Options binaires fiable means to an end
The field of the invention is that the data transmission systems, specifically structured transmission systems, where communication between two terminals are by exchange of binaires data units in Anglo-Saxon: The invention therefore applies in particular to all packet transmission systems system using a standard procedure 3, according to the OSI standard model for seven levels but also to the frames transmission systems level 2 or transport protocol data units level 4. More generally, the invention is applicable to any type of transmission system, regardless of the level in the terminology of OSI standards communication protocol associated with the system considered. On French territory, the invention therefore applies for example to communications made over the public network Transpac trademark. It is known that, especially when you need to borrow a public communication network, the cost is generally based, among others, the number of protocol data units or subsets of these units issued and airtime. One known way to reduce the volume of data transmitted is to compress the data to be transmitted. Many data compression techniques are known to effect, especially in the case of point to point links. For example, an adaptive compression algorithm, said algorithm "Ziv-Lempel," was held in the CCITT recommendation V42a for asynchronous modems. However, in the case of systems operating by exchange of protocol data units, for example according to the X25 standard, several difficulties are typically encountered. Thus there is known a system called DATAMISER trademarkwhich compresses systematically, byte by byte, the entire data stream issued by a terminal. When this data stream organized into protocol data units, end comprising descriptive areas of communication header end and data areas data fieldssuch a compression system does not meet this structure. In other words, it does not recognize or preserves headers. It can not therefore be used for a communication through a packet switched network. In order to overcome this drawback, it is likely to selectively compress data areas. In this case, the headers are maintained in full, and the amount of information to be transmitted is substantially reduced. In other words, means source protocol data unit is associated with a compressed protocol data unit, of which the header is identical to that of the source protocol data unit. In practice, this technique is a non-optimal efficiency. Indeed, means number of packets PDUs transmitted by definition remains the same. Le gain est donc minime, puisqu'on transmet des paquets partiellement vides. The gain is minimal, since it transmits partially empty packets. Also known from EP-A, a method for interconnecting two networks using data units of different sizes bits 8 and 9. The proposed technique is to group 9th bit control bit for 8 consecutive words in one control word c. On peut ainsi transmettre simplement des mots de 8 bits, sans perte d'informations. We can simply send words of 8 bits, no loss of information. The invention particularly aims to overcome these drawbacks of the state of the art. More specifically, an objective of the invention is to provide a communications efficiency of the optimization process carried out on a network for exchanging data, reducing the number of protocol data units or subassemblies of transmitted units. Another object of the invention is to provide such a method, as regards the structure of the transmitted protocol data units. In other words, an object of the invention is to provide such a process that is transparent from the point of view of the terminal and from the point of view of the transmission network. More generally, the invention aims to provide such a method that reduces at least one of the following: Another object of the invention is to provide such a method, which can be implemented on any structured data transmission system. The invention also aims to provide a device implementing such a method, which can be inserted between a terminal and a communication network in a transparent manner, that is to say, does not require any adaptation or terminal or the network. These objects, and others which will become apparent hereinafter are achieved according to the invention with the aid of a data compression method according to claim 1. In this way, the performance of the communication can be greatly increased, but the initial structure of the data is retained. It is possible, on reception, to reconstruct the source sequence, using the reconstruction of data added to the transmitted data. Thus, according to the invention, it momentarily alters, during transmission, the data structure, which reduces the number of protocol data units to be transmitted, a further additional stage is provided, compared to known techniquesnamely the inclusion among the compressed data to be transmitted to data reconstruction. Such a method advantageously comprises the steps of: Under this method, so we added in some end, a network of binaires transparently to the network. It should be noted that the concatenation steps, cutting and creating the second sequence, which are distinguished here in order to facilitate understanding of the invention may actually be carried out simultaneously one begins to create a package from as data to be concatenated are received. In practice, the steps of slicing and concatenation are generally conducted simultaneously: Whether a data sub-field has been cut to complete the packet data field to be transmittedthe remaining part of this sub-field forms the beginning of the packet data field to be transmitted next. Advantageously, said reconstruction data generating step includes a step of associating 26 each of said sub-fields 30 of compressed data of a sub-header 29 having at least one information options representative of the length of said sub-field 30 of compressed data. In other advantageous embodiments of the invention, said