NL2003664C2

NL2003664C2 - Mobile communication device, system, and method of communicating.

Info

Publication number: NL2003664C2
Application number: NL2003664A
Authority: NL
Inventors: Oswald Ortiz; Peter Erik Fuijk
Original assignee: 6Gmobile B V
Priority date: 2009-10-16
Filing date: 2009-10-16
Publication date: 2011-04-19
Also published as: WO2011046439A1

Description

MOBILE COMMUNICATION DEVICE, SYSTEM, AND METHOD OF

COMMUNICATING

The invention relates to a mobile communication device for sending and receiving 5 voice data coded in a digital format over a packet-switched network. The invention further relates to a mobile communication system. The invention further relates to a method for receiving voice data from a packet-switched network and rendering it on an audio output unit of a mobile communication device. The invention further relates to a method for receiving voice data from a audio input unit of a mobile communication 10 device and sending it over a packet switched network. The invention further relates to a method for setting up a call from a calling party to a mobile communication device.

The invention further relates to a method for setting up a call from a mobile communication device to a called party.

15 During a telephone conversation between users, wherein one or more users use for example a mobile communication device such as a handheld GSM phone or a smartphone, voice traffic or voice data is generated by each user and exchanged between users over a communication network (herein also referred as the network or the data network).

20

The voice data is based on analog voice data recorded by a microphone device, after which the analog data is usually digitized by an analog-to-digital converter (ADC) and subsequently coded into a specific digital format by a coder/decoder unit (codec) as digitized coded voice data, hereafter voice data.

25

These voice data can be transferred from the sender to the receiver over a telecommunication network in a number of ways. When transferring data over a network, network resources such as available network bandwidth and network routing facilities are used. For voice communication, two types of networks are common: 30 circuit-switched networks and packet-switched networks.

The first type, a circuit-switched network, uses circuit switching to transfer said voice data, in which the network resources are statically allocated from the point of origin of 2 the data to the point of destination. These resources are allocated irrespective of their use. For example, if there is momentarily no voice data to be transported, or only some microphone noise, still the full transfer capacity is allocated. Although circuit-switched networks work quite well, because of this inefficiency the bandwidth requirements are 5 usually higher than in alternative packet-switched networks.

The second type, a packet-switched network, uses a technique in which the message is subdivided into packets, each of which can take a different route to the destination. At the destination the packets are recompiled into the original message. Packet switched 10 networks may be an attractive choice for voice transport, for reasons including cost, ability to scale, ability to make efficient use of the available total bandwidth, and the possibility to integrate of voice and general data applications (e.g. e-mail, video telephony, web browsing, etc).

15 For mobile communication devices, such as for example GSM phones or multi-purpose mobile devices, a number of packet-switched network services are available at present, such as GPRS (General Packet Radio Service), EDGE (Enhanced Data Rates for Global Evolution), WCDMA (Wideband Code Division Multiple Access), UMTS (Universal Mobile Telecommunications System), or HSDPA (High-Speed Downlink 20 Packet Access) network service. GPRS and EDGE may be considered a second generation (2G) service, and are also sometimes called a 2.5G service. WCDMA and UMTS are examples of third generation (3G) network services, and HSDPA is considered 3G or 3.5G technology. Other wireless packet-switched network services currently available are for example based on IEEE 802.11 variants, and are sometimes 25 called Wi-Fi.

VoIP (Voice over IP) is the transport of voice traffic or voice data using a packet-switched Internet Protocol (IP). For voice data traffic, VoIP systems have a protocol layer on top of the Internet Protocol layer. The standard protocol for VoIP voice data 30 traffic is RTP (Real-time Transport Protocol). The Real-time transport protocol is a protocol that comprises a control layer called RTCP (Real-time Transport Control Protocol). RTCP is used to monitor transmission statistics and quality of service (QoS) information. The Real-time Transport Protocol has error-correction features, to correct 3 for errors in the transmission.

For other types of communication, such as setting up out-going calls and handling incoming calls, the standard protocol used by VoIP applications is SIP (Session 5 Initiation Protocol), a text-based signaling protocol maintained by the IETF Network Working Group and described in RFC 3261. Using a session protocol such as SIP, a user of a VoIP device may set up a VoIP telephone conversation or connection with a user of another VoIP device that supports the same protocol using an available network such as the internet. Because the VoIP telephone network and the standard Public 10 Switched Telephone Network (PSTN) use different protocols for setting up connections and for exchanging voice data, in principle a VoIP device is not able to connect to a telephone device connected to the PSTN, even if the VoIP device in question is a PSTN-connected mobile smartphone making use of a mobile internet connection service such as the aforementioned GPRS. A VoIP device has it’s own VoIP identifier 15 (sometimes called a Uniform Resource Identifier, URI), which is not a standard telephone number.

A drawback of current Packet Switched voice communication implementations like VoIP is, that the communication protocol, for example RTP, has a considerable 20 network usage overhead that consumes a part of the available network bandwidth, reducing said bandwidth available for the voice data. The overhead is due to the error-correcting features of the protocol which are needed to handle errors introduced in the data as it is transported over the network. Another drawback is that the communication protocol features can introduce additional latency, which is undesirable in real-time 25 applications such as voice communications. A further drawback of VoIP devices is that they can not by themselves connect to a device on the PSTN. This means that a subscriber having a telephone number through which he or she can be reached using the PSTN, cannot be reached using the same number via a VoIP connection. Therefore, a user having a landline telephone, a GSM mobile communication device, and a VoIP 30 capable device, will have three distinct contacting numbers which he needs to distribute, along with an explanation which is which, to his contacts. A further drawback of VoIP devices is that they often have a user interface (UI) which is very unlike a traditional telephone user interface, and thus confusing to people who are used 4 to standard telephone devices. A further drawback of VoIP devices is that they are dependent on the availability of a packet switched network service. In many areas in the world, only standard circuit-switched GSM service is available, so in such an area a standard VoIP device is unreachable. A drawback of GSM devices is that GSM 5 services, especially international GSM connections, are typically more expensive than VoIP services. A drawback for a user having both a VoIP service subscription and a circuit-switched mobile telephony subscription, is that the user will typically have value added services, such as voice mail, in duplicate. A drawback of VoIP services is that some network providers actively discourage, for a variety of reasons, VoIP data 10 traffic on their networks which may lead to a reduced quality of the VoIP service.

It is an object of this invention to provide a mobile communication device that overcomes at least one of the drawbacks. A still further object of this invention is to provide a mobile communication device that makes more efficient use of a packet 15 switched network than a standard VoIP device. A further object of this invention is to provide a mobile communication device that has an improved user interface. A further object of this invention is to provide a mobile communication device that can be more easily contacted through a number. A further object of this invention is to provide a mobile communication device that can be reached in more areas. A further object of 20 this invention is to provide a mobile communication device that allows to make connections at a lower cost.

