Optimum Data Rates and Packet Sizes for a Wireless Integrated Emergency Medical Services Environment

By

Sowmya L Mulukutla

Under the Guidance of Prof.David G. Green

A Project Report

Submitted to the graduate faculty of
The University of Alabama at Birmingham , in partial fulfillment of the requirements for the degree of Master of Science

BIRMINGHAM , ALABAMA

2005

ABSTRACT OF THE PROJECT

GRADUATE SCHOOL , UNIVERSITY OF ALABAMA AT BIRMINGHAM
Degree Masters

Program Electrical and Computer Engineering

Name of Candidate Sowmya L Mulukutla

Committee Prof. David G.Green (Committee Chair), Dr. Gary J. Grimes, Dr. Greg

Vaughn

Title Optimum Data Rates and Packet Sizes in a Wireless Integrated    Emergency Medical Services Environment

This project aims at building a standalone application which would detect connectivity and explore the protocols and data-rates that would be optimum in an application involving wireless medium for data or any type of information transfer. This application can be implemented and tested in order to work for a 911 Emergency Medical Service setting, dealing with Pre-Hospital Patient Care. The Pre-Hospital Patient Care can be enhanced by making the patient data available to the physicians or the Emergency Doctors (ED) even before the patient arrives at the hospital. The above mentioned objective could be met by integrating wireless technologies into the Emergency Medical Services (EMS). The scenario considered for implementing and testing my application, is of an ambulance uploading patient information while driving through a hotspot (wireless access point) keeping in mind the constraints of speed and accuracy, from the event location to the hospital. Various protocols and packet sizes have been taken into consideration. This project gives an overview of performance of different wireless protocols at different signal strengths. The project analyzes the data rates obtained for different wireless selection standards namely, 802.11 b and 802.11 b/g at different distances from the hotspot or the wireless access point and thus proposes a best combination of the wireless selection standard, signal strength and packet size to be chosen to upload data, so that there is minimal data loss

ACKNOWLEDGMENTS

I would sincerely like to thank the committee members Dr. Gary J. Grimes, Dr.Greg Vaughn and Prof.David G. Green (Committee Chair) for taking time off for my project. I would specially like to thank Prof.David G.Green, my advisor for being a great source of help and a driving force for me to complete my project.

I would also offer my special gratitude to Dr.Helmuth F.Orthner, for providing me a chance to do my Masters Project as a part of his grant project, the Advance Networking Infrastructure for Health and Disaster Management (ANIHDM), funded by the National Library of Medicine (NLM) under contract (N01-LM-3-3513) . I would also like to thank his crew for offering me the help and resources for my project. Last but not the least I would like to thank everyone who has been a part of my efforts to complete the project.

TABLE OF CONTENTS

Title Page …………………………………………………………………………..... i

Abstract …………………………………………………………………………....... ii

Acknowledgments ………………………………………………………………...... iii

Table of Contents ………………………………………………………………....... iv

•  Introduction ……………………………………………………………………………1

•  Overview Of Standards And Protocols……………………………...............................3

•  802.11 Networking Protocol

•  Transport Layer Protocol …………………………………………………………..... 3

•  Experiment Design …………………………………………………………………. 4

4.1 Components

  4.1.1 Client

  4.1.2 Server

  4.1.3 Log Files

4.1.4 Tools

•  Experiment Protocol……………………………………………………………...... 13

•  Data Analysis……………………………………………………………………... 16

6.1 Data Collection Process

6.2 Overlay of Data Collection Process

6.3 Results and Analysis

6.3.1 Results of Test I

6.3.2 Ethereal Results

6.3.3 Results of Test II

6.3.4 Results of Test III

•  Outcomes…………………………………………………………………………….23

•  Discussion……………………………………………………………………………27

•  References

•  Appendix

•  INTRODUCTION

The use of wireless technology is quickly becoming an established part of many societies today. There are millions of people around the globe who are using wireless technology. Wireless technology has greatly improved ones ability to work and communicate at home or at work in our local and global communities. In the jargon of the wireless world, hotspots are defined as the backbone to any wireless set-up. Hotspots are the Wi-Fi access points or areas, in particular, for connecting to the Internet. These are designated access points, advertised as ‘ wireless hotspots ' and most use Wi-Fi technology (also called IEEE 802.11) to connect one to the Internet. Hotspots are fixed and have a limited range — one needs to be within about 100 meters of a location. To use a Wi-Fi hotspot, one needs:

•  A Wi-Fi-enabled laptop or PDA.

•  To set up and pay for a wireless account with the ISP supplying the access.

Many fields today merge with the wireless mode of operation and have benefited enormously from this step. Medical field is one of those most benefited areas where the integration of wireless technologies has made a huge difference in the total outlook of modern medical treatment. The Medical field has equipped itself with many of the latest technologies to deliver better healthcare, but the area of Pre-hospital medical care is still trailing behind. From a communications perspective, Emergency Medical Services (EMS) personnel are, for the most part, isolated from the rest of the health care delivery system. In a medical emergency, prognosis of a condition depends highly on the quality of treatment delivered in the “Golden Hour”. “Golden Hour” is the time when EMS personnel reaches the patient and with some preliminary treatment transports him to a hospital relevant to the emergency and some initial time after admitting the patient. There are two important places where an improvement in the Pre-Hospital medical care can be achieved; one of them being acquiring additional professional help from the Emergency Physicians who would take the patient in for treatment, and the other being updating and giving some control to the Emergency department (ED) Physician where the patient would be transported. The cost and current practices rule out availability of a physician at the incident location and thus alternate ways of improving the patient's treatment in the “Golden Hour” before he arrives to the hospital are to be sought after. One solution with a maximum probability of success is the integration of wireless technologies with the Emergency Medical Services offered now.

The grant project under Dr. Helmuth F Orthner , Dr.Gary J Grimes and Dr. Thomas E. Terndrup named the “Advance Networking Infrastructure for Health and Disaster Management” (ANIHDM), funded by the National Library of Medicine (NLM) under contract (N01-LM-3-3513) , looks at the prehospital (or out-of-facility) medical emergency and public safety information environments that are at a threshold of revolutionary change. The change is driven, in part, by several emerging technologies such as secure, high-speed wireless communication in the local and wide area networks (wLAN, 3G), Geographic Information Systems (GIS), Geopositional Systems (GPS), and powerful hand-held computing and communication devices. The ANIHDM project aims at enhancing the present Pre-Hospital patient care, from the time the patient is attended by the EMT to the time when he is taken to the hospital. The wireless part of the ANIHDM project deals with uploading patient data collected at the event location to a central database, from where the Emergency Physicians or doctors can access it before the patient arrives at the hospital. For this, an application needs to be built which will test the network in different settings and with different configurations and determine an optimum packet size that will be supported in that setting and with a particular standard of wireless selection mode namely, 802.11 b, 802.11 g, and 802.11 b/g.

