||||||||||||||||||||||l|||l|||||||||||ll||||||||||||||||||||||||||||||||l|||||||||||||l|||
`
`US 200502?6259A1
`
`(19} United States
`(12) Patent Application Publication (10) Pub. No.: US 2005/0276259 A1
`
`
` Nakabayaslti ct al. (43) Pub. Date: Dec. 15, 2005
`
`(54} WIRELESS PACKET TRANSFER
`APPARATUS AND METHOD
`
`Publication Classification
`
`(76}
`
`Inventors: Sumie Nakabayashi, Kokubunji (JP);
`Masam Adachi, Kodaira (JP)
`
`Correspondence Address:
`MATTINGLY STANGER MALUR &
`BRUNDIDGE, PLC.
`’
`1800 DIAGOI'GAL ROAD
`SUITE 370
`ALEXANDRIA VA 22314 (US)
`’
`"
`111145388
`
`(2” Appl. No;
`
`{23
`
`Filed:
`
`Jun. 7‘ 2005
`
`(30}
`
`Foreign Application Priority Data
`
`Jun. 9, 2004 (JP) 2004471153
`
`Int. CL" ....................................................... H04Q 7;“00
`{51)
`(52) U.S. CI. 370349
`
`(57)
`
`ABS'I'RAC'I'
`
`.
`'
`. a
`.
`,
`.
`.
`i
`,
`»
`,
`A ,
`. n error correctionlcncodmg rate selection table is proVIdecl
`In an error correction processmg LII'Jll ot a packet transier
`apparatus, and tire table stores an error correction encoding
`rate preset to maintain a destred Q08 in correspondence wrth
`a protocol type and an application type. When a transmission
`packet
`is. transferred to a wireless transmission path, an
`encoding control unit jpdges the protocol type and applica-
`tion type of
`a transmission packet
`from a header of the
`transmission packet, in accordance with a judgement result
`and the error correction encoding rate selection table, an
`error correction encoding rate is selected, and the transmis—
`sion packet
`is subjected to error correction encoding and
`transferred.
`
`WIRELESS
`APPARATUS
`
` RD2
`WIRELESS
`APPARATUS
`
`RDt
`
`PT1
`
`
`
`
`PACKET
`
`TRANSFER
`APPARATUS
`
` PACKET
`
`TRANSFER
`APPARATUS
`
`PT2
`
`
`
`
`
`NW1
`
`NW2
`
`Microsoft
`
`Ex. 1028 - Page 1
`
`Microsoft
`Ex. 1028 - Page 1
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 1 of 14
`
`7..
`
`11A0152(07..
`
`:0;
`
`z<._
`
`
` .._2u.S.....................
`m«32
`
`
`nu‘--.....-,,,“--
`
`._.wv_O<n_
`
`wDF<m<mm<Imum25”....
`
`._.m_x0¢m
`
`Emu—mzfiut.
`
`w3h¢¢<mm<
`
`NOW—wwmem—S;
`
`EOE
`
`wqur—(n—nE
`
`wwmgmmS:
`
`m3._.<m_<n_m<
`
`
`
`FOE
`
`Ch—
`
`Microsoft
`
`Ex. 1028 - Page 2
`
`Microsoft
`Ex. 1028 - Page 2
`
`
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 2 of 14
`
`US 2005;0276259 A1
`
`FWCEJZ
`
`PTLPTZ
`
`I:
`
`2
`
`WIRELESSINTERFACEUNIT
`
`ROUTECONTROLUNIT
`
`ERRORCORRECTIONPROCESSINGUNIT
`
`Z:
`
`3 8qL
`
`L
`CE
`UJ
`’—
`
`WIRELESS
`APPARATUS
`
`Microsoft
`
`Ex. 1028 - Page 3
`
`EX(
`
`I O 3I
`
`...
`LL!
`
`Microsoft
`Ex. 1028 - Page 3
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 3 of 14
`
`US 200501276259 A1
`
`FIG3
`
`ERROR CORRECTION
`ENCODNG UNIT
`
` CV.22:
`
`mugmmhzfimmdmg
`
`
`
`ow:23onmeEmmmdm?
