Share this:

Throughput – A Factor to Establish a Point to Point Link

Measuring actual throughput

Customers feel because they get 100Mpbps or a Gigabit Ethernet in their network, the wireless world will be the same. This is as fake as a brand called Adibas (Adidas remade in China). The wireless world takes years to develop and it has a lot of time and money spent on engineering, testing and then finally manufacturing. Take 3G for example; it was actually launched in 2001 on a commercial scale. It took 10 years before 4G could be released as the next new standard. On another note, WIMAX was also launched at the same time with the hype of being the next big thing after wireless but within 10 years it’s almost completely obsolete. There are quite a few things to consider in wireless and I’m hoping that this article will help put light on some subject formerly unknown to most people in the Middle East.

I have 5 users in my remote site and 20 users in my main office. I need a 100Mbps!
Firstly, if this is airspeed, it won’t be achieved practically in the field.

Secondly, although this question depends on what application is being used, for 20 users 100Mbps is way too high. That’s almost 5Mbps per user. It something an ISP would provide to a home user. Unless the goal is something like disaster recovery, real time replication of database or backup of complete data to another location, such throughput is not required. You could even put IP Phones in this scenario and it’s still too high. A normal IP phone requires 64Kbps (per user) of throughput with the right codecs, 128kbps (per user) if the IP Pbx system is not properly optimized. What an IT manager should be targeting is 1Mbps (per user) 2Mbps max( keeping this is as the upper limit). Actual throughput is what should be targeted not total bandwidth or airspeed. This will significantly help reduce cost and to mention learn some new stuff in the process. (I mean optimize what you already have in wireless and use QOS for voice in the network)

The other vendor is providing me with 100Mbps. How come you are providing low bandwidth?
The answer to this is – simply because Netronics provides actual throughput not airspeed. A lot of manufacturers focus on airspeed, which will never be achieved in practical situations. They mostly charge less for it because in actuality it’s probably no more than 10% of their claim i.e. 10Mbps. If the same manufacturer is asked what the actual throughput is, he won’t or shouldn’t be able to answer.

Another vendor is giving 300Mbps, which is higher than yours for the price that you are offering!
Most customer’s think that 300Mbps (airspeed) is actually what he’ll get, which is not true. Additionally most vendors don’t do much to correct that ideology because it helps them sell. An unsuspecting customer ends up then falling for the trap of; products being cheap to buy giving high bandwidth like 300Mbps as opposed to alternatives. And right after that a price war occurs in the market of 300Mbps radio.

These are all imaginary because…..well keep reading below and you’ll find out why.

What is the difference between airspeed and throughput?
Airspeed (bandwidth) is theoretical
Throughput is actual. (After all calculations are done and when the device is installed on the field)

You’ll need to know what the actual throughput is to make a proper comparison. Otherwise, you’ll be comparing an apple and an orange together. Both may be fruits but 1 is a citrus fruit and the other is not.

What does channel size have to do with throughput?
Usually manufacturer’s datasheets talk about max bandwidth that their product can achieve. Unfortunately, this is only true if the channel size is at 40Mhz. The bandwidth will be halved when a 20 MHz channel size is used and a ¼ when a 10 MHz channel size is used. So… think about what bandwidth can be achieved at a 5 MHz channel size.

E.g.  – 100Mbps radio is (a rule of thumb)
@ 40Mhz channel size – 100Mbps
@ 20Mhz channel size – 50Mbps
@ 10Mhz channel size – 25Mbps
@ 5Mhz channel size – 12.5Mbps

There may be times however, when this is not important and it’s more important to establish the link at whatever channel size. There has to be a business goal (like cost reduction compared to other methods) for this.

What does modulation have to do with throughput?
A rule of thumb is, more modulation = more bandwidth. This is because higher modulation allows more bits (data) to be sent at a given moment.

E.g – 100Mbps radio is (a rule of thumb)

@64QAM – 100Mbps
@32QAM – 50Mbps
@16QAM – 25Mbps
@8QAM – 12.5Mbps

@QPSK – 6.25Mbps
@BPSK – 3.125Mbps

So, it’s important to note, how a mixture of modulation and channel size can change the throughput levels of the radio. Again, there may be times however, when this is not important and it’s more important to establish the link at whatever channel size or modulation.

What does distance have to do with throughput?
There is an inverse relationship with these 2. More distance = lesser throughput. This is because of the air itself. Unfortunately there’s no real thumb rule here. A link budget calculation will give you an idea. Just for information sake though, this is called Fade Margin.

Does LOS or nLOs or NLOS affect my throughput?
You’ll need to review my previous article in the LOS section. I’ve mentioned there what LOS is and how we measure it.

What you need to know is that the percentage of the Fresnel zone being blocked will reflect the amount of throughput you will be losing. This is again a rule of thumb.

At 120km how much throughput I can get?
Like I’ve mentioned earlier, there are too many factors to get a black and white answer. If you want to get an estimate for this, try doing a link budget calculation. I’ve mentioned this in my earlier posts.

What is half duplex and full duplex? How does it affect my throughput?
A half-duplex system provides communication in both directions, but only one direction at a time (not simultaneously)
A full-duplex (FDX), or sometimes double-duplex system, allows communication in both directions and simultaneously.

So its easy to discover that this will do to throughput.