The present course of project builds an application that determines or measures the data rate in the above mentioned scenario of uploading patient data, for a particular wireless selection standard, and for a particular amount of time. This project is written in C Sharp (C#), a .NET programming language using the Microsoft Visual Editor. The application is a client/server application which was originally planned for an open environment simulating the scenario of an Emergency Medical Trainee (EMT) collecting patient data, but due to constraints of wireless network availability outside any building in the campus , it was limited to testing inside a building. Finally the project analyzes the data rates obtained for different wireless selection standards namely, 802.11 b, 802.11 g, and 802.11 b/g at different distances from the hotspot or the wireless access point and thus proposes a best combination of the wireless selection standard, distance from the hotspot and signal strength to be chosen to upload data, so that there is minimal data loss.

•  OVERVIEW OF STANDARDS AND PROTOCOLS

The project deals with 802.11 protocols. In specific, the experiment was conducted with the 802.11b, and b/g protocols. Also at the transport layer, the project uses User Datagram Protocol (UDP) as against the Transmission Control Protocol (TCP). A brief overview of each of the above mentioned protocols and standards is given below. A detailed explanation of the 802.11 standard working and its protocols is given in Appendix-A, 802.11 Wireless Networking Protocol and detailed explanation of UDP/TCP is given in Appendix-B.

2.1 802.11 Networking Protocol

802.11 is a family of wireless networking protocols out of the IEEE. The IEEE 802.11 specifications are wireless standards that define an "over-the-air" interface between a wireless client and a base station or access point, as well as among wireless clients. Task groups within the 802.11 WG enhance portions of the 802.11 standard. A particular letter corresponding to each standard/revision, such as 802.11a, 802.11b, and so on, represents the different task groups. The most popular of these, 802.11b, has been in commercial use since 1999. It has a maximum theoretical throughput of 11 Mbps, which is only about one-tenth the speed of common Ethernets, but much faster than broadband solutions like DSL or cable modems. Other standards include 802.11g, an upcoming wireless networking protocol with speeds up to 22 Mbps and 802.11a, with speeds up to 54 Mbps.

•  TRANSPORT LAYER PROTOCOL

There are two main classifications of the transport layer protocol, namely

•  Transmission Control Protocol (TCP)

•  User Datagram Protocol (UDP)

Rational for UDP selection in this project: UDP as mentioned above is an unreliable protocol, which does not guarantee delivery of packets. The project makes use of this feature of UDP, to measure the packet loss during the transmission of packets from the mobile client to the static server, in order to propose the protocol best suited to upload data as well as the optimum packet size, so that there is minimal data loss.

•  EXPERIMENT DESIGN

•  AIM: To analyze and come to a conclusion regarding an optimum packet size that can be transmitted over a wireless medium, given a particular protocol i.e., UDP with 802.11b and b/g and for a particular set up.

•  Client/Server application.

•  Environment: . NET environment with C Sharp (C#) as the coding language

The project is basically a client/server application. It is a standalone application which would detect connectivity and explore the protocols and data-rates that would be supported in a scenario where an ambulance has to upload patient information while driving thru a hotspot with the constraints of speed and accuracy. The client and server log all the transactions going on between them to excel sheets. This data is later processed and studied in order to come to a conclusion regarding the best combination of wireless standard, and data rate in wireless mode of data transfer. The project was written in C Sharp (C #), a .NET environment programming language. The Microsoft Visual Editor was used as the editor for doing the coding in C #. Also adding to the tools are the components used to conduct the experiment,

• i. Laptop, which acts as a client, on which the client application will run.

• ii. Another laptop or a desktop which is static and will serve as server, on which the server application runs.

• iii. Wireless Network Adapter, which supports 802.11 b and b/g on the client

4.1 Components

  This project comprises of some major components, namely:

•  Client,

•  Server,

•  Log Files, and

•  Tools, like the C# language, Editor, .NET Environment.

A brief explanation of the basic components of the project is given below.

4.1.1 Client

•  Client is mobile.

•  Sends data packets of different sizes to the server through a wireless medium, to decide on an optimum data rate for a particular set of parameters like protocol 802.11b, or b/g, signal strength and time for which they are sent.

•  Features:

•  Once started the client keeps sending different amounts of data to the server for a predefined amount of time.

•  The data packet will be in the format:

ID No # 1

DATE and TIME: 3/11/2005 6:28:30 PM

Packet Size: 256bytes

Protocol used: UDP with 802.11b/g

Signal Strength: Excellent

•  The details like the protocol and signal strength will vary for every new instance because they would be manually entered as they change when the experiment is run again and again. Each of the packet contents

•  Identification number : Required in order to keep track of the packets. This feature would help in studying the packet transmission, whether there was any packet loss.

•  Time Stamp : For E.g.: 3/11/2005 5:38:37 PM :

•  In order to know when the packet was sent, if sorted according to the time.

•  Size of the packet : Can be used for logging purposes. To compare the packet size sent and received. Sizes considered are 45, 256, 500, 1000, 1500 bytes.

•  Protocol : 802.11 b/g/ or b to use for later analysis.

•  Signal Strength : In order to analyze the data rate according to the signal strength at a particular instant. This value is entered manually depending on the signal strength at that particular run of the experiment. The signal strength value is obtained from the value shown by the wireless Network Adapter of the client (laptop).

•  Variables: These values are fed to the code from separate text files, so that only the text file will be changed instead of making changes everywhere in the program, whenever needed.

•  Server IP address

•  Protocol i.e., UDP with 802.11b/g&b.

•  802.11b

•  802.11b/g

•  Signal Strength

•  Excellent

•  Very Good

•  Good

•  Low

•  Results: The client after each run of the experiment displays :

•  Total number of packets sent

•  Total number of bytes sent for a particular amount of time.