reconstruction data generation step may include: Preferably, said data for reconstructing said first end of source protocol data units include at least one of the information present in the header field of each of said source protocol data units. In fact, besides the length of the compressed data, it may be necessary to reconstitute the source sequence, store information describing the source protocol data units. In a particular embodiment of the invention, said sub-header includes at least a first byte, bearing at least one of the information present in the header field of the source protocol data unit corresponding to said sub-header, and at least one second byte bearing the sub-field length associated compressed data. Advantageously, said compression algorithm is an adaptive compression algorithm to manage the transmission and reception of the data series of dictionaries, said dictionaries being updated identically on each transmission fiable one of said protocol data units. This adaptive algorithm turns out particularly interesting effect, in terms of compression efficiency. Preferably, the method includes a data compressing step comprising the steps of: It is clear that, in the case where the compression is not effective, it is preferable to transmit data in uncompressed form. In this case, said sub-header includes information representative of said selection of the source data field or of the compressed data field. The protocol data units considered may be of any type, for example data packets, data frames or level 4 data units transport. Conventionally, these may have a protocol data unit length chosen from a set of at least one possible length. Advantageously, the method also comprises a step of initializing a transmission, consisting in particular to verify that the receiver has means for implementation of a method corresponding to said compression decompression process. The invention makes it possible to transmit only full data units, since their data fields include concatenated data from several source units. However, the filling of a unit should not be a mandatory requirement, at the risk of blocking the process. Indeed, in particular when it comes to an end sequence, it must be possible to transmit an incomplete data unit. Pour binaires, deux temporisations sont envisageables. For this, two timers binaires possible. Thus, the method advantageously comprises a first delay step, fixing a maximum waiting time D1 before the creation of a protocol data unit to be transmitted not full. In addition, it may also include a second time delay step, fixing a maximum waiting time D2, D2 is less than D1, before the creation of a protocol data unit to be transmitted comprising at least one data item length predetermined full, and no non-solid element. The invention relates of course also a decompression process symmetric to the compression method described above, comprising the steps of: In this way, the device is quite transparent. Other features and advantages of the invention will become apparent from the following description of a preferred embodiment of the invention, given as illustrative and not restrictive, and the appended drawings fiable which: The embodiment described below is particularly applicable to networks with a consistent interface to the X25 standard, especially for networks using a pricing principle by segments, such as the French public network Transpac trademark. It is clear, however, that the invention is not limited to this type of networks, but may instead be applied in all cases where data exchanges are by protocol data units. As already mentioned, the invention aims to improve the efficiency of communications on a given network, while maintaining the output data stream structure. Figure 1 shows an example of implementation of the invention. These terminals 11A and 11 C show any end system that can connect to a network, such as for example a computer, a database server, an input terminal, etc Two terminals 11A and 11C can be associated to a particular communication. Both terminals can possibly manage multiple communications simultaneously. For example, the terminal 11 A can simultaneously maintain a first communication with the terminal 11 B and a second communication with the terminal C In other words, the device behaves 13A 14, 15seen from the terminal 11 A, as the network, and, seen from the network 12, such as a terminal 16, Each device 13A, 13B comprises compression means for data 18 A, 18 B, implementing the method which will be described later, and symmetrical decompression means 19 A, 19 B. Essentially be described subsequently the compression process. The decompression process can easily be deduced by symmetry. Note that the compression of a communication data is not systematic. First, it is necessary that the receiving terminal is associated or understand a pressure relief device. Ainsi, toute communication impliquant le terminal 11 C se fera sans compression. Thus, any communication involving the terminal 11 C will be without compression. Figure 2 is a simplified block diagram of options compression method of the invention. This can be done either passively, each compression device managing a list of devices that can perform decompression, either actively, through an interactive query procedure. In a second step, you should check that the receiving device has decompression means available. On aborde ensuite si la compression est possible une phase de compression 22 proprement dite. Then addresses if compression is possible a compression stage 22 itself. This stage 22 comprises a receiving step 23 of source packets to be compressed. It should be noted that the method of the invention makes no assumptions on the sequence of received packets, nor on the incorporation of these packets. These can be filled or not be of fixed or variable size, from a variety of origins, etc En revanche, ainsi qu'on le verra par la suite, il assure la conservation de cette