WEKO3
アイテム
{"_buckets": {"deposit": "dd0655e9-fc66-46e1-bb55-b6a97a6958fb"}, "_deposit": {"id": "4069", "owners": [], "pid": {"revision_id": 0, "type": "depid", "value": "4069"}, "status": "published"}, "_oai": {"id": "oai:repository.dl.itc.u-tokyo.ac.jp:00004069", "sets": ["280", "329"]}, "item_7_alternative_title_1": {"attribute_name": "その他のタイトル", "attribute_value_mlt": [{"subitem_alternative_title": "T-complexityの解析と応用"}]}, "item_7_biblio_info_7": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2010-03-24", "bibliographicIssueDateType": "Issued"}, "bibliographic_titles": [{}]}]}, "item_7_date_granted_25": {"attribute_name": "学位授与年月日", "attribute_value_mlt": [{"subitem_dategranted": "2010-03-24"}]}, "item_7_degree_grantor_23": {"attribute_name": "学位授与機関", "attribute_value_mlt": [{"subitem_degreegrantor": [{"subitem_degreegrantor_name": "University of Tokyo (東京大学)"}]}]}, "item_7_degree_name_20": {"attribute_name": "学位名", "attribute_value_mlt": [{"subitem_degreename": "博士(科学)"}]}, "item_7_description_5": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "T-codes are variable-length self-synchronizing codes introduced by Titchener in 1984. T-code codewords are constructed recursively from a finite alphabet using an algorithm called T-augmentation, resulting in excellent self-synchronization properties. An algorithm called T-decomposition parses a given sequence into a series of T-prefixes, and finds a T-code set in which the sequence is encoded to a longest codeword. There are similarities and differences between T-decomposition and the conventional LZ78 incremental parsing. The LZ78 incremental parsing algorithm parses a given sequence into consecutive distinct subsequences (words) sequentially in such a way that each word consists of the longest matching word parsed previously and a literal symbol. Then, the LZ-complexity is defined as the number of words. By contrast, T-decomposition parses a given sequence into a series of T-prefixes, each of which consists of the recursive concatenation of the longest matching T-prefix parsed previously and a literal symbol, and it has to access the whole sequence every time it determines a T-prefix. Alike to the LZ-complexity, the T-complexity of a sequence is defined as the number of T-prefixes, however, the T-complexity of a particular sequence in general tends to be smaller than the LZ-complexity. In the first part of the thesis, we deal with our contributions to the theory of T-codes. In order to realize a sequential determination of T-prefixes, we devise a new T-decomposition algorithm using forward parsing. Both the T-complexity profile obtained from the forward T-decomposition and the LZ-complexity profile can be derived in a unified way using a differential equation method. The method focuses on the increase of the average codeword length of a code tree. The obtained formulas are confirmed to coincide with those of previous studies. The magnitude of the T-complexity of a given sequence s in general indicates the degree of randomness. However, there exist interesting sequences that have much larger T-complexities than any random sequences. We investigate the maximum T-complexity sequences and the maximum LZ-complexity sequences using various techniques including those of the test suite released by the National Institute of Standards and Technology (NIST) of the U.S. government, and find that the maximum T-complexity sequences are less random than the maximum LZ-complexity sequences. In the second part of the thesis, we present our achievements in terms of application. We consider two applications -- data compression and randomness testing. First, we propose a new data compression scheme based on T-codes using a dictionary method such that all phrases added to a dictionary have a recursive structure similar to T-codes. Our data compression scheme can compress any of the files in the Calgary Corpus more efficiently than previous schemes based on T-codes and the UNIX compress, a variant of LZ78 (LZW). Next, we introduce a randomness test based on the T-complexity. Recently, the Lempel-Ziv (LZ) randomness test based on the LZ-complexity was officially excluded from the NIST test suite. This is because the distribution of P-values for random sequences of length 106, the most common length used, is strictly discrete in the case of the LZ-complexity. Our test solves this problem because the T-complexity features an almost ideal uniform continuous distribution of P-values for random sequences of length 106. The proposed test outperforms the NIST LZ test, a modified LZ test proposed by Doganaksoy and Göloglu, and all other tests included in the NIST test suite, in terms of the detection of undesirable pseudorandom