method outperforms wavelet packet transform method.
Experimental results verified that support vector machine is an efficient classifier for
vehicles.Other spectrum based feature extraction can be found in [11], [12], [13], [14], and [15].
Similarly, Discrete Wavelet Transform (DWT) is used in [16] and [17] to extract features using
statistical parameters and energy content of the wavelet coefficients.
This paper describes an algorithm to classify audio signals of vehicles. Quadratic discriminant
analysis uses feature vectors of periodic segments with elements: short time energy, average zero
cross rate, and pitch frequency to distinguish between vehicles’ signals. If we separate feature
vectors with high energy from others properly and train classifier with these vectors, better
classification accuracy will be achieved due to the correspondence of low energy regions and
noises of the background, as simulation results confirm. Obviously high energy feature vectors
have less zero cross rate. Considering this point is the basic idea for separation criterion. Our
innovation in separation is assuming vectors by energy larger than ܧ[ܧ], average of all
segments’ energy, and ZCR less than ܼ[ܧ],average of all segments’ ZCR, are high energy
elements. We have used three features which are easy to be calculated in time domain, which
enable practical implementation of efficient classifier. The cost paid is lower, but still satisfactory,
3. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
55
classification accuracy than other classification applications reported in literature for this
problem. Sections of paper are as follow: After introduction first,we determine important
parameters and features.Then we study quadratic discriminant analysis and high energy element
separation criterion in section III. Simulation results show improvement in classification accuracy
for separation criterion. Finally, the paper ends by conclusion.
2. FEATURE EXTRACTION
Changing the characteristics of an audio signal over time seems normal. For example, Changes in
the signal’s peak value or rate of sign changes of amplitude are such a time varying
features.Although these changes exist in signals, short time analysis is a method to illustrate them
asobvious way. Basic assumption of this method is based on these changes occur slowly; then
wedivide signal into smaller parts and features are extracted.
2.1. Short Time Energy
This feature indicates how the signal amplitude changes over the time which is defined as
follows:
ܧ = ∑ [ݔሺ݉ሻݓሺ݊ − ݉ሻ]ଶ∞
ୀି∞ (1)
Where ݔሺ݊ሻis the audio signal and ݓሺ݊ሻis the window that slides along the audio sequence
selecting the interval to be involved in the computation.
2.2. Short Time Average Zero Cross Rate
This feature indicates how the signal sign changes over the time which is defined as follows:
ܼ = ∑ |ݔ[݊݃ݏሺ݉ሻ] − ݔ[݊݃ݏሺ݉ − 1ሻ]|∞
ୀି∞ ݓሺ݊ − ݉ሻ (2)
Where
ݓሺ݊ሻ = 1 2⁄ 0 ≤ ݊ ≤ ܰ − 1
= 0 ݐℎ݁݁ݏ݅ݓݎ (3)
The rate at which signal sign changes occur is a simple measure of the frequency content of a
signal. This is particularly true for narrowband signals. For example, a sinusoidal signal of
frequency ܨ, sampled at a rateܨ௦, has 2 ܨ ܨ௦⁄ average rate of zero-crossings.Thus, the average
ZCR gives a resonable way to estimate the frequency of a sine wave. Audio signals are
broadband signals and the interpretation of average ZCR is therefore much less precise. However,
rough estimates of spectral properties can be obtained using a represantation based onthe short
time energy and zero cross rate together.This representation was proposed by Reddy, and studied
by Vicens [18] as the basis for a large-scale speech recognition system.
4. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
56
2.3. Pitch Frequency
Pitch period estimation (or fundamental frequency) is one of the most important problems in
speech processing. Pitch detectors are used in vocoders [19], speaker identification and
verification systems [20, 21]. Because of its importance, many solutions to this problem have
been proposed [22]. In this paper we use the fact thatthe autocorrelation function for periodic
segments attains a maximum at samples 0, ±ܶ,±2ܶ,… where T is pitch period. Since, there may
be exist other autocorrelation peaks except those due to the periodicity, the simple procedure of
picking the largest peak at sampleܶin the autocorrelation will failto estimate Pitch period. To
avoid this problem it is useful to process the signal so as to make periodicity more prominent
while suppressing other distracting features of the signal. Spectrum flattener with the objective to
bring each harmonic to the same amplitude level as in the case of a periodic impulse trainhas
been used. Specifically, center clipped signalhas been used in computing autocorrelation function.
In the scheme proposed,similar to [23], the center clipped signal is obtained by following
nonlinear transformation
For samples above ܥ, the output is equal to the input minus the clipping level. For samples below
the clipping level the output is zero.We set clipping level, ܥ, equal to 68% of the minimum of
two maximum amplitudes found in both the first third and last third of the audiosegment;denoted
respectively by݉ܽݔଵ and ݉ܽݔଶ. However, for the un-periodic segment, there are no strong
autocorrelation periodicity peaks and it seems to be a high frequency noise-like waveform [22].