According to an aspect of the invention the object is achieved in a mobile communication device, in particular a device integrated with a mobile telephony 25 handset, more in particular a mobile telephony handset configured to communicate using the GSM standard, for sending and receiving voice data coded in a digital format over a packet-switched network, the mobile communication device comprising: - a codec unit, arranged to encode voice data to be sent by a communication unit and arranged to decode voice data received from the communication unit, wherein the 30 encoded voice data to be sent is provided with information on the basis of which errors in voice data can be detected, and wherein the codec unit is configured to, while decoding received voice data, detect errors in the received voice data and at least partially compensate for said errors, 5 - the communication unit, arranged to receive a stream of data packets from the packet-switched network, said stream of data packets comprising valid and invalid data packets and to directly send at least the received valid data packets as voice data to the codec, the communication unit being further arranged to receive voice data from the 5 codec unit and to subsequently send the voice data as a stream of data packets over the packet-switched network.

It is advantageous to integrate the mobile communication device with a known communication device, such as a GSM handset, for a number of reasons. Integration 10 may involve sharing some components and the house containing said components between devices. It is advantageous if the user has to carry around less physical devices. Also, the integration can reduce cost as a number of components (for example, battery, display, keypad, a processing unit, speaker, microphone, codec) can potentially be shared by the devices. The integration also allow the user interfaces to be integrated, 15 so the user has to learn to use only one user interface. Furthermore, value added services, such as voice mail can be integrated as well.

The communication unit may be configured to detect invalid data, for example using a checksum algorithm. When a packet arrives out of order, for example first packet 8 20 arrives, then packet 10, then packet 9, packet 9 may be marked as invalid, even though internally the data may be valid. Packet 10 can also be marked invalid, since it arrived in stead of the expected packet 9. In an embodiment, all received packets, valid and invalid, are sent to the decoder. In a further embodiment, some received invalid packets are discarded by the communication unit, for example all packets that arrive too late. In 25 an embodiment, packets that are detected as invalid by the communication unit may be discarded. It is a known characteristic of packet-switched networks that packets may take different routes to the destination and thus arrive in a different order than in which they were sent. Not waiting for late packages has the advantage that the latency of connection, meaning the time between sending the data and receiving it, is reduced.

It is advantageous if a codec can detect and compensate errors in received voice data.

In this context, compensation can comprise decoding a damaged packet in such a way that the essence of the recorded voice message is kept. It may also comprise 30 6 interpolating missing data by analyzing received voice data before and after the moment for which voice data is missing. It is advantageous to integrate the error correction features for the voice data transfer in the codec rather than in the packet-switched network, since integration in the codec allows to account for the human 5 perception of voice data. In other words, it is possible to judge the sensitivity of bits, meaning the impact that the error will have on the perception of the decoded signal by the human user. In essence, not all bits are equally important (sensitive) and an error correction algorithm in the codec can advantageously utilize this by ignoring errors in less important bits, thus reducing the overhead of asking for a retransmission of new 10 data. In contrast, in a packet-switched network layer, all bits are assumed equally sensitive, and a network layer cannot advantageously use this sensitivity to choose to not request a retransmission of invalid but insensitive data.

In an embodiment of a mobile communication device according the invention, the 15 arrangement to directly send received data to the codec unit involves sending the received data without requesting a retransmission of data associated with an invalid or missing data packet.

This advantageously reduces the network overhead of retransmitting damaged or 20 missing packets. Integration of an error correction feature in the audio codec means that the network no longer needs to supply this feature. As such, more bandwidth of the network can be allocated to the voice data rather than to overhead.

In an embodiment of a mobile communication device according the invention at least 25 the part of the packet-switched network connecting the mobile communication device to the rest of said packet-switched network, or the entire packet-switched network, is a wireless network.

In an embodiment of a mobile communication device according the invention, the 30 mobile communication device is a mobile telephone, such as a GSM telephone or a WCDMA telephone.

A GSM or WCDMA telephone comprises some of the components that are needed for 7 a mobile communication device according the invention, so it is advantageous from a cost perspective to integrate the device with such a telephone (also called a handset). Furthermore, it allows the possibility to present the user with a GSM UI for the functionality of the device. It is also advantageous to reduce the number of devices that 5 the user needs to carry around with him.

In an embodiment of a mobile communication device according the invention, at least the part of the packet-switched network connecting the mobile communication device to the rest of said packet-switched network, is a packet-switched network according to 10 a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), or newer family of standards for mobile telecommunications as defined by the International Telecommunication Union (ITU), the European Telecommunication Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

15

In an embodiment of a mobile communication device according the invention, the voice data is selectively sent and received via the packet-switched network or via a circuit-switched network, wherein the device is configured to select either the packet-switched network or the circuit-switched network depending on local network services 20 availability.

Circuit-switched mobile (for example GSM) networks are widespread throughout the world, but are usually relatively expensive to use. Packet-switched mobile network services are less widespread, but usually less expensive to use as carriers for voice data. 25 It is advantageous to be able to use either service, if available, with a single mobile communication device.

In an embodiment of a mobile communication device according the invention, the device is configured to select either the packet-switched network or the circuit-switched 30 network depending on information stored in a memory of the mobile communication device.

The information stored in the memory may comprise a user preference, such as a 8 preference for a particular type of network. It may also comprise an operator preference, such as a preference for a particular type of network. It may comprise a preference, such as a preference for a particular type of network, from an interested party, such as a representative of the business that pays for the contract of the 5 subscriber using the mobile communication device. The information may comprise rules, for example the rule to select the available network with the lowest cost, or the available network with the best audio quality. The information may comprise data that can be used to evaluate such rules, for example a table indicating cost of different types of connections over different types of networks.

10

In an embodiment of a mobile communication device according the invention, the information comprises user preference information.

In an embodiment of a mobile communication device according the invention, the 15 device is configured to replace the information with updated information, wherein the updated information is wirelessly received from a remote server.

This advantageously allows a party, for example the M(V)NO to update the information dynamically, for example when conditions have changed.

20

In an embodiment of a mobile communication device according the invention, the codec unit is configured to use unequal bit-error detection (UED) and/or unequal biterror protection (UEP) to detect errors and/or to partially compensate for said errors.

25 The UEP/UED mechanisms sort the bits of voice data into perceptually more and less sensitive classes. A frame is only declared damaged and not delivered if there are errors found in the most sensitive bits. If the voice frame is delivered with one or more bit errors in the less sensitive bits (here “sensitive” is based on human audio perception), the voice frame is kept as it was received. The decoded digital audio signal created 30 from such a frame will not be a perfect copy of the encoded digital signal, but should be a “good enough” approximation. As such, a codec unit supporting the UED/UEP mechanisms can often compensate for bad data without requiring a retransmission of the data.