sequences generated by a multiplicative congruential generator (MCG) and non-random byte sequences Y = Y0, Y1, Y2, ···, where Y3i and Y(3i+1) are random, but Y(3i+2) is given by Y3i + Y(3i+1) mod 28.", "subitem_description_type": "Abstract"}]}, "item_7_dissertation_number_26": {"attribute_name": "学位授与番号", "attribute_value_mlt": [{"subitem_dissertationnumber": "甲第26141号"}]}, "item_7_full_name_3": {"attribute_name": "著者別名", "attribute_value_mlt": [{"nameIdentifiers": [{"nameIdentifier": "9382", "nameIdentifierScheme": "WEKO"}], "names": [{"name": "濱野, 健二"}]}]}, "item_7_identifier_registration": {"attribute_name": "ID登録", "attribute_value_mlt": [{"subitem_identifier_reg_text": "10.15083/00004060", "subitem_identifier_reg_type": "JaLC"}]}, "item_7_select_21": {"attribute_name": "学位", "attribute_value_mlt": [{"subitem_select_item": "doctoral"}]}, "item_7_subject_13": {"attribute_name": "日本十進分類法", "attribute_value_mlt": [{"subitem_subject": "007", "subitem_subject_scheme": "NDC"}]}, "item_7_text_22": {"attribute_name": "学位分野", "attribute_value_mlt": [{"subitem_text_value": "Science (Kagaku)(科学)"}]}, "item_7_text_24": {"attribute_name": "研究科・専攻", "attribute_value_mlt": [{"subitem_text_value": "Department of Complexity Science and Engineering, Graduate School of Frontier Sciences (新領域創成科学研究科複雑理工学専攻)"}]}, "item_7_text_27": {"attribute_name": "学位記番号", "attribute_value_mlt": [{"subitem_text_value": "博創域第558号"}]}, "item_7_text_36": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"subitem_text_value": "Thesis"}]}, "item_7_text_4": {"attribute_name": "著者所属", "attribute_value_mlt": [{"subitem_text_value": "東京大学大学院新領域創成科学研究科基盤科学研究系複雑理工学専攻"}, {"subitem_text_value": "Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, The University of Tokyo"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Hamano, Kenji"}], "nameIdentifiers": [{"nameIdentifier": "9381", "nameIdentifierScheme": "WEKO"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2017-06-01"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "K-02429.pdf", "filesize": [{"value": "3.0 MB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_free", "mimetype": "application/pdf", "size": 3000000.0, "url": {"label": "本文(fulltext)", "url": "https://repository.dl.itc.u-tokyo.ac.jp/record/4069/files/K-02429.pdf"}, "version_id": "6d4d2343-d0ac-4a18-a252-019c18d0b4d2"}]}, "item_keyword": {"attribute_name": "キーワード", "attribute_value_mlt": [{"subitem_subject": "T-code", "subitem_subject_scheme": "Other"}, {"subitem_subject": "T-complexity", "subitem_subject_scheme": "Other"}, {"subitem_subject": "T-decomposition", "subitem_subject_scheme": "Other"}, {"subitem_subject": "Lempel-Ziv complexity", "subitem_subject_scheme": "Other"}, {"subitem_subject": "NIST SP 800-22", "subitem_subject_scheme": "Other"}, {"subitem_subject": "data compression", "subitem_subject_scheme": "Other"}, {"subitem_subject": "complexity profile", "subitem_subject_scheme": "Other"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "thesis", "resourceuri": "http://purl.org/coar/resource_type/c_46ec"}]}, "item_title": "Analysis and Applications of the T-complexity", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Analysis and Applications of the T-complexity"}]}, "item_type_id": "7", "owner": "1", "path": ["280", "329"], "permalink_uri": "https://doi.org/10.15083/00004060", "pubdate": {"attribute_name": "公開日", "attribute_value": "2012-11-01"}, "publish_date": "2012-11-01", "publish_status": "0", "recid": "4069", "relation": {}, "relation_version_is_last": true, "title": ["Analysis and Applications of the T-complexity"], "weko_shared_id": null}
Analysis and Applications of the T-complexity
https://doi.org/10.15083/00004060
https://doi.org/10.15083/00004060d06d46f0-eda2-4649-b776-d83814a1f4fc
名前 / ファイル | ライセンス | アクション |
---|---|---|
本文(fulltext) (3.0 MB)
|
|
Item type | 学位論文 / Thesis or Dissertation(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2012-11-01 | |||||
タイトル | ||||||
タイトル | Analysis and Applications of the T-complexity | |||||
言語 | ||||||
言語 | eng | |||||
キーワード | ||||||
主題 | T-code | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | T-complexity | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | T-decomposition | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | Lempel-Ziv complexity | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | NIST SP 800-22 | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | data compression | |||||
主題Scheme | Other | |||||
キーワード | ||||||
主題 | complexity profile | |||||
主題Scheme | Other | |||||
資源タイプ | ||||||
資源 | http://purl.org/coar/resource_type/c_46ec | |||||
タイプ | thesis | |||||
ID登録 | ||||||
ID登録 | 10.15083/00004060 | |||||
ID登録タイプ | JaLC | |||||
その他のタイトル | ||||||
その他のタイトル | T-complexityの解析と応用 | |||||
