The essence of communication remains a critical aspect in the modern society. Different technologies are emerging every single day that are geared to make it cheaper and efficient. The end users are presented with a variety of communication mechanisms for the different forms of communications already existing. In the midst of all these technology is the aspect of communication accuracy. The accuracy is defined by the level of distortion every communication channel can introduce to the original message (Hong et al., 2014).

The aspect of accuracy is considered to not only affect the reliability of the communication channel but also increases the communication cost. Different communication channels have different methods of ensuring highest level of communication accuracy. Further, there are many algorithms that have been developed to facilitate accurate delivery of the sender message. It can, however, be noted that the accuracy of the message is affected by a number of factors that are channel related. The channel is considered to introduce considerable amount of noise which has the potential of distorting the original message. Most of the existing algorithms are meant to evaluate and compare the original message against the delivered message. Any slight deviation from the expected standard automatically prompts for the retransmission of the original message. The mode of retransmission and the decision of retransmitting the message is normally made and evaluated on the basis of the packets delivered and their sequence (Hong et al., 2014).

Retransmission of the message over the satellite communication is more or less the same scenario. In this case the satellite is perceived to be part of the communication channel. As such, there is sufficient probability of noise combinations of the data signal. Point to multipoint communication over the satellite is perceived to have relatively high chances of noise distortion as a result of the impending confusion on the information flow. The directional flow of the information is critical in the satellite communication since it defines the accuracy and reliability of the transmission channel (Tsuda and Ha, 2009).

Satellite are considered to be among very successful transmission channels for digital broadcast. It is further perceived that the satellite transmission potential has not been explored fully. This paper provided a discussion on the retransmission aspect that is observed on the point to multipoint satellite communication system. The paper will explore the transmission errors that are generated in this mode of communication. Further, discussion on the existing error correction mechanisms will also be discussion. The paper will pay special attention on the go-back-n error recovery strategy. The strategy will be simulated on the Matlab software in order to evaluate the underlying efficiency and opportunities for improvement (Towsley, 2009).

The paper intends to explore the error correction strategies that are existing on the satellite communications. Satellite communication remains to be critical channel for relaying digital broadcast information across the globe. The paper intends to provide a discussion on the retransmission aspect that is observed on the point to multipoint satellite communication system. The paper will additionally explore the transmission errors that are generated in this mode of communication. Further, discussion on the existing error correction mechanisms forms part of the objectives of this paper. At the end of the day a simulation on go-back-n error recovery strategy will be undertaken. The strategy will be simulated on the Matlab software in order to evaluate the underlying efficiency and discover opportunities for improvement.

Many studies on the satellite communication system has been undertaken in the recent past. The studies have realized different communications systems that are currently implemented. Significant efforts have also been directed toward improving the already existing satellite communication systems. The studies have made it possible for the satellite digital broadcast possible and also introduced the asymmetric nature of the IP traffic associated with the satellite communication systems. At the moment satellite is considered to be the most reliable IP multicast communication provider with high level of reliability (Singh and Chandran, 2012).

Perhaps the study on the retransmission aspect of the satellite communication is an aspect that has gained popularity in most of the recent studies. The retransmission aspect has been regarded as the only solution for the transmission errors realized in the satellite communications (Singh and Chandran, 2012). The following figure can be used to illustrate the point-multipoint communication over the satellite;

Figure 1 Point-to-multipoint transmission over the satellite

The above communication scheme can either be designed for single or bi-directional systems where the information is relayed from a central antenna to an array of antennas by use of time-division multiplexing. In this communication scheme same message is delivered to an array of receiving antennas at a very high speed. As such, it is preferred for establishing real-time communication in the teleconferences systems. This is considered a unique mode of communication since the central connection endpoint is expected to establish a specific link for transmission to multiple remote peripherals. Any data originating from the central antenna is received by all the peripheral remote antennas at their dedicated sessions. It is then becomes clear that the idea of retransmission of the signal is complex since the central antenna has to establish the specific message transmitted at a given time to the specific remote peripheral. There are several algorithms that have been developed to establish the retransmission mechanism whenever there is such a need (Singh and Chandran, 2012).

