WO2015192452A1

WO2015192452A1 - Header compression method, decompression method and device

Info

Publication number: WO2015192452A1
Application number: PCT/CN2014/084315
Authority: WO
Inventors: 杨飞飞
Original assignee: 深圳市中兴微电子技术有限公司
Priority date: 2014-06-17
Filing date: 2014-08-13
Publication date: 2015-12-23
Also published as: CN105208596A

Abstract

The present invention relates to the technical field of communications. Disclosed are a header compression method, decompression method and device, for solving the problems in the prior art of high memory consumption by header compression, low compression speed and low compression efficiency; the header compression method comprising: configuring a corresponding index for a list item to be compressed by determining whether optional idle indexes exist; and if not, then selecting, from the indexes configured for the list items, an index with frequency of use lower than a preset threshold, and replacing the list item corresponding to the index with the list item to be compressed, such that the list item to be compressed is compressed by employing the replaced index.

Description

TECHNICAL FIELD The present invention relates to the field of communications technologies, and in particular, to a header compression method, a decompression method, and an apparatus. Background Art Due to limitations of physical conditions, a wireless link in a mobile communication system has a lower transmission rate and a higher bit error rate than a wired link. In order to effectively utilize limited wireless channel bandwidth resources, Robust Header Compress (ROHC) technology has been introduced. The core of ROHC is to transparently compress and decompress the information in the packet header between directly connected nodes by using information redundancy between packets of the traffic flow. The ROHC technology is described by the Internet Engineering Task Force (IETF) RFC 3095 document, but the compression and decompression of the IP header is not defined in this document, so in June 2004, the ROHC working group had to be IP alone in RFC 3843. The header compression decompression defines a framework, and in February 2007 the IETF revised the ROHC documentation, which is RFC 4815.