E.g. 100Mbps is –

@Half Duplex: 100Mbps in one direction with delay and then 100Mbps in the other direction. (Behaves like a 100Mbps; 50 upstream, 50 downstream at the same time)
@Full Duplex: 100Mbps in one direction and then 100Mbps in the other direction. (Behaves like a 200Mbps; 100 upstream, 100 downstream at the same time)

What is MIMO and Diversity modes? How does it affect my throughput?
MIMO = Multiple In Multiple out
MiMO means that there are multiple radios and antennas to send and receive a signal.

E.g 1 Radio’s airspeed is 54Mbps. If use 2×2 MIMO mode I can achieve (54+54) 108Mbps. 1 radio can send and receive at the same time and by using 2 the capacity has doubled. But do remember that this is still airspeed not actually throughput. However, it will affect throughput the same way.

Diversity mode is used when there is too much interference in an environment. I have 2 radio here as well as 2 antennas but he capacity is halved instead of doubled. This is because 1 antenna and 1 radio is sending and receiving data (signal) in vertical polarization and the other is sending the exact same data (signal) in the horizontal polarization. The receiving radio will then choose what is the best signal received by comparing the signals together.

Related articles

• LOS – A Factor Of PtP (lesleyanthony.wordpress.com)

Share this:

Line of Sight – A Factor To Establish a Point to Point Link

What is LOS? What is the effect of my link or radio?

LOS – is Line of Sight; meaning a visual line of sight between the 2 locations. Although this is conceptually correct and a lot of people go with this understanding it is not the exact definition. The right definition is – when the Fresnel zone is clear there is a clear LOS.

In this case, the best signal can be achieved. It is ideal.

What is nLOS? What is the effect of my link or radio? 

nLOS – is near Line of Sight; meaning a visual line of sight is slightly obstructed between the 2 locations. Here, the Visual Line of sight is clear but the Fresnel zone is partially obstructed.

In this case, a moderate signal can be achieved. Some throughput will be lost. This is not ideal but it’s manageable.

What is NLOS? What is the effect of my link or radio?

NLOS – is Non Line of Sight; meaning a visual line of sight is almost completely obstructed between the 2 locations, which means the Fresnel zone is almost completely obstructed as well.

In this case, a poor signal is achieved. Most of the signal is lost and throughput is really low. Despite this, usually 6-7kms NLOS for Netronics is considered possible. But again, this defers from place to place as it is purely environment dependent.

“I have Location A and Location B that I want to connect together with a distance of 120Km. Will there be LOS?”
This is both a distance and a Line of Sight (LOS) question. The answer to this is not simple. There are other questions that need to be asked and answered first but with this kind of distance, you most likely won’t be able to achieve a link easily. Let me explain. After every 50km the earth has a bulge after which the human eye can’t see anything anymore. This effect is similar on wireless PtP links as well. This bulge effect of the earth is called Earth Curvature.

E.g. Imagine having a tennis ball in front on you. With a marker mark 1 dot on 1 side of the ball and mark the 2^nd dot on the exact opposite side. Now, it’s quite evident that both dots can’t see each other but if you were to stick a long toothpick into either dots then 1 toothpick could see each other to some extent based on the height of the toothpick. Right?

So in simple words longer distances, in wireless, can be reached when you have tall towers on each side that can see each other. This establishes a LOS and that can solve the distance problem. Now the next that immediately pops up is – How much tower height do I need? Well there is a general rule of thumb to this. For every 1km (horizontal distance) you need a 1m (vertical distance) tower height.

What is the difference between nLOS and NLOS?
nLos: Near Line of Sight and NLos: Non line of sight. nLOS is very small obstruction in the Fresnel zone and NLOS could be a complete obstruction.

What is an Obstruction or what do you consider as an obstruction?
An obstruction can be anything that blocks the Fresnel zone of the radio wave.

Is a tree considered to be an obstruction?
Yes, even a tree can be considered as an obstruction.

What does Fresnel zone mean?
A Fresnel zone is a zone in the air in which the air waves are being propagated both vertically and horizontally by the antenna.

So if I have a NLOS situation I can’t establish a link?
A link a link can be established but this is based on 2 things – Distance of both locations from each other and the amount of the Fresnel zone being blocked. A site survey is better suited to get a real idea.

So how can I measure the Fresnel zone?
An 8.5Dbi antenna provides a 11 degree Fresnel zone (both vertical and horizontal)
A 24Dbi antenna provides a 3.7 degree Fresnel zone (both vertical and horizontal)

There is a simple rule to understand here and that’s based on the antenna being used.

A low gain antenna provides wider Fresnel zone and a high gain antenna provides thinner Fresnel zone. This means that low gain is good for NLOS situations and high gain is good for longer distances with clear LOS.

What is a site survey and what is a link budget? Are both the same thing?
A site survey can be conducted using 2 methods

1. Software method.  (here one can use the link budget to get an idea of the link establishment)
2. Physical method (this is going on foot and physically confirming)

So, to answer the question; a link budget is a software method for an initial site survey.