9

In an embodiment of a mobile communication device according the invention, the codec unit is configured to interpolate voice data to replace invalid or missing voice data.

5 A codec supporting interpolation can compensate for one or more missing frames without requiring a retransmission of the data, since the codec can use interpolation to reconstruct a missing frame from the data in received frames. Alternatively, a interpolating codec may interpolate data inside a frame that is partially invalid, using 10 valid data from the same frame and possible data from other frames.

In an embodiment of a mobile communication device according the invention, the codec unit implements an Adaptive Multi-Rate (AMR) codec according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), 15 or newer family of standards for mobile telecommunications as defined by the International Telecommunication Union (ITU), the European Telecommunication Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

20 Advanced Multi-rate (AMR) voice frames contain voice data, typically 20 milliseconds of voice, encoded by an AMR codec. An AMR codec supports unequal bit-error detection and protection (UED/UEP). Furthermore, the AMR codec is able to interpolate voice data in case a whole AMR frame is either not received at all, or received with errors in the most sensitive bits so that the frame is discarded.

25

Another benefit of AMR is adaptive rate adaptation which allows smooth switching between codec modes. Most implementations of AMR codecs have modes for a number of different bitrates, such as 12.2, 10.2, 7.95, 7.4, 6.7, 5.9, 5.5 and 4.75 kbit/s. In addition, a wideband (WB) mode AMR WB at 12.65 kbit/s is available. This gives 30 the designer of the mobile communication device unit freedom to implement fallback scenarios in case of for example a persistently bad connection.

In an embodiment of a mobile communication device according the invention, the 10 communication unit is arranged to send and receive voice data according to the User Datagram Protocol (UDP) or to a similar protocol.

The User Datagram Protocol advantageously has a low overhead, and does not 5 automatically request a retransmission of invalid received data or missing data. It is also an internet standard, so that the voice data of the mobile communication device may be routed via the internet.

In an embodiment of a mobile communication device according the invention, the 10 communication unit is arranged to send and receive control data via a control protocol, wherein the control protocol is an error-correcting protocol.

In order to set up connections and to provide other services, the device needs to exchange control messages with a central part, such as a core network. The bandwidth 15 used for these messages is typically much less than the bandwidth used for voice data. It is therefore advantageous to use an error-correcting protocol, such as a protocol that checks for invalid or missing data and request a retransmission of said data, since it is important that the control messages arrive intact and the overhead of the error correction is of less importance, since the total bandwidth used is relatively small, 20 compared to the bandwidth allocated to voice data.

In an embodiment of a mobile communication device according the invention, the control protocol is the Transmission Control Protocol (TCP) or a similar protocol.

25 TCP is a standard internet protocol, thus making it possible to route the control messages via the internet.

A further object of the invention is met by providing a mobile communication system, comprising: 30 - at least a mobile communication device according to any of the previous claims, - a core network device, the core network device comprising a gateway unit to be connected or connected to a Public Switched Telephone Network (PSTN), wherein the mobile communication device is arranged to exchange voice data with the 11

Public Switched Telephone Network (PSTN) via the gateway unit.

By allowing access to the PSTN via a gateway unit, devices on the PSTN, such as devices connected to a landline or mobile communication devices using GSM or other 5 mobile standards, may be contacted using the mobile communication device. Hereafter, the core network device will also be called core network.

In an embodiment of a mobile communication system according the invention, the gateway is a Gateway Mobile Switching Center (GMSC) or equivalent arranged to act 10 as a gateway by exchanging voice data with the mobile communication device via the packet-switched network and by exchanging voice data with the Public Switched Telephone Network (PSTN).

A Gateway Mobile Switching Center (GMSC) is a standard gateway for GSM 15 networks. As such, the enhanced functionality of a mobile communication device according to the invention can be used to connect to an other device on a hosted network of the GMSC or to an other device on the PSTN in a manner that is similar to the way a GSM device on the hosted network of the GMSC can connect to these other devices.

20

In an embodiment of a mobile communication system according the invention, the gateway is connected to a Home Location Register (HLR) or equivalent, of an existing telephone network, in particular of an existing GSM network, wherein the HLR is arranged to store information relating to redirecting incoming connections to the mobile 25 communication device via the packet-switched network.

A mobile communication device will send data about how it may be reached to the HLR of the operator where the user of the mobile communication device has a service contract, hereafter called subscription. In case of a GSM device, this data comprises 30 details of the GSM network the device is currently on. In case of a packet-switched device, the details may comprise the IP address and port number of a mobile communication device according the invention. Once this data is stored in the HLR, incoming connections may advantageously be redirected to the communication device.

12

In an embodiment of a mobile communication system according the invention, the gateway is configured to redirect incoming connections for the mobile subscriber to the mobile communication device via the packet-switched network or via an existing 5 telephone network, in particular via an existing GSM network, depending on data stored in the HLR.

Advantageously, the party attempting to contact the user/subscriber does not know whether the connection is made using for example a GSM network or a packet-10 switched network. Thus the user can be reached in multiple ways using a single contact identifier such as a telephone number.

The invention further provides a method for receiving voice data from a packet-switched network and rendering it on an audio output unit of a mobile communication 15 device, the method comprising - receiving a stream of data packets from the packet-switch network, said stream of data packets comprising valid and invalid data packets, - directly converting at least the received valid data packets into voice data to be decoded 20 - decoding said voice data, thereby detecting errors in said voice data and at least partially compensating for said errors, - converting the decoded voice data into an analog voice signal, and rendering the analog voice signal on the audio output unit.

25 The invention further provides a method for receiving voice data from an audio input unit of a mobile communication device and sending it over a packet switched network, the method comprising: - receiving an analog voice data signal from an audio input unit, - converting said analog voice data signal into digital voice data, 30 - encoding said digital voice data as voice data, thereby providing the voice data with information on the basis of which errors in the voice data can be detected, - sending the voice data as a stream of data packets via the packet-switched network, wherein each data packet is sent only once.

13

According to an embodiment of the invention, the voice data is selectively sent as a stream of data packets via the packet-switched network, or as voice data via a circuit-switched network, wherein the selection is based on information stored on a memory of 5 the communication device.