We assume segments with weak peaks below 30 % clipped signal’s energy, ܧ
=
∑ [ݔሺ݉ሻݓሺ݊ − ݉ሻ]ଶ∞
ୀି∞ ,are un-periodic. Finally pitch frequencies of frames smoothed
through median operation. Following figures present features.
−ܥܥ
Fig.1: center clipping function
5. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
57
Fig.2: Energy for Bus in first row for three differentsounds, for Car in the second row, for Motor in the
third row, and for Truck in the fourth row.
Fig.3:ZCR for Bus in first row for three different sounds, for Car in the second row, for Motor in the third
row, and for Truck in the fourth row.
6. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
58
Fig.4: Pitch Frequency for Bus in first row for three different sounds, for Car in the second row, for Motor
in the third row, and for Truck in the fourth row.
Features are extracted from rectangular windowed segment of length 15 Milliseconds and
overlapping of 5 Milliseconds between consecutive windows. Feature vectors of periodic
segments are used for training and classification. After determining the important features and
parameters for audio signal modeling, we will review quadratic discriminant analysis and high
energy separation criterion will be described.
3. CLASSIFIER AND SEPARATION CRITERION
Now the discriminant boundaries between classes should be determined. Quadratic
discriminantanalysis is used to discriminate the classes. The classifier that separate K classes has
K quadratic functions as follow [24]:
ݕሺܺሻ = ்ܺ
ܳܺ + ்ܸ
ܺ + ݒ (4)
The coefficientݒis constant coefficient,ܸis vector of linear coefficients, and ܳ is matrix
containing quadratic coefficients. Feature vectorX will belong to class k, ifݕሺܺሻ > ݕሺܺሻ; ∀݆ ≠
݇.In classification step, after feature vectors were extracted, we are able to define class of each
vector separately. Finally, signal will allocate to a classwhichachieved great supply in feature
vector classification phase. In other words, probability of belonging to each class is proportional
to abundance of that class in allocated labels to vectors by quadratic discriminant analysis.
7. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
59
3.1. High Energy Vector Separation Criterion
Up to now all periodic vectors are considered in training and classification. However, low energy
regions corresponding with noises of the background consist of high frequency variations may
not be suited for vehicle classification. Hence by ignoring mentioned vectors, an improvement in
classification accuracy can be achieved.On the other hand, it can be observed that when
amplitude is high zero cross rate decreases. So, high energy elements have low zero cross rate,
and low energy vectors have fast sign alternation. A criterion for separation elements with high
energy and low zero cross rate can be defined as follow:
ܧ > ߙ. ܧ[ܧ]Andܼ < ߞ. ܼ[ܧ] (5)
Where ܧ[ܧ]and ܼ[ܧ]denote average of all segments’ energy, ZCR. The coefficientα is
more than 1, andߞis less than 1. The more α and the less ߞ than unit, the less data
coincides in condition. In this paper α = ߞ = 1are chosen. Figure 5 presents a block
diagram of our method.
Fig.5: Block diagram of the proposed scheme for vehicle classification
8. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
60
4. RESULTS AND DISCUSSION
Simultaneous arrival of different types vehicles, overtaking at the study location, and random
back ground noises were the different challenges encountered in the data collection. Sound
emitted by vehicles was captured for a divided road carrying moderate traffic. Roadside recording
was performed on clear sunny working days. The recordings’ sites were selected to capture the
sound of moving vehicles without much acceleration change.Environmental back ground noise is
at moderate level. The conditions under which data set is recorded are realistic.Our algorithm is
implemented on the Matlab7.10.0 (R2010a) platform.
Preprocessing: Each signal is normalized such that −1 ≤ ݔሺ݉ሻ ≤ 1. Windowed signal segments
are filtered by 4௧
degree Butterworth low pass filter.
The data set includes four classes; bus, car, motor, and truck. All signals sampled at 11025 rate.
Each signal is segmented by rectangular window of length 15 milliseconds and overlapping of 5
milliseconds between consecutive windows. Vectors containing short time energy, average zero
cross rate, and pitch frequency are calculated for each segment and those with 0 pitch frequencies
are removed. We used k-Fold cross validation with k=10 in our applications. In this method data
is randomly divided into a test and training setk different times [25] and the average error across
allk trials is computed. Table 1 provides the values of some important parameters employed in
this work. Classification accuracies for following classifier have been provided in table 2.
