DE202014007214U1 - Device for bundling multiple Internet access media with forward error correction - Google Patents

Abstract

Device (5), in particular router, for bundling a plurality of Internet access lines (4) to a virtual Internet access line for providing the sum of the bandwidths of the plurality of Internet access lines (4) for data transmission over the virtual Internet access line, wherein the device (5) comprises a data packet to be transmitted a plurality of data packets for separate transmission over the plurality of Internet access lines (4), characterized in that the device is configured to calculate and transmit redundant information from which lost data packets are recoverable, so that packet losses on an Internet access line (4) not to packet loss on the bundled virtual lead.

Description

FIELD OF THE INVENTION

The invention relates to devices according to the preamble of claim 1 or 13 and an arrangement, a transmitting auxiliary device and a receiving auxiliary device according to the preamble of claim 16, 17 and 18, respectively.

In particular, the invention relates to a device such as an Internet access router, which can use several different Internet access at the same time, while the capacity of the connected lines in total for the user provides, with the peculiarity that by a forward error correction even with disturbances with significant packet losses on one Portion of the lines from the user's point of view All IP packets can be transmitted without loss and delay.

TECHNOLOGICAL BACKGROUND

The bundling of several different wired or wireless Internet access lines for the purpose of parallel data transmission and thus increase the bandwidth is z. B. off US 8,125,989 B2 or the EP 1 976 202 A2 known. The methods and devices described therein are incorporated by reference into this application.

A typical application of this technique is an installed in a vehicle (car, train, aircraft) Internet router that combines several different mobile Internet access - z. B. all mobile service providers of a country. A single mobile network will have dropouts repeatedly while driving, z. B. if the signal path to each cell tower of the mobile operator blocked or the transmitter tower is disturbed. For this reason, a single mobile line is not usable for mission-critical applications where timely and lossless delivery of data packets is required (eg, in healthcare).

Existing solutions for bundling, as they are known from the documents mentioned above, although the bandwidths of all connected Internet access bundled provide. However, if packets are lost during data transmission on one of the lines, they must be retransmitted via one of the other lines. Signal propagation delays in the transmission of user data.

The invention is therefore based on the object to provide devices for data transmission over bundled network access lines, which are more stable in particular to packet losses on one or more of the network access lines.

PRESENTATION OF THE INVENTION

This object is achieved by a device and a system according to claim 1 or 13 and by an arrangement, a transmitting auxiliary device and a receiving auxiliary device according to claim 16, 17 and 18, respectively.

By utilizing forward error correction, dropouts on one or some of the lines can be fully compensated for without retransmissions being required. So can be z. For example, combining 4 Internet access lines (wireless or wired) can compensate for packet losses of up to 25% without requiring retransmission.

Further embodiments and features of the invention will become apparent from the following description, the drawings and the claims.

DESCRIPTION OF THE FIGURES

1 illustrates the principle of the invention underlying the bundling of Internet access lines.

2 illustrates an embodiment.

3 illustrates the inventive system for data redundancy.

DETAILED DESCRIPTION OF EMBODIMENTS

In the 1 illustrated arrangement for transmitting a data stream a from a transmitting device 1 to a receiving device 2 via a packet-based network, especially the Internet 3 , Transmitter side and receiver side each comprises a device 5 . 6 expediently in the form of a respective router, with a plurality of network access lines being used to increase the data transmission rate and / or to increase the reliability 4 . 4 ' can be provided transmitter and / or receiver side. The device 6 may optionally also be on the Internet and the composite data stream a 'via a line to the receiving device 2 transferred, which can then be configured conventionally. Such arrangements and the associated transmission methods are known from the US 8,125,989 B2 or the EP 1 976 202 A2 The applicant is known and will not be further described here, but by incorporating part of the present disclosure.

So can the sending device 1 a computer or a medical device, cf. 2 which is in an ambulance 7 about the device according to the invention 5 , z. B. one with multiple wireless networks 4 connected radio router, as well as radio masts 8th coupled to the Internet and ultimately with a remote station, z. B. a telemedicine device in a clinic 9 , preferably also via the device according to the invention 6 , z. A router connected to wireless and / or wired network access lines. Possibly. is the remote site 6 on the internet and in the clinic is a simple router 2 with only one internet access line 4 ' intended.