How is a site survey conducted?
A site survey can be conducted using 2 methods –

1. Software method (initial method)
2. Physical method (final confirmation to establish the link)

Here’s how you can conduct it in the preliminary stages (software method)

Step 1 (software method)

1. Use Google Earth and mark the GPS coordinates of both sites.
2. Then use the ruler and find out the distance of the 2 sites
3. Save the line and change the color of the line to make more evident
4. Along the line view if there are any obstructions
5. Take a print out of this

Step 2 (software method)

1. Use a Link budget calculator or radio planner(either 3^rd party or specific vendor software)
2. Input all values, link distance, antenna type, radio power etc
3. Save and print out the output

Step 3 (physical method)

1. Take the earlier print out and first visit the possible obstruction that was observed on the Google Earth print out
2. Observe whether this obstruction is higher than both sides by going to the roof the obstruction (if it’s a building).
3. If the obstruction is higher (NLOS), then ensure that the distance between 2 sites is short or ensure that the tower height can be increased to overcome the NLOS
4. If the obstruction is in the same line nLOS, then increasing the tower height a little would help overcome this.
5. If there is no obstruction, nothing extra maybe required.
6. Based on the situation on the field, take the results back to the link budget calculator and observe how much of the Fresnel zone would be lost if there is an obstruction and estimate that much throughput loss.
7. Tweak the results by changing the antenna or increasing tower height etc. to get desired results.

Related articles

• Distance – A Factor of PtP (lesleyanthony.wordpress.com)

Share this:

 Distance – A Factor To Establish a Point to Point Link

“I have Location A and Location B that I want to connect together with a distance of 120Km. Can you do this?”
This is both a distance and a Line of Sight (LOS) question. The answer to this is not simple. There are other questions that need to be asked and answered first but with this kind of distance, you most likely won’t be able to achieve a link easily. Let me explain. After every 50km the earth has a bulge after which the human eye can’t see anything anymore. This effect is similar on wireless PtP links as well. This bulge effect of the earth is called Earth Curvature.

E.g. Imagine having a tennis ball in front on you. With a marker mark 1 dot on 1 side of the ball and mark the 2^nd dot on the exact opposite side. Now, it’s quite evident that both dots can’t see each other but if you were to stick a long toothpick into either dots then 1 toothpick could see each other to some extent (based on the height of the toothpick). Right?

So, in simple words, longer distances in wireless can be reached when you have tall towers on each side that can see each other. This establishes a LOS and that can solve the distance problem. Now the next that immediately pops up is – How much tower height do I need? Well there is a general rule of thumb to this. For every 1km (horizontal distance) you need a 1m (vertical distance) tower height. But there are still other factors to consider.

So less than 50 km, a wireless link (PTP) can be established?
Not necessarily. Again its dependent on LOS and some other factors on the Radio. If there is LOS then the correct answer is – “It is possible”.

OK, I have a 10 km link, can a wireless link (PtP) can be established?
The answer again is not necessarily. Again its dependent on LOS and some other factors on the Radio. If there is LOS then the answer still remains – “It is possible”.

So what if I don’t have LOS? I can’t establish a link?
Unfortunately this answer is not straightforward either. If you don’t have LOS there are limits to distance. There are different levels of LOS, which I will explain in the LOS section.

Besides LOS, there is a mathematical calculation to know the correct answer but even these are rough estimates. Basically, the calculation is just physics. The factors are  –

The power of the radio (Tx power)
Length of the antenna cable (a cable that connects the radio and the antenna together. Longer the length, more signal you will lose)
And the antenna gain (output of the antenna).

This is usually calculated by a manufacture using their software. Thumb rule is a low gain, wide angle antenna should do the trick to achieve it a link if there is no line of sight. Also the radio should be based on OFDM technology. I’ll be covering this in more detail in the LOS section.

Then, what factors affect my distance? Are these sub factors to Distance?

This is the right question to be asking. Yes, they are sub factors.  Apart from the 5 main factors a PTP link, 3 sub factors are required on a technical level of a wireless PtP (radio) solution.

Radio Tx Power – This is the actual power coming out of the radio and its measured in DBm. The higher the Tx Power, more potential distance the radio can go upto.

Length of the antenna cable – Usually this will be a N type connection between the radio and the antenna. The smaller the length the better.

Antenna Gain – Higher the gain means longer the distance. This is usually measured in DBi.

How much distance can I get from 25 DBm TX power radio?
Like I’ve mentioned earlier, TX power is not the only factor that determines the distance. Even after having all 3 factors (mentioned above) on radio well balanced, the result will still be an estimate of how much distance can be achieved. Radio planning or link budget is required for a proper calculation.

Give me the highest gain antenna so that I can get the best quality link.
Although this sounds logical, going with a high gain antenna is good but NOT for every situation. High gain works best for long distances. In short distances, it will cause interference or a lot of noise. A simple way to explain this –

E.g. Imagine you have a person standing at a 10m distance from you. If he speaks softly (low gain antenna) you may not hear him very well. He needs speak a little loud (high gain antenna) for you to hear him clearly. Now, if he is standing less than 1m away from you and speaks loudly (high gain), like as if he is 10m away, what do you think would happen to your ear? That would be the same way high gain antennas work in short and long distance situations.

Also, more noise will be added if the TX power of the radio is low and you use a high gain antenna. There has to be a balance.

Then, give me the highest TX power so that I can get the best quality link.
Having more TX power on the radio will ensure that the signal is cleaner but there are regulations to maintain. So manufacturers can’t increase the TX power of Radio willy-nilly. 25DBm to 26DBm is considered the max. currently across the globe for PtP links. Yet, certain countries have their Telecom Regulatory Authority (TRA) set TX powers lower than this despite international standards. Having high TX power increase radiation which is harmful for human when exposed directly in front for long periods of time in the long run.

What is Link Budget? Can this help me calculate distance?

This is finally the right question.

“A link budget is the accounting of all of the gains and losses from the radio transmitter, through the medium (free space, cable, waveguide, fiber, etc.) to the receiver in a telecommunication system. “

This is a textbook definition of link budget. Clearly, it will help you calculate the distance for a wireless system. Each manufacturer has their own link budget calculation or you could use several free link budget calculators online. The only difference will be you will need to manually enter a lot of technical details for the wireless radio being used. The manufacturer, on the other hand, would have a lot of this done automatically when choosing the radio model.

