PID Controller Based on a Self-Adaptive Neural Network to Ensure QoS Bandwidth Requirements in Passive Optical Networks

N. Merayo; D. Juárez; Juan C. Aguado; I. de Miguel; R. J. Durán; P. Fernández; R. M. Lorenzo; E. J. Abril

doi:10.1364/JOCN.9.000433

Journal of Optical Communications and Networking
Vol. 9,
Issue 5,
pp. 433-445
(2017)
•https://doi.org/10.1364/JOCN.9.000433

PID Controller Based on a Self-Adaptive Neural Network to Ensure QoS Bandwidth Requirements in Passive Optical Networks

N. Merayo, D. Juárez, Juan C. Aguado, I. de Miguel, R. J. Durán, P. Fernández, R. M. Lorenzo, and E. J. Abril

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
N. Merayo, D. Juárez, Juan C. Aguado, I. de Miguel, R. J. Durán, P. Fernández, R. M. Lorenzo, and E. J. Abril, "PID Controller Based on a Self-Adaptive Neural Network to Ensure QoS Bandwidth Requirements in Passive Optical Networks," J. Opt. Commun. Netw. 9, 433-445 (2017)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Check for updates

Abstract

In this paper, a proportional-integral-derivative (PID) controller integrated with a neural network (NN) is proposed to ensure quality of service (QoS) bandwidth requirements in passive optical networks (PONs). To the best of our knowledge, this is the first time an approach that implements a NN to tune a PID to deal with QoS in PONs is used. In contrast to other tuning techniques such as Ziegler–Nichols or genetic algorithms (GA), our proposal allows a real-time adjustment of the tuning parameters according to the network conditions. Thus, the new algorithm provides an online control of the tuning process unlike the ZN and GA techniques, whose tuning parameters are calculated offline. The algorithm, called neural network service level PID (NN-SPID), guarantees minimum bandwidth levels to users depending on their service level agreement, and it is compared with a tuning technique based on genetic algorithms (GA-SPID). The simulation study demonstrates that NN-SPID continuously adapts the tuning parameters, achieving lower fluctuations than GA-SPID in the allocation process. As a consequence, it provides a more stable response than GA-SPID since it needs to launch the GA to obtain new tuning values. Furthermore, NN-SPID guarantees the minimum bandwidth levels faster than GA-SPID. Finally, NN-SPID is more robust than GA-SPID under real-time changes of the guaranteed bandwidth levels, as GA-SPID shows high fluctuations in the allocated bandwidth, especially just after any change is made.

Full Article | PDF Article

More Like This

Implementation of a PID Controller for the Bandwidth Assignment in Long-Reach PONs

Tamara Jiménez, Noemi Merayo, Patricia Fernández, Ramón J. Durán, Ignacio de Miguel, Rubén M. Lorenzo, and Evaristo J. Abril
J. Opt. Commun. Netw. 4(5) 392-401 (2012)

Fair Bandwidth Allocation Algorithm for PONs Based on Network Utility Maximization

N. Merayo, P. Pavon-Marino, J. C. Aguado, R. J. Durán, F. Burrull, and V. Bueno-Delgado
J. Opt. Commun. Netw. 9(1) 75-86 (2017)

Dynamic Bandwidth Allocation With SLA Awareness for QoS in Ethernet Passive Optical Networks

Abhishek Dixit, Bart Lannoo, Goutam Das, Didier Colle, Mario Pickavet, and Piet Demeester
J. Opt. Commun. Netw. 5(3) 240-253 (2013)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (12)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (3)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (15)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Notation	Definition
$k_{p}, k_{i}, k_{d}$	Tuning parameters of the PID
$e [k] = \| B_{guarantee}^{sla \in {onu}_{i}} - \bar{B_{alloc}^{{onu}_{i}} [n]} \|$	Committed error; difference between the guaranteed bandwidth of each ONU and its mean allocated bandwidth
$n$	Number of iterations when the tuning parameters are updated
$m$	Number of iterations when weights of the NN are updated
$I_{i}^{(L)} [k]$	Input of neuron $i$ at layer $L$
$O_{i}^{(L)} [k]$	Output of neuron $i$ at layer $L$
$f (x) = \tanh (x) = \frac{e^{x} - e^{- x}}{e^{x} + e^{- x}}$	Sigmoid function; activation function for the hidden layer
$g (x) = x$	Linear function; activation function for the output layer
$Δ w_{l i}^{(a, b)} [k]$	Increments in the last weight modification between neuron $i$ at layer $a$ and neuron $j$ at layer $b$
$J_{1} = (\frac{1}{2}) e^{2} [k]$	Quadratic performance index of error, used to modify the network weights
$α$	Inertia term in the NN
$η$	Learning rate coefficient in the NN

Number of Neurons	Standard Deviation ${SLA}_{0}$ (Mbps)	Standard Deviation ${SLA}_{1}$ (Mbps)	Standard Deviation ${SLA}_{2}$ (Mbps)
2	0.046169	0.0378122	0.02310
4	0.04882	0.030392	0.023456
5	0.0497220	0.0297659	0.0234553
6	0.067423	0.02963	0.02432
8	0.05567	0.028723	0.022764

Time (s)	Guaranteed Bandwidth ${SLA}_{0}$ (Mbps)	Guaranteed Bandwidth ${SLA}_{1}$ (Mbps)	Guaranteed Bandwidth ${SLA}_{2}$ (Mbps)
0–150	100	70	50
Higher than 150	70	40	60

Notation	Definition
$k_{p}, k_{i}, k_{d}$	Tuning parameters of the PID
$e [k] = \| B_{guarantee}^{sla \in {onu}_{i}} - \bar{B_{alloc}^{{onu}_{i}} [n]} \|$	Committed error; difference between the guaranteed bandwidth of each ONU and its mean allocated bandwidth
$n$	Number of iterations when the tuning parameters are updated
$m$	Number of iterations when weights of the NN are updated
$I_{i}^{(L)} [k]$	Input of neuron $i$ at layer $L$
$O_{i}^{(L)} [k]$	Output of neuron $i$ at layer $L$
$f (x) = \tanh (x) = \frac{e^{x} - e^{- x}}{e^{x} + e^{- x}}$	Sigmoid function; activation function for the hidden layer
$g (x) = x$	Linear function; activation function for the output layer
$Δ w_{l i}^{(a, b)} [k]$	Increments in the last weight modification between neuron $i$ at layer $a$ and neuron $j$ at layer $b$
$J_{1} = (\frac{1}{2}) e^{2} [k]$	Quadratic performance index of error, used to modify the network weights
$α$	Inertia term in the NN
$η$	Learning rate coefficient in the NN

Number of Neurons	Standard Deviation ${SLA}_{0}$ (Mbps)	Standard Deviation ${SLA}_{1}$ (Mbps)	Standard Deviation ${SLA}_{2}$ (Mbps)
2	0.046169	0.0378122	0.02310
4	0.04882	0.030392	0.023456
5	0.0497220	0.0297659	0.0234553
6	0.067423	0.02963	0.02432
8	0.05567	0.028723	0.022764

PID Controller Based on a Self-Adaptive Neural Network to Ensure QoS Bandwidth Requirements in Passive Optical Networks

Author Affiliations

Abstract

Cited By

Figures (12)

Tables (3)

Equations (15)

Journal of Optical Communications and Networking