`
`ENCODING
`CONTROL UNIT
`
`ERROR CORRECTION
`DECODING UNIT
`
`
`
`ON.:23401.500WSOEON:2355.200wFDOm
`
`mnmmwHa4WMmR2D:D:30mW
`
`W
`
`Microsoft
`
`Ex. 1028 - Page 4
`
`Microsoft
`Ex. 1028 - Page 4
`
`
`
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 4 of 14
`
`US 2005;0276259 A1
`
`FIIGA-
`
`ERROR CONTROL UNIT
`
`332
`
`G1 :1 +DZ+Da +D5 +05
`
`
`
`
`ENCODING
`OUTPUT
`
`FHGB
`
`
`
`
`
`
`
`
`
`
`
`ENCODING AND PUNCTURE PATI'ERN
`
`Microsoft
`
`Ex. 1028 - Page 5
`
`Microsoft
`Ex. 1028 - Page 5
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 5 of 14
`
`US 2005;0276259 A1
`
`FflGfi
`
`
`
`FRAME
`momma
`LENGTH
`RATE
`INFORMATION INFORMATION
`n.“
`
`FCS
`
`\
`
`
`
`SYNCBIT
`
`
`
`ENCODiNG RATEFIELD
`
`FIXED
`ENCODiNG RATE
`
`:
`5
`
`
`
`FRAME DATAFILED
`
`VARIABLE ENCODING RATE
`
`
`
`ERROR CORRECTION ENCODING HATE FRAME FORMAT
`
`Microsoft
`
`Ex. 1028 - Page 6
`
`Microsoft
`Ex. 1028 - Page 6
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 6 of 14
`
`US 2005;0276259 A1
`
`o
`
`FIG.7A _
`
`16
`
`TRANSMISSION SOURCE
`PORT NUMBER
`
`DESTINATION PORT NUMBER
`
`SEOU ENCE NUMBER
`
`ACKNOWLEDGMENT NUMBER
`
`
`
`
`
`
`
`0
`
`
`
`OFFSET
`
`RESERVED CODE BIT
`
`CH ECK SUM
`
`
`
`WINDOWSIZE
`
`URGENT POINTER
`
`TCP HEADER FORMAT
`
`FIG?!
`
`TRANSMISSION SOURCE
`PORT NUMBER
`
`15
`
`
`DESTINATION PORT NUMBER
`
`PACKET LENGTH
`
`CHECK SUM
`
`UDP H EADER FORMAT
`
`FIG.7C
`16
`
`
`SERVICETYPE
`
`
`
`_
`
`PACKET LENGTH
`
`
`
`IDENTIFIER
`
`FLAG
`
`FRAGMENT OFFSET
`
`UFETIME
`
`PROTOCOL
`NUMBER
`
`HEADER CHECKSUM
`
`TRANSMISSION SOURCE IPADDRESS
`
`DESTINATION PADDRESS
`
`HEADERM LENGTH
`
`
`
`
`OPTION
`
`IP HEADER FORMAT
`
`PADDING
`
`31 (BIT)
`
`
`
`
`
`
`31 (BIT)
`
`
`
`
`
`
`31 (BID
`
`
`
`
`
`
`
`
`Microsoft
`
`Ex. 1028 - Page 7
`
`Microsoft
`Ex. 1028 - Page 7
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 7 of 14
`
`US 2005;0276259 A1
`
`FIGS
`
`PROTOCOL
`NUMBER
`PORT NUMBER
`_—
`_—
`m
`17 (UDP)
`
`ENCODING RATE
`
`1l2
`
`DEFAULT
`
`—
`FIRST EXAMPLE OF
`-
`ERROR CORRECTION ENCODING RATE SELECTION TABLE
`
`[F IG . 9
`
`__—
`__—
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`FIRST EXAMPLE OF DECODING CONTROL TABLE
`
`FIGJB
`
`
`
`
`
`——_—
`___“
`———-m
`
`
`
`
`
`
`DEFAULT —— memo
`SECOND EXAMPLE OF DECODING CONTROL TABLE
`
`
`
`
`Microsoft
`
`Ex. 1028 - Page 8
`
`Microsoft
`Ex. 1028 - Page 8
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 3 of 14
`
`US 2005;0276259 A1
`
`FIG.10
`
`10a
`
`
`
`TRANSMISSION PACKET
`INPUT ?
`
`
`YES
`
`READ HEADER
`
`10b
`
`1OC
`
`PORT NUMBER EXIST ?
`
`
`
`
`SELECT
`CORRESPONDING
`ENCODING RATE
`
`
`SELECT
`PROTOCOL NUMBER
`CORRESPONDING
`
`
`EXIST ?
`ENCODING RATE
`
`
`
`
`N0
`
`
`
`SELECT ENCODING RATE
`(DEFAULT)
`
`1 0
`
`9
`
`Microsoft
`
`Ex. 1028 - Page 9
`
`Microsoft
`Ex. 1028 - Page 9
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 9 of 14
`
`US 2005;0276259 A1
`
`FIGH
`
`RECEIVING PACKET
`INPUT ?