Now, on the next factor: Line of Sight (LOS)

Related articles

• Super high frequency (en.wikipedia.org)
• Antennas: Everything You Wanted To Know (avalanwireles.wordpress.com)

 

Share this:

THE 5 FACTORS OF PTP

“I have Location A and Location B that I want to connect together and share data. Can you do this?”

Point to Point Wireless Example

That’s a typical example of the type of question I get from a customer every now and then. Most of the time the reason behind this is that customers don’t know what factors affect establishing a connection between 2 locations using a wireless solution. This article aims to address this repetitive question and provide knowledge that may otherwise be answered in a very diplomatic way by certain vendors and manufacturers in the industry.

Typically there are 5 factors involved in establishing broadband wireless link (PtP link). I will first list the factors and then address them with the type of question I get.

1. Distance – Majority of the questions that I get revolve around this factor.
2. LOS (Line of Sight)– this is the second most common question type I get. Sadly this should be the first question.
3. Throughput requirement – The 3^rd most popular request from customer’s who claim they require high bandwidth.
4. Link availability – This question I’m never asked at all. The main reason being customers don’t even know that it exists.
5. Legal Approvals (frequency and TRA) – this is usually the last question I’m asked.
6. Environmental Factors – Like temperature, humidity etc.

Now for my most popular way of addressing all of this: Questions!

Sadly the first question that is asked is actually price. Unfortunately, unlike Carrefour that has a box moving, high volume mentality wireless is more a solution oriented design approach. A lot of time is spent in designing the solution with all variables in mind to make the solution rigid and reliable. All good ISP’s have this approach and that’s why we choose them to be our service providers for our cell phones. As a user of the wireless, one must have the same approach.

I’ve decided that its a better idea and it would more in-depth if i were to discuss the questions related to each factor separately. Look forward to it in my next blog posts.

Related articles

• PTP Ex-Dividend Reminder – 8/29/13 (forbes.com)
• Can 4G wireless take on traditional broadband? (reviews.cnet.com)

 

Share this:

WHY ARE EQUIPMENT CLASSES IMPORTANT?

As consumers, we are all comparing 1 product to another. Like when buying a Flat Screen TV, all options would almost be on the same level. The only choice to make really is in the technology, whether LED, LCD or plasma.

Unfortunately, most of us in our profession (in purchasing or related) have the same ideology. Potential customers in some countries almost always start a conversation with price. When told an approximate, they come back with – “this is expensive”. Let’s consider this to be an example of a typical customer regardless of purchasing power, market’s price sensitivity and economic conditions.  The reason this is said, more often than not, is – people don’t understand the different classes of a similar line products and believe that the comparison is apple to apple. When in reality, the comparison is of different qualities/grades of the same fruit.

This article is written to educate (with a high level understanding) most buyers, customers and others that there are classes of products and they exist for a particular reason. I will attempt to explain this with the help of an example of cars. When buying cars, there are different classes of cars and that would decide the starting price. So let’s take an example of 4 different categories (classes) of cars:  Toyota Yaris, Nissan Altima, Porsche Carrera and Hummer H2.  (Note: this is only for illustration purposes)

Consumer-Class

…Is like a Toyota Yaris and such products focus on basic functionality. This type of equipment can be picked up from a small shop in the peripheral market anywhere. Those shops are like a “supermarket for IT equipment”. This type of equipment is sold mostly for home use only and doesn’t really require any expertise. They are as simple, basic, cheap and easy to install/maintain as a microwave. Here are some examples –

• Linksys
• NetGear
• U.S Robotics

Most SOHO environments (or even SMB) would also consider using this.

Enterprise-Class
Is like a Nissan Altima. This type of equipment is sold by resellers with a focus on functionality for the corporate/enterprise businesses. It requires reseller specialization from manufacturers, is more complex and more expensive than consumer class (Yaris) and comes with a list of features and benefits for businesses and their networks. Some businesses are interested in simple data while others are looking for triple play services of voice, video and data over the wireless. Hence, the manufacturers focus more on the stability of the product and IT elements (like no. of applications supported). This makes it more a networking product on an indoor level. Its downfall is OPEX. For its upkeep, enterprises have to pay for licenses, whether it is no. of access points required or support in general. Here are some examples –

• Cisco
• Aruba
• Nortel
• Rukus

They are well-known for being on an indoor level and have some AP’s to extend some limited level of coverage on an outdoor level (like a parking lot). Some applications of this class would be; providing conference rooms with wifi or a hotel looking to provide access to guests.

Carrier-Class
Here’s the Porsche Carrera. This type of equipment runs in a Carrier’s network on an outdoor level. They are meant to support large no. of consumers and enterprises across a larger geographical area. They focus more on performance, support all international standards, are extremely reliable and provide high uptime. Such equipments are made to handle harsh operating conditions and have strong electrical supply regulations. They can be really complex to design and install. All manufacturers of such type of equipment have a centralized software platform called NMS or Network Management Software sitting at the carriers NOC (Network Operations Center) for management and monitoring purposes round the clock. For such type of products, the vendor focuses on a tiered level for support and since carriers have business critical services, they like knowing how to support their own equipment. For support, manufacturers are considered Tier3, an SI/partner is considered Tier2 and the carrier itself is considered Tier1. Tier1 also keeps some stock of parts and units with them.