structure d'origine. However, as we shall see later, it ensures the conservation of the original structure. It has already been explained in the preamble that it was not possible to compress without distinction data fields and headers. According to the invention, only the data field 24 is compressed A particular method advantageously compression is a compression algorithm of the type called "Ziv-Lempel". Encoding basic principles, decoding and updating dictionaries are similar to those described in Recommendation V42a. Of course, any other compression algorithm can be used. Eventually, several algorithms can be implemented simultaneously, the most effective result being alone retained. Moreover, creating 26 a sub-header, or information area 27 in particular comprising an information representative of the length of the compressed data field issued by the step 25 of compression. This header contains over 28 data necessary for the reconstruction of the sequence of original packets, that is to say, in particular information indicating in the header of each original packet not compressed type package Qualifying informationthe structuring into complete messages according to the X25 protocol, etc A particular example of sub-header is described later in connection with Figure 4. The sub-header 29 and the compressed data field 30 corresponding to each source packet are then concatenated 31 in a series of 32 continuous data. This result 32 is then cut 33 arbitrarily, that is to fiable without respect of the original packet, at a length corresponding to a data field length allowed by the network, or submultiple length This allowed length. Each element 34 thus cut is used to create 35 a new package, becoming the data field or a portion of that field. This new package has instead a header again of its own. The efficiency of this process will be best seen in the example shown in Figure 3 compression. Consider fiable source sequence of three consecutive packets 41A, 41 B and 41 C. Each packet consists conventionally of a header 42 A, 42 B and 42 C in the example considered, in accordance with the X25 standard and a data field 43 A, 43 B and 43 C. Here we consider the case of full packets but this is not a mandatory requirement and a format consistent with that admitted by the communication network. Each packet 41 i is then compressed Thus, a compressed packet sequence 45 A, 45 B, 45 C. Of course, the number of packets is preserved, and the headers 42 A to 42 C of each of them remain unchanged. This compression can be very interesting, especially when the network uses a segment by the charging system. This is particularly the case of the Transpac network, which uses its pricing segments of 64 bytes. This cut into segments, which is different from the packet structure, a basic means to encourage users to issue full data packets. Indeed, each begun segment that is to say containing at least one byte is recorded, and therefore has the same cost to a full segment 64 bytes. Thus, the package 41 A, if released, would be recorded for two segments 47 1 and 47 2. In contrast, the package 45b includes only a recorded segment 48 1. Indeed, whatever the quality of compression achieved, it will always have at least one segment billed per pack. The invention provides a new solution, and very effective for this problem. Indeed, after conducting compression 44, an additional processing step compaction 50 is performed. According to the invention, a new sequence of packets 51 D, 51 E is created. These packages are quite independent of the sequence of packets 41 A to 41 C. In particular, their headers 52 D, 52 E correspond to no headers 42 A to 42 C. However, all the data useful for the reconstruction of the process of source packages is available in the data fields 53 D and 53 E 51 new packages D and E More specifically, and it is the main feature of the invention, a data field 53 D includes not only the concatenated data of the end fields 46 A and 46 B, but also in the form of a sub -in header 47 A, of the original sequence reconstruction end. This sub-header 47 A is transparent, the network's point of view. However, decompression, it allows one hand to extract the data field 53 D data sequence corresponding to the options 46 A 45 A, the length information Moreover, it contains the necessary options reconstruction, at least in part, of the header 42A. The data fields 53 D and 53 E are considered continuously. Thus, the compression data 46 B can be smoothly divided into two subsets 46 B 1 and 46 B, 2 in two separate packages. The sub-header 47 B indicates the end of the subassembly 46 B 2. A sub-header can also be split into two packets. Figure 4 shows an example of a sub-header may advantageously be used in the X25 protocol. This sub-header options of two bytes 61 and The first byte contains the one hand two bits Q and M, which are the direct reproduction of the bits Q "Q-bit" and M " M-bit " existing in the header of a packet according to the X25 protocol. These bits are needed to respectively describe the packet and indicate the beginning or end of a message. Furthermore, it includes a bit C, which indicates means the compression was implemented for the particular package. Five bits 63 are available for future developments. For example, they can carry information indicating the compression method used in the case where several modes can be used. Le second octet comprend, means 8 bits, la longueur L 54, fig. The second byte comprises 8 bits, the length L 54, Fig. This length can, if required, be coded on two bytes, or more. Source packages 71 are received in a module 72 for receiving and separation, which delivers one hand the data field 73 and the other header 74 of each packet. The data field 73 is then compressed by the compressor 75, which implements an adaptive algorithm such as "Lempel Ziv". In this case, a dictionary is updated continuously in a memory The compressed data field 77 is then transmitted to a module 78 for