Relay Network is most suited wireless communication technique for future Smart Grid, because it has a lot of advantages such as reliability against failures and offers redundancy, self-configuring and self-healing. For example if one nod or path is blocked, it can find a neighbour node, by different algorithms schemes. The purpose of this paper is to simulate wireless relay channel model in presence of impulse noise and computing the BER performance of channel data. It makes to be useful for a designer to predicting error and behaviour of a channel for optimum design, to be able choose a robust channel.
Impulse noise is a noise which often generates in power line and can be a source of error in transmission of data in a channel and it is a major contributor to the increasing of the error rate. In some cases e.g. in Smart Grid [1,2], errors are not tolerated and it may cost big damages to infrastructure of a Grid or jeopardize live. The data error may also degrade capacity. Source of some type impulse noises are natural such as lightning, but some are manmade such as mechanical switches, breakers, even light switches or breakers, power lines, car ignitions, fluorescent light, which calls industrial impulse noise increasing day by day [
In case of impulse noises caused by lightning discharges occur randomly in time and geographical location. Nature of these discharges yield random pulses of electromagnetic field at a given receiving antenna [
Future of Smart Grid will be based on intelligence, which is directly related to ability of bidirectional communication through whole Grid for different kinds of traffic during different situations, from Grid failures, natural catastrophes to attacks. There are several ways to handle two way data communication, which cover the Smart Grid requirements.
One of most suitable communication ways for smart grid is co-operative networks. The advantages of cooperative networks are: reliability, mobility, easy installation, cost effective, high capacity and dynamical architecture.
Realization of co-operative relay networks depends on efficiency and reliability of channel capabilities. The measurements [
Co-operative communication has recently drawn tremendous attraction to the research community. It is a new communication paradigm which generates independent paths between the user and the base station by introducing a relay channel [
There are several schemes used in co-operative relay network namely Amplify and Forward (AF) [
Physical Network Coding (PNC) [
Impulse noise is sporadic, non-contiguous, consisting of irregular pulses or noise spikes of short duration and of relatively high amplitude as high as several hundred micro-volts and may occur in bursts or discrete impulses, as shown in
In order to desire impulse noise characteristic, a mathematical model needs. We have applied BernoulliGaussian impulse noise model of Poisson arriving delta Rayleigh probability density function [
Bernoulli distribution is a discrete probability distribution, which takes value 1 with success probability p and value 0 with failure probability, and mean p.
The characteristic function of the Bernoulli distribution is:
Signal S, transmitted over a channel with impulsive and White Gaussian noise N and signal R received, as described by the following equation. See
N is AWGN with mean zero and variance and is the impulse noise. is product of AWGN and
Bernoulli process, also:. All these parameters assumed to be complex and independent of each others. This means that each data symbol independently can be hit with probability of p and with random amplitude N. The characteristic function of total noise of From (1) is:
This equation has been to a great degree used by others [19,20] and gives a realistic characterization of real impulse noise outlook.
SystemWireless communication is a well known and is very popular technique, because of advantages like; mobility (wireless devices easily move within the wireless range), easy installation (no cable needs), cost effective (cost of cabling, running and maintaining is minimal), high capacity (cable has less capacity), dynamical architecture (Easley can redesign). Most popular wireless technique for smart gird is cooperative networks or Wireless Relay Network (WRN) [
Let us consider a bidirectional relay network consisting of two sources and a relay. Each node has a single antenna and it operates in half duplex mode. S1, S2 and R are used to denote source-I, source-II and the relay respectively as shown in the
Let denotes binary information from the source. The BPSK modulated signal of will then be where and for i =
[1,2]. Let us denote the complex channel coefficient between and R by hi during first time slot and during the second time slot. Channels can be modelled as for i = 1, 2 where indicates h is circularly symmetric complex valued Gaussian random variable with mean m and variance. For simplicity it can be assumed that channel coefficients are constant in two consecutive time slots. It means that. It is assumed that the phase of the transmitted signals from S1 and S2 are synchronized at the relay node.
During first time slot, both S1 and S2 transmit their information and respectively to the relay node R. The signal received by the relay node can be shown by
where is the transmission power at, is the additive white Gaussian Noise (AWGN) and i is impulse noise which can be defined as Bernoulli-Gaussian Model [
From
Once the mapping is done at the relay node, the relay forwards the XORed symbol z towards the sources over the broadcast channel. The signal received by is given by
where means transmission power at relay, means AWGN at and means impulse noise at. After receiving XORed signal from the relay node, source nodes S1 and S2 decode the received signal and detect the signal sent from peer node by executing XOR operation with the detected signal from relay node and its own signal.
In this section performance evaluation will be shown for
cooperative relay communication system under Rayleigh fading condition with AWGN as well as IN environment. It will be shown in two steps. At first Signal to Noise Ratio (SNR) will be varied and corresponding Error Rate will be observed to analyze the system performance. In that case a fixed amount of Impulse Noise will be in effect for each value of SNR.
Secondly Signal to Impulse noise Ratio (SIR) will be varied and similarly corresponding Error Rate will be observed to study the overall performance. In this case SNR will be fixed to a certain value.
In this step error rate is measured while varying the value of SNR. The value of SIR is kept constant. Later the value of SIR is changed to another value to visualize how the performance changes. For a single value of SIR, different curves are considered by changing the value of p.
Figures 6-8 represent the performance graphs of cooperative relay system in impulsive noise environment considering the value of SIR as 5 dB, 10 dB and 15 dB respectively. For each case, when the value of p is increasing, the performance deteriorates. Basically p is the probability of adding impulse noise with the system. In other words p implies how frequently impulse noise is adding to the received signal.
For example, when the value of p is equal to 1, it means that Impulse Noise is considered for each bit of the information stream.
When p is equal to zero, it indicates that there is no impulse noise is added with the received signal. As expected when the value of p is increasing from 0 to 1, the performance gradually deteriorates. Similarly when the value of SIR is increased, the performance improves since the since strength is increasing or, in other words, impulse noise power is decreasing. It is also observed that while p has a value greater than 0 the graph saturates after a certain value. It is happening due to the fact that although the value of SNR is increasing which indicates less effect of noise, there is still a certain amount of impulse noise adding to the received signal. Since the increment of SNR does not have any hold on that, the performance curves saturate after a certain SNR value.
Performance of the co-operative communication system is studied in this section while varying the value of SIR and putting the SNR into a fixed value. The value of SNR is then changed to another value to compare the system performance.
For each scenario, the value of p is changed from 0 to 1 to analyze the impact of impulse noise to the system. Three different scenarios are considered to investigate the overall system performance while varying SIR. Scenarios are selected for different values of SNR which are for 5 dB, 10 dB and 15 dB shown in Figures 9-11 respectively. For each case the value of p is varied from 0 to 1 which, in fact, describes the effect of impulse noise to the system. From the figures, it is clear that as the value of p is increasing, the performance of the system is getting worse. One interesting fact can be observed from here that whenever the value of p is equal to zero, the
performance curve is just parallel to X-axis. When the value of p is zero, it implies that there is no impulse noise adding to the system. So in such case, there is no impact of increasing the value of SIR and the system will only experience a fixed value of AWGN which makes the curve parallel to the X-axis. Saturation effect is similar to the effect described for varying SNR.
In this paper it is tried to investigate how the system performs when it is required to exchange the information among different nodes through co-operative communication in a smart grid system. It is known that impulse noise can deteriorate the overall system performance drastically in communication in a smart grid system [