There are many different kinds of carrier class products– like point to point, point to multipoint, Outdoor Wi-Fi and Microwave. It is important to note that in this space there are different levels as well; Entry level, Mid-range and “Full Options” and pricing differs on those levels. Here are some examples:

Entry Level: Ubiquity, Microtek, Proxim, Ligowave
Mid-range: Redline, Netronics, Dragonwave, Alvarion
Top level: Motorola (now Cambium), Alcatel-Lucent, Huawei, Nokia-Siemens

A kind of drawback is that technology releases are slow in this class.

Industrial-class
is like a Hummer H2. This type of equipment is a specialized for industrial environments. A very unique class made to fulfill unique demands of applications like mining or oil & gas. These applications can be dangerous and the use of wireless networks helps makes things safer for personnel. The equipment is designed to handle extreme operating conditions and temperatures, electrical discharge and vibrations and deliver highly available networks. Due to its rugged form factor and other features it is very expensive and doesn’t require any support personnel since the equipment understands that its purpose is mission-critical. To design such a network an engineer who understands such an environment is not mandatory but a preferred requirement. A good example would be Moxa, or a SCADA manufacturer like GE.

Understanding this well means recognizing in crystal clear why 1 class/type of equipment can’t be used for a purpose outside of its class.

Share this:

Licensed or Unlicensed Spectrum

The Telecommunication Regulatory Authority of each country has the right to decide what is considered licensed and unlicensed within that particular country. Without this body radio’s would be propagated by everyone everywhere and would eventually result in what has happened to China; overpopulation. For this reason TRA breaks the entire spectrum into little chunks called Bands and provides it to the public. Not only does the TRA regulate the frequency but it also regulates the Transmit Power, EIRP and certifications of the product if needed to allow a particular product to be used within the country. Some countries are strict about this while others are more relaxed.

This article explains what the considerations under each band are.

Unlicensed Band
2.4Ghz and 5.8Ghz are considered worldwide standard for unlicensed bands. There are some other countries where 5.0Ghz – 5.9Ghz is considered unlicensed. This means it’s Sub 6Ghz frequency. This is also called Free-band or license exempt and its meaning is self explanatory. But in some countries the TRA doesn’t assign a free band and you still have to pay for the frequency usage whether you are an ISP or an enterprise. Parts of GCC, Afghanistan, Nigeria, Egypt are some examples of countries that don’t have free band. So it can be confusing and different between even neighboring countries.

Licensed Band
Some agencies have a reserved space in this band like Police, Military and public safety within the country. Besides these industries, corporate (in license band) must apply for a license and pay for a license to use the spectrum on an annual basis. This is separate from the cost of the equipment itself. The license grants the user of the license exclusivity over the use of that spectrum and hence why this is a paid service. This can be in 6Ghz and above called Mircrowave frequencies or below 4.9Ghz and Pre-Wimax or Wimax frequencies.

Which one do I choose?

The advantage of choosing the unlicensed spectrum is it’s a quick and easy way to get started with almost no OPEX. Decently high throughputs can be achieved with relatively low cost equipments and quick installations. For startup ISP’s and general corporations this is a great solution. Its greatest disadvantage is that there is no barrier to entry meaning the probability of interference is high (since everyone can start this operation) especially in urban areas.

The major reason for choosing licensed spectrum is the lower probability of interference, especially in urban areas.  Now the flip side is cost and time. So if you are a corporate with mission critical services, Oil & gas Company or anyone else in an urban environment looking for alternative communication methods and don’t want to worry about interference for now or future (and cost in comparison to any other alternative is not a deterrent), this is a good choice.

Related articles

• Good Luck Auctioning Spectrum (businessweek.com)
• Coverage Maps for New Wireless Spectrum Available to Fixed Services (aviatnetworks.com)
• Carriers ambivalent about sharing airwaves with the feds (gigaom.com)
• Difference Engine: Wireless’s colliding worlds (economist.com)

Share this:

I get a lot of customers who are looking for point to point and point to multipoint systems and don’t know how to decide what is best to be chosen. There are some business and technical factors to consider. The idea behind this article is to familiarize everyone with what would be the right choice for them and I will attempt to explain this with the help of an example which I present to most of my customers.

Think of a public park. There is a lot of greenery so there has to be a watering/sprinkler system.  If you’ve ever seen them running you would have noticed that there are 2 different types of hose’s that are used to water the grass.

POINT TO POINT CONCEPT

In 1^st type of hose, the nozzle is only 1 which is small and thin. This type only shoots out 1 concentrated beam. This type of nozzle is used to reach long distances and can carry a lot of water while doing so. So, the hose is the radio, the nozzle is the antenna and the water is the data. This is how wireless point to point technology works; concentrated pressure in a single direction for longer distances and higher throughput.

ADVANTAGE & DISADVANTAGE
Wireless Point to Point is best used for higher throughput and longer distances. Interference will always be there however. Smaller the beam of the antenna the longer will be the distance and lesser the interference (noise) from external sources. The disadvantage is that point to point wireless can be more expensive than a point to multipoint system (if you require many point to points). You also need to ensure a Line-of-sight from source to destination. Point to Point isn’t very good at NLOS but it is capable.

POINT TO MULTIPOINT CONCEPT

In the 2^nd type of hose, the nozzles are many and covers a 360 degree area. This type shoots the same amount of water in total but it is spread in all directions. Hence, it doesn’t go very far and only covers close proximity. This is how point to multipoint system works; single point for a wider area to be covered in shorter distances.

