Investigation and Comparison of Generalization Ability of Multi-Layer Perceptron and Radial Basis Function Artificial Neural Networks for Signal Power Loss Prediction

Ebhota, Virginia C.; Isabona, Joseph; Srivastava, Viranjay M.

doi:10.15866/irecap.v9i1.15330

Cited by 6 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Understanding the rudiments of the algorithm and the involved processes in the computational simulation of learning within the neural network models is very essential. Artificial neural networks such as VNN are the basic building blocks of AI technology and the foundation of ML models, which simulate processes of learning identical to those in the human brain [27], [28].…”

Section: Gradient Descent and Back-propagation In Machine Learning Al...mentioning

confidence: 99%

“…Selection of an adequate model for a given problem complexity as well as tuning suitable parameters require a thorough understanding of the problem domain, and the algorithms must be adequately considered to ensure correct application of the models and interpretation of results. The gradient descent algorithm facilitates the speed of learning the ANNs from the dataset, particularly where modifications to the network's parameter values occur as a result of operations that involve data points and the neural network [28].…”

Section: Gradient Descent and Back-propagation In Machine Learning Al...mentioning

confidence: 99%

See 1 more Smart Citation

Neural Network Optimization and Implications of High and Low Gradient Results in Training Vanilla and Hybrid Adaptive Neural Network Models for Effective Signal Power Loss Prediction

Ebhota,

Shongwe

2023

IREMOS

View full text Add to dashboard Cite

This work investigated the gradient performances during Neural Network (NN) model training: the Vanilla NN (VNN) model and the designed hybrid adaptive NN model using measurement datasets from a Line-of-Sight (LOS) area 1 and a Non-Line-of-Sight (NLOS) area 2 from a Long-Term Evolution (LTE) micro-cell environment. The NN model training results from tables 1 to 4 for training with the dataset from LOS Area 1 show that the gradient stopped value after the NN model training of the VNN model using the Levenberg Marquardt (LM) training algorithm gave 1.29 and trained 13 epochs, while training the same VNN model with the Bayesian Regularization (BR) training algorithm gave a gradient stopped value of 0.084 with 32 epochs. Training the NN model with a hybrid adaptive NN model using the LM training algorithm gave a stopped gradient value of 3.23 with 11 epochs, while training the same hybrid adaptive NN model using the BR training algorithm gave a stopped gradient value of 0.055 with an epoch of 20. The same pattern of results was seen when training the NN models with datasets from NLOS Area 2. The training results showed effective training of the NN with a low gradient stopped value using a designed hybrid adaptive NN model as well as the exertion of the BR training algorithm over the VNN model and the LM training algorithm. This is because the hybrid adaptive NN model combines both the merits of ADALINE and VNN models for optimal performance, and also because the BR training algorithm serves as an efficient training technique that compensates for the exaggerated amount of the network parameters during network training, thereby avoiding over-fitting.

show abstract

Section: Gradient Descent and Back-propagation In Machine Learning Al...mentioning

confidence: 99%

Section: Gradient Descent and Back-propagation In Machine Learning Al...mentioning

confidence: 99%

Neural Network Optimization and Implications of High and Low Gradient Results in Training Vanilla and Hybrid Adaptive Neural Network Models for Effective Signal Power Loss Prediction

Ebhota,

Shongwe

2023

IREMOS

View full text Add to dashboard Cite

show abstract

“…The processing elements are known as the neurons which are processor that operates on the inputs they are fed through the connections. The processors are vast parallel-spread out and comprises of simple processing units with natural tendency of storing and making available exponential knowledge [18]. Artificial neural networks acquire knowledge from the environment by means of learning process and the acquired knowledge is stored in the network by inter-neuron connection strength called synaptic weights [17], [18].…”

Section: Ii3 Concepts Of Artificial Neural Networkmentioning

confidence: 99%

“…The processors are vast parallel-spread out and comprises of simple processing units with natural tendency of storing and making available exponential knowledge [18]. Artificial neural networks acquire knowledge from the environment by means of learning process and the acquired knowledge is stored in the network by inter-neuron connection strength called synaptic weights [17], [18]. It derives its computing power from its architecture and its learning ability to generalize well [18].…”

Section: Ii3 Concepts Of Artificial Neural Networkmentioning

confidence: 99%

“…Artificial neural networks acquire knowledge from the environment by means of learning process and the acquired knowledge is stored in the network by inter-neuron connection strength called synaptic weights [17], [18]. It derives its computing power from its architecture and its learning ability to generalize well [18]. The neurons are distinct mathematical functions which captures and organizes information in accordance to the neural network architecture.…”

Section: Ii3 Concepts Of Artificial Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Designing and Investigating the Effect of Single-Intermediate and 2-Intermediate Layers of Vanilla Neural Networks in Electromagnetic Path Loss Prediction

