辅导data留学生编程、讲解Java，c++程序设计、Python编程讲解
- 首页 >> Java编程 Project 2: Reliable Data Transfer over UDP (RDT-UDP)
1 / 4
Project 2
Reliable Data Transfer over UDP (RDT-UDP)
1. Objectives
Transport Layer Protocols are central piece of layered architectures. They provide the
logical communication between application processes. UDP (User Datagram Protocol) is a
connectionless, unreliable protocol. It uses a simple transmission model without implicit
handshaking techniques for providing reliability and ordering of packets. Thus, UDP provides
an unreliable service and datagrams may arrive out of order, appear duplicated, or go missing
without notice.
UDP is non-reliable protocol, so to use it you must add reliability features at the
application level. In this project, you will build a reliable transport protocol “RDT-UDP” over
unreliable UDP. Your protocol must provide in-order, reliable delivery of UDP datagrams, and
must do so in the presence of packet loss, delay, corruption, duplication, and re-ordering.
2. Requirements
Clearly describe the network protocol specifications and service definations.
The reliability is achieved using the Sliding Window protocol. You will first
implement the Go-Back-N version with no buffer at the receiver. Then, you will
implement the Selective Repeat version.
Use TCP-style connection maintenance, including establishment, maintains and
release.
Define the format of data carried as the payload of each UDP datagram. The header
length should be as short as possible. In general, the header includes the following
fields:
(1) Sequence Number (32 bits): The sequence number of the first data octet in this
packet (except when SYN is present). If SYN is present, the sequence number is
the initial sequence number (ISN) and the first data octet is ISN+1. The sequence
number is given in the unit of bytes.
(2) Acknowledgement Number (32 bits): If the ACK control bit is set this field
contains the value of the next sequence number the sender of the segment is
expecting to receive. Once a connection is established this is always sent. The
acknowledgement number is given in the unit of bytes.
(3) Connection ID (16 bits) (if exists): A number representing connection identifier.
(4) A (ACK, 1 bit): Indicates that there the value of Acknowledgment Number field
is valid.
(5) S (SYN, 1 bit): Synchronize sequence numbers (connection establishment).
(6) F (FIN, 1 bit): No more data from sender (connection termination)
Avoid losing messages.
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
2 / 4
In case of lost message, detect and recover it (time that takes to recover a message is
related to both reliability and low latency requirements, so the protocol should pay a
special attention to this issue)
Handle messages reordering.
(Optional) Add Congestion Control. The control algorithm is AIMD (like TCP)
The maximum UDP packet size is 524 bytes including a header (512 bytes for the
payload).
The maximum sequence and acknowledgment number should be 102400 and be reset
to zero whenever it reaches the maximum value.
The retransmission (and appropriate congestion control actions) should be triggered
when no data was acknowledged for more than time-out value (e.g. 5 or 10 seconds,
the fixed or dynamic retransmission timeout).
Initial and minimum congestion window size (CWND) should be 512.
The maximum congestion window size (CWND) is 51200.
The initial slow-start threshold (SS-THRESH) should be 10000.
If ACK field is not set, Acknowledgment Numberfield should be set to 0.
FIN should take logically one byte of the data stream (same as in TCP).
FIN and FIN | ACK packets must not carry any payload.
3. Test and Evaluation
(1) First, assume there is no packet loss. Just have the server send a packet, the receiver
responds with an ACK, and so on.
(2) Second, introduce a large file transmission. This means you must divide the file into
multiple packets and transmit the packets based on the current flow and congestion
window size.
(3) Third, introduce packet loss. Now you have to add a timer at the first sent and
unacked packet. There should be one timeout whenever data segments are sent out.
Also congestion control features should be implemented for the successful file
transmission.
4. Performance Analysis
The common performance metrics:
(1) Delay: the time difference between the transport layer accepting data from upper
layer at the sender and the transport layer passing the data to upper layer at the
receiver.
(2) Goodput (useful throughput): Throughput is defined as the number of bits arriving
at the receiver per unit of time. One problem of using throughput as a performance
measure is that it does not take re-transmissions into account. For example: in
Go-Back-N, we may retransmit many redundant packets, which are still counted
toward the traditional throughput definition. These redundant bits are not useful for
the application. A more meaningful measure is known as “Goodput”, which counts
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
3 / 4
the number of bits passed to upper layer here. This basically means that we ignore all
corrupted packets, un-necessary retransmissions, and headers.
(3) Overhead: There are three aspects of overhead: communication, processing, and
storage. Communication overhead has direct relationship with Goodput measurement.
Processing and storage (memory) overhead comes with the additional complexity of
the Selective Repeat protocol, which is its main drawback.
Evaluation is done based on key parameters such as Loss rate, Window size,
Retransmission Timeout, Packet size, total data sent and Network Delay, Channel bandwidth,
Round trip time.
5. Programming
Language: any (C, C++, Java, platform independent)
No high-level network-layer abstractions are allowed in this project. You are allowed
to use some high-level abstractions for parts that are not directly related to
networking, such as string parsing, multi-threading. Use of C++ or Java is preferred.
6. Submission
(1) Project Report
Cover Page
• Project name
• Student Number
• Name
• Date
• School and University
Contents
• Objectives
• Description of Distance Vector Algorithm
• Requirement of experiment
• Programing language/Developing platform and tools
• Design ideas
• Data structures
• Implementation (development tools, model, definitions of objects/methods and
processes and threads, processing flows, etc.)
• Test, Verification, Result Analysis, Performance Analysis, Screenshot with
explanation
(2) Readme.txt
• Plain text format.
• Contain any bugs or issues that you know of in your code.
• It should also indicate how you run your code using command line.
(3) Source Codes
• The source of the program (Must have a program comments).
