google-site-verification=VAmea75XZEowVbRm09fPMJIIRF_llgYpguw4XveyuVM CW Testing Continuous Wave Testing | Pranab Hazarika

CW Testing- Continuous Wave Testing

CW testing is Continuous Wave Testing.

CW testing is the First Phase of Telecom/Mobile Network/Radio Frequency (RF) Planning.

Network Planning/ RF Planning is always start with Continuous Wave Testing Method for all GREENFIELD project.

The main aim of doing CW testing is to gather information to design a propagation model of transmitted signal for mobile communication.

It is carried out to gather actual propagation of mobile signals in different terrain and clutter.

Basic of doing this test is to measure or to know the actual distance, a transmitted signal would travel from a Mobile Base Station that we simply say MOBILE TOWER or SITE.

Propagation Model primarily depends on geography and terrain of a circle.

Secondly it also depends on clutter type e.g. Dense Urban (High rise congested buildings), Urban or Rural area.

This is the traditional method and best method for Model Tuning/Model designing.

This is applicable for all technologies that may be 2G, 3G, LTE or 5G. It is mainly Spectrum dependent test.

With the help of CW testing and Model tuning we can design a model for our network to generate coverage prediction and network planning.

Quality of any network depends on the accuracy of the Propagation Model designed based on CW testing and Model Tuning.

Basic concept of CW testing is very simple-

  1. It has one transmitter along with Omnidirectional antenna which will transmit frequency of corresponding licensed frequency (e.g. 1800Mhz, 900MHz band)
  2. A receiver (Drive test kit) will move around the transmitter and checked for signal strength. How far the Receiver can receive signal from the static transmitter.

Set up of CW Testing:

The CW testing equipment contains-

  1. Transmitter
  2. Omnidirectional Antenna.
  3. Power supply to provide power to the transmitter.
  4. Drive test Kit as a receiver.

Now there are two Scenarios for CW Testing.

Scenario-1:

Suppose in a telecom circle or LSA (usually a state e.g. Assam Circle, Maharashtra Circle) there is no Operator is having a new technology (e.g. 4G-LTE or 5G)

When we are planning to start this new technology (e.g. 4G-LTE or 5G) or new frequency band (e.g. 2300MHz, 2100MHz band) in that telecom circle we first have to

Start with Continuous Wave Testing (CW) method.

Then we will use the above mentioned equipment for doing the CW testing.

In transmitter we can set our required frequency to radiate (e.g. 2300MHz band).

Scenario-2:

Suppose in a Telecom Circle Operator-A is already giving service of 4G in 2300MHz band.

Operator-B is a newcomer and planning to start 4G network in same frequency band 2300MHz in that same circle.

In this case we do not require to set up the transmitter for radiating the licensed frequency band.

Instead we can use a SIM of Operator-A and do the drive test to check actual coverage level.

In that case we do not require the CW testing set up as the existing operator’s SITE (BTS/nodeB/eNodeB) will do the actual transmission of signals.

Drive test will be carried out by using existing operator’s SIM card.

But if Operator-A is providing services in 2300MHz band and Operator-B is going to operate 4G in 2500MHz band then this method will not work.

We can use SIM card method only for same frequency band.

For different frequency band we must go for normal Continuous Wave testing method with Transmitter+Omnidirectional antenna.

Data Collection by Drive test Method:

The transmitter should place in such a location that it is at the top of a high rise building or some high rise locations from where Omni-directional antenna can radiate to a maximum distance.

Drive test need to carry out in different clutter e.g. Dense Urban, Urban and rural area.

Propagation of signal will be different in different clutter based on penetration losses.

Collection of Samples and Plotting:

In CW testing we would collect samples with the help of drive test kit.

Collected samples can be easily imported in MapInfo Tool. For Final Model Tuning we will use Atoll Planning Tool.

For initial check of collected samples and to analyse the signal quality it is better to use MapInfo as it is fast and simple to operate.

We will discuss about MapInfo tool separately. You can visit my youtube videos to know basic of MapInfo functionality.

https://www.youtube.com/user/hazarikap  (Pranab Hazarika)

In the above picture it is showing that Transmitter is placed in a fixed position and transmitted a frequency band say 2100MHz (Suppose 5MHz frequency range in 2100MHz band)

The Drive test car with kit were roaming in the town on the motor-able roads.