`
`EXECUTE
`CORRESPONDING
`ERROR FRAME
`PROCESSING
`
`
`
`PROTOEX?é—TN.}JMBER
`
`
`
`EXECUTE
`CORRESPONDING
`ERROR FRAME
`PROCESSING
`
`EXECUTE
`
`
`ERROR FRAME PROCESSING
`
`(DEFAULT)
`
`
`
`Microsoft
`
`Ex. 1028 - Page 10
`
`Microsoft
`Ex. 1028 - Page 10
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 10 0f 14
`
`US 2005;0276259 A1
`
`FIG.12
`
`ESTIMATED BER
`
`ENCODING RATE
`
`
`
`Amman
`M
`
`aomowan
`
`SECOND EXAMPLE OF
`ERROR CORRECTION ENCODING RATE SELECTION TABLE
`
`FIG.13
`
`TRANSMISSION PACKET
`INPUT ?
`
`YES
`
`ACQUIRE ESTIMATED BER
`
`13c
`
`SELECT
`ENCODING RATE 1 I2
`
`
`
`
`
`
`SELECT
`ENCODING RATE 7 i 8
`
`
` SELECT
`
`
`ENCODING RATE 7 18
`
`Microsoft
`
`Ex. 1028 - Page 11
`
`Microsoft
`Ex. 1028 - Page 11
`
`

`

`Patent Application Public
`
`ation Dec. 15, 2005
`
`Sheet 11 0f 14
`
`US 20030276259 A1
`
`02—Doozm02—DOUZM 36$
`
`GZWDFMMZN"OZ—DOOZM
`
`
`
`
`
`
`
`EafimemomuEm:$30..mz<1&3Euro:mEEmo“EmmamEmmwmwmzJMwmgummm
`
`
`
`
`
`m;m<._.ZOFDmfiwmw._.<mOZEOUZm—ZOFOMIEOOION—mm“.0m_._n=2<xwDEEP
`
`
`
`
`
`
`
`.
`
`Microsoft
`
`Ex. 1028 - Page 12
`
`Microsoft
`Ex. 1028 - Page 12
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 12 0f 14
`
`US 200501276259 A1
`
`158
`
`TRANSMISSION PACKET
`INPUT ?
`
`
`YES
`
`
`
`
`
`READ HEADER
`
`ACQUIRE ESTIMATED BER
`
`FIG-15
`
`15h
`
`BERzA ?
`
`BER>B ?
`
`
`SELECT
`ENCODING RATE 1 f2
`
`YES
`
`
`
`
`
`
` SELECT
`
`ENCODING RATE 7 I 8
`
`
`SELECT
`COFIRESPOND
`PORT NUMBER EXIST '?
`
`ENCODING FIATE
`
`
` ING
`
` ING
`
`
`
`SELECT
`
`PROTOCOL NUMBER
`
`CORRESPOND
`
`EXIST ?
`ENCODING RATE
`
`
`
`NO
`
`SELECT ENCODING RATE
`(DEFAULT)
`
`15m
`
`Microsoft
`
`Ex. 1028 - Page 13
`
`Microsoft
`Ex. 1028 - Page 13
`
`

`

`Patent Application Publication Dec. 15, 2005 Sheet 13 0f 14
`
`US 20030276259 A1
`
`5.9m
`
`
`
`lmqgamgomxw
`
`
`
`Ems—:258.55
`
`oExm
`
`oBrackmm8Bmmmo
`
`at.
`
`m
`
`.5onOzfimomm
`
`
`
`m..51;
`
`DE.
`
`
`
`mum$.52meIDEN—
`
`wm>
`
`
`
`wtmm>
`
`
`
`lEQmEmSomvm.inmgqu2.20%
`
`
`
`
`
`
`
`
`
`Oznow—mmrEwmoa81.30wm>
`
`
`
`3585whnomxm.
`
`mumngkm506m.
`
`
`
`UK._.