•  Data rate for that particular wireless standard, quality of signal strength and the amount of time (See Fig 4 )

Fig . 4. Client Window

4.1.2 Server

•  Server is static.

•  Keep listening to data packets as the client sends them. The server is used to statically receive data and log the received data.

•  Features:

•  Once started keeps receiving different amounts of data from the client.

•  The data packet will contain:

ID No # 1

DATE and TIME: 3/11/2005 6:28:30 PM

Packet Size: 256bytes

Protocol used: UDP with 802.11b/g

Signal Strength: Excellent

•  The details like the protocol and signal strength will vary for every new instance because they change when the experiment is run again and again. Each of the packet contents and their significance are the same as in the client, as the same contents are displayed at the server.

•  Variables: The Server IP address is fed to the code from separate text file, so that only the text file will be changed instead of making changes everywhere in the program.

5) Results: The server after every run of the client: connecting and sending packets displays the last received packet and waits for next client connection. The server logs all the data it receives along with time stamps. (See Fig . 5)

Fig . 5. Server Window

4.1.3 Log Files

Both the client and the server log all the transactions. Later these log files are analyzed for studying and deciding the data rates supported in a wireless environment, under different wireless standards. The log files include the data logs indexed with time stamps and exceptions log also indexed with time stamps.

•  Client side log file contains:

•  ClientLog.txt (See Fig 6)

•  Time Stamp For E.g.: 3/11/2005 5:38:37 PM :

•  Identification/Sequence number of the packet sent.

•  Size of the packet E.g.: 256 bytes, 1500 bytes

Format e.g.:

Fig . 6. Client Log

•  ClientExceptions.txt (See Fig 7)

• i. Exceptions occurred as a result of any hindrances while connecting or sending data packets to the server.

• ii. Logged with date and time as serial numbers, for indexing

Format e.g.:

Fig . 7. Client Exceptions

•  Server side log file contains:

•  ServerLog.txt (See Fig 8 , Fig. 9)

•  Time Stamp

•  Identification/Sequence number of the packet sent.

•  Size of the packet.

•  Protocol used, i.e., UDP with 802.11b/g/b&g.

•  Form of Movement: i.e., walking, moving at a rate slighter higher than the former walking speed.

Format e.g.:

Fig . 8. Server Log

Fig . 9. Server Log continued

•  ServerExceptions.txt

• i. Exceptions occurred as a result of any hindrances while receiving data from the client, for E.g.: if the sent data is larger than the buffer, then an exception is thrown.

ii. Logged with date and time as serial numbers, for indexing

4.1.4 TOOLS

The project was done in the .NET Environment using the C Sharp (C #) programming language.

•  .NET Environment: The .NET initiative is a Microsoft project to create a new software development platform focused on network transparency, platform independence, and rapid application development. According to Microsoft, .NET includes many technologies that are designed to facilitate rapid development of Internet and Intranet applications. .NET is a software platform, which was released in 2002. It presents a platform-independent target for software development.

•  Microsoft Visual Editor (Visual Studio .Net): .NET is a collection of development tools specifically developed for use with the .NET platform. The principal example being Visual Studio .NET, an integrated development environment provided by Microsoft. .NET languages

Visual Studio . NET: Visual Studio .NET is an IDE developed (2002) by Microsoft. It is for the Microsoft Windows operating system and is aimed primarily, but not exclusively, at development for Win32 platforms. The latest version in their line of IDEs, Visual Studio .NET supports the new .NET languages C#, Visual Basic .NET and Managed C++ in addition to C++. You can use Visual Studio .NET to make applications targeting Windows (using Windows Forms), Web (using ASP.NET and Web Services) and portable devices (using .NET Compact Framework ). [8]

•  C Sharp (C #): C# (pronounced see-sharp ) is an object-oriented programming language developed by Microsoft as part of their .NET initiative. Microsoft based C# on C++ and the Java programming language . C# is a strongly-typed object-oriented language designed to give the optimum blend of simplicity, expressiveness, and performance. The .NET platform is centered around a Common Language Runtime (similar to a JVM) and a set of libraries which can be exploited by a wide variety of languages which are able to work together by all compiling to an intermediate language (IL). C# and .NET are a little symbiotic: some features of C# are there to work well with .NET, and some features of .NET are there to work well with C# (though .NET aims to work well with many languages).

C# and Java: Below is a list of features C# and Java share, which are intended to improve on C++.

• Compiles into machine-independent language-independent code which runs in a managed execution environment.
• Garbage Collection coupled with the elimination of pointers (in C# restricted use is permitted within code marked unsafe)
• Powerful reflection capabilities
• No header files, all code scoped to packages or assemblies, no problems declaring one class before another with circular dependencies
• Classes all descend from object and must be allocated on the heap with new keyword
• Thread support by putting a lock on objects when entering code marked as locked/synchronized
• Interfaces, with multiple-inheritance of interfaces, single inheritance of implementations
• Inner classes
• No concept of inheriting a class with a specified access level
• No global functions or constants, everything belongs to a class
• Arrays and strings with lengths built-in and bounds checking
• The "." operator is always used, no more ->, :: operators
• null and boolean/bool are keywords
• All values are initialized before use
• Can't use integers to govern if statements [9]

C# carries over the programming simplicity of Java, but strides in leaps
and bounds to improve code legibility and to make class and method use much more
intuitive.  By treating all data types as objects, C# defines a true object-oriented
programming language.   Allowing the developer to make syntactical shortcuts is
a proven way to reduce maintenance and testing costs, and C# makes great advances
in language structure.  Best of all, C# allows a developer to express complex
thoughts in a natural way, not to bogged down by complicated API calls and the
paranoia over improper garbage collection. Try Blocks can have a finally clause. [10]

5. EXPERIMENT PROTOCOL

•  Server and client are on the same network connected by a wireless connection.

•  Server is static and does not move.

•  Client is mobile.

•  Start server : UDPServer.exe on port 80.

•  Server started and waiting for client to send data packets.

•  Start client : UDPClient.exe.

•  When the client starts, the user has to provide the protocol or mode of operation i.e., 802.11b, or 802.11b&g, signal strength, server IP and the time interval for which the client is expected to run, in their respective text files, from which they are fed to the code, so that they can be kept constant for a set of tests for a particular instance.