In the illustrated embodiment, the location of the user (eg in the vehicle 7 ) a device according to the invention 5 , here in the form of a router (Transceiver + Receiver) to which multiple wireless and / or wired Internet access lines 4 available (eg mobile radio modems). IP data packets to and from the Internet are not sent directly from the sending device 1 via the connected internet access lines 4 but instead via the device 5 in particular encapsulates preferably via a VPN tunnel to the counterpart 6 the device, the z. B. can be configured as a router on the Internet (sender device + receiver auxiliary device), in which the packets may be decapsulated and then to the actual destination 6 to get redirected. Packets from the Internet to the user also initially travel via a corresponding router before they are expediently sent encapsulated to the user's router.

It is now the object of the invention that IP data packets from / to the user in the transmitting aids / receiving aids are divided into several fragments depending on the number of connected lines available, the number of fragments being, for example, the number of lines currently corresponding to the quality requirements minus 1 and at least one further fragment is generated, which contains redundancy data from the remaining fragments, and this fragment is then sent over the remaining wire.

This forward error correction drastically minimizes transmission errors and transmission delays: Especially in mobile networks, each different network, especially when used in a moving vehicle often short dropouts because the connection to a mobile phone mast or the transfer of a mobile phone mast to the next leads to transmission misfires. When using multiple mobile service providers these dropouts are usually not at the same time, but delayed, because each provider has its own mobile mast. Through the transmission of redundancy data, the reception auxiliary device can compensate for such packet losses as well as prevent packet transmission delays, since delayed fragments in most cases no longer need to be maintained, but the missing fragment can be regenerated from the forward error correction data.

system

The system according to the invention is for carrying out a method with reference to 3 described.

An IP data packet P ₁ is sent by the sender with the destination address of the recipient. It is forwarded to the router, which serves as a transmitter auxiliary device.

The transmission aid device checks whether the IP data packet belongs to an existing connection flow. If this is not the case, a new unique connection flow number F _{1 is} generated and assigned to the IP data packet. If so, the IP data packet is assigned the existing connection flow number.

On the basis of quality rules stored in the transmission aid device, the transmission aid device now determines which of the connected Internet access media is suitable for transporting this packet. The following criteria may be used, among others: latency of the line (time required for a packet to reach the receiver on this line from the transmitter assistant), packet loss rate of this line between transmitter and receiver, cost of data transmission on this line, and other criteria.

After the transmitting auxiliary device has determined which lines can be used for the IP data packet P ₁ , the determined lines are stored in the list P ₁ L. The number of entries in this list is stored in NP ₁ L.

If the value NP ₁ L = 0, no lines are currently available. In this case, the IP data packet is either discarded or cached for a later time.

If the value NP ₁ L = 1, the data packet is completely stored in the encapsulated packet KP ₁ . The encapsulated packet receives a mark in the header data of the encapsulation that it is complete. Likewise, the connection flow number is stored. Subsequently, the encapsulated packet is provided with the source IP address of the single line stored in P ₁ L, and over this Line sent to the destination IP address of the receiving auxiliary device.

In the following, on the other hand, it is assumed that NP ₁ L = 6, ie more than one line is available for transmission and meets the quality requirements.

Fragmentation of the data packets

If the value NP ₁ L> 2, the data packet P ₁ in NP ₁ L must be fragmented into 1 part of the same size. In order to do this, the expected size of the fragments is first determined. For a size of P ₁ of 1452 bytes would result in a division by NP ₁ L - 1 = 5 290.4 bytes. This size value is rounded up to the next value divisible by 4, in this case 292 bytes.

Subsequently, 5 fragment packets of 292 bytes are generated. The fragment packet 1 (FP ₁ F ₁ ) contains the bytes 1-292 of the data packet P ₁ , the fragment packet 2 (FP ₁ F ₂ ) contains the bytes 293-584, and so on. Thus, only 284 bytes from the packet P1 are available for the last fragment packet FP ₁ F ₅ . The missing 8 bytes in FP ₁ F ₅ are padded with 0 bytes.