`
`Microsoft
`
`Ex. 1028 - Page 14
`
`Microsoft
`Ex. 1028 - Page 14
`
`
`
`
`
`
`

`

`Patent Application Public
`
`ation D
`
`cc. 15
`
`, 2005
`
`She
`
`ct 14 0f 14
`
`US 20030276259 A1
`
`GZEOOZm
`
`WEE
`
`6VEmmy
`
`m_..O_n_
`
`w
`
`AOOOPONE
`
`
`wzaoozmGzEOozmwzfiouzmzo_h<o_._mn_<
`ae<"Em:2Amum:Enfime20.2582.mmmssz
`
`
`
`
`
`mqu5.5..m2:OEOmmm.EOm
`
`mmmEDz
`
`FJdewD
`
`Microsoft
`
`Ex. 1028 - Page 15
`
`
`
`
`
`mind?20_._.Om_._m_mwhimOZEOOZMZOEOMEEOOEOmEm“*0mjasixm.IHEDOu
`
`
`
`
`
`Microsoft
`Ex. 1028 - Page 15
`
`

`

`US 2005/0276259 A1
`
`Dec. 15, 2005
`
`WIRELESS PACKET TRANSFER APPARATUS AND
`METHOD
`
`INCORPORATION BY REFERENCE
`
`[0001] The present application claims priority from J apa-
`nese application JP2004-171153 filed on Jun. 9, 2004, the
`content of which is hereby incorporated by reference into
`this application.
`
`BACKGROUND ()I’ THE INVENTION
`
`[0002]
`
`1.
`
`l-‘ield of the Invention
`
`[0003] The present invention relates to a circuit switched
`wireless communication system such as an anti-disaster
`wireless communication system, and more particularly to a
`wireless packet transfer apparatus for transferring a packet
`by wireless by using Internet Protocol (IP).
`
`[0004]
`
`2. Description of the Related Art
`
`[0005] With the recent advancement of It) networks, vari-
`ous applications are provided by using IP. It is expected that
`seamless communications of IP are realized even among
`different
`types of networks such as wired and wireless
`networks, by adopting new technologies such as mobile
`IPv6 (IP version 6).
`
`[0006] A public utility wireless communication system
`such as an anti-disaster wireless communication system
`generally adopts a circuit switching system. Using IP for
`wireless communication systems of this type is associated
`with the following problems. The first problem is that
`it
`takes a long time for data transmission because a transmis~
`sion bandwidth is narrow {about 25.6 Kbitis at a maximum
`transmission speed). The second problem is a throughput
`lowered by re-transmission caused by packet loss and a
`reduction in a window size of Transmission Control Protocol
`
`('I‘Cl’) and the Like, because of a high bit error rate (10'3 to
`10“) of a transmission line. The third problem is a fixed
`error correction encoding rate of each call because of circuit
`switching. Since a predetermined error correction encoding
`is performed for data transmission even in a good quality
`state of a wireless propagation path, the transmission band-
`width cannot be used elficiently. If moving image transmis-
`sion is performed by using, for example, Moving Picture
`Experts Group 4 (MPEG4), this transmission is processed
`like file transfer which does not permit any error, although
`MPEG4 itself has an error proof function of about 10‘3 to
`10‘". The communication service quality such as a moving
`image display rate is therefore degraded. The fourth problem
`is that if a router or the like is used as an interface between
`wireless apparatus and a network, an error packet is uncon-
`ditionally discarded because of a data link protocol of the
`router. Therefore, even if an application having a bit error
`permission function, such as MPEG4, is used, the whole
`packet is lost because of a bit error so that the image quality
`is degraded.
`
`[0007] As technologies capable of solving the first prob-
`lem, header compression technologies are known. The
`header compression technologies are described in the docu-
`ments Request for Comment (RFC) compiling standardized
`specifications of Internet Engineering Task Force (IETF).
`TCPt'IP header compression technologies are written in
`Rl"C1144, and lPt’UDPi’R’l'P header compression technolo-
`gies are written in RFCZSUS. Although these header com-
`
`pression technologies are effective for Voice over IP (VoIP)
`packets having a short packet
`length,
`they provide less
`compression effects for long packets used in most of com~
`municalions. As the technologies for solving the second
`problem, wireless 'I‘CP has been proposed.
`In a gateway
`apparatus disposed between a wired section and a wireless
`section,
`the wireless TCP performs division of a TCP
`connection and proxy re-transmission of TCP by mounting
`an agent. However, the wireless TCP cannot provide a TCP
`connection of endsto-end and the mount of an agent
`is
`complicated.
`
`In mobile communications, the technologies have
`[0008]
`been proposed in which wireless parameters for error cor-
`rection, modulation and the like are adaptively set in accor-
`dance with Quality of Service (QoS) (e.g., refer to JP-A-
`2003-2594417).
`
`[0009] The technologies described in JP—A~2003—259447
`are directed to wide band mobile communications over 2
`Mbps or faster ultra high speed mobile packet transmission
`systems, and do not consider a circuit switched wireless
`packet transmission system having a relatively low trans-
`mission speed to which the present invention is applied.
