CC BY
15
Ч7
PUBLICATIONS IN ENGLISH
COMMUNICATIONS
ANALYSIS OF ISSUES WITH CONNECTING DEBT COLLECTION AGENCIES TO COMMUNICATIONS PROVIDER NETWORK
Ivan V. Uglov,
expert of Public Join Stock Company "Mobile TeleSystems",
Moscow, Russia, uglov_ivan@mail.ru Keywords: ¿eht coHectbn agendes,
telephone traffic, provider network, VOIP, TDM.
Since 2014, stable growth of overdue debt in mortgage and other types of credits to individuals within the Russian Federation has been observed [1, 2]. As the amount of overdue credit payments grows, debt collection agencies become very popular in the bank industry [3]. For their activities, debt collection agencies use automatic devices that call the numbers from the debtor database and create anomalous traffic in the operator's network: the answer seizure ratio (ASR) usually does not exceed 7-10%, the average call duration is 10 s. The load intensity from a single client like this may reach several hundreds of call attempts per second and change unevenly. Due to low ASR and short average call duration, the number of busy voice channels is not high. Such traffic may create overloads in the operator's network unless the network performance is specially boosted on purpose taking into account the peculiar traffic profile.
Debt collection agencies are a new type of clients, standard approaches to connecting them to a network being not quite applicable to them. Low ratio of the number of chargeable (successful) connection to the load on the network equipment testifies to a need for special rates for this type of clients. The article presents a short analysis of telephone traffic features attributable to debt collection agencies. The results of this analysis were used to propose recommendations on calculation of the throughput of network elements in order to decrease the failure rate when transmitting traffic from debt collection agencies.
Для цитирования:
Углов И.В. Анализ проблем подключения коллекторских агентств к сети оператора связи // T-Comm: Телекоммуникации и транспорт. - 2016. - Том 10. - №9. - С. 73-75.
For citation:
Uglov I.V. Analysis of issues with connecting debt collection agencies to communications provider network. T-Comm. 2016. Vol. 10. No.9, рр. 73-75.
IL
73
COMMUNICATIONS
Calling debtors: scenarios and features
As stated above, debt collection agencies use automatic calling systems in their activities, therefore it is incorrect to use standard rules for load calculation in this situation [4]. The scenario of generating calls coming from collection agency equipment can be briefly described as follows:
- data on debtors, including their telephone numbers (mainly mobile), are loaded into an automatic calling system with a built-in CRM;
- a significant part of the numbers in debtor profiles is either powered off' or cut off for nonpayment {the subscriber changed the number due to often calls from collectors);
- the system starts the automatic calling process;
- if the connection is established, the call is transferred to one of the employees from the agency's contact center, who speaks to the debtor.
As the collector buys a wide range of numbers from the operator, and random numbers of the calling side are chosen for calling, the debtor can't block calls from collection agencies (for example, by using white or black lists). This scheme promotes the increase of ASR when calling.
The geography of calls depends on the region where the collection agency operates. Large agencies working across Russia call to mobile numbers of debtors all over the country.
Traffic features
At the moment, the total traffic intensity from collection agencies working in Russia is, according to various estimates, from 4 to 15 million busy hours call attempts (BHCA). Such load, especially on modules that process signals, is significant and comparable to the load from real subscribers.
ASR varies from 3 to 10%. The mean hold time (MHT) is from 5 to 7 s. The load intensity generated by collection agency equipment depends on the client and usually varies from 100 KBIICA from a small agency and up to 1.5 MB11CA and over for a large one. The average number of SIP messages when making a call does not exceed 5-6 [5| because of the large number of unsuccessful calls. Load intensity changes unevenly, changes are significant - up to 100% and more. Fig. 1 shows an example dependency of load intensity from one collection agency in call attempts per second (CAPS) from the time oTday. Load outbursts are observed at 11 o'clock, where the intensity increased by over 120 calls per second.
2SC-, UtL___
20G- '(' i__ 1___1_
160 " |ji y 111 1 .