The invention provides a method for setting up a call from a calling party to a mobile communication device, the method comprising: - receiving a call setup request from the calling party for the mobile communication, 10 - querying an information storage, in particular a database of a Home Location

Register, wherein information relating to redirecting incoming connections to the mobile communication device via the packet-switched network is stored, for said information, - receiving the information from the information storage, 15 - routing the call setup request to the mobile communication device via the packet- switched network based on the information received from the information storage, - receiving a call setup acknowledgement from the mobile communication device via the packet-switched network, - establishing the call, 20 - receiving a stream of data packets, said data packets comprising voice data encoded by the mobile communication device, wherein said voice data has been provided with information on the basis of which errors in the voice data can be detected, and said received stream of data packets comprising valid and invalid data packets, - directly converting at least the received valid data packets into voice data to be 25 decoded, - decoding said voice data, thereby detecting errors in said voice data and at least partially compensating for said errors, - sending the decoded voice data to the calling party.

30 According to an embodiment of the invention, the method further comprises: - encoding voice data received from the calling party, thereby providing the voice data with information on the basis of which errors in the voice data can be detected, - sending the voice data as a stream of data packets to the mobile communication 14 device via the packet-switched network.

The invention provides a method for setting up a call from a calling party to a mobile communication device selectively according to the method mentioned above, or 5 according to a method for setting up a call from a calling party to a mobile communication device over a circuit-switched network, such as a GSM or WCDMA network, where the selection is based on information stored in an information storage, in particular a database of a Home Location Register.

10 The invention provides a method for setting up a call from a mobile communication device to a called party, the method comprising: - receiving a call setup request from the mobile communication device for a called party, - routing the call setup request to the called party, 15 - receiving a call setup acknowledgement from the called party - establishing the call, - receiving a stream of data packets, said data packets comprising voice data encoded by the mobile communication device, wherein said voice data has been provided with information on the basis of which errors in the voice data can be detected, 20 and said received stream of data packets comprising valid and invalid data packets, - directly converting at least the received valid data packets into voice data to be decoded, - decoding said voice data, thereby detecting errors in said voice data and at least partially compensating for said errors, 25 - sending the decoded voice data to the calling party.

According to an embodiment of the invention, the method further comprises: - encoding voice data received from the calling party, thereby providing the voice data with information on the basis of which errors in the voice data can be detected, 30 - sending the voice data as a stream of data packets to the mobile communication device via the packet-switched network.

According to an embodiment of the invention, a method further comprises: 15 the mobile communication device is a mobile mentioned in the method is a communication device as described in the preceding text.

According to an embodiment of the invention, directly converting at least the received 5 valid data packets into voice data to be decoded involves converting said data packets without requesting a retransmission of data associated with an invalid or missing data packet.

According to an embodiment of the invention, the packet-switched network comprises 10 a wireless network.

According to an embodiment of the invention, the packet-switched network comprises a packet-switched network according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), or newer family of standards for 15 mobile telecommunications as defined by the International Telecommunication Union (ITU), the European Telecommunication Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

According to an embodiment of the invention, decoding or coding comprises the use of 20 unequal bit-error detection (UED) and/or unequal bit-error protection (UEP) to detect or be able to detect errors and/or to partially compensate or be able to partially compensate for said errors.

According to an embodiment of the invention, decoding comprises interpolating voice 25 data to replace invalid or missing voice data.

According to an embodiment of the invention, decoding or coding is performed using an Adaptive Multi-Rate (AMR) codec according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), or newer family of 30 standards for mobile telecommunications as defined by the International

Telecommunication Union (ITU), the European Telecommunication Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

16

According to an embodiment of the invention, call setup request and call setup acknowledgments are exchanged using a control protocol which is an error-correcting protocol.

5 According to an embodiment of the invention, the control protocol is the Transmission Control Protocol (TCP) or a similar protocol.

BRIEF DESCRIPTION OF THE FIGURES

10 Figure 1 schematically shows a mobile communication device 21 and core network 10.

Figure 2 schematically shows an overview of a mobile communication device according to an embodiment of the invention.

15 Figure 3 schematically shows an overview of a method for receiving voice data from a packet-switched network and rendering it on an audio output unit of a mobile communication device.

Figure 4 schematically shows an overview of a method for recording voice data using 20 an audio input unit of a mobile communication device and sending it over a packet switched network.

Figure 5 schematically shows an overview of a method for setting up a call from a calling party to a mobile communication device.

25

Figure 6 schematically shows an overview of a method for setting up a call from a mobile communication device to a called party.

DETAILED DESCRIPTION 30

Figure 1 schematically shows a mobile communication device 21 and core network 10. Said core network 10 comprises a gateway mobile switching center (GMSC) 15, Session Initiation Protocol (SIP) to Time-Division Multiplexing (TDM) converter unit 17 11, Home Location Register (HLR) unit 14, Billing unit 16, Provisioning Unit 17,

Other Services unit 18, UDP unit 12, and TCP unit 13. The TCP unit 13 can be connected to a memory storage 22. The UDP unit 12 comprises a codec unit 23, which is compatible with the codec unit 35 of the mobile communication device 21. A core 5 network 10 is typically operated by a mobile network operator (MNO) or an mobile virtual network operator (MVNO).

The GMSC 15 can receive incoming and initiate outgoing connections with a device connected to or at least reachable via the PSTN 19. The GMSC is connected to an own 10 hosted radio network 20, for example comprising a system of base transceiver stations (BTS), base station controllers (BSC) (not shown).

Inside the core network 10, the GMSC 15 is connected to the HLR 14, and to the Billing unit 16, the Provisioning unit 17, and the Other Services unit 18. The HLR 14 15 comprises a database that contains information about the MNO’s subscribers, and the current known way of reaching a mobile communication device 21 of a subscriber. For example, the HLR 14 may store information indicating which location area of the own hosted radio network 20 is currently in contact with the mobile communication device 21 of a subscriber.

20

The Billing unit 16 is used to store information related to subscriber billing. The Other Services unit 18 provides miscellaneous services, such as for example Short Messaging Service (SMS).

25 The core network 10 can receive data from and send data to a packet-switched network, for example the internet or an intranet, via UDP unit 12, accepting the UDP protocol, and via TCP unit 13, accepting the TCP protocol. Data received at the UDP unit 12 or TCP unit 13 can be forwarded to a SIP/TDM converter 11, which can forward converted data to the GMSC. The SIP/TDM converter 11 can translate voice data 30 coded according to a packet-switched voice communication standard, such as UDP or VoIP, to a Time-Division Multiplexing standard that is typically used in a GMSC 15. The TCP unit 13 can also forward received data to the HLR 14. Vice versa can the SIP/TDM converter 11 receive TDM data from the GMSC 15 and send this to the UDP

18 unit 12 and/or the TCP unit 13 as VoIP, SIP, RTP, or according another packet-switched protocol. The UDP unit 12 comprises a codec unit 23, which can be used to encode or transcode voice data received from the SIP/TDM converter 11 to be sent over the packet-switched network. The codec unit 23 can also be used to decode or 5 transcode voice data received from the packet-switched network into a format, such as a standard VoIP format such as PCM, suitable for the SIP/TDM converter 11. With transcoding is meant converting from one data format to another data format. This may involve first decoding from the one data format to an uncompressed, so-called raw data, and then encoding said raw data to the other data format. The TCP unit 23 can read 10 data from a memory storage 22 and transmit said data over the packet-switched network. The reverse is also possible.