`
`SUMMARY OF THE INVENTION
`
`[0010] The present invention has bee made in the above—
`described circumstance, and an object of the present inven-
`tion is to provide a wireless packet transfer apparatus and
`method capable of solving the above-described third and
`fourth problems in which although a circuit switched system
`is used, a packet can be transmitted by wireless at a high
`throughput while a desired Quality ol" Service (008) is
`maintained for each packet.
`
`In order to achieve the above object, a first inven-
`[0011]
`tion provides a wireless packet transfer apparatus to be used
`in a circuit switched wireless communication system for
`transmitting a packet among a plurality of networks via a
`wireless transmission path has a table for storing a plurality
`of transmission conditions expected when a packet is trans—
`mitted and a corresponding error correction encoding rate
`preset to each of the conditions to obtain a desired commu-
`nication service quality. Transmission conditions of each
`transmission packet are judged. An error correction encod-
`ing rate is selected in accordance with the judged transmis-
`sion conditions and the storage information in the table, and
`the transmission packet
`is subjected to error correction
`encoding at the selected error correction encoding rate and
`transferred to the wireless transmission line.
`
`In one example, the table uses, as expected trans-
`[0012]
`mission conditions, at least one of a packet protocol type, a
`packet application type and a transmission quality of the
`wireless transmission path, and stores the expected trans-
`mission conditions and a corresponding error correction
`encoding rate preset to each of the expected transmission
`conditions to obtain a desired communication service quality
`under the expected transmission conditions.
`
`[0013] According to the first invention, therefore, when a
`packet sent form a network is transferred to a wireless
`transmission path, the transmission conditions are judged in
`the packet unit. An error correction encoding rate corre—
`sponding to the judged transmission condition is selected
`from the table. The packet is subjected to error correction
`
`Microsoft
`
`Ex. 1028 - Page 16
`
`Microsoft
`Ex. 1028 - Page 16
`
`

`

`US 2005/0276259 A1
`
`Dec. 15, 2005
`
`encoding at the selected error correction encoding rate and
`transferred to the wireless transmission path. Namely, each
`packet is transmitted by wireless after being subjected to
`error correction encoding at an error correction encoding
`rate necessary for maintaining a desired communication
`service quality and corresponding to the transmission con-
`dition of each packet, e.g., the packet protocol
`type, the
`packet application type or the transmission quality of the
`wireless transmission line. Accordingly, a wireless packet
`transmission of a high throughput can be realized even in a
`circuit switched system, while a desired communication
`Service quality of each packet is maintained and the limited
`narrow transmission band is efiectively utilized.
`
`[0014] A second invention provides a wireless packet
`transfer apparatus to be used in a circuit switched wireless
`communication system for transmitting a packet among a
`plurality of networks via a wireless transmission path has a
`table for storing a plurality of transmission conditions
`expected when a packet is transmitted and a corresponding
`error processing method preset to each of the transmission
`conditions to obtain a desired communication service qual-
`ity,
`the method processing a reception packet having an
`error. Transmission conditions of each packet received via
`the wireless transmission path are judged. An error process-
`ing method is selected in accordance with the judged trans—
`mission condition and the storage information in the table,
`and the reception packet is subjected to error processing by
`the selected error processing method and transferred to the
`wireless transmission path.
`
`In one example, the table uses, as expected trans
`[0015]
`mission conditions, at least one of a packet protocol type and
`a packet application type, and stores the expected transmis-
`sion condition and corresponding information for designat-
`ing whether a reception packet having an error is transferred
`or discarded.
`
`the table uses, as expected
`In another example,
`[0016]
`transmission conditions, a packet protocol type, a packet
`application type and a desired communication quality set to
`each of the types, and stores the expected transmission
`conditions and corresponding information for designating
`whether a reception packet having an error is transferred or
`discarded.
`
`[0017] According to the second invention, therefore, when
`a packet sent form a network is transferred to a wireless
`transmission path, the transmission condition is judged in
`the packet unit. Error processing for an error packet
`is
`performed in correspondence with the judged transmission
`condition. Therefore, for example, if a reception packet has
`an error and if a protocol or application itselt‘of the reception
`packet has an error correction ability or an error permission
`ability,
`the reception packet is transferred directly to the
`network. If the protocol or application itself of the reception
`packet does not have an error correction ability or an error
`permission ability, the reception packet is discarded. In this
`manner, the error processing becomes possible by consid-
`ering the function of the protocol or application of a packet.
`Therefore, also in the second invention, a wireless packet
`transmission of a high throughput can be realized even in a
`circuit switched system, while a desired communication
`service quality of each packet is maintained.