Exception: First package of a new connection flow

If this is the first packet (P ₁ ) of a new connection flow number, the division ensures that the size of the fragments is at least large enough for the first fragment to contain the complete header data (IP header and possibly UDP or TCP header data) of the IP data packet P ₁ . In practice, the fragments of the first IP data packet of a connection flow number must therefore be at least 40 bytes in size. If the packet size of P _{1 is} divided by NP ₁ L - 1 <40, fragment sizes are now divided into 40/40. If there is a remainder in the division, then (fragmentsize / 40) + 1 fragments are split. The last fragment then again contains 0-bytes as fill data. This procedure ensures that, at the time of obtaining the fragment P ₁ F ₁ , the reception auxiliary device is able to associate the transmitted connection flow number F _{1 with} the destination address information of the original IP packet.

Calculation of the redundancy information

The transmitting auxiliary device thus now has the fragment packets FP ₁ F ₁ to FP ₁ F ₅ . In addition, a fragment redundancy packet FP ₁ F _R with the same length as the fragment packets FP ₁ F ₁ to FP ₁ F ₅ is now generated. This fragment redundancy packet is filled with redundancy data, which are calculated from FP ₁ F ₁ to FP ₁ F ₅ . For this purpose, the first byte of FP ₁ F _{1 is} linked by exclusive-OR (XOR) with the first byte of FP ₁ F ₂ . The result is then in turn linked by exclusive-OR (XOR) to the first byte of FP ₁ F ₃ , this result with FP ₁ F ₄ , this result with FP ₁ F ₄ . The result is now stored as the first byte of the fragment redundancy packet FP ₁ F _R. The same procedure is used for the remaining 291 bytes.

As a result, the packet FP ₁ F _R thus contains the result of all XOR operations of all the bytes of the fragment packets FP ₁ F ₁ to FP ₁ F ₅ .

Optional calculation of the double redundancy information

If, due to a high packet loss rate on more than one of the lines used, the generation of a simple redundancy is not sufficient to ensure that the reception auxiliary device can recover the original data even in the event of packet loss, the transmitting auxiliary device may decide to calculate double redundancy information. This is possible and useful if at least 4 lines are available, ie NP ₁ L> 3. In this case only NP ₁ L - 2 fragments of the data packet P _{1 are} generated. The first fragment redundancy packet FP ₁ F _R is then generated as above by XOR operation. A second fragment redundancy packet FP ₁ F _R2 is generated based on Galois field multiplications. This method is known from redundant storage on data carriers (eg patent DE 19922253 A1 ), but is hereby first described for use in real-time data transmission of IP data.

The further statements relate to the transmission of simple redundancy information, are analogous to use when using redundancy redundant information.

Encapsulation and dispatch of fragment packets

Each fragment packet FP ₁ is now encapsulated in a new IP data packet KFP ₁ . In the header of the encapsulated packet, the IP source address of the corresponding line of the transmission helper as well as the destination IP address Address of the receiving auxiliary device stored. Also stored is the connection flow number F ₁ , the total number of fragments NP ₁ L - 1, and the fragment packet number FP ₁ Fn. Subsequently, each packet KFP ₁ Fn is sent via the associated line In.

The fragment redundancy packet FP ₁ F _R is encapsulated in KFP ₁ F _R. In the header of the encapsulated packet, the IP source address of the transmission line NP ₁ L ₆ and the destination IP address of the reception auxiliary device are stored. Also stored is the connection flow number F ₁ , the total number of fragments NP ₁ L-1, as well as a flag indicating that it is a fragment redundancy packet. This is done by using a reserved number in the fragment packet number field. The encapsulated fragment redundancy packet KFP ₁ F _R is then sent via the line NP ₁ L ₆ .

If double redundancy information is used, the procedure is the same - the fragment packets FP ₁ to FP _{4 are sent here} via the lines L ₁ to L ₄ , the first fragment redundancy packet FP ₁ F _R via the line L ₅ , the second fragment redundancy packet FP ₁ F _R2 via the line L ₆ .

Misordering and optional speculative retransmission of fragment packets

Due to the different transit times of the encapsulated fragment packets on the lines or the infrastructure of the network operator located behind them, the fragment packets will not reach the receiving auxiliary device in the order in which they were sent. Not only the individual fragments can arrive in the wrong order (eg KFP ₁ F ₃ , KFP ₁ F ₂ , KFP ₁ F ₄ , KFP ₁ F _R , KFP ₁ F ₄ , KFP ₁ F ₁ ), also the actual ones Package numbers may arrive in the wrong order due to overhauls (eg KFP ₂ F ₁ , KFP ₁ F ₂ , KFP ₂ F ₅ , KFP ₁ F ₄ ...).

