a second end device

Description

The present invention relates to a communication system, an end device, and a method for establishing a communication connection between a first end device and a second end device.

In some industry or business branches, the phone service belongs to the business critical environment. Such branches are, for example, trading (e.g. brokers), dispatching (e.g. flight control) and emergency services (e.g., US 911 or EU 112). Some central components of telecommunications systems are usually operated with redundancies. But the phone, as for example, a turret or even a communication application for every established call, usually receives just one voice stream, for example, transmitted according to Real-Time Transport Protocol (RTP) or Secure Real-Time Transport Protocol (SRTP), and one data stream, for example, transmitted according to Session Initiation Protocol (SIP). In case any of these streams fails, the call may break down or at least be interrupted for a certain period of time.

Another issue which may happen is that the streamed voice quality for the call gets worse. A reason for this could be just network issues because the user is using a mobile device and changes his location during the call. Also, the traffic in the network used for the call may change. In this case, the use of another network would solve this problem. However, this again may cause a break-down of the communication connection or at least an interruption.

Therefore, the present invention is based on the object to overcome the problems described above. This object is solved by a communication system having the features according to claim 1 , an end device having the features according to claim 7, and a method for method for establishing a communication connection between a first end device and a second end device having the features according to claim 9. Preferred embodiments of the invention are specified in the respective dependent claims.

Accordingly, a communication system for performing a call comprising a first end device and a second end device in a communication network is provided, the first end device being adapted to establish a first communication connection to the second end device, the first communication connection being performed via a first communication path and a second communication path, the first communication path being used for a voice and/or video stream, and the second communication path being used for a data stream, wherein the first end device is adapted to establish a second communication connection simultaneously to the first communication connection to the second end device, the second communication connection comprising at least a third communication path being used for the voice and/or video stream, the third communication path redundantly transmitting the same voice and/or video stream as the first communication path to the second end device, wherein during the call to the second end device, the first end device further is adapted to continuously evaluate the quality of both voice and/or video streams and to select from the first and third communications paths the one with the better quality of the voice and/or streams as an active voice and/or video stream for output and to mute the other voice and/or video stream with the poorer quality in both communication directions.

According to the inventive communication system, since a first end device is equipped with the functionality to establish two calls via two communication connections or via two different communication paths in parallel which contain actually the same information, namely, the same voice and/or video stream, so the voice and/or video stream is transmitted redundantly end-to-end, the call quality may always be maintained and will not be interrupted or break down. The voice streams, which may be RTP streams, of the two parallel communication connections for the two parallel calls, which basically are mainly relevant for the call quality, are likely to differ with respect to call quality since they are transmitted via different paths. The end device compares both voice and/or video streams and evaluates the quality of the two streams, e.g. latency and jitter by means of network Quality of Service (QoS) statistics gathered, for example, by means of Real-Time Control Protocol (RTCP) reports. The voice and/or video stream with the better quality will forwarded and presented to the user at the handset or speaker and only this communication path will be used as the upstream voice channel for the call. The voice and/or video stream transmitted via the path with the respectively poorer quality, thereby, is always muted in both communication directions.

The user of the end device will not recognize switching to the voice and/or video stream with the better quality at all. Rather, the call always appears like a normal call, since the stream with the respectively poorer quality always is muted in both communication directions so that the user will not realize that, from a technical point of view, there are two calls. The call handling isn’t affected at all. The user is able to set up calls, put the calls on hold, monitor lines or release the call etc. With every action of the user, the two calls will be handled automatically, whereby the user is always speaking and listening only to the call with the best voice quality.