`
`In summary, aceording to the inventions, the table
`[0018]
`stores in advance the packet transmission conditions and a
`
`corresponding error correction encoding rate or error pro-
`cessing method preset to each of the expected transmission
`conditions to obtain a desired communication service quality
`under the expected transmission conditions. The error cor-
`rection encoding or error processing is executed for each
`transmissionireception packet in accordance with the trans-
`mission conditions and the storage information in the table.
`
`[0019] Therefore, according to the present invention, a
`wireless packet transfer apparatus can be provided which
`can transfer a wireless packet efliciently even in a circuit
`switched system, while a desired communication service
`quality of each packet is maintained.
`
`features and advantages of the
`[0020] Other objects,
`invention will become apparent from the following descrip-
`tion of the embodiments of the invention taken in conjunc-
`tion with the accompanying drawings.
`
`BRIEF IJESCRIPI'ION OF THE DRAWINGS
`
`[0021] FIG. 1 is a schematic diagram showing the con-
`figuration of an anti-disaster wireless communication sys-
`tem using wireless packet transfer apparatuses according to
`a first embodiment of the present invention.
`
`[0022] FIG. 2 is a functional block diagram illustrating
`the functions of the wireless packet transfer apparatus of the
`first embodiment.
`
`[0023] FIG. 3 is a block diagram showing the structure of
`an error correction processing unit of the wireless packet
`transfer apparatus shown in FIG. 2.
`
`[0024] FIG. 4 is a diagram showing an example of the
`structure of an error correction encoding unit of the error
`correction processing unit shown in FIG. 3.
`
`[0025] FIG. 5 is a diagram showing the relation between
`an encoding rate and a puncture pattern of the error correc-
`tion encoding unit shown in FIG. 4.
`
`[0026] FIG. 6 is a diagram showing the format of an error
`correction encoding frame generated by the error correction
`processing unit shown in FIG. 3.
`
`[0027] FIGS. 7A, 7B and 7C are diagrams showing the
`formats of a TCP header, a UDP header and an IP header.
`
`[0028] FIG. 8 is a diagram showing the structure of an
`error correction rate selection table provided in the error
`correction processing unit shown in FIG. 3.
`
`[0029] FIG. 9 is a diagram showing the structure of a
`decoding control
`table provided in the error correction
`processing unit shown in FIG. 3.
`
`[0030] FIG. 10 is a flow chart illustrating an error cor—
`rcction encoding rate selection control procedure and its
`contean by the error correction processing unit shown in
`FIG. 3.
`
`illustrating a decoding
`[0031] FIG. 11 is a flow chart
`control procedure and its contents by the error correction
`processing unit shown in FIG. 3.
`
`[0032] FIG. 12 is a diagram showing the structure of an
`error correction encoding rate selection table provided in a
`wireless packet transfer apparatus according to a second
`embodiment of the present invention.
`
`Microsoft
`
`Ex. 1028 - Page 17
`
`Microsoft
`Ex. 1028 - Page 17
`
`

`

`US 2005/0276259 A1
`
`Dec. 15, 2005
`
`[0033] FIG. 13 is a flow chart illustrating an error cor-
`rection encoding rate selection control procedure and its
`contents by the error correction processing unit shown in
`FIG. 12.
`
`[0034] FIG. 14 is a diagram showing the structure of an
`error correction encoding rate selection table provided in a
`wireless packet
`transfer apparatus according to a third
`embodiment of the present invention.
`
`[0035] FIG. 15 is a flow chart illustrating an error corv
`rection encoding rate selection control procedure and its
`contents by the error correction processing unit shown in
`FIG. 14.
`
`[0036] FIG. 16 is a diagram showing the structure of an
`error correction encoding rate selection table provided in a
`wireless packet
`transfer apparatus according to a fourth
`embodiment of the present invention.
`
`illustrating a decoding
`flow chart
`[0037] FIG. 17 is a
`control procedure and its contents by the error correction
`processing unit shown in FIG. 16.
`
`[0038] FIG. 18 is a diagram showing the structure of an
`error correction rate selection table provided in the error
`correction processing unit of a wireless packet
`transfer
`apparatus according to another embodiment of the present
`invention.
`
`DESCRIPTION OF THE EMBODIMENTS
`
`First Embodiment
`
`[0039] FIG. 1 is a schematic diagram showing the con-
`llguration of an anti-disaster Wireless communication sys-
`tem using wireless packet transfer apparatuses according to
`the first embodiment of the present invention.