•  Client started. It keeps sending data packets of different sizes to the server as long it is in the range of an access point or a hotspot, as of now, it can be assigned a time so that the client will exit close after that pre-set time interval.

•  Conduct the experiments under various signal strengths, with varying wireless standards 802.11b, or 802.11b&g, and time intervals to obtain different data rates.

•  For E.g.:

Wireless Standard -802.11b

DATE and TIME: 3/11/2005 6:28:30 PM

Signal Strength – Low, Good, Very Good, Excellent

Packet Sizes: 45, 256, 500, 1000, 15000

Observe the number of packets lost and data rate obtained. Then repeat the experiments for next wireless standard.

   NOTE: The signal strength here is implicitly tied to the distance the client is from the access point. The different signal strengths map the client's distance from the access point.

•  Conduct the tests keeping the wireless network adapter in different modes as in, only 802.11b, or 802.11b&g.

•  The data packet will be in the format:

ID No # 1

DATE and TIME: 3/11/2005 6:28:30 PM

Packet Size: 256bytes

Protocol used: UDP with 802.11b/g

Signal Strength: Excellent

•  Server and client create log files.

•  Client side log file contains:

•  ClientLog.txt

•  ClientExceptions.txt

•  Server side log file contains:

•  ServerLog.txt

•  ServerExceptions.txt

•  Stop Client : Client stops after the pre-set time interval.

•  Server keeps running.

•  Stop server after sometime.

•  Compare after each round the client/server log files to track any lost packets and refresh the log files for new values after a few runs.

•  Analyze the log files and make case studies, i.e., for each protocol of 802.11b/g&b.

•  Depending on the analysis decide which combination of the wireless standard, signal strength and data rates best suited for this environment, keeping in mind the constraints under which the experiment was conducted.

Graphically representation of the experimental protocol is as shown in Fig 10

Fig . 10. Diagrammatic representation of the experiment protocol

6. DATA ANALYSIS

The aim of this project was to acquire data using the application developed, analyze them in order to produce an optimum combination of wireless standard, the packet size under particular signal strength, and within a certain time interval. This section deals with the study of the collected data, analysis and then tries to draw a best conclusion from the final comparison.

6.1 Data Collection Process

The data collection was done with the 802.11 b and b/g protocols. The 802.11 g was initially considered for the experiment but because of lack of a router supporting 802.11g, it was dropped. However, this application is transparent to the underlying protocol, so it can be used to test even with 802.11 g, if a router supporting 802.11g is available. Hence, the overall data collection was concentrated on the 802.11 b and b/g with different signal strengths and packet sizes. The data rates, number of packets transmitted were the main areas of concentration and hence the results also revolve around them. The backbone to this data collection and comparison is the logging property of the application, as there is no way to know any packet loss or coming to a conclusion about the protocol to be followed.

6.2 Overlay of the Data Collection Process:

TEST I – Packet Loss based on the wireless standard and signal strength

•  The total process was divided into two parts, according to the wireless standard modes, namely the 802.11 b and 802.11 b/g, and again in each of the modes, the data collection was done on the basis of the signal strengths, namely,

•  Excellent

•  Good or Very Good

•  Low or Very Low.

•  In each wireless mode, say 802.11b, packets of sizes 45 bytes, 256 bytes, 500 bytes, 1000 bytes, 1500 bytes were transmitted for a time interval of one minute, i.e., sixty seconds, each and in each of the signal strength regions individually.

•  These values were logged simultaneously both on client and server into files, with self-explanatory names to facilitate referring back later to compare or analyze them.

•  In The Fig 11 the file names are in the format of “ Bx”, where ‘x' denotes the particular test, for e.g., BEx is for test conducted with 802.11 b and with “Excellent” signal strength, similarly BG and BL denote the signal strengths “Good” and “Low”. Also the log values are classified under different folders namely “B” and “BG”, for ease of use.

•  Same is the case with the server side

Fig . 11. Client Log Files

TEST II—Packet loss according to wireless standard only, not considering the signal strength- most approximating to the real life scenario

•  This test was performed with the data packets, without any constraints of signal strengths. In this case the packets of different sizes were transmitted as the client was approaching and receding from the wireless access point at equidistance, which was about 50 to 60 feet on both sides. This test was conducted for both the wireless standards, just to compare the packet drops, based on the packet size and also the influence of the wireless standard used.

•  The results of this test are also logged at both the server and the client.

•  In the Fig 11 the files with a number suffix, say ‘B256', denote the tests conducted with the packets of that size singly without any constraint on the signal strengths. More emphasis was here given to the distance and the wireless standard than to the signal strengths.

TEST III –Just to observe loss of packets within a fraction of a time interval

•  Pull out the wired network from the server and replace it back.

6.3 Results and Analysis

6.3.1 Results of TEST I – Packet Loss based on the wireless standard and signal strength

Table.1 represents the streamlined and final comparison of the two wireless standards in different signal strengths. The top most row represents the packet sizes, 45, 256, 500, 1000 and 1400 bytes. The vertical columns show the percentage of data transmitted, i.e. performance of the respective wireless standard in that particular signal strength and for that packet size.

Table. 1 . 802.11b/g Performance Results

Table. 1.1 Number of Packets Transmitted

From the table one can conclude that there is a tradeoff between the two wireless standards, basing on the different circumstances and demands of an application, for which they are used. The graphical representation of this table is given in Fig 12 and Fig. 13 for Table 1 and Fig. 14 and Fig. 15 for Table. 2.

802.11 b/g Characteristics: From the Table. 1, it seems that

1. Transmission performance decreases as the packet size increases beyond 1000 bytes and also as the signal strength decreases.
2. It can be observed that 802.11b/g can be used to,

1. Transmit efficiently packet sizes of upto 1000 bytes under any type of signal strength,
2. Low signal strength, shows smaller data rates as from Table. 1.1 one can observe that the number of packets transmitted are significantly less than the number sent when in other signal strengths, failing to assist the evaluation of optimum packet size when the signal strength is ‘Low'. The reason for this is not only because of the low transmission performances, but also because of the Medium Access Layer (MAC) below is not able to send many packets.

Table. 2 . 802.11b Performance Results

Table. 2.1 Number of Packets Transmitted

802.11 b Characteristics : From the Table. 2, it seems that

•  Has almost the same characteristics as the 802.11b/g, regarding the performance.