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
4 / 4
(4) Running Codes
• The compiled program that can run on Windows or Linux system.
1 / 4
Project 2
Reliable Data Transfer over UDP (RDT-UDP)
1. Objectives
Transport Layer Protocols are central piece of layered architectures. They provide the
logical communication between application processes. UDP (User Datagram Protocol) is a
connectionless, unreliable protocol. It uses a simple transmission model without implicit
handshaking techniques for providing reliability and ordering of packets. Thus, UDP provides
an unreliable service and datagrams may arrive out of order, appear duplicated, or go missing
without notice.
UDP is non-reliable protocol, so to use it you must add reliability features at the
application level. In this project, you will build a reliable transport protocol “RDT-UDP” over
unreliable UDP. Your protocol must provide in-order, reliable delivery of UDP datagrams, and
must do so in the presence of packet loss, delay, corruption, duplication, and re-ordering.
2. Requirements
Clearly describe the network protocol specifications and service definations.
The reliability is achieved using the Sliding Window protocol. You will first
implement the Go-Back-N version with no buffer at the receiver. Then, you will
implement the Selective Repeat version.
Use TCP-style connection maintenance, including establishment, maintains and
release.
Define the format of data carried as the payload of each UDP datagram. The header
length should be as short as possible. In general, the header includes the following
fields:
(1) Sequence Number (32 bits): The sequence number of the first data octet in this
packet (except when SYN is present). If SYN is present, the sequence number is
the initial sequence number (ISN) and the first data octet is ISN+1. The sequence
number is given in the unit of bytes.
(2) Acknowledgement Number (32 bits): If the ACK control bit is set this field
contains the value of the next sequence number the sender of the segment is
expecting to receive. Once a connection is established this is always sent. The
acknowledgement number is given in the unit of bytes.
(3) Connection ID (16 bits) (if exists): A number representing connection identifier.
(4) A (ACK, 1 bit): Indicates that there the value of Acknowledgment Number field
is valid.
(5) S (SYN, 1 bit): Synchronize sequence numbers (connection establishment).
(6) F (FIN, 1 bit): No more data from sender (connection termination)
Avoid losing messages.
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
2 / 4
In case of lost message, detect and recover it (time that takes to recover a message is
related to both reliability and low latency requirements, so the protocol should pay a
special attention to this issue)
Handle messages reordering.
(Optional) Add Congestion Control. The control algorithm is AIMD (like TCP)
The maximum UDP packet size is 524 bytes including a header (512 bytes for the
payload).
The maximum sequence and acknowledgment number should be 102400 and be reset
to zero whenever it reaches the maximum value.
The retransmission (and appropriate congestion control actions) should be triggered
when no data was acknowledged for more than time-out value (e.g. 5 or 10 seconds,
the fixed or dynamic retransmission timeout).
Initial and minimum congestion window size (CWND) should be 512.
The maximum congestion window size (CWND) is 51200.
The initial slow-start threshold (SS-THRESH) should be 10000.
If ACK field is not set, Acknowledgment Numberfield should be set to 0.
FIN should take logically one byte of the data stream (same as in TCP).
FIN and FIN | ACK packets must not carry any payload.
3. Test and Evaluation
(1) First, assume there is no packet loss. Just have the server send a packet, the receiver
responds with an ACK, and so on.
(2) Second, introduce a large file transmission. This means you must divide the file into
multiple packets and transmit the packets based on the current flow and congestion
window size.
(3) Third, introduce packet loss. Now you have to add a timer at the first sent and
unacked packet. There should be one timeout whenever data segments are sent out.
Also congestion control features should be implemented for the successful file
transmission.
4. Performance Analysis
The common performance metrics:
(1) Delay: the time difference between the transport layer accepting data from upper
layer at the sender and the transport layer passing the data to upper layer at the
receiver.
(2) Goodput (useful throughput): Throughput is defined as the number of bits arriving
at the receiver per unit of time. One problem of using throughput as a performance
measure is that it does not take re-transmissions into account. For example: in
Go-Back-N, we may retransmit many redundant packets, which are still counted
toward the traditional throughput definition. These redundant bits are not useful for
the application. A more meaningful measure is known as “Goodput”, which counts
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
3 / 4
the number of bits passed to upper layer here. This basically means that we ignore all
corrupted packets, un-necessary retransmissions, and headers.
(3) Overhead: There are three aspects of overhead: communication, processing, and
storage. Communication overhead has direct relationship with Goodput measurement.
Processing and storage (memory) overhead comes with the additional complexity of
the Selective Repeat protocol, which is its main drawback.
Evaluation is done based on key parameters such as Loss rate, Window size,
Retransmission Timeout, Packet size, total data sent and Network Delay, Channel bandwidth,
Round trip time.
5. Programming
Language: any (C, C++, Java, platform independent)
No high-level network-layer abstractions are allowed in this project. You are allowed
to use some high-level abstractions for parts that are not directly related to
networking, such as string parsing, multi-threading. Use of C++ or Java is preferred.
6. Submission
(1) Project Report
Cover Page
• Project name
• Student Number
• Name
• Date
• School and University
Contents
• Objectives
• Description of Distance Vector Algorithm
• Requirement of experiment
• Programing language/Developing platform and tools
• Design ideas
• Data structures
• Implementation (development tools, model, definitions of objects/methods and
processes and threads, processing flows, etc.)
• Test, Verification, Result Analysis, Performance Analysis, Screenshot with
explanation
(2) Readme.txt
• Plain text format.
• Contain any bugs or issues that you know of in your code.
• It should also indicate how you run your code using command line.
(3) Source Codes
• The source of the program (Must have a program comments).
Project 2: Reliable Data Transfer over UDP (RDT-UDP)
4 / 4
(4) Running Codes
• The compiled program that can run on Windows or Linux system.