We have bifurcated all the collected samples in two different parts.

  1. Good Quality received signal : received level from 0 to -93 dBm (Color GREEN)
  2. Poor Quality received signal : received level from -93 to -120 dBms (Color RED)

Now we can see that till 500m distance we are getting GOOD signal level (Color GREEN) in that particular area of a Town.

Beyond 500m distance in some areas we got POOR signal (Color RED).

If we consider this part of the town as URBAN, then we can come to a conclusion that for URBAN area we will consider coverage of a site till 500m distance in 2100MHz band.

If we do 4, 5 sample test in 4, 5 different URBAN areas of different towns then accuracy of the conclusion of coverage prediction will be improved.

But still with 1 sample test also we can conclude to a good value.

Clutter wise sample collection:

So we have completed CW testing and sample collection in 1 clutter, i.e. URBAN

Now we will identify a location of the same town or in different town with high rise congested building area as DENSE URBAN.

In DENSE URBAN we may get a GOOD signal level travelling to a distance of 300m instead of 500m.

In RURAL area this distance may increase upto 1.5Km, 2Km or may be upto 4Km depending on the band of Frequency.

More Frequency band less travelling distance because of more losses in high frequency.

E.g. 900MHz band signal may travel upto 4Km in rural area, whereas 1800MHz band can travel upto 2 to 2.5 Km.

So now we are having collected data/ samples for all 3 category of clutter i.e. DENSE URBAN, URBAN and RURAL.

We also come to a conclusion that with suppose 2100 MHz band we are getting GOOD signal strength

For DENSE URBAN-300m

URBAN-500m

RURAL-1.5 Km

Drawback of CW Testing:

Drawbacks of CW testing are

  1. CW testing model tuning is based on outdoor coverage only. We don’t collect samples inside a building to measure indoor coverage.
  2. Time taking method.
  3. Lot of iterations.
  4. Analogous Transmitter.

Rest it is the best method of sample collections for Model tuning.

In next blog we will discuss about MODEL TUNING which is the 2nd Phase of Network/ Radio Frequency (RF) Planning.

What do you think about the points discussed? And do you think there are some points I missed here? Leave a comment below.

Thank you for reading so far. You are important for me, please leave your footprint here:

<script async src="https://pagead2.googlesyndication.com/pagead/js/adsbygoogle.js?client=ca-pub-2754967439384345"
     crossorigin="anonymous"></script>
4.7 11 votes
Article Rating

Leave a Reply

5 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Max Nail
Max Nail
5 years ago

HI Pranab, I tried to tune a model using the benchmark data. We selected two points/locations/setups to do so. We got the non-make sense results/prediction after the model has been calibrated. 1st point gave extremely optimistic prediction. As per the calibration report, K1 and K2 values are so much reduced specially in K2. K1 is reduced from 2 to 1.5 and K2 is reduced 32.45 from to 0.1. The material losses are changed as well. The following is the mean error and standard deviation before and after: Mean Error: 35.73 è -18.999 STD: 40.32 è 28.268 RMS.Error: 53.868 è 34.055… Read more »

Pramod Doke
4 years ago

hi Pranab. even I’m facing the same issue.. calibrating model is too hectic task. can u please confirm that 1)what values we have to check while calibration.. are only ME= 0.25/0,5 & SD= <7/8 or some more too.. & 2)can k values of some operator A of band 3 (1800MHz) be used by some other operator B having band 3 (different channel)

syed ahsan
syed ahsan
4 years ago

HI Pranab, I am going to design a Fixed wireless greenfield network. I needed to which existing ATOLL propagation models can I use for my CBRS B48 design/planning. I will be having Fixed CPE’s as received end nodes(subscribers placed at 15 feet)

Steve
Steve
3 years ago

Hi Pranab,
What would be the power setting for the CW transmitter? In other words, if I want to model tune a 100 MHz 5G at 3.5 GHz, what would be the setting to tune the model for that band?

Regards
Steve

5
0
Would love your thoughts, please comment.x
()
x