In operation, the system works as follows. First the user of mobile communication device 21 making an outgoing connection to a user on the hosted radio network will be 15 discussed, then the case of a connection coming in from the PSTN for the user of the mobile communication device 21. The mobile communication device 21 has an identification means (not shown) such as a SIM card, which makes the mobile communication device a subscribed device and the user of the device 21a known subscriber of the M(V)NO operating the core network 10. Unless otherwise stated, the 20 terms “user”, “subscriber”, and “user of the device” are used interchangeably.

In the example, the mobile communication device 21 has access to the internet via a GPRS service. Generally, any packet-switched connection to the internet with sufficient available bandwidth can be used, such as EDGE, UMTS, etc. The user of the 25 mobile communication device 21 wants to make a phone call, and decides (or the device automatically decides for the user) that a connection via a packet-switched rather than the default GSM circuit-switched network is preferable. A motive can be the lower cost of a packet-switched data connection.

30 The mobile communication device 21 is arranged to exchange packets with the UDP unit 12 and TCP unit 13 of the core network 10. To be able to make a connection, the 32-bit IP numbers of said units 12, 13 are stored in a memory of the mobile communication device 21 or otherwise accessible to it. Typically the mobile 19 communication device 21 will be preconfigured by the M(V)NO that operates the core network 10. Alternatively, this information can be dynamically sent to the mobile communication device 21 “in the field”, using any of the mobile telecommunications device software update mechanisms available to a skilled person.

5

To initiate the connection, the mobile communication device 21 sends TCP packets to the TCP unit 13 containing data indicating an outgoing connection with an other device (not shown) having a certain telephone number is desired. This information is routed by the TCP unit 13 to the relevant unit inside the core network, and treated essentially in 10 the same way as a similar request coming in from a mobile communication device on the own hosted radio network 20 would be handled.

The connection is set up, comprising two parts. The first part is between the mobile communication device 21 and the GMSC 15, and the second part is between the GMSC 15 15 and the other device. In this example, the user is on the hosted radio network 20, so the second part of the connection involves GMSC 15 and hosted radio network 20, and voice data is transferred using a standard TDM switched-circuit network. The case where the other device is a device somehow connected to the PSTN is very similar from the point of view of the GMSC 15. The first part of the connection involves 20 GMSC 15, SIP/TDM converter 11, UDP unit 12, and mobile communication device 21. Voice data is sent to and from the mobile communication device using a packet-switched network, using the UDP protocol. The data passes through the UDP server, which exchanges the data with the SIP/TDM converter 11, thereby decoding or transcoding it using the codec unit 23. The SIP/TDM converter 11 makes the 25 translation between the “VoIP world” of the UDP unit 12 and the “GSM world” of the GMSC 15.

The voice data between the mobile communication device 21 and the UDP unit is preferably sent in a manner utilizing substantially all available bandwidth for data 30 communication, and as little bandwidth as possible for overhead such as error- correction related information. The rationale is that the bandwidth between the mobile communication device 21 and the nearest radio tower of the mobile packet-switched wireless network may be very limited, for example if the GPRS service is used. The 20

User Datagram Protocol (UDP) is an advantageous protocol, since it has very limited overhead. The lack of error-correction related information at the level of the communication protocol is at least partially compensated for by the error-correcting features of the codec used to generate the voice packets. This codec will be discussed 5 further in reference to figure 2.

Once the voice data signal has arrived at the core network 10 via the UDP unit 12, bandwidth has become relatively inexpensive and subsequently there is no longer a need to reduce overhead. It is then advantageous to use a standard VoIP protocol, such 10 as RTP, to transport the voice data inside the core network 10, since this has the twin benefits of providing error-correcting features and allowing inter-operability with standardized VoIP components. Also, a standard VoIP voice data format, such as PCM, may be used. The codec unit 23 of the UDP unit 12 may be arranged to decode the received voice data as PCM voice data, and to encode PCM voice data into coded voice 15 data suitable for the codec unit 35 of the mobile communication device 21.

Now the example of an other device on the PSTN connecting to the mobile communication device 21 will be explained. As before, the user or mobile communication device 21 itself has decided that the preferred network for voice 20 conversations is currently the packet-switched network. Via a control protocol, in this case TCP, this information is sent to the TCP unit 13, which supplies it to the HLR.

The HLR stores the preference, along with related information such as an IP address and port number of the mobile communication device 21 to connect to in order to establish a voice connection with the mobile communication device 21 over a packet-25 switched network such as the internet.

When the user of the other device on the PSTN 19 dials the PSTN number of the subscriber of the mobile communication device, the operator responsible for the other device on the PSTN will attempt to initiate a connection. At some point, the HLR 14 of 30 the core network 10, being the HLR of the M(V)NO responsible for the mobile

communication device’s 21 subscription, will be contacted with a request about how mobile communication device 21 can currently be reached. The HLR 14, together with the GMSC 15 will arrange that the GMSC 15 will act as a gateway between the PSTN

21 19 and the mobile communication device. Internally, the voice data received from the PSTN 19 will be converted to SIP/RTP data by the converter 11 and forwarded to the mobile communication device 21 by UDP unit 12, after it has been coded into an appropriate form by the codec unit 23. Reversely, data received from the mobile 5 communication device 21 will be received by the UDP unit 12, decoded into a suitable format for use inside the core network 10 by the codec unit 23, converted to TDM data by the converter 11, and forwarded to the PSTN by the GMSC 15.

Figure 2 schematically shows an overview of a mobile communication device 21 10 according to an embodiment of the invention. The device 21 comprises a microphone 31, which can convert an audible acoustic signal (a sound) into an electrical signal, which it delivers to the analog-to-digital converter (ADC) 33, which in turn creates a digitized voice data signal. It further comprises a digital-to-analog converter (DAC) 34 which can receive a digitized voice data signal and can covert it into an analog 15 electrical signal. The DAC is connected to a speaker or headphone device 32 which can receive an analog electrical signal and transform it into an audible acoustic signal (a sound). Both the ADC 33 and the DAC 34 are connected to a codec unit, in this example an Adaptive Multi-Rate (AMR) codec unit 35. The AMR codec 35 can receive a digitized voice signal from the ADC 33 and convert it into coded voice data, or data 20 coded in a digital format, that is sent to the connected processing unit 36. Alternatively, it can receive coded voice data, or data coded in a digital format, from the processing unit 36 and decode the voice data into a digitized voice signal to be delivered to the DAC 34. The processing unit 36 is connected to a memory 39. The processing unit 36 is also connected to a GSM communication unit 37 which is capable of sending and 25 receiving voice data according the standard GSM protocol and to a TCP/UDP

communication unit 38 which is capable of sending standards-compliant TCP and UDP packets over a packet-switched wireless network such as for example GPRS, EDGE, UMTS, or WiFi (including any of the IEEE 802.11 based standards).

