Mobile TV

How the Service deployed

We first have to download the Mobile TV application and then register to a particular TV channel. For example imagine that you are going to register for the channel TenCricket. Content providers main server is in India alongside with a customer database. Our customer attached to the SGSN and the normal procedure going to the internet will take place as usual. When he clicks “subscribe for a channel”, his request will reach the server located at India as a HTTP request. To do that authentication part should be handled by our Fire Wall and according to the request server will take necessary actions. Then the server will send an HTTP/XML request to our External Charging Gateway (ECG) to check the details of our customer. Refer the image below for further details.

ECG will direct it to OCS to check for the balance of the customer. We can directly charge prepaid customers in OCS if they have got enough balance. For postpaid customers OCS will generate a CDR (Call Detail Record) and send it to OFCS for billing him later. After that OCS will send a response back to ECG indicating the status. It will forward again to Main server.

If the charging process of the customer succeeded he will be given separate RTP (Real Time Protocol) addresses for video and audio (i.e. RTP://video, RTP://audio) for that channel. 

            E.g.: RTP://203.xxx.x.xxx

You can’t see above addresses or you can’t identify them. Because when the addresses

assigned you will see a “Play” button in the Mobile TV interface. When you press the button above mentioned two addresses will be opened through mobile’s default player. 

Company prefer customers under 3G coverage but even with a 2G coverage (will depend on traffic, weather, etc.) you may able to get an uninterrupted service. This application requires at least 248kbps (3G) or 64kbps (2G). There are 3 profiles defined for the convenience of the customer.

     1.      MPEG4 (video),AAC (audio) – specially for older mobiles

          2.      H.264,AAC – high (i.e. android phones)

          3.      H.264,AAC – low (for customers in 2G coverage)

Giver URLs were all internal private IPs. Because we don’t need to use over internet bandwidth for streaming. That is handled using an Encoder as shown below. After successful subscription from the main server side Encoder will send a unicast RTP addresses (mentioned above) for the customer. Now customer can watch the TV channel. For streaming only bandwidths between mobile and Encoder will be used.

 

When we get the content from a content provider (i.e. Sky TV) we have to get it as shown in the image. Because the switching room is isolated from outside world to protect it from lightning and etc. It will only get power from outside. So we connect roof top cabin and switch using fiber. This much of security measurements taken in order to protect all the high cost equipment located inside the room.

How the charging happens

To charge a particular customer we need to get his mobile number through the main server in India. In order to pass that information; we send a HTTP enriched header request to the server which consist of following details.

•          MSISDN
•          Bearer (2G or 3G)
•          SGSN ID (We don’t allow this service to roaming customers. If a customer is roaming his SGSN ID will be different from ours. So we can restrict him by looking at this ID)

Why the Main Server is not in Sri Lanka (Can’t we set up it in Sri Lanka)

Major distribution points of the content provider are not located in Sri Lanka. If we are going set up those facilities here in Sri Lanka, the cost will be high. It is not hard (very easy) to configure and get connected with existing facilities. We only have to get the authentication done because we have to connect through public internet. Authentication will be done using IPs, ports, usernames, passwords, etc. only when we subscribe we have to communicate with the Main Server. Especially as I mentioned above while we stream, it is controlled inside our network. 

ANC TRE Connection

 

 

*Note: Similar color ports are connected

For this ANC we can connect up to 4 TREs.

Operator dependent Configurations:

·         Connect with coupling and it will also add a loss. With this configuration we can add 2 TREs per one ANC.

 

Longitude and Latitude

Ohhh I should have mention this before. Because this is very important…… without knowing this....

Latitude (parallels) - Horizontal line

This is the angular distance, in degrees, minutes, and seconds of a point north or south of the Equator.

 

 

 

 

Longitude (meridians) - Vertical line

This is the angular distance, in degrees, minutes, and seconds, of a point east or west of the Prime (Greenwich) Meridian.

Distance between Lines

By dividing the circumference of the earth (apprx. 25,000 miles) by 360 degrees, the distance on the earth's surface for each one degree of latitude or longitude is just over 69 miles (111 km). 

You know how to press it but do you know how it works?

Before we move on to the theory behind this, just take a look at these sound clips.