Also, failover is provided by this configuration, since in case either one of the two voice and/or video streams fails, the other one will be used. This is particularly important, for example, for stockbrokers who are receiving and providing instructions via such end devices, which in this case would be configured as so-called turret phones. Here, it is particularly important that the call is not interrupted, because every second may count when processing or settling a transaction via the phone. Moreover, with the inventive configuration, the stockbroker always will always be able to correctly understand the instructions provided via the phone, since he always will listen to the voice and/or video stream having the better quality. But also, the inventive configuration is particularly important for other business critical phone applications, like flight control dispatching, emergency services (911 numbers), or the like. Namely, in all of these scenarios beside telephones and soft clients, for example, so-called turrets or turret phones are in use capable of handling several simultaneous and/or bridged calls For such business critical phone systems and applications, it is not acceptable that a call gets dropped due to network failures or the voice quality gets degraded due to network impairment. This, however, will be efficiently avoided by using the inventive communication system.

According to a preferred embodiment, the first communication connection, in particular, the first communication path of the first communication connection, is established via a network or infrastructure which is different from the one used for the second communication connection, in particular, the third communication path from the second communication connection. Thus, since the voice and/or video stream is transmitted via different paths, always the one with the better call quality will be used for output and in case the selected path degrades with respect to call quality, the end device will switch to the other one for being output, for example, at a handset of the end device.

According to a further preferred embodiment, the communication network further comprises geo-separated central units, in particular, a plurality of media servers, gateways, or session border controllers, wherein the first communication connection is directed over a first central unit and the second communication connection is directed over a second central unit. Namely, the use of different network infrastructures can be facilitated by routing the corresponding calls via geo-separated central networks entities like media servers, gateways, or session border controllers.

Further, according to still another preferred embodiment, the second communication connection comprises a fourth communication path used for the data stream. Thus, signaling may also be made redundant, in particular, end-to-end, whereby the two established communication connections also use different signaling systems (for example, SIP, Web Real-Time Communication (WebRTC), or the like) as long as they traverse the same network element routing the calls across different network infrastructures.

Preferably, the voice and/or video stream uses RTP or SRTP, and wherein the data stream uses Session Initiation Protocol SIP.

It is noted that simultaneous call handling also applies for feature invocations for call hold or the like of the corresponding calls. SIP-based phone systems, e. g. Private Branch Exchange, can support the call route distribution by the so called Multi-line Appearance (MLA) normally used for keyset device groups.

According to still a further preferred embodiment, the first end device is a multi-line dispatch phone, in particular, a turret phone, adapted to handle a plurality of calls simultaneously. The second end device may also be a multi-line dispatch phone, and in particular, a turret phone. It may also be another type of phone, but in any case, it needs to be adapted to handle a plurality of calls, or at least two calls, simultaneously, too.

Moreover, according to the present invention, an end device is adapted to perform a call and to establish at least a first communication connection to a second end device, the first communication connection comprising a first and a second communication path, the first communication path being used for a voice and/or video stream, and the second communication path being used for a data stream, wherein the end device is adapted to establish a second communication connection simultaneously to the first communication connection to the second end device, the second communication connection comprising at least a third communication path being used for the voice and/or video stream, the third communication path redundantly transmitting the same voice and/or video stream as the first communication path to the second end device, wherein end device during the call to the second end device, further is adapted to continuously evaluate the quality of both voice and/or streams and to select from the first and third communications paths the one with the better quality of the voice and/or streams as an active voice and/or video stream for output and to mute the other voice and/or video stream with the poorer quality in both communication directions.

The end device according to the present invention provides the advantages already outlined above. In particular, the end device is able to establish a redundant call or “shadow call” routed via different network infrastructures seamlessly for the user sustaining call stability and voice quality for business critical voice services in case of network impairments or failures. According to the inventive end device, always two calls are established by the device seamlessly for the user, each routed, for example, by the multi-line controller via different network infrastructures. The end device, which either may be the sending or the receiving device in a communication connection for a call, is adapted to continuously analyze the network using existing Quality of Service (QoS) measures (network latency and jitter), e.g. from RTCP feedback metrics analysis, of both voice and/or video streams and renders only the voice and/or video stream to a user with the estimated sufficient or better voice quality experience. The end device is also able to switch over to the other media of the non- active and muted“shadow” call in case that one of the two calls breaks down due to network failure.