. . Ij 1 i m
BO.__.WJ* IfljH 1__ L
lit. L 1.. L'll -j— ^
H144M4 WtM OfcM liM It;)] ¡£¡0 MlHi-li
— ^lKtfcSif »iyiS'OFS; i'figris;
Fig. 1. Example dependency of load intensity from one collection agency in CAPS from the time ofday
PUBLICATIONS IN ENGLISH
Collection agency equipment generates only outgoing load and very rarely process incoming calls. The profile of the incoming traffic is very similar to the Poisson profile, as it is generated by a large number of independent subscribers. The profile of the outgoing traffic of the collection agency can't be described by the Poisson law due to presence of the automatic system that generates calls with a predetermined or a floating intensity specified manually by the equipment tuner.
Main problems with traffic from collectors
and recommendations on how to solve them
There are several problems with connecting collection agencies to the operator's network. Some of them are engineering, the main ones are described in this article. Other ones are regulatory and legal, they arc beyond the scope of this research.
Problem I - anomalous had on equipment
For many telephone stations manufactured ten years ago and earlier, the load of I MBHCA is significant. Nowadays, such equipment predominates in networks of Russian operators of fixed line telephone.
Software switches of leading global manufacturers could cope with 3-5 MBHCA only in maximum configuration [6]. Software switches Huawei and Genband (SX3000 and C20, respectively) based on ATCA platform, which were announced several years ago and recently certified for Russian networks, can cope with such intense traffic [7], but Russian operators only begin to adopt such solutions.
When connecting collection agency equipment to the operator's network, a well-designed calculation of network performance is needed. Main bottlenecks of the operator's network, which should be given close attention:
- performance of signal processing modules throughout the calling route (at the station where connection to collector's equipment is made, internal interfaces between network switches, area and intercity communication nodes);
- performance of the transit signal line (if used);
- performance of signal processing modules at interfaces with other operators that will process traffic from the collector.
The calculation should take into account the uneven nature of the load or, if the contract with the client allows, limit the maximum intensity of the incoming load.
However, even if all the peculiarities were taken into account in the calculation of the network for collection agency traffic, there is one problem that can't be solved by the operator itself-transmission of such traffic through interoperator interfaces. If equipment of the connected operator is not designed for load from collectors, it can be overloaded, and ASR will be reduced due to activation of protection mechanisms.
Problem 2 - call blocking
Nowadays, all large operators use call analysis platforms to identify various fraud calls. Usually, these platforms analyze calls coming from the connected operators. It is likely that collection agency traffic forwarded through an interoperator interface will be blocked as fraud due to its anomalous features.
This problem is solved in several ways:
- by mixing collector traffic with traffic from regular subscribers, which does not require additional expenses from the
MV
ПУБЛИКАЦИИ НА АНГЛИЙСКОМ ЯЗЫКЕ
operator, but may lead to interface blockage if features of the mixed call stream at the connected operator will be considered as "fraud traffic".
by transmitting collector traffic through separate interoperator interfaces, which will likely lead to changing rates for clearing payments between operators for this dedicated interface.
Problem 3 - billing calls from collection agencies
As ASR in the call stream from a collection agency is anomalously low, and operators charge only for successful calls, there is a discrepancy between the large amount of network resources dedicated for transmitting traffic from collectors and a relatively low amount of paid telephone traffic. A solution to this problem is reviewing rate plans and taking into account not only the total duration of successful calls, but also their intensity.
Conclusion
In this article, the author tried to formulate the main features of connecting equipment of collection agencies to communication operator networks. Estimated features of outgoing call stream from collection agencies were given. A detailed statistical analysis is required to give an exact description of its features. It should be noted that the features
СВЯЗЬ
mentioned in the article are attributed not only to traffic for collection agencies, but also for telemarketing entities, thus making the detailed research even more topical.
References
1. Mortgages could not stand // RBC Newspaper. - Moscow, 28.10.2014.-№201 (1976) (2810)//3158,
2. Desperate Russian debt increased for the year twice // [Electronic resource] // URL: littp://www.rbc.ru/finances/26/11/2015/ 5655c5d79a7947c9e07e5ee8.