There exist three modes of communication over the satellite under the IP multicast; unicast, multicast and broadcast. The unicast form of communication is established between a single antenna against a specific receiver. The receiving antenna will acknowledge received of the message when the transmission has been completed. Multicast transmission is established when the central connection endpoint establishes a specific link for transmission to multiple remote peripherals. Any data originating from the central antenna is received by all the peripheral remote antennas at their dedicated sessions when the connection has been established. The broadcast mode of connection does not have a specific receiving antenna. As such, there is a single transmitting antenna that sends data to unknown receiving antennas. The receiving antennas will only be required to request for access to the transmitted data using a specific criteria. Once the access permission has been granted the receiving antenna will continue to receive the data as long as the transmitting antenna continues to transmit (Singh and Chandran, 2012).

Most of the studies are considered to focus on the different communication aspects that are available with the satellite communication. There is little that has been explored in terms of the transmission challenges and the efficiency of the whole communication system. Perhaps the errors introduced during the transmission session have realized insignificant attention. It can however be noted that satellite communication just like any other form of communications suffers the drawbacks of transmission errors. The communication channel is associated with the huge transmission packets thus making it difficult to rectify the flow of information at the lowest level. As such, there remains to be a challenge of determining the correctness and reliability of the transmission channel (Singh and Chandran, 2012).

It can also be noted that point-to-multpoint transmission has the ability of self-correction when a comparison on the received data at the receiving antennas is established. The receiving antennas are expected to receive similar information. An establishment of what was received at any given moment for every single receiving antenna can easily determine and correct occurrence of transmission errors. This mode of error correction however is not reliable since the receiving destinations are normally far distance apart. Establishing an error correction mechanism requires a dedication connection and evaluation algorithm that will monitor the content of each packet received on all the receiving end points. It is therefore critical to explore other mechanisms that can detect and correct the errors generated in the point-to-multipoint communication over the satellite. This is not only a gap in the satellite communication literature but also a challenge to the telecommunication industry (Singh and Chandran, 2012).

There is need to pay attention to the error recovery mechanism that can improve the reliability of the satellite communication systems. There exists a couple of error recovery mechanisms that have not been fully explored. The use of the go-back-n recovery mechanisms has gained popularity as a result of the simplified nature of recovering lost transmission packets. Additional review on this recovery mechanism has been provided in the literature review section (Selig, n.d.).

Satellite communication constitutes three main error detection and correction mechanisms. The use of any of these mechanisms depends on the number of nodes where the transmission is undertaken, the nature and form of data being transmitted and the mode of transmission used. Automatic repeat request is among the common mechanism that is employed in the satellite communication. This approach basically request for the retransmission of the signal that is considered to have been lost or distorted by the transmission channel. Automatic repeat request make use of the error-detection code to acknowledge the reception of the transmitted data. The receiving antenna sends an acknowledgment signal indicating the delivery of the given transmitted data. However, when the transmitter fails to receive the acknowledgment signal from the receiver within a given set time it will automatically retransmit the signal. There are basically three modes of Automatic repeat request; Stop-and-wait Automatic repeat request, Go-Back-N Automatic repeat request and selective repeat (Selig, n.d.).

Stop-and-wait Automatic repeat request

This is considered to be the simplest error correction technique. This technique is considered to be adequate for simple communication protocols. Stop-and-wait Automatic repeat request constitutes transmitting of a protocol data unit (PDU) of information and then waiting for the response from the receiving antenna. The receiver in this case is expected to provide an acknowledgement PDU when it receives all the expected PDUs from the transmitting unit. The receiver can also provide a negative acknowledgement if it receives an incorrect PDU. The technique is designed with an ability of the transmitter to automatically retransmitting the signal when there is a delay in receiving a correct acknowledgment signal from the receiving unit (Selig, n.d.). The following figure can be used to illustrate the Stop-and-wait Automatic repeat request error correction mechanism;

