List 1

Topics of interest eg construction management , legislation , facilities management , costing .

Begin by listing topics of interest in quite broad terms e.g. Concrete prestressing ; Project procuirement approaches ; Price setting mechanisms . Such topics are too broad and insufficiently specific for certain types of research , but topics such as " THe culture of design and build contractors ' or "Effect or water content on bearing capacity of London Clay " could lend themselves to investigation .


List 2 Personal strengths and weaknesses

A second list of personal strengths and weakness may contain self evaluations such as "goodat economics but quite weak in maths and statistics or experienced in new build and in contracting but n experience in refurbishment or design consultancy .

List 3 : Topics of CUrrent Interest in Practice
Over a recent years , " hot topics " have included ; buildabilituy , sick building syndrom , life cycle quality assurance , quality management , refurbishment , facilities management , dispute resolution procedures , public - private partnerships . Other topics , although never getting quite " red hot " , many retain enduring interest " CAD , communications , information flows , payments mechanisms , interparty relationship , price determination , environmental issues , client satisfaction , demolition of prestressed tower blocks .

List 4 Data required

Data requirement are detemrined by the topic of investigation , the research ' question 9s) ' and the varibes involved - especially the independent and dependent variables . What is to be measured must , first , be defined so that apprpriate metrics can be selected ; that is especially important for qualitative data , such as quality of constrction project or component . However , quantitatvie data must be defined also - such as the capital costs of a project ( to whom , when to measure , unit of measurement . etc ) .

List 5 : Sources of Data
Once the data requirements have been identified , the possible sources of data should be noted ; some sources are likely to be better than others regarding ease of collecting the data , quality of the data accuracy , amount of data ) . Access to data should , ideally , match the data required , but is likely to be modified by practicalities of obtaining the data ; especially if the data are 'sensitive ' (e.g. costs , safety , corruption ) . Employers and sponsoring bodies as well as professional institutions may be helful in securing access to the data needed .

List 6 : Research Limitations
Research limitations will be : the constraints on the resources available to execute the research and , hence , the scope of the study , data available , the methods / techniques employed . The time and resources avaialable for the research , especially for dissertation and academic theses , are likely to be well-known , so it is the particular llimitations which relate to the research which should be noted  - especially data avaialability . However , if limitations  should have been foreseen from the nature of the topic - sensitiveit , the literature , experiences of other researches etc . - care  must be taken to explain why such limitations were not taken into account and avoided in the research design . In any event , it is important to note the reasons for what was done to overcome the limitations and to ensure validity of the resulant research .

Waveguides

Above 2GHz , the wavelength is short enough to allow practical , efficient energy transfer by different means . A waveguide is a conducting tube through which energy is transmitted in the form of electromagnetic waves . The tube acts as a boundary that confines the waves in the enclosed space . The Faraday cage effects prevents electromagnetic effects from being evident outside the guide . The electromagnetic fields are propagated thjrough thje waveguide by means of reflections against its inner walls , which are considered perfect conductors . The intensity of the fiels is greatest at the center along the x dimension , and must fiminish to zero at the end walls because the existence of any field parallel to the walls at the surface would cause an infinite current to flow in a perfect conductor . Waveguides , of course , cannot carry RF in this fashion .
The X , Y , and Z dimensions of a rectangular waveguide can be seen in the following figure :

There are an infinite number of ways in which the electric and magnetic fields can arrage themselves in a waveguide for frequencies above the low cutoff frequency . Each of these field configurations is caled a mode . The modes may be separated into two general groups . One group , designated TM reaction of propagation , but has a component of the electric field in the direction of propagation . THe other type , designated TE ( Transverse Electric ) has the electric field entirely transverse , but has a component of magnetic field in the direction of propagation .


