WO2007042352A1 - Method and apparatuses for processing calls - Google Patents

Abstract

The present invention relates to a method and apparatuses for processing calls. More particularly, the present invention relates to a method and apparatuses for determining and refusing potentially unwanted calls. In accordance with the in- vention a call is received from a calling party (140). An announcement is played to the calling party (140), the announcement requesting the calling party (140) to respond, the announcement comprising enough information for a human caller to determine a correct response. The response provided by the calling party (140) is compared with the predefined correct response, and the call is connected to a called party (110) only if the response provided by the calling party (140) matches the correct response.

Description

Description

Method and Apparatuses for processing calls

This application is related to and claims the benefit of com^¬ monly-owned U.S. Provisional Patent Application No. 60/726,716, filed October 14, 2005, titled "Method and Appa^¬ ratus for a Spit Blocker Service" which is incorporated by reference herein in its entirety.

The present invention relates to a method and apparatuses for processing calls. More particularly, the present invention relates to a method and apparatuses for determining and re^¬ fusing potentially unwanted calls.

Telecommunications networks are currently evolving from tra^¬ ditional circuit based networks (PSTN = Public Switched Te^¬ lephony Network) to packet based networks, wherein communica^¬ tion is facilitated by well-known voice-over-packet (VoP) mechanisms. A prominent example of VoP is voice over Internet Protocol (VoIP) , wherein the well-established Internet Proto^¬ col (IP) is used as network layer protocol for conveying both signaling and voice.

In general, phone service via VoIP is free or costs less than equivalent service from traditional sources. Some cost sav^¬ ings are due to using a single network to carry voice and data.

The rise in the popularity of VoIP, and in particular that of SIP (Session Initiation Protocol) which is most commonly used for the signaling task in VoIP, makes a phenomenon called SPIT (Spam over IP Telephony) a serious concern.

Telephony Spam, or SPIT, consists of computers making telemarketing voice calls or sending SIP instant messages. Such calls consume subscriber bandwidth, waste time, are irritable and generally unwanted by the called party. Many of these calls are generated using computers. A computer application employed by a telemarketer may send a SIP invite message, wait for the called party to answer/accept the call and then re-direct the call to a sales associate or a recorded mes^¬ sage. If a called party employs voice mail, his/her voice- mail box could become clogged with unwanted advertising mes^¬ sages, just like email inboxes are today.

Assuming that several automated call agents ("SPIT bot") at^¬ tempt to reach a particular called party simultaneously or in rapid succession, e.g. during the early evening hours normally preferred by telemarketers, the same effect as what is known as a distributed denial-of-service attack may occur, namely a service interruption of the called party's phone and even internet service.

A known attempt to reducing the number of unwanted calls is to maintain black lists having entries of caller IDs known to initiate unwanted calls. This is easily rendered ineffective by masquerading the caller ID of a black-listed caller, in particular in the VoIP realm where spoofing an IP address poses no problem at all.

Another approach to avoiding unwanted calls is known as Selective Caller Accept (SCA) . This service provides the sub- scriber with a PIN that each caller must enter to be connected. Callers who do not enter the correct PIN are diverted either to voice mail or are rejected. A disadvantage with SCA is that the PIN has to be pre-shared among all allowed call^¬ ers. Communicating (i.e., pre-sharing) the PIN, and changing the PIN, is a time-consuming process.

It is therefore an object of the present invention to provide a novel method and apparatuses for processing calls, the method and apparatuses capable of determining and refusing potentially unwanted calls.

In accordance with the foregoing objects, there is provided by the invention a method for processing calls, comprising: receiving a call from a calling party; playing an announcement to the calling party, the an^¬ nouncement requesting the calling party to respond, the announcement comprising enough information for a human caller to determine a correct response; - comparing a response provided by the calling party with the predefined correct response; and connecting the call to a called party only if the response provided by the calling party matches the correct re^¬ sponse .

By this method, a non-human caller, e.g. an automated call agent such as a SPIT bot, will be confronted with a request it cannot respond to without analyzing the announcement and comprehending the information contained therein, pointing to the correct answer. Thereby, most automated call agents will be prevented from being connected to the called party.

In an embodiment of the invention, a white list screening may precede the step of playing the announcement, thereby allow- ing known caller IDs (friends, family) to bypass the request- response steps and directly reach the called party.

To prevent automated call agents having automated speech rec^¬ ognition capabilities from correctly recognizing the an- nouncement, the announcement may be composed of an instruc^¬ tion or question with underlying noise or music.