Figure 2 Stop-and-wait error correction technique

It is important to note that the sender will always wait for the acknowledgement signal before transmitting another PDU. The sender is considered to be in the idle mode whenever it is waiting for the acknowledgment signal. In the above figure the blue allow points to the sequence of data PDU that is being transmitted cross the communication channel. The Stop-and-wait Automatic repeat request can implement either full or half duplex form of communication (Selig, n.d.). The green allow points to the acknowledgment signals that is sent by the receiver to the transmitter. There is normally a small delay that is witnessed between the reception of the last byte of PDU and the generation of the corresponding ACK. Perhaps the most important aspect of Stop-and-wait Automatic repeat request is the reliance on the time for the transmitter to note the loss of the data PDU. In this case the transmitter will wait for the acknowledgement signal within a specified time. The lack of the arrival of the acknowledgement signal within the expected time will prompt the transmitter to resent the data PDU again. It is important to note that the receiver does not have the capability of detecting data loss (Pujolle and Puigjaner, 2011). The transmitter will always receive an acknowledgement signal that has unique code to the sent signal. The transmitter will then forward the same data PDU over the transmission channel and wait again for the acknowledgement signal. The following figure can be used to illustrate the retransmission mode of the Stop-and-wait Automatic repeat request

.

Figure 3 Retransmission mechanism of the Stop-and-wait Automatic repeat request

In the above diagram the second data PDU is considered corrupt. As such, the receiver will discard it and then provide the negative acknowledgement signal to the sender. Alternatively, it will fail to provide an acknowledge signal. As such, the lack of the acknowledgement signal will prompt the transmitter to resend the data packet again over the channel. The efficiency of the stop and wait technique can be computed as shown below (Lam, 2009). Assuming S is the time taken between the transmission of the packet and the reception of its acknowledgment and DTP is the transmission time of the packet, the efficiency of the system can be illustrated as below;

Then,

And assuming TO is the timeout interval and X is the amount of time taken for the transmission of the data PDU and reception of the acknowledgment signal,

And

Go-Back-N automatic repeat request

Go-Back-N is an error recovery procedure that is preferred for point-to-multipoint in many communication protocols. Go-Back-N is an error recovery procedure is also applied on the satellite communication to ascertain the delivery of the data packets over the transmission channel. The procedure is able to detect and retransmit I-frames that have already been computed in the underlying algorithm. This technique relies on the sequence of the frames that are sent from the sender to the receiver. The receiver will continuously acknowledge the reception of a given number of data PDUs. Additionally, it will keep the record of the sequence of such data PDUs (Idawaty Ahmad. and Mohamed bin Othman., n.d.).

All the acknowledgement regarding the received data PDU will be sent to the transmitter. However, upon loss of a data PDU, the receiver will miss to acknowledge its reception. It will however continue accepting higher sequence number of the data PDUs keep coming from the transmitter. The transmitter will then automatically detect the missing acknowledgement of the lost PDU. At this point the receiver will request the transmitter to stop transmission and go back to the last known correctly sent and received data PDU. The transmitter will then be required to retransmit all the data PDUs starting from the sequence number of the lost PDU (Idawaty Ahmad. and Mohamed bin Othman., n.d.). The following figure can be used to illustrate the Go-Back-N automatic repeat request error corrective technique;

Figure 4Go-Back-N error recovery technique

It is important to note that there are three stages through which the corrupted PDU is recovered. In the first stage the corrupted PDU is discarded at the receiver’s end. The receiver only retains the sequence of the corrupted PDU. The sequence number of the corrupted PDU is then sent back to the transmitting node. The transmitter will then go back to the sequence number and retrieve the correct data PDU. At this point the receiver will continuously discard all the PDUs sent from the transmitter and do not contain the already requested sequence number. When the transmitter sends the correct PDU with the sequence number, it will then be received by the receiver. At this point the receiver will start accepting the higher sequence number of the other PDUs. It is thus clear that the transmission of the PDUs momentarily stops until the correct PDU is sent and then the sequence is proceeded. This technique is associated with delays emanating from the time lost when the corrupt PDU is received at the receiving node (Hong et al., 2014).