 GOTO clip1: http://www.genave.com/audio/dtmf-page-normal-S00P0501P99.wav                                                                     
GOTO clip2: http://www.genave.com/audio/dtmf-page-slow-CP4D34SS.wav

DTMF- Dual Tone Multi Frequency

Hz	1209	1336	1477
697	1	2	3
770	4	5	6
852	7	8	9
941	*	0	#

Did you ever think of how your response (pressing a particular number) is identified by the receiving side? For an example let’s consider a situation where you are in an Interactive Voice Recording (i.e. calling to customer care).  They asked you to press 1 or 2 or any number to identify your response. Pressing a key will send a sinusoidal tone for each of the two frequencies (i.e. 697 and 1209Hz for “1”). See the image below. The multiple tones are the reason for calling the system multi frequency and it is called as dual tone because of the use of two frequencies. Two frequencies are there to make sure that frequencies generated accidently will not be identified as a press. Columns will have higher frequencies than rows. In normal dialing also we can use DTMF, but only for special purposes (if needed MSC can identify the pressed keys).

Interactive Voice Response (IVR) - is a technology that automates interactions with telephone callers.

Figure 1: DTMF

Pulse dial system was used earlier and it is a slow and older method. For optimum use, push-button phones utilize DTMF signaling, which was also known as touch-tone dialing. 

                                                            Figure 2: Push Button Phone

Now you know the theory, and let’s see how it is implemented

Mark and Space

Mark refers to the duration a DTMF tone is produced (the time which a DTMF digit tone is actually producing sound), and Space refers to the duration of the silence between individual digits.
Notations

• Mark/Space : Pronounced as "Mark and Space”. Refers to the durations of mark and space.

•  40/40: The decoders expect the DTMF tones to exist for at least 40 milliseconds and in silence for 40 milliseconds between each DTMF digit.

References: Some of the images, details and sound clips were taken from the following website. http://www.genave.com

What happened before you made a Call

First of all I would like to ask you a question. How many different modes are there in a mobile phone? Most probably your answer will be two (on/off). But when we look at it technically there are three modes and they are switched off, idle and dedicated mode. When we take a call we can say that our phone is in dedicated mode. If our phone is switched on and not taking any calls we could say it is in idle mode. When we switch on the phone it will immediately goes in to idle mode. Depending on the signal strengths present MS (Mobile Station) will decide to which cell it should camp. There are two important things should happen before setup a call.

                1.       Cell Selection

Immediately after MS switched on Cell Reselection occurs. It refers to the initial registration that a MS will make with a network. C1 is the parameter used for selection. The most favorable cell is indicated by the so called C1 parameter for a MS. MS will search all RF channels and decide which cell to camp using C1 values. A basic equation is given below,

              

                                 C1  = Rx level – Rx level Access min

Rx level: Received signal power for the MS or in other words what signal strength the MS sees the tower at.

Rx level Access min: Minimum access level of the tower. This is a parameter of the antenna of the site.

Normally MS will receive signals from many cells. For those cells MS will calculate the C1 and find out the maximum value. Then it will camp into that cell. That’s not the end and it will keep gathering information of nearby cells. Normally it will keep tow lists named as Active set and Merge set. In active set we have a primary cell (the cell mobile has already camped at the moment) and another four cells which are having highest signal strengths next to primary cell. And there are several cells in the merge set. If anyone in the merge set gets better than the active set, it will replace the active set cell. This will keep happening. 

                2.       Cell Reselection

Cell reselection is performed as MS traverses through a network in idle mode. MS will keep details of strongest BCCH cells as I mentioned above. As in previous case we have another parameter known as C2 for Reselection. It can be also called as an improved version of C1.

              

                                             C2= C1 + CRO – TO(t)

     

[CRO (Cell Reselection Offset), TO (Temporary Offset)]

The mobile shall regularly search for a better cell according to the cell reselection criteria. If a better cell is found, that cell is selected. But before we select a one we should check for the neighbors. That means for each and every cell we define neighbors. Neighbors are defined using their locations, directions and distances to a particular cell. If it is a neighbor cell we can camp on to it. But if it isn’t MS will not camp into that.

If we want to reduce the number of customers per a cell we can adjust CRO parameter. Or we can say we will give higher priority to a cell by giving a high value to that cell than its neighboring cells. TO is used to avoid unnecessary re-selections. For example consider a person moves along a highway. As he moves along there may be many cells serving him at different points. But there may be one or two major cells which can serve him for a longer distance. In such occasions we prefer him to stay camp in to that major serving cell. To do that we can give a proper timing value for TO. That means MS will not camp into that cell until it stays in that cell for a particular time period. Normally in high ways we go fast and it will avoid unnecessary re-selections and keep camp on to a major cell.