The mode of propagatyion is identified by the group letters followed by two subscripte numerals . For example , TE 10 , TM 11 , etc . The number of possible modes increases with the frequency for a given size of guide , and there is only one possible mode , called he dominant mode , for the lowest frequency that can be transmitted . In a rectangular guide , the critical dimension is X . This dimensions must be more than 0.5 diumlar at the lowest frequency to be transmitted . in practice , the Y dimension usually is made about equal to 0.5 X to avoid the possibility of operation in other than the dominant mode . Cross - sectional shapes other than the rectangle can be used , the most important being the circular pipe . Much the same considerations apply as in the rectangular case . Wavelength dimensions for rectangular and circular guides are given in the following table , where X is the width of a rectangular guide and r is the radius of a circular guide . All figure apply to the dominant mode .

Cables

RF cables are , for frequencies higher than HF , almost exclusively coaxial cable ( or coax for short , derived from the words of common axis ) . Coax cables have a core conductor wire surrounded by a non-conductive material called dieletric , or simply insulation . THe dielectric is then surrounded by and encompassing shielding which is often made of braided wires . The dielectric prevents and electrical connection between the core and the sielding . Finnally , the coax is protected by an outer casing which is generally made from a PVC material . The inner conductor carries the RF signal , and the outer shield prevents the RF signal from radiating to the atmosphere , and also prevents outside signals from interfering with the signal carried by the core . Another interesting fact is that high frequency electrical signal always travels along the outer layer of a conductor : the larger the central conductor , the better signal will flow . This is called the "skin effect" .

eVEN TRHOUGH the coaxial construction is good at containing the signal on the core wire , there is some resisance to the electrical flow : as the signal travels down the core , it will face away . This fading is known as attenuation , and for transmission lines it is measured in devibels per meter ( dB/m) . The rate of attenuation is a function of the signal frequency increases , so does its attenuation . Obviously , we ned to minimize the cable attenuation as much as possible by keeping the cable very short and using high quality cables .

Here are some points to consider when choosing a cable for use with microwave devices .

1. The sorter the better : The first rule when you install  a piece of cable is to try to keep it as short as possible . The power loss is not linear , so doubling the cable length means that you are going to lose much more than twice the power . In the same way , reducing the cable length by half gives you more than twice the power at the antenna . The best solution is to place the transmitter as close as possible to the antenna , even when this means placing it on a tower .

2. "The cheaper the worse !" The second golden rule is that any money you invest in buying a good quality cable is a bargain . Cheap cables are intended to  be used at low frequencies , such as VHF . microwaves requires the highest quality cables available . All other options are nothing but a dummy load .

3. Always avoid RG-58 . It is intended  for thin Ethernet networking , CB or VHF radio , not for microwave .
4. Always avoid RG-213 . it is intended for CB and HF radio . In this cas he cable diameter does not imply a high quality , or low attenuation .

5. Whenever possible , use Heliax ( also called "Foam") cables for connecting the transmitter to the antenna . When Heliax is unabavailable , use the best rated LMR cable you can find . heliax cables have a solid or tubular center conductor with a corrugated solid outer conductor to enable them to flex . Heliax can be built in two ways , using either air or foram as a dielectric . Air dielectric heliax is the most expensive and guaranttess the minimum loss , but it is very difficult to handle . Foam dielectric heliax is slightly more lossy , but is less expensie and easier to install . A special procedure is required when soldering connectors in order to keep the foam dielectric dty and uncorrupted . LMR is a brand of coax cable available in various diamter that works well at microwave frequencies . LMR-400 and LMR-6000 are a commonl used alternative to Heliax .

6.  Whenever possible , use cables that are pre-crimped and tested in a proper lab . Installing connectors to cable is a tricky business , and is difficult  to do properly even with the proper tools . Unless you have  access to equipment that can verify a cable you make yourself ( such as a spectrum analyzer and signal generator , or time domain reflectometer ) , troubleshooting a network that uses homemade cable can be difficult .

7. Don;t abuse your transmission line . never step over a cable , bend it too much , or try to unplug a connector by pulling directly the cable . All of those behaviour may change the mechanical characteristic of the cable and therefore its impedance , short the innner conductor to the shield , or even break the line . Those problems are difficult to track and recognize and can lead to unpredictable behavior on the radio link .