In ROHC header compression, different profiles are used to identify different packet flows in order to use the corresponding data compression decompression framework. For example, the Profile 0 identifies the IP packet flow using the uncompressed method. The Profile 1 identifies the Real-time Transport Protocol (RTP). The User Data Protocol (User Data Protocol, UDP for short). / Internet Protocol (IP) packet flow is compressed, Profile 2 identifies the UDP/IP packet flow, and Profile 3 identifies the ESP/IP (IP Encapsulated Security Payload ESP) packet flow. The Profile 4 identifier compresses the IP packet stream. When Profilel compresses the RTP header, it also needs to compress the list of the Contributing Source (CSRC) contained in the RTP header. The basic principle of list compression is that the compressor allocates a unique index for each list item, maintains a translation table with all list items, and uses the list item index to identify each list item. The list item index and the corresponding list item are simultaneously sent in the compressed list. When the compressor determines that the decompressor has obtained the mapping relationship between the index and the list item, the compressor can separately send the list index in the compressed package to represent Corresponding list items achieve the purpose of reducing the amount of data for compression. However, RTP list compression is more complicated. In a real RTP packet stream compression transmission, there are thousands of list entries to be processed, so it takes a lot of memory space to maintain the correspondence between list items and indexes. It takes a long time to find the corresponding index, so the compression speed is slow and the compression efficiency is low. SUMMARY OF THE INVENTION The embodiments of the present invention provide a method for compressing a header, a method and a device for decompressing, to solve at least the problem that the compression memory of the header is large, the compression speed is slow, and the compression efficiency is low. In one aspect, the present invention provides a method for compressing a header, including: determining, if a corresponding index is configured for a list item to be compressed, whether there is an available idle index; in the case where there is no idle index, And selecting an index whose frequency is lower than the preset threshold from the index that has been configured for the list item, and replacing the list item corresponding to the index with the list item to be compressed, so that the list item to be compressed is replaced by The index is compressed. Optionally, the method further includes: determining whether a corresponding index needs to be configured for the list item to be compressed before determining whether there is an available idle index. Optionally, the determining whether to configure the corresponding index for the list item to be compressed includes: determining whether the list item to be compressed has been recorded in a translation table; wherein the translation table is used to record a list item Corresponding relationship with the index; in the case that the list item to be compressed has been recorded in the translation table, it is determined that the corresponding index is not required to be configured for the list item to be compressed; If the list item is not recorded in the translation table, it is determined that the corresponding index needs to be configured for the list item to be compressed. Optionally, the determining whether there is an available idle index includes: determining whether there is an idle index in the translation table that does not specify a corresponding list item. Optionally, the selecting, from the index configured for the list item, the index whose usage frequency is lower than the preset threshold includes: determining, in the process of compressing the list, the total number of uses of all indexes and the number of times of using each index separately; If the ratio of the number of individual uses to the total number of uses is lower than the preset threshold, determining that the frequency of use of the index corresponding to the number of individual uses is lower than the preset threshold. In another aspect, the present invention further provides a method for decompressing a header, including: determining whether a received index is configured to be sequentially configured to a different list item; and the same index is sequentially configured to a different list item. In the case, the compressed package is decompressed with the correspondence between the last configured list item and the index. In another aspect, the present invention further provides a header compression apparatus, including: a compression determining unit, configured to determine whether there is an available idle index in a case where a corresponding index needs to be configured for a list item to be compressed; a unit, configured to: when the compression determining unit determines that there is no idle index, select an index whose usage frequency is lower than a preset threshold from an index that has been configured with the list item, and replace the list item corresponding to the index with the Compressing the list item to be compressed, so that the list item to be compressed is compressed by using the replaced index. Optionally, the compression determining unit is further configured to determine whether a corresponding index needs to be configured for the list item to be compressed before determining whether there is an available idle index. Optionally, the replacing unit is configured to: determine, in the process of compressing the list, the total number of uses of all indexes and the number of times of using each index separately; if the ratio of the single use times to the total usage times is low Determining, by the preset threshold, that the frequency of use of the index corresponding to the number of times of use is lower than a preset threshold, and replacing the list item corresponding to the index with the list item to be compressed, so that the list item to be compressed is The compression is performed using the replaced index. In another aspect, the present invention provides a header decompression apparatus, including: a decompression determining unit, configured to determine whether a same index is sequentially configured to a different list item in a received compressed packet; and the decompression unit is configured to be in the When the decompression determining unit determines that the same index is sequentially configured to different list items, the compressed package is decompressed with the correspondence between the last configured list item and the index. The header compression method, the decompression method, and the device provided by the embodiment of the present invention can select a usage frequency lower than a preset from an index that has been configured for a list item, if an index needs to be configured for the list item but no idle index exists. The index of the threshold is replaced with the list item corresponding to the index whose usage frequency is lower than the preset threshold, so that the list item to be compressed is compressed by using the replaced index. In this way, the reuse of the index can significantly reduce the memory space usage of the compression end, shortening the time taken for searching the translation table, thereby effectively improving the compression efficiency. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a flow chart of a method for compressing a header according to an embodiment of the present invention; FIG. 2 is a flowchart of a method for decompressing a header according to an embodiment of the present invention; FIG. 4 is a detailed flowchart of a step in FIG. 3; FIG. 5 is a schematic structural diagram of a header compression device according to an embodiment of the present invention; It is a schematic structural diagram of the meter decompression device provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION The present invention will be described in detail below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. As shown in FIG. 1 , an embodiment of the present invention provides a method for compressing a header, including: S11. If a corresponding index needs to be configured for a list item to be compressed, determining whether there is an available idle index;