•  ‘Low' signal strength is an exception, as the results were random even though the experiment was conducted more than once to verify.

•  It can be said that, 802.11 b can be used to,

•  Transmit efficiently packet sizes upto 1400 bytes under any signal strength except when ‘Low'.

•  Low signal strength, shows smaller data rates as from Table. 2.1 one can observe that the number of packets transmitted are significantly less than the number sent when in other signal strengths, failing to assist the evaluation of optimum packet size when the signal strength is ‘Low'. The reason for this is not only because of the low transmission performances, but also because of the Medium Access Layer (MAC) below is not able to send many packets as been sent when the signal strengths are ‘Excellent' and ‘Good'.

6.3.2 Ethereal Test Results

In order to verify the results obtained from my experiment, Ethereal was used for the same conditions as my experiment. Ethereal runs on the network layer. The Ethereal was run on both the server and client for all the tests.

Table. 3. Ethereal Results

Ethereal Test Outcomes:   

1. The results in Table.3 are mostly identical with the ones obtained in Table 1 and Table. 2.

Table. 4 Amount of data that is being transmitted for each packet size (in Mega Bytes)

1. From Tables. 1,2 and 3, maximum data transfer,
• For ‘Excellent' Signal Strength,1400 bytes
• ‘Good' signal strength: 1000 bytes best
• Though there is 100 % transmission of 45 bytes packets, amount of data is less
• Similar problems as mentioned for Tables. 1 and Tables. 2 repeat even for the Ethereal results, when the signal strength is ‘Low'.

6.3.3 Results of TEST II—Packet loss according to wireless standard only- most approximating to the real life scenario

Table. 5. TEST II Results

On individual packet basis one can say, from the Table. 5 that the behavior of both the wireless standards is almost the same. From the table, it is shown that irrespective of the signal strength, 802.11 b/g is more preferable to transmit packets of larger size as opposed to the observation made in the previous one, taking into consideration the signal strength.

6.3.4 Results of TEST III –Just to observe loss of packets within a fraction of a time interval

The results were not significantly different so are not provided here; the logs just showed some disturbance during the transition period, i.e., the numbering of packets started off from start again when reconnected.

7. OUTCOMES

These outcomes are strictly made from the results obtained as a result of the set-up for the experiment and may or may not relate to the real time applications.

•  From the observations tabulated in Table. 1, it can be said that while the adapter is in 802.11 b/g mode, the best packet size to be used to transmit data would be that of 1000 bytes. The transmission performance improves from lowest packet size (45 bytes) to larger packet sizes i.e. till 1000 bytes, after which it again deteriorates for the packet of 1400 bytes. Consistent repetition of this performance is shown in all three categories of the signal strength, i.e., Excellent, Good and Low. Initially 1500 byte packets were used but, there was some fragmentation that occurs at the next lower level in the OSI stack and loss of these fragmented packets while commuting for the 1500 byte packets. Thus it is advisable to use maximum of 1400 byte packets and not got for 1500 byte packets. Thus while using the 802.11b/g wireless standard; one can conclude that in order to gain highest transmission performance with minimal loss in packets, it is best to plan the packet size to be around 1000 bytes.

•  The transmission performance when the network adapter was in 802.11 b mode, as seen in Table. 2 shows a consistent increase with the increase in packet size similar to 802.11b/g. The case where the signal strength is ‘Low' is an exception as there was no consistent pattern observed in the results.

•  Low signal strength in both the cases, shows smaller data rates, as from Table. 1.1 and Table. 2.1, one can observe that the number of packets transmitted are significantly less than the number transmitted when in other signal strengths, failing to assist the evaluation of optimum packet size when the signal strength is ‘Low'. The reason for this is not only because of the low transmission performances, but also because of the Medium Access Layer (MAC) below is not able to send many packets as been sent when the signal strengths are ‘Excellent' and ‘Good'. This is a critical issue to be considered while transmitting packets, if ‘Low' signal strength is involved. One may have to keep in mind that while dealing with ‘Low' signal strengths, that not only is the transmission performance less, but also the MAC layer slows down very significantly and so are the number of packets to be transmitted.

•  Operating in 802.11b/g also has the option to connect to 802.11g networks, which may provide higher data rates.. Connecting in 802.11b mode to an access point which supports maximum data rate of the 802.11 b, would yield best results as there would be maximum throughput, than connecting in a 802.11b/g mode. Thus, if larger packets are to be sent then it is better to go for 802.11 b as they sustain good percentage of transmission. Otherwise it is wise to use 802.11 b/g as it would have the ability to connect to both b and g networks and may enhance packet transmission.

•  The data obtained collected as a result of this experiment was then grouped, by comparing the server and client logs. This data was further processed to get the transmission performance of each standard, by finding out the percentage values of each. Those values are plotted in Fig. 12 and Fig. 13 . A better view of the graphical representations of Table. 1 . and Table. 2 is available in APPENDIX-C.

7.1 The performance values and data supported in 802.11b/g as given in Table. 1 are graphically represented in the Fig 12 and Fig 13.

Fig. 12. Graphical representation of values from Table. 1

Fig. 13. Graphical representation of Data supported values from Table. 1

7.2 The performance values and data supported in 802.11b as given in Table. 2 are graphically represented in the Fig 14 and Fig 15.

Fig. 14. Graphical representation of values from Table. 2

Fig. 15. Graphical representation of Data supported values from Table. 2

8. DISCUSSION

Wireless medium for data transfer, is one of the hottest areas of research and also mostly used medium today. Especially talking about medical field, one can say that the credit for the most utilization of the wireless data transfer is taken by them. Emergency Medical Services (EMS) is one of them. In this course of the project an application was built which would test the network under different constraints and provide the optimum data rates and packet sizes to be used while uploading data in an EMS environment, for minimum data loss.