30 When the user or subscriber of the device wishes to make a telephone call, the user uses the user interface (UI) of the device to enter a identifier of the party to be contacted, for example a telephone number. Preferably, this UI is familiar to the user, for example by being similar to a typical GSM UI. The processing unit will process the information 22 entered via the UI. It will also gather information on the current network availability, such as the availability of a standard GSM signal (the GSM communication unit 37 will provide this information) and the availability of one or more packet-switched networks (the TCP/UDP unit 38 will provide this information). The processing unit will also read 5 information stored in the memory 39. This information can comprise a user or operator preference, for example the preference to make a connection using the network that has the lowest cost, along with cost indications of various networks. Other criteria can be the network that provides the best audio quality, or the least chance of interception by a third party. The processing unit will combine this information with the network 10 availability information, and then select a network for making a connection. For example, if both GSM and GPRS are available and the user wants to make an international telephone call, the processing unit may determine that the GPRS packet-switched network has the lowest cost for this call, and thus select it. In case of a packet-switched network, the connection will be set up as explained in reference to figure 1. In 15 case the GSM network is selected, the call will be set up as it is normally done for GSM networks. Knowledge on how to arrange this is available to a skilled person.

When a call has been set up, the mobile communication device 21 functions as follows. The voice of the user is recorded as an electrical signal by the microphone 31, and the 20 electrical signal is converted into digitized signal by the ADC 33, for example a 44.1 kHz digitized audio signal. This signal is provided to the AMR codec, which codes it into a stream of coded voice data frames, each frame for example representing 20 ms of voice data. The UEP/UED mechanisms of the AMR codec sort the bits of voice data in the frame into perceptually more and less sensitive classes.

25

If the standard GSM service is used, the AMR codec generated voice frames are delivered to the GSM communication unit 37, and send over the GSM network according to the GSM standard. If, on the other hand, for example the packet-switched GPRS service is used, the voice frames are delivered to the TCP/UDP communication 30 unit. Here every voice frame is “packaged” as a UDP packet. Although many ways are possible in which a stream of AMR codec packets can be sent as UDP packets, it is advantageous to package each entire frame as one UDP packet. The packets are sent over the packet-switched network to the UDP communication unit 12 of the core 23 network 10. The UDP packets are initially transported over the GPRS network. After that, they may be sent over other wireless radio networks or via landlines that are part of the internet or a corporate intranet, before they are received, if all went well, at the UDP communication unit 12 of the core network 10.

5

Voice data sent by the other party of the conversation is received at the gateway 15 of the core network 10 and transmitted by the UDP communication unit 12 of the core network 10 as UDP packets to the mobile communication device 21 over the packet-switched network, as was explained in reference to figure 1. Again, it is advantageous 10 to package one AMR codec frame in one UDP packet. After traversing a part of the internet, an intranet, and/or a GPRS network connection, the UDP packets are received by the TCP/UDP communication unit 38. The packets are stripped of the UDP “container” and directly sent as AMR frames to the AMR codec 35 via the processing unit 36. The TCP/UDP communication device may elect to discard UDP packets that 15 are obviously damaged. It may also send damaged packets to the codec anyway. It will not ask the UDP unit 12 of the core network 10 to re-send invalid or missing packets. The AMR codec 35 uses the UEP/UED mechanisms to check the bits for errors. If no errors are found, or only errors in less sensitive bits, the data is decoded using the AMR decoding algorithm. If errors are found in the most sensitive bits, the frame is 20 discarded. If the frame is discarded, or the AMR codec detects that a frame is missing, the frame will be interpolated. Thus, the AMR codec can compensate for errors in the transmission of the voice data over the packet-switched network.

The decoded digitized voice data is converted to an analog electrical signal by the DAC 25 34. It is subsequently made audible on an audio output device such as a speaker or headphone 32.

The information on user and operator preferences, including the rules to be applied in order to determine the optimal network for connections, may be dynamically updated. 30 This may be implemented as follows: the M(V)NO stores the updated information or data in the memory 22 connected to the TCP unit 13 of the core network 10. The TCP unit 13 reads the information from the memory, contacts the TCP/UDP communication unit of the mobile communication device 21 and transmits the updated information 24 along with an authentication signal. The TCP/UDP device sends the received information and authentication signal to the processing unit, which will check the authentication signal to make sure the source of the information has the right to update information on the device. A skilled person has access to standard authentication 5 schemes that enable this. If authenticated, the processing unit 36 will store the updated information in the memory 39.

It is a known strength of packet-switched networks, compared to a circuit-switched network, that they can better operate under severe stress, albeit at a reduced quality of 10 service. The TCP/UDP communication unit can be configured to detect statistical data on frequency of invalidly received or missing UDP packets. Alternatively, the AMR codec can collect statistical data on the frequency of invalid or damaged frames. This statistical data may be used by the processing unit to select an AMR codec mode. For example, if invalid packets are frequent, the AMR codec may be switched to a mode 15 using a lower bitrate. The advantage of this dynamic mechanism is that in case of bad network conditions, for example due to massive use of the network leading to a bandwidth shortage, a reduced quality of service may be maintained rather than no service at all.

20 Figure 3 schematically shows an overview of a method 100 for receiving voice data from a packet-switched network and rendering it on an audio output unit of a mobile communication device.

The method comprises steps 101, 102, 103, and 104. The first step 101 is receiving a 25 stream of data packets from the packet-switch network, said stream of data packets comprising valid and invalid data packets. Step 102 involves directly converting at least the received valid data packets into voice data to be decoded. After that, the third step 103 involves decoding said voice data, thereby detecting errors in said voice data and at least partially compensating for said errors. The fourth step 104 is converting the 30 decoded voice data into an analog voice signal, and rendering the analog voice signal on the audio output unit. This audio output unit can be a speaker or headphone of a mobile communication device.

25

Figure 4 schematically shows an overview of a method 110 for recording voice data using an audio input unit of a mobile communication device and sending it over a packet switched network, the method comprising steps 111, 112, 113, and 114.