The detection of the corrupt data PDU introduces a transmission delay that is manifested by the time taken for the receiver to request the sender to retransmit the corrupted PDU. Further, the transmitter will be forced to restart sending the PDU from the last successfully received PDU. A buffer on the transmitter is activated to keep sequence of the remaining PDUs that have not been transmitted after the interruption (Hong et al., 2014). The following figure can be used to illustrate the retransmission mechanism that is provided by the go-back-n automatic repeat request;

Figure 5Retransmission mechanisms of the go-back-n automatic repeat request

The efficiency of this system can be computed and illustrated as shown below;

In this case the value of N is chosen in such a way that it is large enough to allow continuous transmission while waiting for the acknowledgment signal for the first packet of the window. As such,

Assuming there are no transmission errors, the efficiency of the system will be;

However when an error occurs, the entire window of N packets will have to be retransmitted as such;

Where X is the number of the packets sent per successful transmission.

Advantages of Go-Back-N automatic repeat request

The level of efficiency is more than stop and wait protocol

One acknowledgment signals can be used on more than one data PDUs

It is possible to set the time on a group of data PDUs

It is possible for the sender to send many data PDUs at the same time

Disadvantages of Go-Back-N automatic repeat request

The Go-Back-N automatic repeat request has buffer requirements which may increase the cost of implementing the protocol

The transmitter requires additional memory that is used to store the last N packets

It introduces unnecessary retransmission of data PDUs that may be error free

There is high inefficiency when there is a delay since large data will be required to be retransmitted

Selective repeat ARQ

Selective repeat ARQ is another error correction mechanism that is employed by the point-to-multipoint satellite communication. This technique is considered to be complex since it involves a set of procedures that provide an error recovery. However, it is considered to be the most efficient mechanism of correcting the transmission errors since it does involve the retransmission of the data PDU. The receiver is designed with the ability of correcting the transmission errors noted on the received data frames (Hercog, n.d.).

It can be pointed that both the sender and the receiver maintain a window of acceptable sequence numbers that help them determine the occurrence of transmission errors. The sender window normally starts at zero and increases to some pre-defined maximum number. The receiver on the other hand will always have a constant size that is equal to a given predetermined maximum. Additionally, the receiver has a buffer that is reserved for each sequence number in its fixed window. The following figure can be used to illustrate the mechanism of the selective repeat ARQ;

Figure 6 Selective repeat ARQ

Attached to the buffer at the receiver is a bit that helps to determine if it’s empty. The sequence number of the frame that arrives is normally checked to see if it falls within the window. If the frame falls within the expectations it is accepted and stored. The frame is normally kept in the link layer and not passed to the network layer up to the point where the lower numbered frames have all been delivered. As such, the sequence of the frames is ensured and sustained. It is also important to note that the sender will only transmit frames whose NAK has been received. As such the efficiency of this system is normally very high. It thus clearly demonstrates that there will be fewer retransmissions as compared to go-back-n technique. The only disadvantage with this system is the complexity involved between the sender and the receiver since each frame must be acknowledged individually. Additionally, the receiver is dedicated with the maintenance of the frame sequence (Hercog, n.d.). The probability of delay variance can be computed as illustrated below;

Let N be independent transmission processes that are defined by the matrix P and they start at the same time. All the processes are expected to complete in or before ZN. The probability that jth process will complete last can be defined as below;

As such it is clear that that the probability of the delay is related to the level of redundancy that should be added to the packet stream in order to meet the user’s daily constraints.This technique is therefore considered to be the most efficiency error recovery technique over the other two methods (Galinina, Balandin and Koucheryavy, n.d.).

The strength of the selective repeat ARQ lies on a number of aspects that are not possible on other ARQ protocols. Selective repeat ARQ is considered to be the most efficient since it does not involve retransmission of the data PDUs as other ARQ protocols. Although it is considered to be complex, the underlying algorithms ensure there is high assurance of accuracy and efficiency. The receiver under selective repeat is designed with the ability of correcting the transmission errors noted on the received data frames. The probability of the occurrence of the errors and the time delay of the recovery mechanism has been illustrated in the three methods presented above (Galinina, Balandin and Koucheryavy, n.d.).