Antennas & Transmissions Lines

The transmitter that generates the RF power to drive the antenna is usually locate dat some distance from the antenna terminals . The connecting link between the two is the RF transmission line . its puirpose is to carry RF power from one place to another , and to do this as efficiently as possible . From the receiver side , the antenna is responsbile for picking up any radio signals in the air and passing them to the receiver with the minimum amount of distortion , so that the radio has its best chance to decode the signal . For these reasons , the RF cable has a very important role in radio systems : it must maintain the interity of the signals in both directions .

More Information

While bandwidth optimization is a complex and often difficult subject ,m the techniques in this chapter should help reduce obvious sorces of wasted bandwidth . To make the best possible use of avaialble bandwidt , yu will need to dfine a good access policy , set up comprehensive monitoring and analysis tools , and implement a networ archtecture that enforces desired usage limits .

For more information about bandwidth optimization , see the free book How to Accelerate Your Internet

Performance-enhancing proxy (PEP)

The diea of a performance-enhancing proxy is described in RFC 3135 and would be a proxy server with a large disk cache that has RFC 1323 extensions , among other features . A laptp has a TCO sessuib wutg te OEO at the ISP . That PEP , and the one at the satellite provider , communicate using a different TCP session or even their own proprietary protocol . The PEP at the satellite provider gets the files from the web server . In this way , the TCP session is split , and thus the lik characteristics that affect protocol performance ( long fat pipe factors ) are overcome . Addirtionally , the PEP makes use of procing and pre-fetching to accelerate web access further .

Implications for universities

If a site has a 512 kbps connection to the Internet , the default TCP/Ip settings are likely sufficient , because a 64KB window size can fill up to 984 jbps . But if the university has more than 984 kbps , it might in some cases not get the full bandwidth of the available link due to the "long fat pipe network " factors discussed above . What these factors really imply is that they prevent a single machine from filling the entire bandwidth . This is not a bad thing during the day , because many people are using the bandwidth . But if , for example , there are large scheduled downloads at night , the administrator might want those downloads to make use of the full bandwidth , and the "long fat pipe network " factors might be an obstacle . his may also become critical if a significant amount of your network traffic routes through a single tunnel or VPN connection to the other end of the VSAT link .

Administrator might consider taking steps to ensure that the full bandwidth can be achieved by tuning their TCP / Ip settings . If a university has implemented a netowrk where all traffic has to go through the proxy ( enforced by network layout ) , then the only machines that make connections tot he Internet will be the proxyand mail servers .

Tranmission errors

In order TCP/IP implementations , packet loss is always considered to have been caused by congestion ( as opposed to link errors ) . When this happens , TCP performs congestion avoidance , requiring three duplicate ACKs or slow start in the case of a timeout . Because of the long RTT values  , once this congestion-control phase is started , TCP/IP on satellite links will take a longert time to return to the previous throughput level . THerefore errors on a satellite link have a more seruious effect on the performance of TCP than over low latency links . To overcome this limitation , mechanisms such as Selective Acknowledgment (SACK) have been developed . SACK specifies exactly hose packets that have been received , allowing the sender to retransmit only those segments that are missing because of link errors .

The mircosoft Windows 2000 TCP/IP Implementation Details White Paper states

"Windows 2000 introduces upport for an important performance feature known as Selective Acknowledgment (SACK ) . SACK is especially important for connections using large TCP window sizes' .

SACK has been a standard feature in Linux and BSD kernels for quite some time . be sure that your Internet router and your ISP;s remote side both support SACK .

Large bandwidth-delay product

The amount of data in transit on a link at nay point of time is the product of bandwidth and the RTT . Because of the high latency of the satellite link , the bandwidth -delay product is large . TCP/IP allows the remot ehost to send a certain amount of data in advance without acknowledgment . An achnowledgment is usually required fro all incoming data on a TCP/Ip connection . However , the remote host is always allowed to send a certain amount of data without acknowledgment , which is important to achieve a good transfer rate on large bandwidth delay product connections . This amount of data is called the TCP window size . Th window size is usually 64kb in modern TCP / IP implementations .