ADVANTAGE & DISADVANTAGE
Wireless point to multipoint system are best for many low bandwidth requiring points in short distances. The management is central so it’s easier and is less expensive overall. Unlike point to point, it is NLOS capability is much better because antennas are wider. The disadvantage is that the frequency being used is spread all over. If tomorrow another ISP (Internet Service Provider) tries to provide service to the same area on the same frequency (this is common in unlicensed frequencies) there will be interference for both parties. A second problem is that it is a single point to failure. Even after having redundancy on the main base station, if something goes wrong, your entire network is affected simultaneously. A third problem is that most point to multipoint systems are contention based. Meaning the subscriber unit (customer-side unit) will fight with other subscriber units and the base station for throughput increasing latency and reducing performance.

INTERFERENCE FACTORS IN POINT TO POINT
If there is another hose nearby and is crossing paths then it may interfere with and not reach the intended distance with the planned amount of water. Keeping nozzles thin will reduce the amount of interference caused by any other hose.

INTERFERENCE FACTORS IN POINT TO MULTIPOINT
This is a little more complicated to explain with the example but I have explained that another ISP close by providing service in the same area and on the same frequency can cause interference. The best way to avoid this is to ensure that from the business side, you either have the frequency under your control or the area in your control exclusively otherwise it becomes a solution only for the short run in.

CONCLUSION
There is no right and wrong here. It just depends on what your business is looking for and how to provide service to a particular area. The truth is that in reality, there is no “One size fits all” like approach in wireless. The best is always to be like a Mochachino, a blend of both. The target market will largely define how to choose the right solution and it’s almost always a mix of both.

Related articles

• Broadband is Broad! (lesleyanthony.wordpress.com)

Share this:

Heard of Wireless Broadband? Nice. But it’s still surprising that many people have still only heard of it and don’t know much about it. This section is intended to inform a general audience, Reseller, System Integrator and startup ISPs with more specifics.

A typical WISP installed on a residence (Photo credit: Wikipedia)

Wireless broadband is almost 20 years old and it’s not a new term to mainstream Telecos but it really started gaining momentum in the early 2000s. Usually manufacturers say that the best area for wireless business is in countries that don’t have a telecom infrastructure. This would mean developing countries are best where the cost of infrastructure through cable is so high, it’s unreasonable. There are other countries that have infrastructure in place but again to the end users, it may still quite a high cost. Imagine how many companies have multiple offices or branches. And now imagine this –

1 head office and 1 branch office located a few Kms away from each other. The branch office has no infrastructure except an Ethernet switch. This means no phone lines, no internet connections, no PBX. Yet all the users can access the internet, call people in the head office just by dialing their extensions. HOW?

10 years ago, anything that had the term “wireless” was considered unreliable or still developing. But today, a converged network with no recurring cost is possible only with wireless broadband.

What is wireless point to point ?
Its original name; BWA (Broadband Wireless Access) is used to provide high speed wireless internet/ intranet access from 1 branch (point) to another branch (point) and these speeds are much higher than what you local ISP would provide you. Not to mention cheaper too.

Is wireless stable?
Yes. Much the same way that you’re TV or radio signals is. Wireless has been around for more than 20yrs now and has matured as a technology tremendously. The right product and a proper configuration can lead to stability.

How is it beneficial to me? (What kind of applications is it used for?)
If you are an Enterprise, you can use this as your leased line (or MPLS etc.) replacements, backup line or DR line.
If you are an ISP, you can provide high bandwidth at a very low cost, since there is no civil work of laying cables round the city. It can be even used as a backhaul to bring data back to your network. Again, popular in many countries where there is either no infrastructure or where infrastructure prices are high.
If you are in Safety and Surveillance, it can be used to connect remote cameras for viewing or recording.
If you are in Seaports or Oilfields, it can be used to connect an offshore rig or use it to extend IP networks to the wellhead.
And lots more that you can just mix and match. We have an example of a bank using it to connect their ATM machines together.

What is this LOS? Does Point to Point require LOS?
LOS – is Line of Sight; meaning a visual line of sight between the 2 locations. Yes, it does.

Can you do Non line of sight?
Yes, we can do NLos but the limitation depends on what the obstruction in the middle of the 2 points is and how much is the obstruction. Usually 6-7kms NLOS for Netronics is considered possible. But again, this defers from place to place as it is environment dependent.

I’ve seen nLos and NLos…what is the difference?
nLos: Near Line of Sight and NLos: Non line of sight. 1 is very small obstruction for visual line of sight and the other could be a complete obstruction.

Do you have any point to point wimax solution?
I’ve written this before in my WIMAX FAQ post. Some countries like Egypt, Libya, Sudan and a few others ask me for point to point Wimax. Let me dispel this by saying there is no such thing. These countries, I think, are told to believe that anything that is wireless is called Wimax. This is not true either. On 2.4 GHz and 5.8 GHz are unlicensed Wi-Fi frequencies that are a worldwide standard. Having a Baskin Robbins ice-cream truck right outside your house but going to the supermarket for Basin Robbins just because it’s more expensive just doesn’t make sense. If there was something like unlicensed Wimax all ISP’s would have adopted it already.

What frequency does this run on?
2.4 GHz and 5.XGhz. This is the worldwide standard for open licenses. Netronics even has 4.9 GHz to 6.0 GHz frequency range within their point to point links.