In an embodiment, the response may be provided by a human caller by subsequently pressing one or more keys of a stan- dard telephone keypad, creating DTMF (dual-tone multi- frequency) signals, which are well known in the art.

In a further embodiment there may exist a set of announce^¬ ments from which an announcement will be randomly selected, thereby preventing a one-time human analysis of the announce^¬ ment for the purpose of providing the correct response to the automated call agent for future automated provisioning of the correct response by the automated call agent. In an embodiment, the information required for a human caller to determine a correct response is given directly in the an^¬ nouncement, e.g. in spoken format ("Please press 1 to con^¬ tinue") .

In another embodiment, the information required for a human caller to determine a correct response is given indirectly in the announcement, i.e., in one or more pieces of information that need to be arithmetically or logically combined by a hu- man caller in order to yield the predefined correct response ("Please enter the result of 7 minus 2") .

In accordance with the invention, there is also provided a network element serving a called party having means for per- forming the method of any of claims 1 through 7, and a user terminal attachable to a telecommunications network having means for performing the method of any of claims 1 through 7.

Embodiments of the invention will now be described in more detail with reference the drawings, wherein:

Fig. 1 schematically shows a network arrangement having a network element in accordance with the invention; Fig. 2 shows the network element of Fig. 1 in greater detail; and

Fig. 3 is a flow diagram of a preferred embodiment of the present invention.

In Fig. 1, there is shown a network arrangement 100 compris- ing a subscriber's premises 110 (in the following also: the called party) having multiple terminal devices including telephones 112, 114, 118 and a personal computer 116; a net^¬ work 120; and a telemarketing agency 140 (in the following also: the calling party) .

Network 120 in the preferred embodiment is an IP based net^¬ work and may, for example, be the Internet. In other embodi^¬ ments, network 120 may represent a legacy TDM (Time Division Multiplex) network. In the preferred embodiment, network 120 comprises several network elements, including at least one switching network element 122 serving the called party 110 and access devices such as integrated access device (IAD) 124 providing connectivity for the called party 110, a DSLAM (digital subscriber line access multiplexer) 126, and IAD 128 providing connectivity for the calling party 140.

A database or database server 130 may be provided for storing the white list and/or for storing announcements and expected responses.

In operation of the network arrangement of Fig. 1, subscriber 110 is assumed to have subscribed to a SPIT blocker service which provides the inventive method for subscriber 110 in network arrangement 100. The SPIT blocker service may be con^¬ figured to allow subscriber 110 to provide personal announce^¬ ments along with corresponding expected responses to his/her serving network element 122 which in turn stores a digital representation thereof in database 130. It shall be noted that personal announcements will prove much more difficult to counter by manufacturers of automated call agents than stan^¬ dardized announcements since somewhat advanced automated call agents would, if at all, only be able to anticipate standard^¬ ized announcements and perhaps respond correctly. Subscriber 110 may further be enabled to provide white list entries com^¬ prising caller IDs that are always allowed to bypass the SPIT blocker service.

If an automated call agent residing in some programmable de- vice at the telemarketing agency 140 initiates a call request to called party 110, switching network element 122 determines that called party 110 has SPIT blocker service activated with a white list configured. Switching network element 122 then verifies whether the caller ID of the automated call agent is in the white list. If it is not, then switching network ele^¬ ment 122 will retrieve a personalized announcement from data^¬ base 130, if available. If no personalized announcement is available, a standard announcement may be used. The announce^¬ ment is played to the caller 140, requesting the caller 140 to respond. If no response is received within a timeout, or the response does not match the expected response, then the call may be terminated or redirected in accordance with the called party's stored preference.

The announcement may be randomly selected from a set of available announcements. Of course, an announcement may be chosen to consist of a fixed part and a randomly selected variable part for more efficient memory management. The fixed part may for example be the general request to the calling party ("Please enter the following number using your telephone keypad") and the variable part a randomly generated number ("123") which will be appended to the fixed part, e.g. in the form of synthesized speech. The caller will then be expected to successively press the keys 1-2-3 on his/her telephone keypad to be connected to the called party.

To further obstruct automated call agents, the announcement may comprise a question that requires the caller to perform an arithmetical or logical task in order to arrive at the ex^¬ pected response. Examples include "What is 6 divided by 3?"; "Please type the first three letters of my last name using the keypad.". Again, the announcement may be split into a fixed and a variable part, the variable part preferentially being the pieces of information to be combined in accordance with the general instructions of the fixed part in order to arrive at the expected response.