On satellite networks , the value of the bandwidth -delay product is important . To utilize the link fully , the window size of the connection should be equal to the bandwidth -delay product . If the largest window size allowed is 64kb , the maximum theoretical throughput achievable via satellite is ( window size ) / RTT , or 54KB/520ms . This gives a maximum data rate of 123 KB/s which is 984 kbps , regardless of the fact that the capactiy of the link may be much greater .

Each TVP segment header contains a field called advertised window , whcih specifies how many additional bytes of data the receiver is prepared to accept . The advertised window is the receiver's current avaiable buffer size .

The sender is not allowed to send more bytes than the advertised window . To maximize performance , the sender should set its send buffer size and the receiver should set its receive buffer size to no less than the bandwidth -delay product . This buffer size has a maximum value of 64KB in most modern TCP/IP implementations .

TO overcome the problem of TCP / IP stacks from operating systems that don't increase the window size beyond 64KB , a technique known as TCP acknowledgment spoofing can be used ( see Performance Enhancing Proxy ,below ) .

Long round-trip time (RTT)

Satellite links have an average RTT of around 530ms to the first hop . TCP uses the slow - start mechanism at the start of a connection to find the appropriate TCP/IP parameters for that connection . Time spent in the slow-start stage is propriational to the RTT , and for a satellite link it means that TCP stays in slow-start  mode for a longer time than would otherwise be the case . This drastically decreases the throughput of short-duration TCP connections .This is can be seen in the way that a small website might take suprisingly long to load , but when a large file is transferred acceptable data rates are achieved after a while .

Furthermore , when packets are loss, TCP enters the congestion -control phase , and owing to the higher RTT , remains in this phase for a longer time , thus reducing the throughput of both short- and long duration TCP connections .

TCP / IP factors over a satellite connection

A VSAT is often referred to as a long fat pipe network . This term refers to factors that affect TCP/ IP performance on any network that has relatively large bandwidth , but high latency . Most Internet connections in Africa and other parts of the developing world are via VSAT . Therefore , even if a university gets its connection via an ISP , this section might apply if the ISP's connection is via VSAT . The high latency in satellite networks is due to the long distance to the satellite and the constant speed of light .This distance adds about 520ms to a packet's round - trip time (RTT) , compared to a typicalRTT between Europe and the USA of about 140ms . 

tHE FACTORS that most significantly impact TCP  / Ip performance are long RTT , large bandwidth dela product , and transmission errors . 

Generally speaking , operating systems that supporty modern TCP/IP implementations should be used in a satellite network . These implementations support the RFC 1323 extensions : 

• The window scale option for supporting large TCP window sizes ( larger than 64kb) . 
• Selective acknowledgment (SACK) to enable faster recovery from transmission errors . 
• Timestamps for calculating approperiate RTT and retransmission timeout values for the link in use . 

Internet Link Optimization

As mentioned earlier , network throughtput of up to 22Mbps can be achieved by using standard , unlicensed 802.11g wireless gear . THis amount of bandwidth will likely be at least an order o magnitude higher than that provided by Your Internet link , and should be able to comfortably suport many simultaneous Internet users .

But if your primary internet connection is throught a VSAT link , you will encounter some performance issues if you rely oin dfault TCP / IF parameters . By optimizing ypur VSAT link , you can significantly improve response times when accesing Internet hosts .

Window NT

To install the DNS service on windows NT4 : select Control Pannel - Network - services - add - microsoft DNS server . Insert the Windows NT4 CD when prompted . Configuring a caching-onl server in NT is described in Knowledge Base article 167234 . From the article :

"simply install DNS and run the Domain name System manager . Click on DNS in the menu , select New Server , and type in the IP address of your computer where you have installed DNS . You now have a caching only DNS server."

dnsmasq

One alternative caching DNS server is dnsmasq . It is avaiable for BSD and most LInux distributions , or from http://www/thekelleys.org.uk/dnsmasq/. The big advantage of fnsmasq is flexibility : it easily acts as both a caching DNS prox and an authoritative source for hosts and domains , without complicated zone file configuration . Updates can be made to zone data without even restarting the service . it can also serve as a DHC server , and wll integrate DNS service with DHCP host request . It is very lightweight , stable , and extremely flexible . Bind is likely a better choice for very large networks ( more than a couple of hundred nodes ) , bu tht esimplicity and flexibility of fnsmasq makes it attractive for small to medium sized networks .