The inventive solution is also robust in case of data center failures or related overloaded network segments.

Also, the end device according to the present invention has the advantage to be configured so as to be flexible with respect to changes, for example, in a VoIP architecture. Namely, by the inventive end device, a solution is provided in the application layer, where the VoIP endpoint can create its own redundant calls using already existing protocols.

Moreover, the inventive end device has the advantage that existing security mechanisms and application gateways do not have to be changed or modified so as to be compatible with it as the redundant calls are fully independent calls, with own call IDs and encryption keys.

Also, as already mentioned above, according to a preferred embodiment, different signaling protocols e.g WebRTC, SIP etc. can be used as long as the media servers are able to mediate.

According to another preferred embodiment, the end device is further adapted to compare the voice and/or video stream of the first communications path and the voice and/or video stream of the third communications path, and to evaluate the quality of the respective voice and/or video streams with respect Quality of Service, in particular, with respect to latency and jitter. Further, according to the invention, a method for establishing a communication connection between a first end device and a second end device in a communication network for performing a call is provided, the method comprising the steps of establishing a first communication connection from the first end device to the second end device via a first network path or infrastructure, the first communication connection comprising a first and a second communication path, the first communication path being used for a voice and/or video stream, and the second communication path being used for a data stream; establishing a second communication connection simultaneously to the first communication connection to the second end device, the second communication connection comprising at least a third communication path being used for the voice and/or video stream; transmitting, via the third communication path, the same voice and/or video stream as being transmitted via the first communication path to the second end device; and during the call from the first end device to the second end device, continuously evaluating the quality of both voice and/or video streams and selecting from the first and third communications paths the one with the better quality of the voice and/or streams as an active voice and/or video stream for output and muting the other voice and/or video stream with the poorer quality in both communication directions. The inventive method provides the advantages already outlined above.

The method may further comprise a step of comparing the Quality of Service (QoS), in particular with respect to latency and jitter, of the voice and/or video stream transmitted on the first communication path and the voice and/or video stream transmitted on the third communication path, and forwarding to a handset of the first end device, the voice and/or video stream with the better QoS.

Preferably, the muted voice and/or video stream is maintained as a non-active voice and/or video stream.

According to another preferred embodiment, if the quality of the active voice and/or video stream drops or the active voice and/or video stream stops, the method further comprises a step of unmuting the non-active voice and/or video stream so as to be used for the ongoing call. When performing the step of unmuting the non-active voice and/or video stream, simultaneously the active voice and/or video stream may be muted in both communication directions.

According to still a further preferred embodiment, the voice and/or video stream transmitted on the first communication path is a first RTP stream which is transmitted from the first end device to the second end device via a first media server, a first gateway or a first border controller located in a first network, and wherein the voice and/or video stream transmitted on the third communication path is a second RTP stream which is transmitted simultaneously from the first end device to the second end device via a second media server, a second gateway, or a second border controller located in a second network.

The first end device may be a multi-line dispatch phone, and wherein the first and second media servers respectively are multi line controllers.

The invention and embodiments thereof are described in connection with the drawing.

Fig. 1 illustrates a communication system according to an embodiment of the invention; and

Fig. 2 illustrates a communication system according to a further embodiment of the invention.