S12, in the case that there is no idle index, select an index whose frequency is lower than a preset threshold from the index that has been configured for the list item, and replace the list item corresponding to the index with the list item to be compressed, The list item to be compressed is compressed by using the replaced index. The header compression method provided by the embodiment of the present invention can select an index whose frequency is lower than a preset threshold from an index that has been configured for a list item, if an index needs to be configured for the list item but no idle index exists. The list item corresponding to the index whose usage frequency is lower than the preset threshold is replaced with the list item to be compressed, so that the list item to be compressed is compressed by using the replaced index. In this way, the reuse of the index can significantly reduce the memory space usage of the compression end, shortening the time taken for searching the translation table, thereby effectively improving the compression efficiency. It should be noted that when the data stream is transmitted from the compression end to the decompression end in the form of data packets, the header of each data packet may also be compressed and sent to the decompression end. The method of the header compression can be used for the compression of the RTP data packet, and can also be used for other types of data packets that support the method of the header compression. The embodiment of the present invention does not limit this. Specifically, in the header compression of the data packet, the correspondence between the list item and the index may be established first, and then the simple index is used to represent the more complicated list item data, so as to reduce the compression of the data amount. Optionally, in the embodiment of the present invention, the mapping relationship between the list item and the index may be established by using various methods, such as maintaining a mapping relationship table, or by using a certain mapping relationship function, etc., which is not limited by the embodiment of the present invention. . The following describes the header compression method provided by the present invention in detail by taking the maintenance mapping table in the RTP packet compression process as an example. In an embodiment of the present invention, a correspondence table between a list item and an index is maintained at the compression end, which is called a translation table. In the translation table, all indexes are stored, including both the index of the list item that has been configured, and the idle index of the list item that has not yet been configured. Whenever a new data packet is to be compressed, the step S11 may further include determining whether a corresponding index needs to be configured for the list item to be compressed, which may be specifically implemented by the following steps: Determining whether a list item to be compressed in the data packet has been recorded in the translation table, wherein the translation table is configured to record a correspondence between the list item and the index; and the list item to be compressed is already in the If there is a record in the translation table, it is determined that the corresponding index is not configured for the list item to be compressed; if the list item to be compressed has no record in the translation table, it is determined that the The compressed list item is configured with the corresponding index. However, since all indexes are stored in the translation table, the translation table has a limited capacity, and the number of packets sent is huge, so it is very possible to translate all the indexes in the table when the index items need to be configured for the list items to be compressed. The corresponding list item has been configured. Therefore, in the case that the corresponding index is to be configured for the list item to be compressed, it is also required to determine whether there is an available idle index. Whether there is an idle index in the translation table that does not specify a corresponding list item. If there is such an idle index, the idle index may be configured to the list item to be compressed. For example, the list item may be written into an item in the translation table corresponding to the idle index. If there is no such idle index, the embodiment of the present invention further provides an index reusing mechanism, which removes the index that is not used for a long time or the index of the recently used frequency that is too low, and the original list item. And configure the index to the new list item to be compressed. Thus, although the number of packets sent is huge, for example, tens of thousands, but the index in the translation table does not need to be large, for example, 200, through the reuse of the index can significantly reduce the memory space of the compressed end, shorten the search The time spent in translating the table greatly speeds up the transmission and effectively improves the compression efficiency. Optionally, in the step S12, in the case that the idle index is not present, selecting an index that uses the frequency lower than the preset threshold from the index that has been configured for the list item may specifically include: determining, in the process of compressing the list, respectively The total number of uses of all indexes and the number of times of use of each index; if the ratio of the number of individual uses to the total number of uses is lower than the preset threshold, determining that the index corresponding to the number of times of use is used less frequently The preset threshold. Correspondingly, an embodiment of the present invention further provides a method for decompressing a header. As shown in FIG. 2, the method includes:

521. Determine whether, in the received compressed packet, the same index is sequentially configured to different list items;

522. In a case where the same index is sequentially configured to different list items, the compressed package is decompressed by the correspondence between the last configured list item and the index. The method for decompressing the header provided by the embodiment of the present invention can determine whether the same index is configured to be configured to different list items in the received compressed package, and in the case that the same index is sequentially configured to different list items, The compressed package can be decompressed by the correspondence between the last configured list item and the index, so that the use of the index can significantly reduce the use of the memory space on the decompression end, shortening the time taken for searching the translation table, thereby effectively improving the efficiency of the compression. . Optionally, a mapping table between the list item and the index may also be maintained on the decompression end, which is also called a translation table. When a data packet is sent, if the data packet includes only the index and there is no corresponding list item, it indicates that the list item has appeared before, and the corresponding record is already in the translation table, so The correspondence in the table is decompressed. If the data packet includes not only the index but also the list item corresponding to the index, it can be checked whether the index has been recorded in the translation table, and if so, the index and the list item sent this time are reused for the index. Correspondingly, the corresponding relationship between the newly transmitted index and the list item should be replaced with the corresponding relationship already recorded in the translation table, and the subsequent data packet is decompressed with the new corresponding relationship until a new index and list item are added. Send it together and update the new index correspondence. The header compression method and the decompression method provided by the present invention will be described in detail below by way of specific embodiments. As shown in FIG. 3, the header compression method and the decompression method provided by the present invention include the following steps:

S301, the compression end sets a storage list of the index of the list entry of the moderate size (for example, between 20 and 50); after receiving the data packet, analyzing whether the list item to be compressed in the data packet needs to be compressed by the profilel type, If yes, analyze whether the number of lists in the RTP domain is 0. If not, perform list compression. S302. When the list item is compressed, the list item to be sent is sequentially read. If the list item already exists in the translation table, it is not necessary to create a new mapping item in the translation table to configure an index for the list item, and use the translation table. The corresponding index compresses the list item; otherwise, S303 is performed.