`
`[0040] This system has first and second networks NW1
`and NW2. 'I‘hese networks NW1 and NW2 are installed, for
`example,
`in prefectural or municipal oflices of a self-
`govcrning body, and configured by wired Local Area Net—
`work (LAN) The networks NW1 and NW2 accommodate a
`plurality of information communication terminals TMll to
`'I'Mln, and TMZI to 'I'Mzm,
`respectively. The networks
`NW1! and NW2 are connected to wireless apparatuses RD1
`and RD2 via packet
`transfer apparatuses P11 and [’12,
`respectively. These packet transfer apparatuses PH and PT2
`and wireless apparatuses RD1 and RD2 are used for trans—
`ferring packets between the first and second networks NW1
`and NW2 via a wireless transmission path.
`
`[0041] As shown in FIG. 2, the packet transfer appara-
`tuses PTI and [’12 each have a network interface unit 10, a
`route control unit 20, an error correction processing unit 30
`and a wireless interface unit 40. The network interface unit
`
`10 transfers information data or control data in the packet
`unit between the first and second nervorks NW1 and NW2.
`
`The route control unit 20 judges a destination from the
`header of an input packet. and in accordance with the
`judgement result. transfers the packet to the information
`terminal TMI in the first network NW1 via the network
`interface 10, or to the wireless interface unit 40 via the error
`correction processing unit 30. The wireless interface 40
`transmitsr'receives a packet between the wireless apparatuses
`RD1 and RD2.
`
`[0042] The error correction processing unit 30 is struc-
`tuer as in the following. FIG. 3 is a block diagram showing
`the structure of the error correction processing unit 30. The
`error correction processing unit 30 provides the function of
`a main portion of the present invention, and is constituted of
`a transmission processing unit 310 and a reception process-
`ing unit 320. The error correction processing unit 30 is
`constituted of a micro processor unit {MPU}, a digital signal
`processor (DSP) and a random access memory (RAM). For
`example, MPU and DSP execute the function shown in FIG.
`3 in a software manner by using a control program stored in
`a program memory. Constituent elements may be structured
`in a hardware manner.
`
`[0043] The transmission processing unit 310 has a trans-
`mission input buffer 311, an error correction encoding unit
`312, a transmission output buffer 313 and an encoding
`control unit 314.
`
`[0044] The transmission input buffer 311 temporarily
`stores a transmission packet
`transferred from the route
`control unit 20 for
`the purpose of an error correction
`encoding process. The error correction encoding unit 312
`has three error correction encoders 3121, 3122 and 3123
`having different encoding rates. In the example shown in
`FIG. 3, the error correction encoders have encoding rates
`R=1/1, 11:33 and R= ”(’3. The error correction encoding unit 3 12
`error-correction encodes a transmission packet read from the
`transmission input bulfcr 311 by using one of the error
`correction encoders 3121, 3122 and 3123, and adds a sync
`bit and the like to generate an error correction encoding
`frame. The generated error correction encoding frame is
`output to the transmission output buffer 313. The transmis—
`sion output buffer 313 holds temporarily the packet output
`from the error correction encoding unit 312 and then outputs
`it to the wireless interface unit 40.
`
`[0045] The encoding control unit 314 has an error correc-
`tion encoding rate selection table. This error correction
`encoding rate selection table stores an error correction
`encoding rate necessary for obtaining a desired Quality of
`Service (008}, in correspondence with a protocol number
`representative of the type of a protocol expected to be used
`and a port number representative of an application type.
`
`[0046] For example. assuming that a desired 008 of TCP
`is set to BER-=l0'° or lower, a desired 008 of UDP is set to
`BI?.R=10's or lower and a desired 008 of an UDP applica~
`tion represented by the port number is set to BER=10‘4 or
`lower, the desired 008 of each traflic can be obtained at a
`wireless transmission path having an expected BER=2xltl‘
`3. FIG. 8 shows an example of the error correction encoding
`rate table used in the first embodiment.
`
`[0047] The encoding control unit 314 acquires the proto-
`col number and port number from the header of the trans-
`mission packet tern porarily stored in the transmission input
`buffer 311, and accesses the error correction encoding rate
`selection table by using the acquired protocol number and
`port number to thereby select an error correction encoding
`rate. In accordance with the selected error correction encod-
`
`ing rate, one of the error correction encoders 3121, 3122 and
`3123 is selectively operated to make it execute the error
`correction encoding process.