It can further be pointed out that this error correction mechanism provides room for larger receiving window compared to the other two ARQs. As such, the buffers at the receiving end can comfortably store the out-of-order packets thereby avoiding the situation of retransmission (Galinina, Balandin and Koucheryavy, n.d.).

The other advantage of the selective repeat ARQ involves the buffered packets that are when they are out-of-order. The receiver has the capability of rearranging them and then sending the acknowledgement signal to the sender thus completing the transmission process. This is however not the case with the stop and wait and go-back-n protocols. It is thus possible for the receiver to re-acknowledge the already received out-of-order packets. This scenario however is not possible with the other two ARQ protocols (Galinina, Balandin and Koucheryavy, n.d.).

This technique is therefore considered to be the most efficiency error recovery technique over the other two methods (Bisio, 2016). The probability of the occurrence of the errors and the time delay of the recovery mechanism has been illustrated in the three methods presented above. A comparison on the efficiency of the three methodologies has also been provided. The following figure can be used to illustrate the comparison of Go-back-N and selective repeat automatic repeat request;

Figure 7 Comparison between selective repeat and Go-Back-N protocols

Forward Error Correction

The forward error correction is yet another technique that is used in satellite transmission. This technique involve the transmission of data PDU that are redundant. The sender purposefully sends redundant data PDU over the transmission channel. The receiver is then left with the option of selecting only those data PDUs that have not been corrupted over the transmission channel. In most cases this technique is preferred for sending broadcast form of communication where many receivers are expecting data from a single transmitter.

Forward error correction technique only work well if the errors occurring on the transmission channel are independent. As such, the correction of one error will not lead to modification of the error in the subsequent data PDUs. However, if there are many errors in the system this technique is found to be less efficient.

The forward error correction technique requires an intelligent transmitter that can introduce unique redundant bits using a predetermined algorithm. The longer sequence of bits introduced to the data PDU is then sent over the transmission channel. The receiver is then expected to use a suitable decoder to retrieve the original signal. The following figure can be used to illustrate the concept of the forward error correction technique;

Figure 8 Forward Error Correction Technique

Advantages

It does not involve retransmission of signals whenever an error is detected on the data PDUs
Redundant bits inserted to the original data makes it difficult for the data to be corrupted

Disadvantages

Requires a special type of receiver and transmitter to embrace the algorithm
Misuses transmission channel bandwidth to send redundant data which would have been used to send additional data

Limitation

Requires special form of algorithm to introduce redundant bits
Requires extra bandwidth for transmission of redundant data

This section has provided a discussion on different approaches that are used to error recovery in point-to-multipoint satellite communication. A discussion on stop and wait, Go-Back-N and selective repeat ARQ has been provided below;

Stope and Wait Automatic Repeat Request

Stop-and-wait Automatic repeat request constitutes transmitting of a protocol data unit (PDU) of information and then waiting for the response from the receiving antenna. The receiver in this case is expected to provide an acknowledgement PDU when it receives all the expected PDUs from the transmitting unit. The receiver can also provide a negative acknowledgement if it receives an incorrect PDU. The technique is designed with an ability of the transmitter to automatically retransmitting the signal when there is a delay in receiving a correct acknowledgment signal from the receiving unit (Selig, n.d.).

Advantages

It has incorporated a timer in the transmission process
Very relevant for noisy channels
It constitutes both flow and error recovery and control mechanisms

Disadvantages

The efficiency of this technique is low
No timer set for individual frames
Only a single frame can be set at a time

Limitations

This technique has only one window size
There is no pipelining in stop and wait ARQ
The receiver window is limited to one

Go-Back-N ARQ Protocol

This technique constitutes the retransmission of data PDU once an error has been detected. This technique relies on the sequence of the frames that are sent from the sender to the receiver. The receiver will continuously acknowledge the reception of a given number of data PDUs. Additionally, it will keep the record of the sequence of such data PDUs. All the acknowledgement regarding the received data PDU will be sent to the transmitter. However, upon loss of a data PDU, the receiver will miss to acknowledge its reception. It will however continue accepting higher sequence number of the data PDUs keep coming from the transmitter. The transmitter will then automatically detect the missing acknowledgement of the lost PDU. At this point the receiver will request the transmitter to stop transmission and go back to the last known correctly sent and received data PDU. The transmitter will then be required to retransmit all the data PDUs starting from the sequence number of the lost PDU (Idawaty Ahmad. and Mohamed bin Othman., n.d.)