著者 |
Hamano, Kenji
× Hamano, Kenji |
|||||
著者別名 | ||||||
識別子 | 9382 | |||||
識別子Scheme | WEKO | |||||
姓名 | 濱野, 健二 | |||||
著者所属 | ||||||
著者所属 | 東京大学大学院新領域創成科学研究科基盤科学研究系複雑理工学専攻 | |||||
著者所属 | ||||||
著者所属 | Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, The University of Tokyo | |||||
Abstract | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | T-codes are variable-length self-synchronizing codes introduced by Titchener in 1984. T-code codewords are constructed recursively from a finite alphabet using an algorithm called T-augmentation, resulting in excellent self-synchronization properties. An algorithm called T-decomposition parses a given sequence into a series of T-prefixes, and finds a T-code set in which the sequence is encoded to a longest codeword. There are similarities and differences between T-decomposition and the conventional LZ78 incremental parsing. The LZ78 incremental parsing algorithm parses a given sequence into consecutive distinct subsequences (words) sequentially in such a way that each word consists of the longest matching word parsed previously and a literal symbol. Then, the LZ-complexity is defined as the number of words. By contrast, T-decomposition parses a given sequence into a series of T-prefixes, each of which consists of the recursive concatenation of the longest matching T-prefix parsed previously and a literal symbol, and it has to access the whole sequence every time it determines a T-prefix. Alike to the LZ-complexity, the T-complexity of a sequence is defined as the number of T-prefixes, however, the T-complexity of a particular sequence in general tends to be smaller than the LZ-complexity. In the first part of the thesis, we deal with our contributions to the theory of T-codes. In order to realize a sequential determination of T-prefixes, we devise a new T-decomposition algorithm using forward parsing. Both the T-complexity profile obtained from the forward T-decomposition and the LZ-complexity profile can be derived in a unified way using a differential equation method. The method focuses on the increase of the average codeword length of a code tree. The obtained formulas are confirmed to coincide with those of previous studies. The magnitude of the T-complexity of a given sequence s in general indicates the degree of randomness. However, there exist interesting sequences that have much larger T-complexities than any random sequences. We investigate the maximum T-complexity sequences and the maximum LZ-complexity sequences using various techniques including those of the test suite released by the National Institute of Standards and Technology (NIST) of the U.S. government, and find that the maximum T-complexity sequences are less random than the maximum LZ-complexity sequences. In the second part of the thesis, we present our achievements in terms of application. We consider two applications -- data compression and randomness testing. First, we propose a new data compression scheme based on T-codes using a dictionary method such that all phrases added to a dictionary have a recursive structure similar to T-codes. Our data compression scheme can compress any of the files in the Calgary Corpus more efficiently than previous schemes based on T-codes and the UNIX compress, a variant of LZ78 (LZW). Next, we introduce a randomness test based on the T-complexity. Recently, the Lempel-Ziv (LZ) randomness test based on the LZ-complexity was officially excluded from the NIST test suite. This is because the distribution of P-values for random sequences of length 106, the most common length used, is strictly discrete in the case of the LZ-complexity. Our test solves this problem because the T-complexity features an almost ideal uniform continuous distribution of P-values for random sequences of length 106. The proposed test outperforms the NIST LZ test, a modified LZ test proposed by Doganaksoy and Göloglu, and all other tests included in the NIST test suite, in terms of the detection of undesirable pseudorandom sequences generated by a multiplicative congruential generator (MCG) and non-random byte sequences Y = Y0, Y1, Y2, ···, where Y3i and Y(3i+1) are random, but Y(3i+2) is given by Y3i + Y(3i+1) mod 28. | |||||
書誌情報 | 発行日 2010-03-24 | |||||
日本十進分類法 | ||||||
主題 | 007 | |||||
主題Scheme | NDC | |||||
学位名 | ||||||
学位名 | 博士(科学) | |||||
学位 | ||||||
値 | doctoral | |||||
学位分野 | ||||||
Science (Kagaku)(科学) | ||||||
学位授与機関 | ||||||
学位授与機関名 | University of Tokyo (東京大学) | |||||
研究科・専攻 | ||||||
Department of Complexity Science and Engineering, Graduate School of Frontier Sciences (新領域創成科学研究科複雑理工学専攻) | ||||||
学位授与年月日 | ||||||
学位授与年月日 | 2010-03-24 | |||||
学位授与番号 | ||||||
学位授与番号 | 甲第26141号 | |||||
学位記番号 | ||||||
博創域第558号 |