Turning now to Fig. 2, there is shown a call processing en- gine 220 of network element 122. An incoming signaling manager of call processing engine 220 receives a call setup mes^¬ sage 210 in accordance with any signaling protocol including but not limited to ISUP (ISDN user part), ITU-T Q.931, H.323, SIP, or MGCP (Media Gateway Control Protocol) . Call process- ing engine 220 then determines the called party from the call setup message and further determines if the called party has the SPIT blocker service enabled. If the SPIT blocker service is enabled for the called party, the white list check and an- nouncement retrieval is performed as explained above by in^¬ terfacing with database 130 using an interface 228.

In accordance with the result of the SPIT blocker handling, call processing engine 220 will either terminate the call or output a modified or an unmodified signaling message 230 in order to redirect the call to an answering device (not shown) or connect the call to the called subscriber, respectively, by means of an outgoing signaling manager 224.

Finally, Fig. 3 shows a block diagram of a preferred embodi^¬ ment of the inventive method. In step 300, the operation be^¬ gins with receiving a call. It is then determined, in op^¬ tional step 310, if the calling party is in a white list. If the calling party is in the white list, then the operation of the call processing resumes normally with connecting the call to the called party in step 350. If the calling party is not in the white list, then the called party will be presented with an announcement in step 320, the announcement prompting the caller to respond. If no response is received in step 330 or the response does not match an expected response, the call processing continues with terminating the call or diverting it to voice mail in step 340. If however the response re^¬ ceived matches the expected response, then call processing resumes with connecting the call to the called party in step 350.

In a preferred embodiment, the announcement is composed of an instruction or question with underlying noise or music, thereby obstructing advanced automated caller agents, namely those capable of automated speech recognition, as these might be able to respond correctly to the announcement. Initial re^¬ search shows that classical music may be suitable to prevent automated speech recognition of the announcement. Alterna- tively, the announcement could be unnaturally distorted by modifying the playback timing parameters, inserting delays, etc., such that the announcement is still comprehensible to human callers but unusable for the purpose of automated speech recognition. In yet another alternative, the announce- merit could be a question or announcement in the form of a short song, i.e. the music itself could be the announcement.

It shall be noted that the SPIT blocker service as described above can be implemented in centrally in a network element

122 such as a switch or a signaling gateway or an application server coupled to a network element 122 but may also be im^¬ plemented in a subscriber device such as a residential gate^¬ way or any terminal device.

While the present invention has been described by reference to specific embodiments and specific uses, it should be un^¬ derstood that other configurations and arrangements could be constructed, and different uses could be made, without de- parting from the scope of the invention as set forth in the following claims.

Claims

Claims

1. A method for processing calls, comprising: receiving a call from a calling party (140); - playing an announcement to the calling party (140), the announcement requesting the calling party (140) to re^¬ spond, the announcement comprising enough information for a human caller to determine a correct response; comparing a response provided by the calling party (140) with the predefined correct response; and connecting the call to a called party (110) only if the response provided by the calling party (140) matches the correct response.

2. The method of claim 1, comprising: prior the playing the announcement, querying a white list, the white list comprising a list of caller IDs allowed to bypass the announcement-answer steps; and connecting the call to the called party if the calling party's caller ID is in the white list.

3. The method of any of claims 1 or 2, wherein the announce^¬ ment is composed of an instruction or question with underlying noise or music, thereby obstructing automated speech rec- ognition of the announcement.

4. The method of any of claims 1 through 3, wherein the an^¬ nouncement can be responded to by subsequently pressing one or more keys of a standard telephone keypad.

5. The method of any of claims 1 through 4, wherein the an^¬ nouncement is randomly selected from a predefined set of an^¬ nouncements .

6. The method of any of claims 1 through 5, wherein the in^¬ formation for a human caller to determine a correct response comprises an audible representation of the predefined correct response .

7. The method of any of claims 1 through 5, wherein the in^¬ formation for a human caller to determine a correct response comprises one or more pieces of information that, when arith^¬ metically or logically combined by a human caller, yield the predefined correct response.

8. In a telecommunications network (100), a network element (122, 124) serving a called party having means for performing the method of any of claims 1 through 7.

9. A user terminal (112, 114, 116, 118) attachable to a tele^¬ communications network (100) having means for performing the method of any of claims 1 through 7.