S303, if it is required to configure a corresponding index for the list item to be compressed, determine whether there is an available idle index in the translation table on the compression side; if yes, execute S304, if not, execute S305. ;

5304, configuring one of the idle indexes to the list item to be compressed; for example, a translation table has a total of 50 indexes, wherein the index with the sequence number 0-27 has been configured with the corresponding list item, then the sequence number may be The index is configured to the latest list item to be compressed. And, the list items in the subsequent data packets are compressed using the correspondence in the translation table in the subsequent compression process. 5305, searching for a list entry stored in the compression end to find an index corresponding to the list item that has not been used for a long time (for example, the frequency of use is less than 1/3), and replacing the list item corresponding to the index with the latest list item to be compressed. ;

5306, sending the compressed list item to the decompression end;

5307. If the data packet includes only the index and there is no corresponding list item, the decompression end decompresses according to the corresponding relationship in the translation table. If the data packet includes not only the index but also the list item corresponding to the index, it is checked whether the index has been recorded in the translation table; if yes, execute S308, if no, execute S309.

5308. Replace the corresponding relationship recorded in the translation table with the newly sent index and the corresponding relationship of the list item, and decompress the subsequent data packet with the new correspondence, until a new index is sent together with the list item. When you come over, update the new index correspondence. S309. Establish a correspondence between the list item and the index in the translation table of the decompression end. Specifically, as shown in FIG. 4, in step S305, searching for a list item that has not been used for a long time can be performed as follows:

5401, defines a global variable, which is used to mark the total number of times of all index indexes in the compression of the list item of the packet stream, and defines a single variable for each list item index, and respectively marks the number of times each index is used alone.

5402, using the index in the translation table for each compression, incrementing the global variable by one. Each time an index is used, the number of individual uses corresponding to the index is increased by one;

5403, when it is necessary to find a list item that has not been used for a long time, the traversal list sequentially acquires the number of times of using the index corresponding to each list item, and if the ratio of the number of used times to the total number of used times is lower than the preset threshold, determining the If the frequency of use of the index corresponding to the number of times of use is lower than the preset threshold, the eligible condition may be reused. Correspondingly, as shown in FIG. 5, an embodiment of the present invention further provides a header compression apparatus, including: a compression determining unit 50, configured to configure a corresponding index for the list item to be compressed, Determining whether there is an available idle index; the replacing unit 52 is configured to, when the compression determining unit determines that there is no idle index, select an index whose usage frequency is lower than a preset threshold from an index that has been configured with the list item And replacing the list item corresponding to the index with the list item to be compressed, so that the list item to be compressed is compressed by using the replaced index. In the header compression device provided by the embodiment of the present invention, the compression determining unit 50 can determine whether a corresponding index needs to be configured for the list item to be compressed, and if the index needs to be configured but there is no idle index, the replacement unit 52 can Selecting an index whose frequency is lower than the preset threshold from the index that has been configured with the list item, and replacing the list item corresponding to the index whose usage frequency is lower than the preset threshold with the list item to be compressed, so that the The list item to be compressed is compressed using the replaced index. In this way, the reuse of the index can significantly reduce the memory space usage of the compression end, shortening the time taken for searching the translation table, thereby effectively improving the compression efficiency. Further, the compression determining unit 50 is further configured to determine whether it is necessary to configure a corresponding index for the list item to be compressed before determining whether there is an available idle index. Optionally, the compression determining unit 50, including a list item determining module, is configured to: determine whether the list item to be compressed has been recorded in a translation table; wherein the translation table is used to record a list item and an index Corresponding relationship; in the case that the list item to be compressed has been recorded in the translation table, it is determined that the corresponding index is not required to be configured for the list item to be compressed; In the case that there is no record in the translation table, it is determined that the corresponding index needs to be configured for the list item to be compressed. Optionally, the compression determining unit 50 further includes an index determining module, configured to determine, if the corresponding index is configured for the list item to be compressed, whether there is an idle index of the unspecified corresponding list item in the translation table. . Optionally, the replacing unit 52 is configured to: determine, in the process of compressing the list, the total number of uses of all indexes and the number of times of using each index separately; if the ratio of the number of individual uses to the total number of uses is lower than a preset threshold, determining that the frequency of use of the index corresponding to the number of times of use is lower than a preset threshold, and replacing the list item corresponding to the index with the list item to be compressed, so that the list item to be compressed is utilized The replaced index is compressed. Correspondingly, as shown in FIG. 6, the embodiment of the present invention further provides a header decompression apparatus, including: a decompression determining unit 60, configured to determine whether the same index exists in the received compressed packet and is configured to different list items in sequence. ; The decompression unit 62 is configured to decompress the compressed package with the correspondence between the last configured list item and the index, in a case where the decompression determining unit determines that the same index is sequentially configured to different list items. In the header decompression device provided by the embodiment of the present invention, the decompression determining unit 60 can determine whether the same index in the received compressed packet is configured to be different to the list item, and the same index is sequentially configured to different list items. The decompressing unit 62 can decompress the compressed package by using the mapping relationship between the last configured list item and the index, so that the use of the index can significantly reduce the memory space usage of the decompressing end, thereby shortening the time taken for searching the translation table. Effectively improve the efficiency of understanding pressure. It should be noted that, at the decompression end, a correspondence table between list items and indexes can also be maintained, which is also called a translation table. When a data packet is sent, if the data packet includes only the index and there is no corresponding list item, it indicates that the list item has appeared before, and the corresponding record is already in the translation table, so The correspondence in the table is decompressed. If the data packet includes not only the index but also the list item corresponding to the index, it can be checked whether the index has been recorded in the translation table, and if so, the index and the list item sent this time are reused for the index. Correspondingly, the corresponding relationship between the newly transmitted index and the list item should be replaced with the corresponding relationship already recorded in the translation table, and the subsequent data packet is decompressed with the new corresponding relationship until a new index and list item are added. Send it together and update the new index correspondence. While the preferred embodiments of the present invention have been disclosed for purposes of illustration, those skilled in the art will recognize that various modifications, additions and substitutions are possible, and the scope of the invention should not be limited to the embodiments described above. INDUSTRIAL APPLICABILITY As described above, a header compression method, a decompression method, and an apparatus provided by an embodiment of the present invention have the following beneficial effects: The reuse of an index can significantly reduce the memory space usage of the compression end, and shorten the cost of searching for a translation table. The time, which effectively improves the compression efficiency.