Do I require a license?
If you are an Enterprise then no. But there are countries like Nigeria, where 5.4Ghz is considered licensed. Some countries like UAE, Iraq, Afghanistan, even if you are an enterprise, consider all wireless equipment to have a spectrum approval/usage charge and that is a nominal fee for a year and it needs to be renewed. This needs to be checked with the Regulatory Authority in your country. It is a little ridiculous because the TRA won’t really do anything, so it’s like paying “Air Tax” but in most cases even including this, the wireless point to point solution is still cheaper than other methods.
If you are an ISP, the answer is maybe. But the type of license is different for providing service using Wireless. You will have to check with your Regulatory authority for more info.

What about interference in on the open license frequency?
The same product can act differently in different environments. Inference can be many things like buildings, trees, power stations or lines or even transmission on the same channels by another ISP. If you are ISP, going on an open license frequency interference is a key concern for service. Luckily there are plenty of workarounds built within the system to handle those technically. The bottom line is, you won’t know how much interference you have to deal with until you test it in your own environment. I highly recommend this otherwise rest of the talk is all in the air and won’t be seen, just like the signal.
For an Enterprise, there are many ways to handle interference. Rest assured that they are reliable and technical in nature and beyond the scope of this discussion.

What kind of distances can I cover?
1 of the most common questions I get. Distance depends on the terrain. It’s kind of like the “LAW of the Land”. The law of 1 land is not the same as that of the other but it has the same general idea. What have been achieved are distances of up to 120km in an onshore-offshore connectivity in an oil and gas application. Different product can extend different ranges. Just remember that this measurement is based on an ideal open environment. You should still test it in your environment. You can get different distances based on the different interference levels.

What Throughput can I get?
2^nd most common question I get. Honestly, the previous question and the interference question along with this one are all interconnected. Is the question – With 120km, what throughputs can I get? My answer is – That’s an ideal environment, forget about it. You first need to know the interference in your environment then you can calculate distance and then throughput. That’s the way forward and its best understood with actual testing.

But I get 100mbps in LAN and some vendors provide 108 Mbps in point to point wireless but in reality I’m not getting that speed. Why?
Good Question. This is because they are referring to air speed. When dealing with telecom related stuff, bandwidths are not the same and principals are completely different. Physics and Chemistry are both a part of science but they’re still2 different subjects. So, airspeed is different from actual throughput. Hence, I advise asking your vendor about throughput not airspeed.

What about security? I think its not secure!
Everyone thinks that inbuilt wireless device security is the only way to secure it. The truth is there are so many ways to secure it, I could write a chapter on it separately. Encryption, link lock, mac filtering are only some ways. Now think about the 2^nd way. Think about backend, Routing, VLANing, AAA, Radius. Usually in and Urban environment people use 5.8ghz frequency. The frequency itself is the 3^rd level of security since there are no usb dongles in 5.8Ghz frequency that people can buy. And even if there is such a dongle (and there isn’t) your hacker would have to be Superman flying in mid-air between both your office locations with his laptop for a really long time! So the 4^th level of security is the antennas.

So is there a planning tool that can help?
Google Earth is my friend and can be yours too. Google Earth usually tells quite a lot about LOS and obstructions between 2 points too. Some of my customers use it as a alignment tool for deployment purposes. There are link budget calculators available as well that can help but they calculate only terrain meaning it will only calculate the earth curvature and not the building on the earth, so that’s not a complete solution. There is a paid solution and it’s called “Path Loss”. The alternative to that is to physically test the link in the field.

Hope this post helps you all. Your feedback would be most appreciated.

Share this:

Reducing Interference

I was initially going to name this post as “white noise” because i thought it would sound cool but it would be too misleading. I understand that there are far more technical people reading my posts than i expected. There are quite a lot of customers who buy outdoor Wi-Fi based products and then think that deployment is straightforward. Actually, a good deployment is a mix of a pre-deployment test, positioning, product and a few other factors. One can’t defy physics, so the first understanding required is if you’re running in 2.4Ghz (Wi-Fi) frequency, there are only 13 channels only. Almost all the 13 are overlapping except channels 1, 6 and 13. These tips have been provided for a Beam Forming base station known as the Netpoint Pro for outdoor purposes. Although in theory, it should work for other base stations as well.

So from here on, cater to these tips for improvement –

1. If you are running the Omni-base station version in an area try using a sector. This might reduce the coverage but it will clear out the sources of interference and provide much better performance.
2. There is a trade-off to consider – the higher the base station, is larger the coverage but this also means higher interference and signal degradation. Conversely, lower the base station lower the interference but also the coverage area. The trick is to find the right height and depends from place to place. Usually 15-25mts above the highest object is as ideal as it can get.
3. Don’t compromise on performance. Set your base station to work in G mode only. Anyway, anyone who is working on b mode is already outdated and its time they get something new.
4. Performance can also be increased from the CPE side. So people who have paid for better CPE’s will have much better management and performance. One useful tweak that can double capacity easily is by configuring all a/b/g standard RTS/CTS mechanism (for packets larger than 256 bytes). This will ensure collision avoidance.
5.  Run ACS on the base station. This may sound very trivial but I’ve seen Pro’s who skip/forget this step.
6.  Finally, use a real throughput measuring tool. Do not test a base station simply by signal strength as this can be quite misleading. Yet, I’ve noticed some professionals coming back to me with this.

These are some simple yet very effective techniques for improvement of performance of the base station. I welcome any feedback, suggestions or other improvements from professionals with such implementation experience.

Hope this post helped!

