Change search
ReferencesLink to record
Permanent link

Direct link
Efficient text compression algorithms for signaling protocols
2002 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This paper presents an algorithm, called LZBS, which is especially suited to compress text generated by reply-request protocols such as SIP, SDP and RTSP. These are protocols used to setup and maintain multimedia sessions, such as phone calls. Essential design criterions for the algorithm are minute memory- and CPU-utilization, because those resources are limited in mobile units, where compression takes place. Real data, packets of SIP and RTSP sessions, were the inputs to three candidate algorithms. Each of the algorithms has its own quality. The two competitors to LZBS, the digram coder and LZSS, are fast at compression and decompression, respectively. Neither factor turned out to be decisive of the total CPU-time. Instead, LZBS is favoured by the quick manipulation of a special kind of state, or context. In most other applications, it can be done passively. Here, the context is actively maintained in order to make full use of redundancies among packets. In doing this, improved coding and higher compression ratios are attained.

Place, publisher, year, edition, pages
2002.
Keyword [en]
Technology, Computer science, Datalogi, Compression, Compaction, Decompression, Decoding, Encoding, Texts, Text-based, protocols, Signaling protocols, Packets, Packet reordering, Contexts, Context-handling, UCS, ISO 10646, IEC 10646, UTF, UTF-8, ASCII, SIP, SDP, RTSP, HTML, LZSS, LZ77, LZ78, Arithmetic coding, Statistical coding, Dictionary coding, Digrams, Bigrams, Digram coders, Fast coders, LZBS, Pucrunch, MTF, Move to front, Move-to-front, Escape bits, Prefix bits, Literals, Indexes, Indeces, Phrases, Tokens, Offsets, Lengths, Variable-length codes, Codes, MSB, Most significant bits, Data, structures, Dictionaries, BST, Binary search trees, Linked, lists, Backskips, Backskip-arrays, Last-occurrence-arrays, Hash tables, Compression ratios, Low memory requirements, Time, factors, ROGER, Sigcomp, Greedy parsing, Optimal parsing
Keyword [sv]
Teknik
Identifiers
URN: urn:nbn:se:ltu:diva-57210ISRN: LTU-EX--02/090--SELocal ID: de5ef73c-7af3-4c78-b8ea-4a0512b88473OAI: oai:DiVA.org:ltu-57210DiVA: diva2:1030597
Subject / course
Student thesis, at least 30 credits
Educational program
Computer Science and Engineering, master's level
Examiners
Note
Validerat; 20101217 (root)Available from: 2016-10-04 Created: 2016-10-04Bibliographically approved

Open Access in DiVA

No full text

Search outside of DiVA

GoogleGoogle Scholar

Total: 1 hits
ReferencesLink to record
Permanent link

Direct link