As explained below, the process described makes it very unlikely that unrecoverable packet loss will occur, as a rule packets can be recovered due to fragmentation and forward error correction at the receiving auxiliary device. But so that the case is covered in which simultaneously occur on several lines packet losses in the way that in the receiving auxiliary device a restoration is impossible, the behavior of the line is additionally checked with a heuristic. Immediately (within the round-trip time of the connection between the transmission helper and the reception helper), after sending the fragments on more than one of the lines used, the line has deteriorated to no longer meet the quality requirements of the connection flow (eg, increasing latency , complete disconnect), the affected fragments are retransmitted in a speculative retransmission over a line that meets the quality requirements.

The receiving aid

The list of known connection flows BVF and the list of unknown connection flows UVF are kept in the reception auxiliary device.

If a new encapsulated fragment packet KFP arrives at the reception accessory, it decapsulates and first checks whether the connection flow number is in the list of known connection flows BVF in the header information of the packet. If this is not the case, it is checked whether it is the first fragment of the first packet of this connection, ie FP ₁ F ₁ . If so, a new entry is created in BVF and a newly created reordering heap NSH is added to this entry. The packet FP ₁ F _{1 is} transferred to this reordering heap. In addition, it is now checked whether there are further packets of this connection flow in the list of unknown connection flows UVF, ie in the order of later packets which have already arrived before FP ₁ F ₁ . If this is the case, the packages contained therein are transferred to the reordering heaps in BVF and the entry in UVF is deleted.

If the incoming packet is one whose link flow number does not appear in the list of known link flows BVF and it is not the first fragment of the first packet (FP ₁ F ₁ ) of that link flow, it is checked if the link flow number in the List of unknown link flows UVF emerges. If so, the package will be added there. If this is not the case, an entry in the list of unknown connection flows UVF is generated and the packet is added.

If it trades on the packet received by one whose link flow number appears in the list of known connection flows BVF, the packet is added to the reorder heap HSP of the list entry in BVF.

The reorder heap input

When data is passed to the reorder heap, it first checks to see if it is a complete package or a fragment. If it is a complete packet, it checks to see if the sequence number of the packet is within the expected range (so it is not a late duplicate) and then sorted into the reordering heap. In addition, the sequence number of the package is marked as existing in a hashtable.

If it is not a complete packet, but a fragment, it is first checked if a fragment map already exists for the sequence number of the fragmented packet. If this is not the case, a new fragment card is created. Based on the header data of the fragment package the fields of the number of fragments for this sequence number are entered in the map and the payload of this packet is copied. If it is a fragment redundancy packet, the data is instead stored in a separate parity information array within the fragment map.

If the number of fragments entered in the card now equals the total number of fragments of that sequence number, the now complete package is sorted into the reordering heap and the sequence number in the hashtable is marked as present.

If the number of fragments entered in the card equals the total number of fragments minus 1, ie if only one fragment is missing, and if parity data (ie a fragment redundancy package) has already been received, the parity data is used to recover the missing fragment. For this purpose, all fragments present are XOR-linked sequentially with the content of the parity data field of the fragment card. The result in the parity data field is then copied to the location of the missing fragment. The now complete package is sorted into the reordering heap and the sequence number in the hashtable is marked as present.

In the case of duplicate redundancy information, the original packet can already be re-generated as soon as n-2 fragments have arrived at the receiving aid.

The reorder heap issue

The reorder heap stores the next packet sequence number to be issued. When the heap is generated, it has the value 1. The receiver auxiliary device checks after each entry in the reordering heap whether an output is now possible. The properties of the Binary Heap ensure that the data contained in the re-sort heap is sorted correctly (in ascending order), but there may be gaps due to packets that have not yet arrived. The output function of the reordering header therefore checks whether the sequence number of the first existing packet corresponds to the expected next sequence number. If so, the packet is dispatched and the expected next sequence number increased by one; if not, no packet is issued.