Related articles

• The New Wi-Fi on Steriods (lesleyanthony.wordpress.com)

Share this:

This section has been written with the intention of bringing my audience with an understanding of how far Wi-Fi has gone and why its called Super-Wifi. Traditional Wi-Fi, started somewhere in the 1999 with the release of 802.11a and b standards. What initially started off as 20mtrs of distance to enable a cable free environment became 100m by 2003 with the release of the G standard. Since then Wi-Fi has some revisions and MIMO technology added to it in 2007 and the N standard in 2009 to increase bandwidth.

However, it was made and designed for indoor use. Yet, I see several customers trying to take the indoor solution outdoor and fail. Now, there is a Super Wi-Fi Solution. Surprisingly very few people know that this technology has existed for almost 4 years.

All Questions may not apply to you, so choose the ones you find most relevant to you. The answers are to provide you with a broad understanding of the technology. You may reach me for more specifics.

What is a Super Wi-Fi base station?
That’s simple. A solution specifically for outdoor environments with a technology called beamforming that can you take up 4 times the distances of a traditional Wi-Fi system.

What is Beam forming?
I would like to keep this answer simple and non technical. Usually, in Wi-Fi what we all know is that the signal emitting from the antenna is circular. This limits the distance, causes interference when there are several AP’s (Access Points) and reduce your bandwidth. Beam forming is a technique used to direct the signal from the antenna in the form of a beam that can be manipulated in any direction.

What applications can I use this base station for?
Campuses, schools, rural areas, last mile, hotspots coverage in suburban cities, seaport applications, building coverage etc there are too many to name.

What frequency does this run on?
2.4Ghz and 5.8Ghz. This is the worldwide standard for open licenses.

What about interference in on the open license frequency?
The base stations acts differently in different environments. Inference can be many things like buildings, trees, power stations or lines or even transmission on the same channels by another ISP. If you are ISP, going on an open license frequency this can be a potential problem but there are workarounds built within the system to handle those technically. From a business angle, you need to find your market and by observation; this is great solution for last mile and few applications that have been mentioned above already. Even still, you won’t know how much interference you have to deal with until you test it in your own environment. I highly recommend this otherwise rest of the talk is all in the air and won’t be seen, just like the signal.

But there are other vendors that claim to do Wi-Fi outdoor for cheaper…Why go with this system?
There are 3 classes of Wi-Fi makers, SOHO, enterprise class and carrier class. The SOHO  and enterprise class sit in the indoor space and only come to outdoor environment to cover areas like a small parking lots but they do with it with the same indoor design chipset with an outdoor enclosure. Additionally, they are all controller based. So you will have to run a cable from that base station to a controller and then to your core network. Not the right design or solution if you want to go in larger outdoor spaces. Going with such system also means a single point of failure, i.e- your controller.

What kind of distances can I cover?
1 of the most common questions I get. Distance depends on the terrain. It’s kind of like the “LAW of the Land”. The law of 1 land is not the same as that of the other but it has the same general idea. What have been achieved are distances of upto 1.5km directly connected to a laptop without any signal boosters. The same can be extended upto 20kms with an outdoor CPE. This is significantly bigger than other Wi-Fi systems but remember that this measurement is an ideal open environment. You should still test it in your environment.

What Throughput can I get?
2^nd most common question I get. Honestly, the previous question and the interference question along with this one are all interconnected. With 1.5km, what throughput can I get is the question. My answer is – That’s an ideal environment, forget about it. You first need to know the interference in your environment then you can calculate distance and then throughput. That’s the way forward. The base station itself can provide 30Mbps of actual throughput.

So is there a planning tool that can help?
Truth is that in WIMAX there is but Wi-Fi a planning tool for outdoor purposes with the capabilities of understanding the beam forming technology is not available yet. However, there is a post-deployment tool that can tell you (after you buy a base station and place it in your desired location) the throughput you can get a different distances. I recommend investing in something like this.

But wouldn’t a system like this be expensive?
Let’s compare this solution to the traditional Wi-Fi which only gives you 100m. The amount of civil work to lay cables, bring power points and have a cable connecting back to the controller is higher than paying for a system like this. It is cheaper going for the Super Wi-Fi system. Besides this, you would get benefits of increased coverage range, better throughput and lesser interference. Now you decide.

What about the back end like billing systems?
Some vendors provide and some don’t. But I can recommend a few billing systems –

–          Access controller from www.4ipnet.com  model name HSG120

• Most simple solution
• For less than 2000 users,
• From a coffee shop to a WISP in small town.
• Simple to setup
• Easy to maintain, due to simple account creation and maintenance
• Limited scalability
• Limited billing capabilities
• Around $2000 price

–          Centralized access controller from www.Lok.com model names Lok Box series 300, 500 and 700

• Moderately simple solution
• Can manage thousands of users
• For medium to large WISP
• Controls the bandwidth and access from the core
• More costly than first option
• More scalable

–          Billing software from www.aradial.com and access controllers from www.mikrotik.com

• More sophisticated solution
• Manages few millions of users
• For Large IPS and tier 1 or tier 2 telecom operators
• Controls bandwidth from right behind the access point
• Utilizes a large number of access controllers (one behind each batch of APs or one per hotspot) connected to a centralized billing system
• More costly than the above two options
• Highly scalable

Can you recommend a solution to use?
Sure. Subscribe to me/blog and I shall provide with you updates regularly.

Related articles

• Reducing Wi-Fi Interference (lesleyanthony.wordpress.com)