DNS caching and optimization

Caching-only DNS servers are not authoritative for any domains , but rather just cache results from queries asked of them by clients . Just like a proxy servers that caches popular web pages for a certain time , DNS addresses are cached until their time to live (TTL ) expires . THis will reduce the amount of DNS traffic on your Internet connection , as the DNS cache may be able to satisfy many of the queries locally . Of course , client computers must be configure dto use the caching - only name server as their DNS server . When all clients use this server as their primary DNS server , it will quickly populate a cache of IP addresses to names , so that previously requiested names can quickly be resolved . DNS servers that are uthroitative for a domain also act as cache name - address mappings of hosts resolved by them .

Bind (named)
Blind is the de facto standard program used for name service on Internet . When Blind is installed and running , it will act as a caching server ( no urther configuratuon is necessary ) . Bind can be installed from a package such as a Debian package or an RPM . Instaling from a package is usually the easiest method . In Debian , type

apt-get install bind 9

In addition to running a cache , Bind can also host authroitative zones , act as a slave to authriitative zones , implement split horizon , and just about every-thing else that is possible with DNS .

Proxy Specifications

On a university campus network , there should be more than one proxu server , both for performance and also for redundancy reasons . With today's chea[er and larger disks , powerful proxy servers can be built , with 50GB or more disk space allocated tot he cache . Disk performance is important , therefore the fastest SCSI disks would perform best ( althought an IDE asd cache is beter than none at all ) . RAID or mirroring is not recommended .

It is also recommended that a separate disk be dedicated tot he cache . For example , one disk coulb be for the cache , and a second for the operating system and cache logging . Squid is designed to use as much RAM as it can get , because when data is retrieved from RAM it is much faster than when it comes from the hard disk . For a campus network , RAM memory should be 1GB or more :

• Apart from the memory reqyuired for the operating system adn other applications , squid requioired 10MB of RAM for every 1 gb of disk cache . Therefore , if there is 50GB of disk space allocated to caching . Squid will require 500MB extra memory . 
• The machine would also require 128MB for Linux and 128MB for Xwindows . 
• Another 256 MB should be added for other applications and in order thant everthing can run easily . Nothign in creases a machine's performance as need to use the hard disk . Memory is thousand sof times faster than a hard disk . Modern operating systems keep frequently accessed data in memory if there is enough RAM available . But they use the page file as an extra memory area when they don't have enought RAM . 

Cache hierachies

When an organization has more than  one proxy server , the procies can share cashed information among them . For example , if  aweb page exits in server A's cache , but not in the cache of server B , a user connected via server B might get the cached object from server A via server B .

Preventing users from bypassing the proxy serve

While circumventing Internet censorship and restrictive information access policy may be a laudable polictical effort , proxies and firewalls are necessary tools in aras with extremely limited bandwidth . Wihtout them , the stability and usability of the network are threatened by legitimate users themselves . Techniques for bypassing a proxy serve can be found at . Thus site is useful for administrator to see how their network measures up against these techniques .

Avoiding noise

The unlicensed ISM and U-NII bands represent a very tiny piece of the known electgromagnetic spectrum . Since this region can be utilized without paying license fees , many consumer devices use it for a wide range of applications . Cordless phones , analog video senders , Bluetooth , baby monitors , and even microwave ovens compete with wireless data networks for use of the very limnited 2.4 GHz band . These signals , as well as othr local wireless networks , can cause significant problems for long range wireless links . here are some steps you can use to reduce reception of unwanted signals .