5 The first step 111 involves recording an analog voice data signal using an audio input unit, for example the microphone of a mobile communications device. Step 112 involves converting said analog voice data signal into digital voice data, for example using an ADC. Step 113 is to encode said digital voice data as voice data, thereby providing the voice data with information on the basis of which errors in the voice data 10 can be detected. This can for example be achieved with a codec, in particular an AMR codec. Step 114 involves sending the voice data as a stream of data packets via the packet-switched network, wherein each data packet is sent only once. With sending each data packet only once is meant that no packets are retransmitted in case the original packets do not arrive intact at the destination. It is characteristic of error-15 correcting network protocols that they have mechanisms for triggering retransmissions of data that is invalid or missing at the destination. These mechanisms cost additional bandwidth of the network. The method described above does not require these mechanisms, and can thus avoid the additional bandwidth.

20 Figure 5 schematically shows an overview of a method 120 for setting up a call from a calling party to a mobile communication device. The method comprises steps 121, 122, 123, 124, 125, 126, 127, 128, 129, and 130.

The steps 121-123 are: 25 - receiving a call setup request from the calling party for the mobile communication 121, - querying an information storage, in particular a database of a Home Location Register, wherein information relating to redirecting incoming connections to the mobile communication device via the packet-switched network is stored, for said 30 information 122, - receiving the information from the information storage 123,

At this point, the receiver of the information knows how to connect to the mobile communication device over the packet-switched network. This information may 26 comprise an Internet Protocol (IP) address and a port number. Usually, the user of the mobile communication device will be a subscriber of the M(V)NO that operates the Home Location Register. Having a valid subscription (and proving this using for example a SIM-card based device in the mobile communication device) will typically 5 be a precondition for registering the information in the Home Location Register.

The steps 124-126 are: - routing the call setup request to the mobile communication device via the packet-switched network based on the information received from the information storage 124, 10 - receiving a call setup acknowledgement from the mobile communication device via the packet-switched network 125, - establishing the call 126,

At this point, a call or telephone conversation, is established and parties can begin exchanging voice data. Advantageously, the setup request and acknowledgment 15 messages are exchanged using an error-correcting protocol, for example TCP.

The steps 127-130 are: - receiving a stream of data packets, said data packets comprising voice data encoded by the mobile communication device, wherein said voice data has been provided with 20 information on the basis of which errors in the voice data can be detected, and said received stream of data packets comprising valid and invalid data packets 127, - directly converting at least the received valid data packets into voice data to be decoded 128, - decoding said voice data, thereby detecting errors in said voice data and at least 25 partially compensating for said errors 129, - sending the decoded voice data to the calling party 130.

As such the voice data sent by the mobile communication unit is forwarded to the calling party, in a format suitable for the network that the calling party is on.

30 Figure 6 schematically shows an overview of a method 140 for setting up a call from a mobile communication device to a called party. The method 140 comprises steps 141, 142, 143, 144, 145, 146, 147, and 148.

27

Steps 141-143 are: - receiving a call setup request from the mobile communication device for a called party 141, - routing the call setup request to the called party 142, 5 - receiving a call setup acknowledgement from the called party 143, - establishing the call 144,

At this point a call or telephone conversation is established. Again, a typical condition for an operator to apply the steps of this method will be that the user of the mobile communication device or the device itself is a subscriber of services offered by the 10 operator.

Steps 144-148 are: - receiving a stream of data packets, said data packets comprising voice data encoded by the mobile communication device, wherein said voice data has been provided with 15 information on the basis of which errors in the voice data can be detected, and said received stream of data packets comprising valid and invalid data packets 144, - directly converting at least the received valid data packets into voice data to be decoded 145, - decoding said voice data, thereby detecting errors in said voice data and at least 20 partially compensating for said errors 146, - sending the decoded voice data to the calling party 148.

As such, voice data is sent from the mobile communication device to the calling party.

While preferred embodiments of this invention have been shown and described, 25 modifications thereof can be made by one skilled in the art without departing from the spirit or teaching of this invention. The embodiments described herein are exemplary only and are not limiting. Many variations and modifications of the system and apparatus are possible and are within the scope of the invention. Accordingly, the scope of protection is not limited to the embodiments described herein, but is only limited by 30 the claims which follow, the scope of which shall include all equivalents of the subject matter of the claims.

Claims

A mobile communication device, in particular a device integrated with a mobile telephone, more particularly one with a mobile telephone 5 which is adapted to communicate with the aid of the GSM standard integrated device, for sending and receiving data in a digital format coded voice data over a packet-switched network (packet-switched network), the mobile communication device comprising: - a coding unit, arranged to encode voice data to be sent by a communication unit and arranged to decode voice data received from the communication unit wherein the coded voting data to be transmitted are provided with information on the basis of which errors in voting data can be detected, and wherein the coding unit is adapted to detect errors in the received voting data and at least partially compensate for errors in the received voting data, - the communication unit to receive a stream of data packets from the packet-switched network, the stream of data packets comprising valid and invalid data packets and to send at least the received valid data packets directly as voice data to the codec, the communication unit being further adapted to receive voice data from receive the coding unit to then send the voice data as a stream of data packets over the packet-switched network.

2. Mobile communication device as claimed in claim 1, wherein the devices are adapted to send received data directly to the coding unit comprising sending the received data without requesting a re-sending of data associated with an invalid or missing data package.

3. Mobile communication device as claimed in any of the foregoing claims, wherein at least the part of the packet-switched network that connects the communication device to the rest of the packet-switched network, or the entire packet-switched network, is a wireless network.

The mobile communication device according to any of the preceding claims, wherein the mobile communication device is a mobile telephone, such as a GSM telephone or a WCDMA telephone.

A mobile communication device according to any one of the preceding claims, wherein at least the part of the packet-switched network that connects the mobile communication device to the rest of the packet-switched network is a packet-switched network according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), or newer family of mobile telecommunication standards as defined by the International Telecommunications Union (ITU), the European Telecommunications Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

A mobile communication device according to any one of the preceding claims, wherein the voice data is selectively transmitted and received via the packet-switched network or via a circuit-switched network (circuit-switched network), the device being adapted to either the packet-switched network or select the circuit-switched network depending on the local availability of 20 network services.

The mobile communication device of claim 6, wherein the device is adapted to select either the packet-switched network or the circuit-switched network depending on information stored in a memory of the mobile communication device.

The mobile communication device of claim 7, wherein the information comprises user preference information.

The mobile communication device according to claim 7 or 8, wherein the device is adapted to replace the information with updated information, the updated information being received wirelessly from a remote server.

10. Mobile communication device as claimed in any of the foregoing claims, wherein the coding unit is adapted to use unequal bit-error detection (UED) and / or unequal bit-error protection (UEP) for detecting errors and / or partially compensating of the errors.