comparison, which selects the shortest data field 79 comparison between the compressed field 77 and the uncompressed field 73, and selection of the shortest, and, in the case of equal length of the compressed name field, which require less processing at the reception. In the case where multiple compression means are implemented in parallel, the module 78 performs the same selection more efficient that is to say, the compression mode offering the best compression ratio. If necessary, the selection can be made for each packet. A module 80 for building the sub-headers 81 prepares each sub-header e. These sub-headers 81 and 79 compressed data fields are then concatenated and cut to the size of the segments or packets 82so as to create full segments and more generally, full packets. In practice, this operation of concatenation and cutting is carried out continuously. It compresses the data and concatenates them. As soon as the concatenation fiable the length of a packet, it delivers this full package 83 to 84 transmission module, which is responsible for forwarding it to the recipient terminal. Ainsi, on ne transmet que des paquets pleins. Thus, it transmits only full packets. This principle, however, can introduce a delay in the transmission of a source package expectation that the packet to be transmitted is full. A delay management mechanism is therefore advantageously be implemented. Such a mechanism is also needed when one is at the end of source sequence end of message. Indeed, it is unlikely that the last packet is full, but it must of course be still transmitted. Therefore considers two timers D1 and D2, as follows: Preferably will be chosen very D2 lower than D1, in order to minimize the number of transmitted segments. The effectiveness of the method and apparatus of the invention is evident in Figure options, since it is found that two packages 51 E, 51 D are transmitted, while the source sequence included three full packets 41 A, 41 B, 41 C. This efficiency is even more convincing, of course, when the source packages are not full. In this case, the principle of concatenation allows a significant gain in number of packets or segments, that is to say portions of packets transmitted. In the embodiment described above, the method of the invention is implemented in dedicated compression units, inserted between a terminal and the network. It can for example be a series device COMPLYS trademarksold by the company OST trademark. This method can also be directly integrated in a terminal, such as, for example, a terminal incorporating an X25 communications PCSNET binaires registered trademarkalso marketed by OST. It can also be integrated with an element of the communication network, for example a 25 ECOM trademark OST. Furthermore, as was already mentioned, the method of the invention is not limited to X25 networks. It can in fact be implemented over any protocol layer providing a connectionless service that is to say having a reliable logical fiable without loss or duplication or reordering. In particular it can be a device with a MAGELLAN card registered trademark. The embodiment means was presented proposes to introduce sub-headers indicating the following data sub-field length. It is clear however that the invention can be implemented in many other ways, provided that the packets PDUs transmitted carry information for reconstructing the packets PDUs source. For example, it is not mandatory that a sub-header is inserted between each data subfield. Simply they appear regularly eg every 5 sub-fieldseach sub-header while carrying information to extract more eg 5 subfields consecutive data. Furthermore, the length information or more sub-field data can be replaced with a pointer indicating the absolute position not relative from the end of the sub-field in the transmitted means. According to another mode of implementation, it is also inserted between each data sub-field in a predetermined format of data element or marker that the receiving terminal knows. When it detects this marker, the terminal knows that a new source package begins. It can also be provided at the beginning of each data field of the packet to be transmitted, an area describing the contents of the packet e. Such an area or descriptor corresponds in other words to an extension of the packet header to be transmitted. Try the new Google Patents, with machine-classified Google Scholar results, and Japanese and South Korean patents. Method and apparatus for data compression and decompression for a transmission system EP B1. Method for data compression of the type applied to a system for data transmission by exchange of protocol data units 41 i between at least one transmitting terminal 11 A and at least one receiving terminal 11 Beach of the said protocol data units 41 i comprising a header field 42 i and a data field 43 iwhich method is characterised in that it associates a second sequence of M protocol data units to be transmitted 51 D51 E with a first sequence of N source protocol data units 41 A41 B41 CM being less than or equal to N, and in that means comprises the following steps: Method according to Claim 1, characterised in that it comprises the following steps: Method according to Claim 2, characterised in that the said step for generation of reconstruction data comprises a step for the association 26with each of the said sub-fields 30 of compressed data, of a sub-header 29 comprising at least one information element 27 representative of the length of the said sub-field 30 of compressed data. Method according to any one of Claims 1 to 3, characterised in that the said reconstruction data 47 i ; 29 comprise at least one of the information elements present in the header field 42 i ; 28 of each of the said source protocol data units 41 i. Method according to Claims 3 and 4, characterised in that the said sub-header 47 i ; 29 comprises at least a first byte 61bearing at least one of the information elements Q, M present in the