Fig. 1 illustrates a communication system 1 for performing a call from a first end device 2 to a second end device 3 according to a first use case in a communication network 4. The first end device 2, here, is embodied as a phone, but it may be also be configured specifically as a turret phone or it may be embodied as a communication application. The second end device 3 may also be a phone, a turret phone, or a communication application. In either case, both phones must be able to establish two calls concurrently. The first end device 2 is configured to establish a first communication connection 5 for a call to the second end device 3, whereby the first communication connection is performed via a first communication path 5a and a second communication path 5b. The first communication path 5a is used for a voice and/or video stream, for example, according to RTP. The second communication 5b path is used for a data stream, for example, according to a signaling protocol like SIP. Further, the first end device 2 is adapted to establish a second communication connection 6 simultaneously to the first communication connection 5 for a further call to the second end device 3, whereby the second communication 6 connection also comprises two communication paths, namely, a third communication path 6a which is used for the voice and/or video stream comprising the same information as the voice and/or video stream transmitted via the first communication path 5a, and a fourth communication path 6b which is used for the data stream. Thus, according to the embodiment shown here, both streams, the voice and/or video stream and the data stream, may be transmitted redundantly. Namely, the voice and/or video stream will be transmitted on the first communication path 5b of the first communication connection 5, and at the same time the same voice and/or video stream will be transmitted on the third communication path 6a of the second communication connection 6.

However, it is noted that generally, for the quality of the call, the transmission quality of the voice stream mainly is important so that it actually would not be necessary to transmit the data stream redundantly, too. Nevertheless, in this embodiment the data stream on the fourth communication path 6b is a SIP call control connection and is in this embodiment needs to be established, too.

After the end device 2 has established all communication connections 5, 6, during the call to the second end device 3, the first end device 2 further will continuously evaluate the quality of both voice and/or video streams, and, if necessary, of both data streams, and will select from the first and third communications paths 5a, 6a the one with the better quality of the voice and/or streams as an active voice and/or video stream for output, for example, at the handset 7. As previously explained, the“voice” quality will most likely be different in both voice streams, since they use different transmission paths, as will be outlined further below. The end device 1 compares both voice streams and evaluates the quality or QoS of both streams, e.g., with respect to jitter and latency by means of network quality of service statistics gathered, for example, on the basis of Real-Time Control Protocol (RTCP) reports. The QoS which may be assessed for evaluating the“call quality” is regarded as the performance seen by the users of the end devices 2, 3. It is noted that in order to quantitatively measure QoS, several related aspects of the network service other than the ones mentioned above may also be considered, such as packet loss, throughput, transmission delay, availability, etc.

Then, after the first end device 2 has determined the voice stream having the better call quality, the latter is set as the active voice stream during the call, and communication path transmitting this voice stream is selected and used as the upstream channel for the call. Only this voice stream will be transmitted to the handset 7, as mentioned above, and will be presented to the user of the first end device 2, while the other voice and/or video stream with the poorer quality in this embodiment may be muted. However, it is noted that the outgoing stream must not be muted necessarily in order to enable end device 3 to choose the better one of both streams. Thus, the user of the end device of phone has the impression that only one call comes in, although technically speaking, there are two incoming calls transmitting the same voice and/or video stream.

In case the selected transmission channel for the voice stream, namely, the communication path having the better call quality, degrades or even is interrupted or fails, then the first end device 2, due to continuously checking and monitoring the call quality, will recognize this, and will instantly switch to the other communication path, which now has the better call quality or the connection of which did not break down, and will unmute it while muting the incoming voice stream on the degraded communication path which previously was the active one. This procedure provides for seamless switching between the two communication paths 5a, 6a so as to always keep the ongoing call on the best possible quality standard.

In order to provide different communication paths from the first end device 2 to the second end device 3, both communication connections 5, 6 are directed through geo- separated central units or components interposed in the communication network 4 between the first and second end devices 2, 3, namely, a first central unit interposed between the first and second end devices 2, 3 in the first communication connection 5, and a second central unit 9 interposed between the first and second end devices 2, 3 in the second communication connection 6. The second central unit 9 is located at another place than the first central unit 8. The first and second central units 8, 9 may respectively be embodied as first and second media servers 10, 10’ (see Fig. 2), as first and second gateways 11 , 1 T, or as first and second session border controllers 12, 12’.