• least square classifier,
• k-nearestneighbor classifier,
• quadratic and linear discriminant analysiswhen training and testing have been done just
based on vectors which be able to satisfy the (5) conditions;denoted respectively by
ܳܣܦ∗∗
and ܣܦܮ∗∗
,
• SVM classifier with harmonic line based feature vectors [10],
• SVM classifier with Schur filter coefficients based feature vectors [10],
• SVM classifier with Mel filter coefficients based feature vectors [10].
We see improvement in classification accuracy has been achieved forquadratic discriminant
analysis and separation criterion, ܳܣܦ∗∗
. The correct classification rate depends not only on the
feature extraction method, but also on the type of the classifier. This is the reason that why not we
see a considerable improvement for ܣܦܮ∗∗
.
Table 3 shows computation complexity in terms of feature space dimensionality for our proposed
algorithm and last 3 methods in table 2. In comparison, we should note that the accuracy of
ܳܣܦ∗∗
with 3 short time parameters based feature vectors is comparable with SVM with 16 Schur
coefficients. SVM with 12 Mel filter coefficients accuracy is 80% and accuracy of QDA∗∗
is
68.75%. Although the computation complexity of QDA∗∗
is low, but the classification accuracy is
lower thanSVM with Mel filter coefficients.
9. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
61
Signal sample rate 11025 sample/sec
Length of window for classification 165 sample
Overlap between windows for classification 55 sample
Type of window Rectangular
Preprocessor 4௧
degree ButterworthLPF
LPF’s cut off frequency 4 KHz
Clipping level 0.68 ∗ min {݉ܽݔଵ, ݉ܽݔଶ}
Maximum pitch frequency f<= 1000 Hz
Periodic/Un-periodic threshold 0.3 ∗ ܧ
Number of frames over which median operates 3
α and ζ 1
Number of cross fold validation 10
examples per class: [bus, car, motor, truck] [50, 70, 80, 40]
Table 1: Values of some important parameters
Classifier Accuracy (%)
Least Square 30
kNN, k=25, Cosine 52.09 ± 2.2
kNN, k=25, Euclidian 53.33 ± 3.83
LDA 46.66
ܣܦܮ∗∗ 49.58
QDA 56.25
ܳܣܦ∗∗ 68.75
SVM with harmonic line 83
SVM with Schur coefficients 68
SVM with Mel filter coefficients 80
Table 2: Classification accuracy
10. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
62
Feature extraction method Number of features
Short time analysis 3
harmonic lines 5
Schur coefficients 16
Mel filter coefficients 12
Table 3: Number of features
Additionally confusion matrix, contain information about actual and predicted classifications
done by ܳܣܦ∗∗
classification system, shown in table 4.
Actual Predicted
Bus vehicles Car vehicles Motor vehicles Truck vehicles
Bus vehicles 27 20 3 0
Car vehicles 0 70 0 0
Motor vehicles 16 12 48 4
Truck vehicles 9 11 0 20
Table 4: Representation of confusion matrix
Although recording conditions were realistic, but we should examine proposed algorithm when
environmental noises increase like rainy days.Since we use high energy feature vectors, ܳܣܦ∗∗
can save its effectiveness.
CONCLUSION
In the research reported in this paper, QDA with a criterion to separate high energy feature
vectors was applied on the task of classification of vehicles based on audio signals and the
simplest method with satisfactory accuracy was evaluated.
Short time energy, average zero cross rate, and pitch frequency made feature vectors; then,signals
were classified using quadratic discriminant analysis. We found that classification accuracy
improved from 56.25% to 68.75%just by considering components which have (5) conditionsdue
to the correspondence of low energy regions and noises of the background.At last, the obtained
accuracy was 68.75% by using three features while the 80% correct classification rate was
achievedby SVM with 12Mel coefficients [10].The results strongly suggest that proposed method
can aid the practical implementation.
11. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
63
REFERENCES
[1] J. George, and et al. ‘’ Exploring Sound Signature for Vehicle Detection and ClassificationUsing
ANN’’ International Journal on Soft Computing (IJSC) Vol.4, No.2, May 2013.
[2] A. Aljaafreh, and L. Dong ‘’an Evaluation of Feature Extraction Methods for VehicleClassification
Based on Acoustic Signals’’ International Conference on Networking, Sensing andControl (ICNSC),
2010.
[3] M.P. Paulraj, and et al. ‘’Moving Vehicle Recognition and Classification Based on TimeDomain
Approach‘’ Procedia Engineering, Volume 53, 2013, Pages 405–410.
[4] Y. Nooralahiyan, and et al. ‘’Field Trial of Acoustic Signature Analysis for VehicleClassification’’
Transportation Research Part C: Emerging Technologies, Volume 5, Issues 3–4,August–October
1997, Pages 165–177.