Ebhota,

Shongwe

2023

IRECAP

View full text Add to dashboard Cite

Impact of two designed architectural compositions of Vanilla Neural Networks (VNNs): the single-intermediate layer of VNN and two (2)-intermediate layers of VNN, in the prediction of signal power loss were examined in this work using measured data from a Long Term Evolution (LTE) network collected from a micro-cell built up environment. The effect of different values of learning rate hyper-parameter were also examined on the VNN architectural models. Early stopping method at the ratio of 75%:15%:15% was adopted during the network training process to avoid over-fitting and Bayesian Regularization (BR) mathematical training algorithm was employed for the training process to ensure good network generalization. Both the intermediate layer neurons for the single-intermediate layer of the VNN and 2-intermediate layers of the VNN were carefully selected to ensure adequate and robust training with excellent predictive performance. Statistical performance metrics, the Correlation Coefficient (R), the Standard Deviation (SD) and Root Mean Squared Error (RMSE) were employed for result analysis of the performances of the two VNN architectural structures while the Mean Squared Error (MSE) and Coefficient of Regression (R) were employed for examining the effect of the various selected values of learning rates on the prediction performances of the two VNN models.The overall experimental results under similar circumstances for training both the VNN models while considering their architectural structures and the effect of learning rates show that for an efficient neural network training and prediction of signal power loss, a well-trained singleintermediate layer of VNN architectural structure with an appropriate neuron numbers gives more efficient and optimal prediction results as the results of the actual output is closest to the desired output showing optimal prediction in comparison with training using 2-intermediate layers of VNN model. The best neural networks training result outputs of a single intermediate layer VNN with 52neurons in the intermediate layer gives R of 0.9728, SD of 1.2758 and RMSE of 1.7435 while best training result of the 2-intermediate layers gives R of 0.9531, SD of 1.5407 and RMSE of 2.2754 on application of [16, 20] neurons which gives the highest prediction results of the considered neuron numbers. Training the single-intermediate layer VNN with small learning rate of 0.002 shows very high R of 0.9903 while with 2-intermediate layers VNN gives R of 0.8810 The training time required for training the single-intermediate layer VNN is considerable low in comparison to training time required to train 2-intermediate layers VNN for best prediction results. These prediction results are of utmost importance in the planning, design and upgrading of wireless network for optimal performance especially when applied in similar environment to the environment of data collection as VNNs results show adaptability and robustness.

show abstract

An Adaptive Vector Median Filtering Approach to Enhance the Prediction Efficiency of Signal Power Loss

Ebhota¹,

Srivastava²

2020

jcm

View full text Add to dashboard Cite

This research work designed and implemented an adaptive Artificial Neural Network (ANN) model using Multi-Layer Perceptron (MLP) and Radial Basis Function (RBF) models built on a Vector Median Filter (VMF) for pre-processing of the dataset. Normalized dataset is denoised using VMF and trained with both MLP- and RBF-ANN models. The proposed model has been developed from measurement data collected from two transmitter locations of non-line-of-sight and line-of-sight operating at the 1900MHz frequency band from LTE cellular network over distances of 1800m 1400m respectively. For non-line-of-sight site-1, VMF-MLP gives a correlation coefficient of 0.9600 compared to 0.9490 for VMF-RBF with a Bayesian regularization training algorithm. The VMF-MLP has 2.1380, 1.5000, and 1.4510 for root mean squared error, mean absolute error, and standard deviation compared to 2.3550, 1.5370, and 1.5610 for VMF-RBF network, respectively. The same trend was seen for line-of-sight in site-2 where correlation coefficient for VMF-MLP is 0.9900 and for VMF-RBF is 0.9840. The VMF-MLP has root mean squared error, mean absolute error, and standard deviation as 2.0670, 1.4900, and 1.3180, respectively, compared to VMF-RBF as 2.3470, 1.9010, and 1.3760, respectively. The predictions of these measurement data have been analyzed in this research work.

show abstract

Investigation and Comparison of Generalization Ability of Multi-Layer Perceptron and Radial Basis Function Artificial Neural Networks for Signal Power Loss Prediction

Cited by 6 publications

References 0 publications

Neural Network Optimization and Implications of High and Low Gradient Results in Training Vanilla and Hybrid Adaptive Neural Network Models for Effective Signal Power Loss Prediction

Neural Network Optimization and Implications of High and Low Gradient Results in Training Vanilla and Hybrid Adaptive Neural Network Models for Effective Signal Power Loss Prediction

Designing and Investigating the Effect of Single-Intermediate and 2-Intermediate Layers of Vanilla Neural Networks in Electromagnetic Path Loss Prediction

An Adaptive Vector Median Filtering Approach to Enhance the Prediction Efficiency of Signal Power Loss

Contact Info

Product

Resources

About