Advantages of Go-Back-N automatic repeat request

The level of efficiency is more than stop and wait protocol

One acknowledgment signals can be used on more than one data PDUs

It is possible to set the time on a group of data PDUs

It is possible for the sender to send many data PDUs at the same time

Disadvantages of Go-Back-N automatic repeat request

It introduces unnecessary retransmission of data PDUs that may be error free

There is high inefficiency when there is a delay since large data will be required to be retransmitted

Limitations

The Go-Back-N automatic repeat request has large buffer requirements

Additional memory for storage of N packets is required

Selective Repeat ARQ

This is the approach that has been recommended for implementation in the error recovery for the point-multipoint satellite communication. In reference to the discussion that has been provided in the previous sections it is clear that this protocol has the highest efficiency with little chances of retransmission. The resources required to implement this protocol are also considered to be less. The strength of the selective repeat ARQ lies on a number of aspects that are not possible on other ARQ protocols. Selective repeat ARQ is considered to be the most efficient since it does not involve retransmission of the data PDUs as other ARQ protocols. Although it is considered to be complex, the underlying algorithms ensure there is high assurance of accuracy and efficiency. The receiver under selective repeat is designed with the ability of correcting the transmission errors noted on the received data frames. The probability of the occurrence of the errors and the time delay of the recovery mechanism has been illustrated in the three methods presented above (Bisio, 2016).

There is no much difference between GBN and selective repeat especially when implementing. Actually the efficiency of these two protocols are considered to be same up to the time when an error is detected. The implementation process should start with the establishment of the sender and receiver windows. Sufficient memory capacity should be set aside for these two sections. The window size should however be less than the half sequence number in the selective repeat protocol. This is meant to avoid packets being recognized incorrectly. The sending process can then be initiated with the new packets as long as there is an acknowledgement algorithm in place. A simulation on this process has been done on MATLAB as shown in the next section (ARQ Protocols in Cognitive Decode-and-Forward Relay Networks: Opportunities Gain, 2015).

Advantages

There is no issue of retransmission
Provides chances of recovering corrupt frames
High efficiencies as compared to Go-Back-N protocol

Disadvantages

Can be time consuming especially when the error is detected

Limitations

Require relatively large CPU processing resources

The selective repeat protocol was implemented in MATLAB using distinct models and codes developed to imitate the actual behaviour of the real system. Different models were first developed before enjoining and starting the transmission process. A discussion on the models developed is provided below;

Transmitter model

This model was developed with two sub-modules; packet generation and the encoding. The packet generator was expected to generate the transmitted packets at a fixed rate. A new data is expected to be added to the buffer of the transmitter after a fixed period of time (Beard and Stallings, 2016). Every packet of data was stored in the form of [seqnum, payload] where seqnum is the sequence number of the packet and payload is the size of the packet in bits. The below codes can be used to illustrate the transmitter model used in the simulation;

Figure 9Pseudocode for transmitter model

The generated packets were then forwarded to the Vandermonde matrix encoder. The encoder basically stripes of the packets with headers and trailers and then converts them into data blocks. At this point the packets are ready for transmission over the channel. Note that the Vandermonde model makes it possible to have linear and independent packets that can be recovered easily at the receiver (Beard and Stallings, 2016).

