Call Detail Record Generator
Our call detail records (CDR) generator is developed with GEDIS Studio and is available on-line. To use it, you need to register for a account at http://www.data-generator.com and import the Telecom project available in the shared workspace.
The purpose of this generator is to produce both Customer Data Records and Call Data Records by simulating the behaviour of the users of a carrier's mobile GSM network.
If you want a quick outlook on how to produce test call data records with GEDIS Studio you can watch our video Call Data Record Generator
Beyond the quality of generated CDR, our solution is unique because it provides you with all the data you will need for testing a Telecom Information System component.
Consider testing the Rating system, it requires both CDR and customer data to compute the charge for each CDR. Obviously, the part identifying a customer in our generated CDRs must match the customer data defined in the Customer Database the rating refers to.
So, the process of producing test CDR must consider the way your tested system may refers to existing customers. In this project, you will have the choice of integrating existing customer data or generate them to inject them in the test environment.
- Section 2 The generated data (structures and properties)
- Customer data records for telecom applications,
- Call detail records structure
- Customer's profiles based data event generation
- Section 3 is about the production process
- Section 3.1 Producing the customer data
- Section 3.2 Producing the profiled customer data
- Section 3.3 Producing the Call Detail Records
Test Data Production Process
Producing realistic and controlled CDR test data for a set of customers is a multi step process. At each step a new file is generated to be used as input for the next step.
- Generate or obtain customer data
- Associate each customer to a usage profile
- Generate the CDR sequence for each customer according to its profile
The two firsts generated data file are customer data, they can be either extracted from your production / test environment or generated and injected into your test environment. The point is that some of those customer data, we identify as "Telecom Data" will be included in the generated CDR and will ensure the consistency of generated CDR with the other data managed by the tested application.
Without this management of customer data, your test CDR may be just rejected because referring to unknown customers.
Step 1 : Creating Customer data
Customer data are made of business data and telecom data. Business information data are only needed if your customer data are generated (and not extracted from your test environment) and need specific fields for being injected in the customer repository of your test environment. Besides, the business data may also include one or several fields that will be use to select the appropriate profile
In our demonstration we designed a customer data generator that uses a generic Personal Data generator with fields such as: Gender, Salutation, First Name, Last Name, Age, Date of Birth, Marital Status, Address, etc. Then, we add "Telecom" data such as IMSI, IMEI, MS-ISDN, Subscription Plan.
The values generated for the fields MS-ISDN, IMSI and IMEI take into account the international numbering plans. For example, the prefix of the IMSI gives the country code and the network carrier code. Then the MS-ISDN is also formatted according to the country of the customer's carrier.
Our sample file is based on the 3 carriers operating in France (eg. Orange, SFR and Bouygues Telecom). You can have a look at the 
Customer Data Sample generated by GEDIS Studio.
Step 2 - Maping Customers to Profiles
Then comes the real magic. Because our data generation is profile based, we have to define the profiles and associate each customer to one of those profiles. For the sake of simplicity in this demonstration we have limited ourselves to those usage profiles (value applicable for a 1 day period):
- Profile 1 is for 10 Voice calls only, for a total of 30 min. and towards national correspondents
- Profile 2 is for 5 to 15 Voice calls for a total 45 min. and for 0 to 5 SMS, towards national correspondents
- Profile 3 is for 4 to 17 Voice calls for a total of 1h05 and for 3 to 7 SMS, towards national and a list of foreign countries,
The association between a customer and its profile is achieved by mapping the customer's PLAN field value to a profile. Remember that the PLAN field is to be found in the business data part of the customer data record.
Again to simplify the demonstration we have limited ourselves to 4 different subscription plans in order to illustrate that there's no necessarily a perfect 1 to 1 match between a customer plan and its profile.
The maping table implemented in our demonstration is as follow:
- All customers with subscription plan PLAN1 will have profile #1
- All customers with subscription plan PLAN2 will have profile #2
- 30% of customers with subscription plan PLAN3 will have profile #3
- 70% of customers with subscription plan PLAN3 will have profile #2
- All customer with subscription plan PLAN4 will have profile #3
Want to read more about profiles ? expand
Step 3 - Creating Calls for Each Customer
The third generated file is therefor the CDR file containing the part of the customer data usually available from the network (MS-ISDN,IMSI,IMEI) and some customer data the mediation may get from the CRM (such as the customer plan for example).
The duration of each call is computed from the total duration of the call of the usage pattern. Each duration is randomly chosen around the mean so that the sum of those durations exactly match the duration provided in the usage pattern!
The time of call for each CDR is also computed in order to make sure that two calls for the same customer never overlap!
You can have a look at the CDR Data Sample
CDR Generator Demonstration
To access the simple CDR generator you can follow the link bellow, choose your own generator parameters and GEDIS Studio will generate your test data.