These tests were conducted with a 802.11 b router, so only 802.11 b and 802.11b/g wireless standards were tested. Usually it is the characteristic of the wireless network adapter card to connect to the lowest available wireless modes. So, even if a 802.11 g network was available, the 802.11 b would have been preferred. And also when in an 802.11b/g mode, there is a process of RTS/CTS to be done in order to know the availability of the channel. This is done because 802.11b networks cannot see 802.11g networks and to avoid collision, it needs the channel to be free. Thus there is a significant difference in the performance between the 802.11b and 802.11b/g. The values may change for more number of changes, but the final analysis and observations were observed to be almost the same, except if some noise was introduced into the environment. The correctness of the data collected can be judged comparing the obtained results with the ones obtained by using the Ethereal. The results also included the data rates, obtained by running the ethereal on the server and the client. These values are the ones which would be obtained in ideal conditions with no problems observed in the code, software or the virtual machine. The performance graphs are the graphs of the values obtained without using the Ethereal. It was obtained by taking the percentage of the received to transmitted ratio, of the packets, in order to normalize the results, or to scale the results. Another critical observation that was made during the experimentation is that during ‘Low' signal strengths, not only is the transmission performance low but also the number of packets transmitted lowers significantly, as the Medium Access Layer (MAC) slows down the packet transmission very significantly. So one should keep in mind this issue while transmitting packets of any size in ‘Low' signal strengths. All these results as mentioned before, relate to an environment set-up inside a building, rather than an open environment, due to constraints of wireless network availability outside any building in the campus.

This project is an integration of the networking world with that of the application building software world. In the course of this project, the learning curve to some new environments and languages was overcome. The .NET Environment used as a platform in this project is a framework, integrating many different languages. In order to be compatible with this environment, the application was built using the .NET programming language, C Sharp (C#). C# is almost the same as JAVA. It is also an object-oriented language, with similar classes and packages as in JAVA, though the implementation method was different. From the experience of this project, one can say that C# is not a competitor to JAVA but, an enhanced version of JAVA, with almost all its features and some additional ones too.

Future work

Due to constraint of time and resource availability, the project scope was reduced. Further developments can be made on the application to enhance and provide more secure data collection or data uploading. Some of the features which maybe added in the future are,   

•  Implement Flow control to minimize packet loss.

•  As the application is transparent to the underlying protocol used, testing with wireless standard 802.11g can be conducted the same way as the other standards like the 802.11b and 802.11b/g.

•  Integration of Geographic Information Systems (GIS), Geopositional Systems (GPS), to enable pointing the location and map it to the signal strengths, instead of feeding the signal strengths as of now.

•  Set up the experiment in an open environment, i.e. in a parking lot with own network, and collect data. This way one can also measure the influence of noise in the surrounding, which would be typical of the Emergency Medical Service environment.

•  Building a GUI and also an installer to make it a completely standalone application to run on any machines.

•  This application can also be used to test the IPV6 routers and enabled products once they come into use. In the present case, some credit to packet loss or the weird behavior in the 802.11b/g for the 1500 bytes packet can be given to the fragmentation, but IPV6 does not support fragmentation, thus this concept can be tested using this application.

REFERENCES

•  Pablo Brener (2005 March), A Technical Tutorial on the IEEE 802.11 Protocol, Available: http://www.sss-mag.com/pdf/802_11tut.pdf

•  Bradley Mitchell, (2005 April), 802.11 Standards- 802.11b, 802.11a, 802.11g , Available: http://compnetworking.about.com/cs/wireless80211/a/aa80211standard.htm

•  Virginia Polytechnic Institute and State University , (2005 April), Center For Wireless Telecommunications, Available: http://www.cwt.vt.edu/faq/80211.htm

•  Cyber Science Laboratory, (2005 April), Introduction to the 802.11 Wireless Network Standard, Available: http://www.nlectc.org/pdffiles/introduction_to_802.11_networks.pdf

•  802.11 Standards, (2005 April), Available: http://wlan.nat.sdu.dk/802_11standard.htm

•  Bradley Mitchell, (2005 April), 802.11, Available: http://compnetworking.about.com/od/wireless80211/g/bldef_80211x.htm

•  (2005 April), 802.11 A Vs B Vs G? Available: http://www.alienware.com/product_detail_pages/area-51m_7700/PDFs/Wireless_Chart.pdf

•  Wikipedia, (2005 Results), Available: http://a9.com/Visual+Studio+.NET?factekey=Visual+Studio+.NET&factdsid=2222

•  Ben Albahari , (2005 April), A Comparitive Overview of C #, Available: http://genamics.com/developer/csharp_comparative.htm

•  Jared Miniman , (2005 April), Microsoft C# vs. Java: A Syntactical and Functional Comparison , Available: http://www.devhood.com/tutorials/tutorial_details.aspx?tutorial_id=76

•  Erik Rodriguez, (2005 April), TCP Vs UDP, Available: http://www.skullbox.net/tcpudp.php

APPENDIX –A

802.11 Wireless Networking

•  802.11 WIRELESS NETWORKING

802.11 is a family of wireless networking protocols out of the IEEE. The IEEE 802.11 specifications are wireless standards that define an "over-the-air" interface between a wireless client and a base station or access point, as well as among wireless clients. The 802.11 standards can be compared to the IEEE 802.3 standard for Ethernet for wired LAN's. The most popular of these, 802.11b, has been in commercial use since 1999. It has a maximum theoretical throughput of 11 Mbps, which is only about one-tenth the speed of common Ethernets, but much faster than broadband solutions like DSL or cable modems. Other standards include 802.11g, an upcoming wireless networking protocol with speeds up to 22 Mbps (due in late 2001) and 802.11a, with speeds up to 54 Mbps.

2.1.1 802.11 Extensions/Standards

In June 1997, the Institute of Electrical and Electronic Engineers (IEEE) finalized the initial standard for wireless LANs, IEEE 802.11. This standard specified a 2.4GHz operating frequency with data rates of 1 and 2Mbps. The 802.11 protocol has many extensions, each designated for a separate task. Since the ratification of the initial 802.11 standard, the IEEE 802.11 Working Group (WG) has made several revisions through various task groups.

Task groups within the 802.11 WG enhance portions of the 802.11 standard. A particular letter corresponding to each standard/revision, such as 802.11a, 802.11b, and so on, represents the different task groups. The most popular of these, 802.11b, has been in commercial use since 1999. It has a maximum theoretical throughput of 11 Mbps, which is only about one-tenth the speed of common Ethernets, but much faster than broadband solutions like DSL or cable modems. Other standards include 802.11g, an upcoming wireless networking protocol with speeds up to 22 Mbps and 802.11a, with speeds up to 54 Mbps.

Of the WLAN technologies available now, a few are expected to be dominant over the next five years: 802.11b, g, a. [2] Out of these only the standards b and g are considered in this project. A sneak into each of these standards is given below.