Optional delivery signaling and retransmission

Due to the described invention, the payload data can be provided promptly by means of forward error correction on the receiver auxiliary device, even in the case of severely disturbed lines. However, in the event of a simultaneous disruption of multiple lines that could not be foreseen by the transmission helper by historical behavior, it may still happen that one or more packet fragments would not reach the receiving helper or would be severely delayed. In this case, the receiving auxiliary device would have to wait in the described behavior until the missing fragment has arrived. To avoid this, the reception auxiliary device continuously determines the average time that elapses between receiving the first and last fragment of a packet. This time should normally correspond to the difference between the minimum and maximum latency of the transmission auxiliary device and the reception auxiliary device for this connection flow. If the statistically expected delay is clearly exceeded within the reordering heap within the reordering heap, the loss or significant delay of one of the fragment packets must be assumed.

To address this case, two different approaches are possible based on user choice as part of the service quality settings:

a) skipping missing packets

If the reorder heap fails to recover a gapless connection flow in a timely manner, the missing packets are skipped. The problem of missing user data packets is thus higher transmission layers, ie z. TOP or UDP. The connection flow from user data in this case will rarely have gaps, but will arrive at the final recipient with a constant delay. This is desirable, for example, for services such as telephony or video conferencing.

b) retransmission of missing packets

If the corresponding user option has been selected, the reception auxiliary device of the transmission auxiliary device for each connection flow communicates cyclically which fragment packets have arrived at the reception auxiliary device. Analogously to the analysis described on the basis of the packet transit times (latencies) of the individual lines, it is now the transmitting auxiliary device that recognizes early due to lack of confirmation, if not all fragment packets have arrived in the receiving auxiliary device within the expected time window. In this case, the send helper speculatively retransmits packet fragments via a different line than previously used for that fragment.

Between transmission auxiliary device and reception auxiliary device, the transmitted fragment packets can be compressed and / or encrypted. Depending on the compression and encryption methods used, some data packets may be indispensable to the receiving auxiliary device, so the data stream could not be recovered if such a packet were missing. In this case, even with a user decision for the variant a) the transmitting auxiliary device can mark such indispensable packets separately in the header field. In this case, the receiving auxiliary device will signal to the transmitting auxiliary device only the receipt of such indispensable packets or their fragments, and the transmitting auxiliary device according to b) only transmit such packets. On the other hand, all dispensable packets are handled according to a), that is to say the receiving auxiliary device will accept gaps in the data stream in order to avoid delays in the connection flow.

Forwarding the packages to the actual destination

At the receiving auxiliary device now a connection flow is generated, which corresponds exactly to the input data stream at the transmitting auxiliary device. As soon as the first fragment packet FP ₁ F _{1 has reached} the reception auxiliary device, the original header information of the IP packet is available again so that IP-based routing can now be performed for the data flow. According to configured routing protocols, the connection flow can now be treated as an ordinary IP data stream and forwarded to the actual destination.

Bidirectional data flows

In the transmitting auxiliary device, a receiving auxiliary device is implemented in parallel, and in the receiving auxiliary device, a transmitting auxiliary device. For a bidirectional useful data stream, the receiving auxiliary device now becomes the transmitting auxiliary device in the opposite direction, and the transmitting auxiliary device becomes the receiving auxiliary device. The outward and return direction of a connection flow can be handled independently of each other. It is also possible, however, for the transmitting aid device and receiving auxiliary device to automatically allocate the flow and return flows of the connection flow and then use bundling parameters shared for both directions.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 8125989 B2 [0003, 0013]
• EP 1976202 A2 [0003, 0013]
• DE 19922253 A1 [0031]

Claims (18)