- Increase antenna gain on both sides of a point-to-point link . Antennas not only add gain to a link , but their increased directionalityu tends to reject noise from areas around the link . Two high gain dishes that are pointed at each other will reject noise from directions that are outside the path of the link . Using omnidirectional antennas will receive nolise from all directions

- Use sectorials instead of using an omnidirectional . By making use of several sectorial antennas , you can reduce the overall noise received at a distribution point .
By staggering the channels used on each sectorial , you can also increasethe available bandwidth to your clients .

- Dont use an amplifier . As we will see in Chaptr 4 , amplifiers can make interference issues worse by indiscriminately amplifying all received signals , including sources of interference . Amplifiers also cause interference problems for other nearby users of the band .

- Use the best available channel . Remember that 802.11b/g channels are 22MHz wide , but are only separate by 5 MHz .Perform a site survey , and select a channel that is as far as possible from exisiting sources of interference . Remmeber that the wireless alndscape can change at any time  as people add new devices ( cordless phones , other netwrks , etc . ) if your link suddenly has trouible sending packets , you may need to perfrom an other site survey and pick a different channel .

- Use smaller hops and repeaters , rather than a single long distance shot . Keep your point-to-point links a short as possible . While it may be posible to create a 12km  links that cuts across the middle of a city , you will likely have all kinds of interference problems . if ypou can break that link into two or three shorter hops  , the link will likely be more stable . Obviously this isn't possible on long distance rural links where power and mounting structuress are unabialable , but noise problems are also unlikely in those settings .

- If  possible , use 5.8 Ghz , 900Mhz , or another unlicensed band . While this is only a short term solution , there is currently far more consumer equiment installed in the field that uses 2.4Ghz . Using 802.11a  or a 2.4GHz to 5.8 GH step-up device will let you avoid this congestion altogether . if you can find it  , some old 802.11 equipment uses unlicensed spectrum at 900MHz ( unjfortunately at much lower bit rates 0 / Other technologies , such as Ronja use optical technology for short distance , noise - free links .

- If all else fails , use licensed spectrum . These are places where all available unlicensed spectrm is effectively used . In these cases , itmay make sense t spend the additional mone for proprietary equipment that uses a less congested band . for long distance poin-to-point  links that require very high throughput and maximum uptime , this is certainly an potion . of course , thjese features come at a much higher price tag compared to unlicensed equipment .

To indentify source of noise , you need tools that will show you what is happening in the air at 2.4 GHz . We will see some exaples of these tools in Chapter 6.

Radio Mobile under Linux

Radio Mobile will also work using Wine under Ubuntu Linux . While the application runs , some button labels may run beying the frame of the button and can be hard to read .

We were able to make Radio Mobile work with Linux using the following environment :

- IBM Thinkpad x 31
- ubuntu Breezy ( v5.10) , http :
- Wube versuib

Radio Mobile

Radio Mobile is a tool for the design and simulation of wireless systems . it predicts the performance of a radio link by using information about the equipment and a digital ma of the area . it is public domain software that runs on Windows , or using linux and the Wine emulator .

Radio Mobile uses a digital terrain elevation model for the calculation of coverage , indicating received signal strength at various points along the path . It automatically builds a profile between two points in the digital map showing the coverage area ad first fresnal zone . During stimulation , it checks for line of sight and calculatres the Path Loss , including losses due to obstacles . It is possible to create networks of different topologies , including net master / slave , point-to-point , and point-to-multipoint  . The software calculates the coverage areas from the base station in a point-to-multipoint system . It works for systems having frequencies from 100kHz to 200kHz . Digital elevation maps ( DEM) are avaialable for free from several sources , and are available for most of the world . DEMs do not show coastlines or other readily identidiable landmarks , but they can easily be combined with other kinds of data ( such as aerial photos or toppographical charts ) in several layers to obtain a more useful and readily recognizable representation . You can digitize your own ,maps and combine them with DEMs . The digital elevation maps can be merged with scanned maps , satellite photos and Internet map services ( such as Google Maps ) to produce accurate prediction plots .