A mobile communication device according to any of the preceding claims, wherein the coding unit is arranged to interpolate voice data to replace invalid or missing voice data. 10

A mobile communication device according to any one of the preceding claims, wherein the coding unit is an Adaptive Multi-Rate (AMR) codec according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), or newer family of mobile telecommunication standards as defined by the International Telecommunications Union (ITU), the European Telecommunications Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

13. A mobile communication device according to any one of the preceding claims, wherein the communication unit is adapted to send and receive voice data according to the User Datagram Protocol (UDP) or a similar protocol.

14. A mobile communication device according to any one of the preceding claims, wherein the communication unit is adapted to send and receive control data via a control protocol, wherein the control protocol is an error-correcting protocol.

The mobile communication device of claim 14, wherein the control protocol is the Transmission Control Protocol (TCP) or a similar protocol.

16. Mobile communication system, comprising: - at least one mobile communication device according to one of the preceding claims, - a core network device, the core network device comprising an access point unit connected or to be connected to a Public Switched Telephone Network (PSTN), the mobile communication device being arranged to exchange voice data with the Public Switched Telephone Network (PSTN) via the gateway unit.

The mobile communication system of claim 16, wherein the access port is a Gateway Mobile Switching Center (GMSC) or equivalent, which is arranged to serve as an access port by exchanging voice data with the mobile communication device via the packet-switched network and by voting data with exchange the Public Switched Telephone Network (PSTN). 15

A mobile communication system according to claim 16 or 17, wherein the access point is connected to a Home Location Register (HLR) or equivalent, of an existing telephony network, in particular of an existing GSM network, the HLR being arranged for information relating to store the referral of incoming connections to the mobile communication device via the packet-switched network.

19. Mobile communication system according to any of the claims 16-18, wherein the access port is adapted to refer incoming connections to the mobile communication device via the packet-switched network or via an existing telephone network, in particular an existing GSM network, depending on of data stored in the HLR.

20. Method for receiving voice data from a packet-switched network and generating it on a sound output device of a mobile communication device, the method comprising: - receiving a stream of data packets from the packet-switched network, the stream of data packets being valid and invalid data packets, - directly converting at least the received valid data packets into voice data to be decoded, - decoding the voice data, together with detecting errors in the voice data and at least partially compensating for the errors, 5. converting the decoded voice data in an analog voice signal, and generating the analog voice signal on the sound output device.

21. Method for receiving voice data from a sound input device of a mobile communication device and sending it over a packet-switched network, the method comprising: - receiving an analog voice signal from a sound input device, - converting the analog voice signal to digital voting data, - encoding the digital voting data as voting data, together with providing the voting data with information on the basis of which errors in the voting data can be detected, - sending the voting data as a stream of data packets via the packet-switched network , with each data package being sent only once.

22. Method according to claim 21, wherein the voice data is selectively transmitted as a stream of data packets via the packet-switched network or as voice data via a circuit-switched network, the selection being based on information stored in a memory of the mobile communication device.

23. Method for initiating a call from a calling party to a mobile communication device, the method comprising: - receiving a call start request from the calling party for the mobile communication device, - interrogating an information store, in particular a database of a Home Location Register, in which information relating to the forwarding of incoming connections to the mobile communication device via the packet-switched network is stored, for the information, - receiving the information from the information store, - routing the call start request to the mobile communication device via the packet-switched network based on the information received from the information store, - receiving a call start acknowledgment from the mobile communication device via the packet-switched network, - starting the call, - receiving a stream data packages, where the data packets comprise voice data encoded by the mobile communication device, the voice data being provided with information on the basis of which errors in the voice data can be detected, and the received stream of data packets comprises valid and invalid data packets, - directly converting at least the received valid data packets to voice data to be decoded, - decoding the voice data, together with detecting errors in the voice data and at least partially compensating for the errors, - sending the decoded voice data to the calling party.

24. Method as claimed in claim 23, comprising: - coding voice data received from the calling party, together with providing the voice data with information on the basis of which errors in the voice data can be detected, - sending the voice data as a stream data packets to the mobile communication device via the packet-switched network.

A method for selectively initiating a call from a calling party to a mobile communication device according to the method of claim 23 or according to a method for initiating a call from a calling party to a mobile communication device over a circuit-switched network, such as a GSM or a WCDMA network, the selection being based on information stored in an information store, in particular a database of a Home Location Register.

A method of initiating a call from a mobile communication device to a called party, the method comprising: - receiving a call start request from the mobile communication device for a called party, - routing the call start request to the called party, - receiving a call start acknowledgment from the called party, 5. initiating the call, - receiving a stream of data packets, the data packets comprising voice data encoded by the mobile communication device, and wherein the voice data is provided with information on the basis of which errors in the voice data can be detected, and the received stream of data packets comprises valid and invalid data packets, - directly converting at least the received valid data packets into voice data to be decoded, - decoding the voice data, together with detecting errors in the voice data and at least e partially compensating for the errors; 15. sending the decoded voice data to the called party.

27. Method as claimed in claim 26, comprising: - coding voice data from the calling party, together with providing the voice data with information on the basis of which errors in the voice data can be detected, - sending the voice data as a stream of data packets to the mobile communication device via the packet-switched network.

28. Method as claimed in any of the foregoing claims 20-27, wherein the mobile communication device is a mobile communication device according to any of claims 1-15.

A method according to any of claims 20 or 23-28, wherein directly converting at least the valid data packets to voice data to be decoded comprises converting the voice data without requesting a re-sending of data associated with an invalid or missing data packet.

The method of any one of claims 20 to 29, wherein the packet-switched network comprises a wireless network.

31. A method according to any one of claims 20-30, wherein the packet-switched network comprises a packet-switched network according to a first generation (1G), second generation (2G), third generation (3G), fourth generation (4G) , or newer family of mobile telecommunication standards as defined by the International Telecommunications Union (ITU), the European Telecommunications Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant body.

The method according to any of claims 20-31, wherein the decoding or coding comprises the use of unequal bit-error detection (UED) and / or unequal bit-error protection (UEP) for detecting or being able to detect errors and / or or partially compensating or being able to partially compensate for the errors.

The method of any of claims 20 or 23-32, wherein decoding comprises interpolating voice data to replace invalid or missing voice data. 20

The method of any one of claims 20-33, wherein the decoding or coding is performed using an Adaptive Multi-Rate (AMR) codec according to a first generation (1G), second generation (2G), third generation (3G ), fourth generation (4G), or newer family of mobile telecommunication standards as defined by the International Telecommunications Union (ITU), the European Telecommunications Standards Institute (ETSI), the Third Generation Partnership Project (3GPP) or any other relevant organ.

The method of any one of claims 23-26, wherein the call start request and the call start confirmation are exchanged using a check protocol that is an error correcting protocol.

The method of claim 35, wherein the control protocol is the Transmission Control Protocol (TCP) or a similar protocol.