Contraption ( 5 ), in particular routers, for bundling a plurality of Internet access lines ( 4 ) to a virtual Internet access line to provide the sum of the bandwidths of the multiple Internet access lines ( 4 ) for data transmission over the virtual internet access line, the device ( 5 ) a data packet to be transmitted to a plurality of data packets for separate transmission over the plurality of Internet access lines ( 4 ), characterized in that the device is designed to calculate and transmit redundancy information from which lost data packets are recoverable, so that packet losses on an Internet access line ( 4 ) do not lead to packet losses on the bundled virtual circuit. Device according to Claim 1, characterized in that packet losses on the physical lines are compensated for by virtue of the fact that redundancy information has been transmitted in advance on the lines not affected by packet losses so that the original data can be completely restored at the receiving end. Apparatus according to claim 1 or 2, characterized in that it is in the transmitted data to those in the Internet Protocol Format (IP). Device according to one of claims 1 to 3, characterized in that the redundancy information is generated by distributed parity information based on an XOR operation of the user data. Device according to one of claims 1 to 4, characterized in that the redundancy information is generated by distributed double parity information based on Galois field multiplications. Device according to one of claims 1 to 5, characterized in that the redundancy information is generated and transmitted packet by packet. Device according to one of Claims 1 to 6, characterized in that the redundancy information is generated and transmitted sub-packet by packet, ie a data packet is divided into a plurality of fragments, and additional fragments carry the redundancy information. Device according to one of claims 1 to 7, characterized in that for the selection of the lines and the amount of redundancy information to be transmitted, the respective current quality characteristics of the individual lines (latency, cost, packet loss rate, bandwidth) are used. Apparatus according to claim 8, characterized in that in case of subsequent negative change in the quality characteristics of the line already sent but not yet arrived data this speculative on a further quality management line corresponding retransmitted. Device according to one of claims 1 to 8, characterized in that the device based on the historical behavior of the lines automatically predicts future packet loss behavior of lines, and based on it automatically decides what level of redundancy information must be transmitted to no packet losses in the transmitted payload to suffer. Device according to one of claims 1 to 10, characterized in that it can be selected by user control, if determined in the case of late arrival of packets to a receiving device by the user optionally missing packets are skipped or retransmitted. Device according to one of claims 1 to 11, characterized in that data packets between the transmitting auxiliary device and the receiving auxiliary device are compressed and / or encrypted transmitted, and the transmitting auxiliary device marks packets that are for a continued restoration of the data stream in the receiving auxiliary device forcibly necessary, the receiving auxiliary device of the transmitting auxiliary device only confirms these packets marked as indispensable upon receipt of the transmission aid device, and speculatively retransmits the transmission aid device as having been confirmed as received on the basis of a prognosis. System for Internet data transmission via logical means of a transmitting auxiliary device ( 5 ) bundled to a virtual data transmission line individual transmission lines ( 4 ), in which the data to be transmitted to the individual transmission lines ( 4 ), characterized in that in addition to the data redundancy information is calculated and transmitted from which an optionally lost data packet in a reception auxiliary device ( 6 ) is reconstructable, so that data packet losses on the individual transmission lines ( 4 . 4 ' ) are repairable and do not result in data packet loss on the virtual communication line. A system according to claim 13, characterized in that n individual data transmission lines are used, each virtual data packet is divided into n-1 data packets, a data packet with redundancy information relating to the n-1 data packets is calculated, and the n total data packets are sent over the n individual data transmission lines become. A system according to claim 13, characterized in that n individual data transmission lines are used, each virtual data packet is divided into n - 2 data packets and two data packets are calculated with redundancy information concerning the n - 2 data packets, and the n total data packets are sent over the n individual data transmission lines become. Arrangement for Internet data transmission over a plurality of individual data transmission lines ( 4 . 4 ' ), comprising a transmitting auxiliary device ( 5 ), in particular a device or a router according to one of the preceding claims, which is connected via a plurality of individual data transmission lines ( 4 ) to the Internet ( 3 ) is connected, as well as a receiving auxiliary device ( 6 ), in particular a device or a router according to one of the preceding claims, which is connected via a plurality of individual data transmission lines ( 4 ' ) is connected to the Internet, whereby the transmitting auxiliary device ( 5 ) and the receiving auxiliary device ( 6 ) is designed for carrying out the method according to one of the preceding claims. Transmission aid device ( 5 ) for carrying out the method according to one of the preceding claims, wherein the transmitting auxiliary device ( 5 ) can be coupled to n individual data transmission lines, where n> 1, characterized in that the transmitting auxiliary device ( 5 ) for dividing a data packet to be sent onto n - m data packets and for calculating m data packets with redundancy information, where m <n. Receiving auxiliary device ( 6 ) for carrying out the method according to one of the preceding claims, wherein the reception auxiliary device ( 6 ) can be coupled to a plurality of individual data transmission lines, characterized in that the reception auxiliary device ( 6 ) for receiving a plurality of data packets including at least one data packet with redundancy information and for recovering lost data packets from received data packets by means of the redundancy information (s).

Images