The channel model

The channel model was designed in such a way that it is possible to simulate corrupt frames accurately. The channel was modelled with binary symmetric channel that is considered to be independent and identically distributed. The channel model keeps checking for the available data. Once the data PDU are placed on the channel, it confirms if the data is already acknowledged by the receiver (Beard and Stallings, 2016). If already acknowledged it is passed to the transmitter otherwise it is forwarded to the error sub-module then finally to the receiver. The following figure can be used to demonstrate the pseudocode that was developed for the channel model;

Figure 10Pseudocode for channel model

Receiver model

The receiver model is designed also with an error checker. The received packets are first forwarded to the error checker where they are checked for errors. If errors are found the receiver proceeds to check all the packets in the Vandermonde Matrix. A negative acknowledgement signals is then generated and sent to the transmitter. Additionally, the corrupt packets are stored in the buffer with a hope that it will be able to correct future errors. Corrupt files are discarded after a specified length of time and new packets requested in the system (Beard and Stallings, 2016). The following table can be used to illustrate the parameters that were used in the simulation;

Parameter

Value

Pkt-size

1000 bits

Lampda

1000 packets

Rate

10kbps

Tprop

20 ms

Tsim

3000 ms(Simulation time)

n

7 blocks per code

FER

Forward error rate (0, 0.2, 0.7, 0.8)

Ack-wait-time

0.001ms

MATLAB Codes

The following codes were used to simulate the selective repeat ARQ on MATLAB;

% program for protocol analysis

clc; clear all; close all;

n=input(‘Enter the data sequence length’);

m=input(‘Enter the number of packets ‘);

x=randint(m,n);

% make packet

p=zeros(1,m);

for i=1:m

for j=1:n

a=x(i,j);

b=p(i);

p(i)=bitxor(a,b);

end

pac(i,:)=[x(i,:),p(i)];

subplot(m,1,i); stem(pac(i,:));

end

xlabel(‘Transmitted Data,last bit is the parity bit’);

% send first group of packets

% send packets

figure

ba=m/8;

for k=1:m

% for l=1:8

% g=l*k;

data(k,:)=bsc(pac(k,:),.1);

subplot(m,1,k); stem(data(k,:));

% end

end

xlabel(‘Recieved Data,last bit is the parity bit’);

figure

err = 1;

erf=1;

while (err~=0)

do=data(:,n+1)’;

err=bitxor(p,do);

stem(err);

display(err);

display(‘displaying retransmitted packets’);

for i=1:m

if err(1,i)== 1

display(err);

display(‘error detected in packet no:’);display(i);

% figure

for j=i:m

data(j,:)=bsc(pac(j,:),.1);

display(j);

% subplot(m,1,j);stem(data(j,:));

end

end

do=data(:,n+1)’;

err=bitxor(p,do);

end

end

% figure

% for g=1:m

% subplot(m,1,g);stem(data(g,:));

% end

% xlabel(‘Finally Received data after retransmission, last bit is the parity bit’);

The simulation of the selective Repeat ARQ protocol in Matlab;

Figure 11 Illustration of the simulation of Selective Repeat ARQ protocol

Illustration of the frame generation model;

Figure 12 Packet Generation model

Illustration of transmitter model;

Illustration of a receiver model

Figure 13Receiver model

The following figure can be used to show the results obtained from the simulation;

Figure 14 DAta PDU Transmitted

Figure 15 Data PDU received

The following figure can be used to demonstrate the errors detected during the transmission process;

Figure 16Errors detected during the transmission process

err =

0 1 0 0 0 0 0 0 0 1 0

displaying retransmitted packets

err =

0 1 0 0 0 0 0 0 0 1 0

This is further illustrated in the following figure;

In reference to the results obtained from the above simulation it is clear that selective repeat is very efficient. The number of packets retransmitted as a result of the errors detected during the transmission process is zero. This not only reduced the time delay of the transmission but also avoids the need to invest on buffer resources that would otherwise be used to store the corrupted files. The protocol is capable of correcting errors that are realized in the transmission channel and thus improve the overall performance of the transmission process (Beard and Stallings, 2016). A typical analysis on the transmission channel is provided below;