2.1.2 802.11a Standard

When 802.11b was developed, IEEE created a second extension to the original 802.11 standard called 802.11a . Because 802.11b gained in popularity much faster than did 802.11a, it was believed that 802.11a was created after 802.11b. In fact, 802.11a was created at the same time. Due to its higher cost, 802.11a fits predominately in the business market, whereas 802.11b better serves the home market.

802.11a supports bandwidth up to 54 Mbps and signals in a regulated 5 GHz range. Compared to 802.11b, this higher frequency limits the range of 802.11a. The higher frequency also means 802.11a signals have more difficulty penetrating walls and other obstructions. Because 802.11a and 802.11b utilize different frequencies, the two technologies are incompatible with each other. Some vendors offer hybrid 802.11a/b network gear, but these products simply implement the two standards side by side.

• Pros of 802.11a - fastest maximum speed; supports more simultaneous users; regulated frequencies prevent signal interference from other devices
• Cons of 802.11a - highest cost; shorter range signal that is more easily obstructed

2.1.3 802.11b Standard

IEEE expanded on the original 802.11 standard in July 1999, creating the 802.11b specification. 802.11b supports bandwidth up to 11 Mbps, comparable to traditional Ethernet. 802.11b uses the same radio signaling frequency - 2.4 GHz - as the original 802.11 standard. Being an unregulated frequency, 802.11b gear can incur interference from microwave ovens, cordless phones, and other appliances using the same 2.4 GHz range. However, by installing 802.11b gear at a reasonable distance from other appliances, interference can easily be avoided. Vendors often prefer using unregulated frequencies to lower their production costs.

• Pros of 802.11b - lowest cost; signal range is best and is not easily obstructed
• Cons of 802.11b - slowest maximum speed; supports fewer simultaneous users; appliances may interfere on the unregulated frequency band

2.1.4 802.11g Standard

In 2002 and 2003, WLAN products supporting a new standard called 802.11g began to appear on the scene. 802.11g attempts to combine the best of both 802.11a and 802.11b. 802.11g supports bandwidth up to 54 Mbps, and it uses the 2.4 GHz frequency for greater range. 802.11g is backwards compatible with 802.11b, meaning that 802.11g access points will work with 802.11b wireless network adapters and vice versa. An issue is that the presence of an 802.11b user on an 802.11g network will require the use of RTS / CTS (request-to-send / clear-to-send), which generates substantial overhead and lowers throughput significantly for all 802.11b and 802.11g users. RTS / CTS ensures that the sending station first transmit a RTS frame and receive a CTS frame from the access point before sending data. A mixture of 802.11b and 802.11g requires RTS / CTS to avoid collisions because 802.11b stations can't hear 802.11g stations using OFDM

• Pros of 802.11g - fastest maximum speed; supports more simultaneous users; signal range is best and is not easily obstructed
• Cons of 802.11g - costs more than 802.11b; appliances may interfere on the unregulated signal frequency

A brief comparison of the three most widely used IEEE 802.11 standards is given in Fig 1.

Fig. 1. Wireless Chart 802.11 ABG [7]

With the ratification of the 11 Mbps Wi-Fi standard in the second half of 1999, guaranteed interchangeability and declining costs became a reality and traditional networking players such as Cisco, Lucent, 3COM, Intel and Texas Instruments entered the WLAN market. Another major advantage of 802.11b is the existence of the Wireless Ethernet Compatibility Alliance (WECA), where wireless industry leaders have united outside the standards bodies.

802.11b specification states that chip sets use a modulation scheme known as Complementary Code Keying (CCK) to transmit data signals at 11 Mbps through an unlicensed portion of the spectrum found at 2.4GHz. Considered revolutionary at the time 802.11b gave way to a new generation of products that allowed an Ethernet connection to finally break free of wires but its speed was still only one-tenth that of its wired equivalent.

In order to enhance the standard, the IEEE's overall Working Group that oversaw the development of 802.11 assigned individual tasks to several specialty groups. The mission of 802.11g was to boost the data transmission to rates of 54 Mbps while still maintaining interoperability to earlier specs.

When the original 802.11b specification was approved in 1999, the IEEE concurrently approved the specs for 802.11a. These chip sets are designed to use the OFDM scheme to transmit data at 54 Mbps through a separate portion of the spectrum (located in the 5GHz range). 802.11a is only licensed for usage in North America as opposed to 802.11b which is accepted throughout Europe and Asia as well. A great problem is that 802.11b and 802.11a were never meant to interoperate. Still, several vendors from start-ups like Sunnyvale, Calif.-based Atheros Communications to household names like Intel and 3Com have already announced their support of 802.11a. [5]

2.1 Purpose and general description of the 802.11 standard [5]

The aim of the 802.11 standard was to develop a MAC and PHY layer for wireless connectivity for fixed, portable and moving stations within a local area. The higher OSI-layers are the same as in any other 802.X standard; this means that at this level there is no difference perceptible between wired and wireless media.

The 802.11 standard describes the functions and services required by a compliant device to operate within ad hoc and infrastructure networks as well as the aspects of station mobility. The difference between ad hoc and infrastructure networks will be explained further in the report. The standard permits the operation of an IEEE 802.11 conformant device within a WLAN that may coexist with multiple overlapping IEEE 802.11 WLANs and describes the requirements and procedures to provide privacy of user information being transferred over the wireless medium and authentication of IEEE 802.11 conformant devices. When talking about WLAN, a very critical aspect is the limited throughput. This was a problem of 802.11: it only provided 1Mbps and 2Mbps rates which of course are too slow to support common requirements and explains the not that fast starting process of WLAN.

2.2 The 802.11 operation modes [5]

There are two operation modes defined in IEEE 802.11: Infrastructure Mode and Ad Hoc Mode. (See Fig 2.1 )

•  Infrastructure mode

In infrastructure mode, the wireless network consists of at least one access point (AP) connected to the wired network infrastructure and a set of wireless end stations. Access points that act as routers can also assign an IP address to the PC's using DHCP services. AP's can be compared with a base station used in cellular networks. This configuration is called a Basic Service Set (BSS). An Extended Service Set (ESS) consists of two or more BSSs forming a single subnetwork. Traffic is forwarded from one BSS to another to facilitate movement of wireless stations between BSSs. Almost always the distribution system which connects this networks is an Ethernet LAN.