3. Golden times for collectors // [Electronic resource] // URL: http://www.gazeta.ru/social/2015/03/19/6605273.shtml.
4. Guidance document 45.120-200. Norms of technological design. Urban and rural telephone networks, - introduced 2000 - 10 -12. Moscow: Federal Slate Unitary Enterprise "Center of scientific and technical information "Informsvyaz", 2000. - 178 c.
5. IETF RFC 3665. Session Initiation Protocol (SIP) Basic Call Flow Examples // Johnston A., Donovan S., Sparks R., [etc.], [Electronic resource] // URL: http://tools.ietf.org/litml/rtc3665.
6. Product Specifications CS // [Electronic resource] // URL: http://www.si3000.ru/products/cs/.
7. SX3000 // [Electronic resource] // URL: http://wwwl.huawei.com/en/products/core-network/singlecore/fixed-
so fts w i tch/i n d ex. htm.
АНАЛИЗ ПРОБЛЕМ ПОДКЛЮЧЕНИЯ КОЛЛЕКТОРСКИХ АГЕНТСТВ
К СЕТИ ОПЕРАТОРА СВЯЗИ
Углов Иван Валерьевич, к.т.н., эксперт ПАО "МТС", Москва, Россия, uglov_ivan@mail.ru
Аннотация
Начиная с 2014 г. наблюдается стабильный рост просроченной задолженности по ипотечным и другим видам кредитов для физических лиц на территории Российской Федерации (РФ) [1, 2]. В связи с ростом объемов просроченных платежей по кредитам в банковской среде большую популярность приобрели услуги коллекторских агенств [3]. Коллекторские агентства для своих целей используют устройства автоматического обзвона телефонных баз должников, создавая на сети оператора трафик с аномальными характеристиками: процент вызовов с ответом (Answer Seizure Ratio, ASR) обычно не превышает 7-10%, средняя продолжительность вызова - 10 с. Интенсивность нагрузки от одного такого клиента может составлять несколько сотен попыток установления вызова в секунду и меняться скачкообразно. В силу низкого ASR и малой величины средней продолжительности вызова общее число занятых голосовых каналов не велико. Подобный трафик может создавать перегрузки на сети оператора в случае, если производительность сети не будет специально увеличена c учетом специфического профиля трафика. Коллекторские агентства представляют собой новый вид клиентов, к которым не применимы стандартные подходы к включению в сеть. Низкое соотношение объема тарифицируемых (состоявшихся) соединений к нагрузке на оборудование сети говорит о необходимости ввода специальных тарифов для клиентов данного типа. Проведен краткий анализ характеристик телефонного трафика коллекторских оператора связи. На основе проведенного анализа предложены рекомендации по расчету пропускной способности элементов сети, позволяющие снизить величину отказов при пропуске телефонного трафика коллекторских агентств.
Ключевые слова: коллекторское агентство, телефонный трафик, сеть оператора связи, VOIP, TDM.
Литература
1. Ипотека не выдержала // Газета РБК. - М., 28.10.2014. - № 201 (1976) (2810) // 3158.
2. Безнадежные долги россиян выросли за год в два раза. [Электронный ресурс] // URL: http://www.rbc.ru/finances/26/ll/20l5/5655c5d79a7947c9c07e5ee8.
3. Золотые времена для коллекторов. [Электронный ресурс] // URL: http://www.gazeta.ru/social/20l5/03/l9/6605273.shtml.
4. РД 45.l20-200. Нормы технологического проектирования. Городские и сельские телефонные сети. - Взамен ВНТП ll2-92; введ. 2000 - l0 - l2. Москва: ЦНТИ "Информсвязь", 2000. - l78 с.
5. IETF RFC 3665. Session Initiation Protocol (SIP) Basic Call Flow Examples // Johnston A., Donovan S., Sparks R., [etc.]. [Электронный ресурс] // URL: http://tools.ietf.org/html/rfc3665.
6. Технические характеристики продукта CS. [Электронный ресурс] // URL: http://www.si3000.ru/products/cs.
7. SX3000. [Электронный ресурс] // URL: http://wwwl.huawei.com/en/products/core-network/singlecore/fixed-softswitch/index.htm.
T-Comm Vol.10. #9-2016