[5] M. V. Ghiurcau, C. Rusu, ‘’Vehicle Sound Classification Application and Low PassFiltering
Influence’’ In proceeding of International Symposium on Signals, Circuits andSystems, 2009, ISSCS
2009.
[6] M. Wellman, N. Srour, and D. Hillis, “Feature Extraction and Fusion of Acoustic andSeismic Sensors
for Target Identification,” in Society of Photo-Optical InstrumentationEngineers (SPIE) Conference
Series, ser. Society of Photo-Optical Instrumentation Engineers(SPIE) Conference Series, G. Yonas,
Ed, vol. 3081. SPIE, 1997, pp. 139–145.
[7] M. Wellman, “Acoustic Feature Extraction for a Neural Network Classifier.” DTICDocument, Tech.
Rep., 1997.
[8] M. F. Duarte and Y. H. Hu, “Vehicle Classification in Distributed Sensor Networks,” Journalof
Parallel and Distributed Computing, vol. 64, pp.826–838, 2004.
[9] H. Wu, M. Siegel, and P. Khosla, “Vehicle Sound Signature Recognition by FrequencyVector
Principal Component Analysis,” in Instrumentation and Measurement TechnologyConference, 1998.
IMTC/98.Conference Proceedings.IEEE, vol. 1, May 1998, pp. 429 –434vol.1.
[10] M. Gorski, J. Zarzycki, ‘’Feature Extraction in Vehicle Classification’’ InternationalConference on
Signals and Electronic Systems (ICSES), 2012.
[11] H. Wu, M. Siegel, and P. Khosla, “Distributed classification of acoustictargets wireless audio-sensor
networks,” Computer Networks, vol. 52,no. 13, pp. 2582–2593, Sep. 2008.
[12] ——, “Vehicle classification in distributed sensor networks,” Journalof Parallel and Distributed
Computing, vol. 64, no. 7, pp. 826–838,July 2004.
[13] Y. Seung S., K. Yoon G., and H. Choi, “Distributed and efficientclassifiers for wireless audio-sensor
networks,” in 5th InternationalConference on Volume, Apr. 2008.
[14] S. S. Yang, Y. G. Kim1, and H. Choi, “Vehicle identification usingdiscrete spectrums in wireless
sensor networks,” Journal of Networks,vol. 3, no. 4, pp. 51–63, Apr. 2008.
[15] H. Wu, M. Siegel, and P. Khosla, “Vehicle sound signature recognitionby frequency vector principal
component analysis,” IEEE Trans.Instrum. Meas., vol. 48, no. 5, pp. 1005–1009, Oct. 1999.
[16] C. H. C. K. R. E. G. G. R. and M. T. J, “Wavelet-based ground vehiclerecognition using acoustic
signals,” Journal of Parallel and DistributedComputing, vol. 2762, no. 434, pp. 434–445, 1996.
[17] A. H. Khandoker, D. T. H. Lai, R. K. Begg, and M. Palaniswami,“Wavelet-based feature extraction
for support vector machines forscreening balance impairments in the elderly,” vol. 15, no. 4, pp. 587–
597, 2007.
[18] P. J. Vicens, “Aspects of Speech Recognition by Computer,” Ph.D. Thesis, Stanford Univ., AI Memo
No. 85, Comp. Sci. Dept., Stanford Univ., 1969.
[19] J. L. Flanagan, Speech Analysis, Synthesis and Perception, 2nd
Ed., Springer Verlag, N.Y., 1972.
[20] B. S. Atal, “Automatic Speaker Recognition Based on Pitch Contours,” J. Acoust. Soc. Am., Vol. 52,
pp. 1687-1697, December 1972.
[21] A. E. Rosenberg and M. R. Sambur, “New Techniques for Automatic Speaker Verification,” IEEE
Trans. Acoust, Speech, and Signal Proc., Vol.23, pp. 169-176, April 1975.
[22] L. R. Rabiner, R. W. Schafer, Digital Processing of Speech Signals. Englewood Cliffs, N.J., Prentice
Hall.
12. International Journal on Soft Computing (IJSC) Vol.6, No. 1, February 2015
64
[23] M. M. Sondhi, “New Methods of Pitch Extraction,” IEEE Trans. Audio and Electroacoustics, Vol. 16,
No. 2, pp. 262-266, June 1968.
[24] K. Fukunaga, Introduction to Statistical Pattern Recognition. San Diego, Academic Press,1990, pp.
153-154.
[25] R. Kohavi, F. Provost, Glossary of terms, Editorial for the Special Issue on Applications ofMachine
Learning and the Knowledge Discovery Process, vol. 30, No. 2–3, 1998.