FIG: 2.1 802.11 Operation Modes

•  Ad-Hoc Mode

Ad-Hoc mode is a set of 802.11 wireless stations that communicate directly with each other without using an access point or any connection to a wired network. This basic topology is useful in order to quickly and easily set up a wireless network anywhere a wireless infrastructure does not exist such as a hotel room, a convention center, or an airport. Ad-Hoc Mode is also called peer-to-peer mode or an Independent Basic Service Set (IBSS).

2.3 The 802.11 Physical Layer [5]

Two main technologies are used for wireless communications:

•  Radio Frequency (RF), and

•  InfraRed (IR)

RF in this case is located in the 2.4GHz ISM-band. RF is capable of being used for 'not line of sight' and longer distance situations. IR is not a useful technology for use in a WLAN system since it is used for short distance communications: there is a standard for such products called IrDA.

There are two methods of spread spectrum modulation used within the unlicensed 2.4-GHz frequency band:

•  Frequency hopping spread spectrum (FHSS), and

•  Direct sequence spread spectrum (DSSS)

Spread spectrum is ideal for data communications because it is less susceptible to radio noise and creates little interference; it is used to comply with the regulations for use in the ISM band. FHSS allows for a less complex radio design than DSSS but FHSS is limited to a 2-Mbps data transfer rate, the reason for this are the FCC regulations that restrict subchannel bandwidth to 1 MHz, causing many hops which means a high amount of hopping overhead. For wireless LAN applications, DSSS is a better choice. DSSS divides the 2.4GHz band into 14 channels (in the US only 11 channels are available). Channels used at the same location should be separated 25 MHz from each other to avoid interference. This means that only 3 channels can exist at the same location (figure 2). FHSS and DSSS are fundamentally different signaling mechanisms and are not capable of interoperating with each other.

2.4 The 802.11 Datalink Layer [5]

A 802.11 datalink layer is divided in two sublayers: Logical Link Control (LLC) and Media Access Control (MAC). The LLC sublayer is the same in 802.11 and other 802 LANs and can easily be plugged in into a wired LAN, but 802.11 defines a different MAC protocol.
For Ethernet LANs, the CSMA/CD protocol regulates the access of the stations. In a WLAN collision detection is not possible.

The 802.11 standard defines the protocol and compatible interconnection of data communication equipment via the air, radio or infrared, in a local area network (LAN) using the CSMA/CA medium sharing mechanism. This basic access method for 802.11 is called Distributed Coordination Function (DCF) and its mandatory for all stations.

2.4.1 CSMA/CA Medium Sharing Mechanism

CSMA/CA needs each station to listen to other users. If the channel is idle
the station is allowed to transmit. If it is busy, each station waits until transmission stops, and then enters into a random back off procedure. This prevents multiple stations from owning the medium immediately after completion of the preceding transmission. Packet reception in DCF requires acknowledgements (ACK). The period between completion of packet transmission and start of the ACK frame is one Short Inter Frame Space (SIFS). ACK frames have a higher priority than other traffic. Fast acknowledgement is one of the features of the 802.11 standard, because it requires ACKs to be handled at the MAC sublayer.
Transmissions other than ACKs must wait at least one DCF inter frame space (DIFS)
before transmitting data. If a transmitter senses a busy medium, it determines a random back-off period by setting an internal timer to an integer number of slot times. Upon expiration of a DIFS, the timer begins to decrement. If the timer reaches zero, the station may begin transmission. If the channel is seized by another station before the timer reaches zero, the timer setting is retained at the decremented value for subsequent transmission.

The method described above relies on the underlying assumption that every station can hear all other stations. This is not always the case: this problem is known as the Hidden-Node Problem. The hidden node problem arises when a station is able to successfully receive frames from two other transmitters but the two transmitters can not receive signals from each other. In this case a transmitter may sense the medium as being idle even if the other one is transmitting. This results in a collision at the receiving station.

To provide a solution for this problem, another mechanism is present: the use of RTS/CTS frames (See Fig 1.2 ). A Request To Send (RTS) frame is sent by a potential transmitter to the receiver and a Clear To Send (CTS) frame is sent from the receiver in response to the received RTS frame. If the CTS frame is not received within a certain time interval the RTS frame is retransmitted by executing a backoff algorithm. After a successful exchange of the RTS and CTS frames the data frame can be sent by the transmitter after waiting for a SIFS.

FIG 2.2 RTS/CTS Method

The drawback of using RTS/CTS is an increased overhead which may be very important for short data frames; the efficiency of RTS/CTS depends upon the length of the packets. RTS/CTS is typically used for large-size packets, for which retransmissions would be expensive from a bandwidth viewpoint.

APPENDIX –B

Transport Layer Protocol

•  TRANSPORT LAYER PROTOCOL

There are two main classifications of the transport layer protocol, namely

•  Transmission Control Protocol (TCP)

•  User Datagram Protocol (UDP)

Transmission Control Protocol (TCP) is the most commonly used protocol on the Internet. The reason for this is because TCP offers error correction. When the TCP protocol is used there is a "guaranteed delivery." This is due largely in part to a method called "flow control." Flow control determines when data needs to be re-sent, and stops the flow of data until previous packets are successfully transferred. This works because if a packet of data is sent, a collision may occur. When this happens, the client re-requests the packet from the server until the whole packet is complete and is identical to its original. (See Fig 2)

Fig . 2. Transmission Control Protocol Implementation

User Datagram Protocol (UDP) is another commonly used protocol on the Internet. However, UDP is never used to send important data such as web pages, database information, etc; UDP is commonly used for streaming audio and video. Streaming media such as Windows Media audio files (.WMA), Real Player (.RM), and others use UDP because it offers speed! The reason UDP is faster than TCP is because there is no form of flow control or error correction. The data sent over the Internet is affected by collisions, and errors will be present. It is important to remember that UDP is only concerned with speed. This is the main reason why streaming media is not high quality. (See Fig 3 )

Fig . 3. User Datagram Protocol

On the surface, an "unreliable" network protocol may not seem very worthwhile or desirable. But in fact, UDP can be very useful in certain situations, and it enjoys one key advantage over TCP -- speed. The reliability features built into TCP can be expensive in terms of overhead at execution time. This is the classic tradeoff between UDP and TCP, and it appears both transports still have their place in today's networking world. [11]