Claims

The claim method and the method for compressing the header include: determining whether there is an available idle index when the corresponding index needs to be configured for the list item to be compressed;

In the case that there is no idle index, an index whose frequency is lower than a preset threshold is selected from the index that has been configured for the list item, and the list item corresponding to the index is replaced with the list item to be compressed, so that the The list item to be compressed is compressed using the replaced index. The method according to claim 1, wherein the method further comprises: determining whether a corresponding index needs to be configured for the list item to be compressed before determining whether there is an available idle index. The method according to claim 2, wherein the determining whether the corresponding index needs to be configured for the list item to be compressed comprises:

Determining whether the list item to be compressed has been recorded in the translation table; wherein the translation table is used to record a correspondence between the list item and the index; and the list item to be compressed is already in the translation table If there is a record, it is determined that the corresponding index is not configured for the list item to be compressed; if the list item to be compressed has no record in the translation table, it is determined that the list to be compressed needs to be The index corresponding to the item configuration. The method according to claim 3, wherein the determining whether there is an available idle index comprises: determining whether there is an idle index in the translation table that does not specify a corresponding list item. The method according to claim 1, wherein the selecting an index from the index that has been configured for the list item to use the frequency below a preset threshold comprises:

During the compression of the list, the total number of uses of all indexes and the number of individual uses of each index are determined separately;

If the ratio of the number of individual uses to the total number of uses is lower than the preset threshold, determining that the frequency of use of the index corresponding to the number of individual uses is lower than the preset threshold. And a method for decompressing the header, comprising: determining whether the same index is configured to be sent to different list items in the received compressed package;

In the case where the same index is sequentially configured to different list items, the compressed package is decompressed with the correspondence between the last configured list item and the index. And a header compression device, comprising: a compression determining unit, configured to determine whether there is an available idle index if a corresponding index needs to be configured for the list item to be compressed;

a replacement unit, configured to: when the compression determining unit determines that there is no idle index, select an index whose usage frequency is lower than a preset threshold from an index that has been configured with the list item, and replace the list item corresponding to the index with The list item to be compressed, so that the list item to be compressed is compressed by using the replaced index. The apparatus according to claim 7, wherein the compression determining unit is further configured to determine whether it is necessary to configure a corresponding index for the list item to be compressed before determining whether there is an available idle index. The device according to claim 7, wherein the replacing unit is configured to: determine, in the process of compressing the list, the total number of uses of all indexes and the number of times of using each index separately;

If the ratio of the number of times of use and the total number of times of use is lower than a preset threshold, determining that the frequency of use of the index corresponding to the number of times of use is lower than a preset threshold, and replacing the list item corresponding to the index with the a list item to be compressed, so that the list item to be compressed is compressed by using the replaced index. 0. A meter decompression device, comprising: a decompression determining unit, configured to determine whether the same index exists in the received compressed packet is configured to be different list items;

The decompression unit is configured to decompress the compressed package with the correspondence between the last configured list item and the index, in a case where the decompression determining unit determines that the same index is sequentially configured to different list items.