running SRprotocolX …

Tx channel P[frameSuccess] = 0.8

sender-to-receiver communication delay = 5 Comm frames

Rx channel P[AckSuccess] = 1

receiver-to-sender communication delay = 10 Ack frames

sending 5 frames …

tic: 0

RxQ: 0 0 0 0 0 0 0 0 0 0

TxQ: 0 0 0 0 0

tic: 1

RxQ: 0 0 0 0 0 0 0 0 0 0

Selective repeate ensures that the frame is kept in the link layer and not passed to the network layer up to the point where the lower numbered frames have all been delivered. As such, the sequence of the frames is ensured and sustained. It is also important to note that the sender will only transmit frames whose NAK has been received. As such the efficiency of this system is normally very high. It thus clearly demonstrates that there will be fewer retransmissions as compared to go-back-n technique. The only disadvantage with this system is the complexity involved between the sender and the receiver since each frame must be acknowledged individually (Ali, n.d.).

The strength of the selective repeat ARQ lies on a number of aspects that are not possible on other ARQ protocols. Selective repeat ARQ is considered to be the most efficient since it does not involve retransmission of the data PDUs as other ARQ protocols. Although it is considered to be complex, the underlying algorithms ensure there is high assurance of accuracy and efficiency. The receiver under selective repeat is designed with the ability of correcting the transmission errors noted on the received data frames. The probability of the occurrence of the errors and the time delay of the recovery mechanism has been illustrated in the three methods presented above (Ali, n.d.).

The paper has explored the different error recovery mechanisms that are used in the point-to-multipoint communication over the satellite. A variety of literature review has been undertaken to establish the most efficient and effective error recovery strategy. Different technical aspects of the stop and wait, GBN and selective repeat protocols have been evaluated. Selective repeat protocol has been found to be the most effective and efficient mode of error recovery. A simulation on the selective repeat protocol has been done using MATLAB. The probability of occurrence of the errors and PDU retransmission has been computed. It has been established that the selective repeat strategy requires relatively less amount of resources compared to other error recovery protocols.

Even though it has been established that the efficiency of selective repeat is high, there is still many opportunities for improvements. There is need to develop mechanisms of preventing and detecting occurrence of errors in the transmission system. If this work is done the efficiency of PDU transmission will be improved greatly.

Ali, I. (n.d.). Architectural exploration of carrier synchronization for TDMA based satellite communication systems =.

ARQ Protocols in Cognitive Decode-and-Forward Relay Networks: Opportunities Gain. (2015). Radioengineering, 24(1), pp.296-304.

Beard, C. and Stallings, W. (2016). Wireless Com Net & Systems. Pearson Australia Pty Ltd.

Bisio, I. (2016). Personal Satellite Services. Next-Generation Satellite Networking and Communication Systems. Cham: Springer International Publishing.

Galinina, O., Balandin, S. and Koucheryavy, Y. (n.d.). Internet of things, smart spaces, and next generation networks and systems.

Hercog, D. (n.d.). Selective-repeat protocol with multiple retransmit timers and individual acknowledgments.

Hong, T., Kang, K., Ku, B. and Ahn, D. (2014). HARQ-ARQ interaction method for LTE-based mobile satellite communication system. International Journal of Satellite Communications and Networking, 32(5), pp.377-392.

Idawaty Ahmad. and Mohamed bin Othman. (n.d.). Performances of Go-Back-N ARQ schemes with block transmission.

Lam, S. (1979). Satellite Packet Communication–Multiple Access Protocols and Performance. IEEE Transactions on Communications, 27(10), pp.1456-1466.

Pujolle, G. and Puigjaner, R. (1991). Data communication systems and their performance. Amsterdam: North-Holland.

Selig, M. (n.d.). Interference Mitigation with Selective Retransmissions in Wireless Sensor Networks.

Selig, M. (n.d.). Interference Mitigation with Selective Retransmissions in Wireless Sensor Networks.

Singh, A. and Chandran, H. (2012). Low complexity FEC Systems for Satellite Communication. Network Protocols and Algorithms, 4(1).

Towsley, D. (1979). The stutter go back-N ARQ protocol. Estados Unidos: The Institute of Electrical and Electronics Engineers, Inc-IEEE.

Tsuda, D. and Ha, T. (1989). Adaptive Go-Back-N. Monterey, Calif.: Naval Postgraduate School.