Further, a first SIP call controller 13 and a second SIP call controller 13’ is provided, whereby the first SIP call controller 13 is used in the first communication connection 5 and the second SIP call controller 13’ is used in the second communication connection 6. By this configuration, it is possible that the voice and/or video stream is transmitted through different paths in the network or different infrastructure, as different networks, for example a Public Switched Telephone Network (PSTN) 14, 14’ or Service Switching Point (SSP) 15, 15’ network, are used for its transmission so that if one connection is interrupted or degrades, there is a high possibility that the other one is still stable.

Fig. 2 illustrates a communication system 1 according to a further embodiment of the invention which basically has the same components as the ones described with respect to Fig. 1. Flere, the first end device 2 again is embodied as a phone, namely, as a multi-line dispatch phone which can handle several calls at the same time. In this embodiment, the end device 2, namely, the dispatch phone is configured to always establish communication connections or to connects calls over a media server 10, 10’ which in this case is configured as a multi-line controller, in which media is anchored to provide centralized voice recording and advanced line sharing capabilities. As can be seen, there are two media servers 10, 10’ in the communication system 1 , whereby the first media server 10 is located in the first communication connection 5 and the second media server 10’ is located in the second communication connection 6. The two media servers 10, 10’ are placed in different geographic locations or data centers so that the calls can run through different networks. Both communication connections 5, 6 between the first end device 2 and the second end device 3, before arriving at the second end device 3 are directed to a SIP PBX or SIP Call Controller 13 which supports shared lines / MLA (Multi-Line Appearance) normally used for keysets. It is noted that according to the use case of this embodiment, the redundant calls are transparent to the outside world. Namely, redundancy is built up by connecting the two media servers 10, 10’ into the SIP PBX’s media server for the two calls 5a, 6a. The PSTN or SSP will not be aware of the redundancy and only has to handle one single call.

This line sharing feature is used to ensure that the two calls transmitted simultaneously via the first and second communication connections 5, 6 as already outlined with respect to Fig. 1 towards the dispatch phone, namely, to the end device 2 which establishes the backup call via the media server 10’ after the first call has been established, contain the same RTP information. The SIP protocol doesn’t need to be changed. Namely, where this additional redundancy or quality adjustment is needed the (dispatch) phone, here, the first end device 2, builds up a second call over another media server, namely, the second media server 10’ for every call. The first and second media servers 10, 10’ will utilize the line sharing of the SIP call controller 13 to barge in (SIP keyset bridging) the original call between the first media server 10 and the PBX. Therefore two calls will be established which contain the same voice information in the RTP stream.

The end device 2 or phone compares both voice and/or video streams and evaluates the quality of the streams against e.g. latency and jitter (QoS, as mentioned above). The voice and/or video stream, or here specifically the RTP stream with the higher quality will be forwarded in the first end device 2 to the handset 7 and therefore can be heard by the user of the phone, this is the active RTP stream. In the use case according to this embodiment, the other voice and/or video stream, or here specifically the other RTP stream will be muted in both directions (non-active stream). If the QoS of the“active” RTP stream drops or even the RTP stream stops, then the non-active stream will be unmuted and it will be used for the call, whereas the previously active RTP stream will be muted so that the user of the phone has the impression that there is only one call although technically speaking, there are two calls one of which being active or unmuted and the other one of which being non- active or muted. Thus, for the user the call will continue even if one data center fails or the network to one data center will be overloaded. Reference numerals

1 communication system

2 first end device

3 second end device

4 communication network

5 first communication connection

5a first communication path

5b second communication path

6 second communication connection

6a third communication path

6b fourth communication path

7 handset

8 first central unit (e.g. Private Branch Exchange)

9 second central unit (e.g. Private Branch Exchange)

10, 10’ first and second media servers

11 , 11’ first and second gateways

12, 12’ first and second session boarder controllers

13, 13’ first and second SIP call controllers

14, 14’ Public Switched Telephone Network

15, 15’ Service Switched Point