header field 42 i of the source protocol data unit 41 i corresponding to the said sub-header, and at least a second byte 62 bearing the length L of the associated sub-field 46 i ; 30 of compressed data. Method according to any one of Claims 2 to 5, characterised in that the said step for generation of reconstruction data comprises a step for the association 26with each of the said sub-fields binaires of compressed data, of a sub-header 29 comprising at least one pointer indicating the location of the end of the said sub-field in the corresponding data field of the protocol unit to be transmitted 51 i. Method according to any one of Claims 2 to 6, characterised in that the said step for generation 26 of reconstruction data comprises a step for generation of a marker of predetermined content, known to said receiving terminal. Method according to any one of Claims 2 to 7, characterised in that the said step for generation 26 of reconstruction data comprises a step for association, with end protocol data unit to be transmitted 51 iof a descriptor of the content of the data field of the said protocol data unit. Method according to any one of Claims 1 to 8, characterised in that the said compression algorithm 25 is an adapted compression algorithm managing dictionaries 76 of data sequences on transmission and on reception, the said dictionaries 76 being updated identically on end transmission of one of the said protocol data units. Method according to any one of Claims 1 to 9, characterised in that it comprises a step 25 for data compression comprising the following steps: Method according to Claim 10 and any one of Claims 2 to 9, characterised in that the said sub-header 47 i ; 29; 81 comprises an information element C representative of the said selection 79 of the source data field 73 or of the compressed data field Method according to any one of Claims 1 to 11, characterised options that the said protocol data units belong to the group comprising: Method according to any one of Claims 2 to 12, characterised in that the said protocol data units 41 i ; 51 i have a protocol data unit length chosen from a set of at least one possible length, and in that the length of the said data elements is a submultiple of the said chosen protocol data unit length. Method according to any one of Claims 1 to 13, characterised in that it comprises a step 28 for initialisation of a transmission, options, in particular, of verifying that the receiver 11 B possesses means 19 B for implementing a decompression method corresponding to the said compression method. Method according to any one of Claims 1 to 4, characterised in that it comprises a first time-delay step, fixing a maximum waiting time D1 before the creation of a non-full protocol data unit to be transmitted. Method according to Claims 13 and 15, characterised in that it comprises a second time-delay step, fixing a maximum waiting time D2, D2 being shorter than D1, before the creation of a protocol data unit to be transmitted comprising at least one full data element fiable predetermined length and no element that is not full. Method for decompression of compressed protocol data units in accordance with the method of any one of Claims 1 to 16, characterised in that it comprises the following steps: Apparatus for data compression, of the type used in-a system for data transmission by exchange of protocol data units 41 i between at least one transmitting terminal 11 A and at least one receiving terminal 11 Beach of the said protocol data units 41 i comprising a header field 42 i and a data field 43 icharacterised in that it comprises means 18 i for compression, in at least one direction of data transmission and for at least one communication, associating a second sequence of M protocol data units to be transmitted 51 D51 E with a first sequence of N source protocol data units 41 A41 B41 CM being less than or equal to N, the said compression means 18 i comprising: Apparatus according to Fiable 18, characterised in that it comprises means 19 i for decompression of data compressed by the said compression means 18 ithe said decompression means comprising: Apparatus according to either of Claims 18 and 19, characterised in that it is integrated with one of the said terminals or with an element of a communications network interconnecting the said terminals. Figure 5 is a simplified diagram of a device implementing the method of Figure 2. Kind code of ref document: THE PATENT HAS BEEN ANNULLED BY A DECISION OF A NATIONAL AUTHORITY. LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES. Country of ref document: Date of ref document: Ref legal event code: Year of fee payment: DED1DET2EPA1USWOA1. Jean-Marc ErnaultThao LaneBernard Thepault. Ouest Standard Telematique S. BiBTeXEndNoteRefMan. Non-Patent Citations 1 binaires, Classifications 8Legal Events EspacenetEP Register. A1 Designated state s: B1 Designated state s: Lapsed in a contracting state announced via postgrant inform. ES Free format text: THE PATENT HAS BEEN ANNULLED BY A DECISION OF A NATIONAL AUTHORITY Effective date: Binaires Free format text: AT Free format text: NL Free format text: GR Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: AT Date of ref document: CH Ref legal event code: DE Date of ref document: PT Free format text: DK Free format text: SE Free format text: LU Free format text: BE Free format text: CH Free format text: LI Free format text: GB Free format text: FR Free format text: DE Free format text: FR Ref legal event code:
Marla states these symptoms are beginning to cause problems for her at her job.
In lieu of perfection, I sent my most profound sincerity and pray it will be well received whether I hear from them or not.
Cheap CustomWritings (20% OFF) Professional custom written Essay, Research Papers, Reports, Power points, Capsules, Dissertation, Term paper etc.
DISCLAIMER: The creator of this page takes no responsibility for the way you use the information provided on this site.