`
`[0048] For example, convolution encoders are used as the
`error correction encoders 3121, 3122 and 3123. FIG. 4
`
`Microsoft
`
`Ex. 1028 - Page 18
`
`Microsoft
`Ex. 1028 - Page 18
`
`

`

`US 2005/0276259 A1
`
`Dec. 15, 2005
`
`shows an example of the structure of a convolution encoder
`having a constraint length of 7. The convolution encoder is
`constituted of a shift resister 331 and two adders 332 and
`333 for synthesizing respective tap outputs. In FIG. 4, data
`is input to the shift register 331 and encoded by generator
`polynomials G0=1+D+D2+D3+D"and Glu1+D2+D3+D5+
`D6. The calculation results of the generator polynomials are
`alternately output from the adders 332 and 333 as the
`encoded data.
`
`[0049] As seen from a correspondence table between an
`encoding rate and a pu ncturc pattern shown in FIG. 5, the
`error correction encoding rate can be changed by deleting an
`output bit at a position of (30:0 for the 00 output and
`deleting an output bit at a position of (1:0 for the (31
`output.
`
`[0050] The reception processing unit 320 has a reception
`input bu lfer 321, a sync processing unit 326, an error
`correction decoding unit 322, a reception output hulfer 323,
`a decoding control unit 324 and a propagation path estimat-
`ing unit 325.
`
`[0051] The reception input buffer 321 temporarily stores
`reception frame data transferred from the wireless interface
`unit 40 for the purpose of an error correction decoding
`process. The sync processing unit 326 executes a frame sync
`establishing process for the reception frame data read from
`the reception input buffer 32 I. The error correction decoding
`unit 322 has three error correction decoders 3221. 3222 and
`3223 corresponding to three error correction encoding rates
`R-‘fz, R-‘E’i and R-ir’s of the error correction encoding unit
`312. The error correction decoding process and a frame
`disassembling process are executed for the reception frame
`data subjected to the frame sync processing, by using one of
`the three error correction decoders 3221, 3222 and 3223 to
`thereby recover the reception packet. The recovered recep-
`tion packet is output to the reception output buffer 323. The
`reception output buffer 323 temporarily holds the recovered
`reception packet and then transfers it to the route control unit
`20.
`
`[0052] The decoding control unit 324 has a decoding
`controi table. This decoding control table stores information
`on an error frame processing method suitable for each
`protocol and each application,
`in correspondence with a
`protocol number representative of the type of a protocol
`expected to be used and a port number representative of an
`application type. The storage contents are created by con-
`sidering the desired 008 of each traffic and each application.
`For example, an error frame is discarded for the protocol not
`permitting an error such as TCP, and an error frame is
`transferred for the application permitting an error. FIG. 9
`shows an example of the decoding control table.
`
`[0053] The decoding control unit 324 selects one of the
`error correction decoders 3221, 3222 and 3223 in accor—
`dance with encoding rate information derived from the start
`of the reception frame data, and makes the selected error
`correction decoder execute the error correction decoding
`process for the reception frame data. It is judged whether
`there is an error in the reception frame data subjected to error
`correction decoding by the error correction decoding unit
`322. If it is judged that there is an error, the decoding control
`table is accessed by using the protocol number and port
`number for the reception frame data to select an error frame
`processing method. In accordance with the selected process-
`
`a discard process is
`ing method, a transfer process or
`executed for the reception frame data.
`
`If the error correction encoding unit 312 uses
`[0054]
`convolution encoders, the error correction decoding unit 32
`generally uses corresponding Viterbi decoders, Viterbi
`decoding is described in detail
`in the document “Error
`Correction Code and its Application" by Etch and Kaneko,
`Ohmsha, Ltd. pp. 16L to 164 (1997}. Encoded data with
`deleted bits by the puncture pattern can be decoded by
`inserting dummy hits at bit positions removed by the punc-
`ture pattern, in the same method as that for no bit deletion
`by the puncture pattern.
`
`[0055] The propagation path estimating unit 325 estimates
`a bit error rate (BER) in accordance with an error correction
`decoding result of the reception frame by the error correc-
`tion decoding unit 322. The relation between BER and an
`error correction ability of error correction code is already
`determined. For example,
`if the generator polynomials
`shown in FIG. 4 are used and soft-decision Viterhi decoding
`is used as error correction decoding, errors up to BER—10‘2
`can be reduced to BER-10'5 or smaller at the encoding rate
`R-lx'z, errors up to BER-3x10'3 can be reduced to BER-
`10'5 or smaller at the encoding rate R=-i’i, and errors up to
`BER=3><10‘3 can be reduced to BER=10'S or smaller at the
`encoding rate 13:373. If an error is not detected in the Viterbi
`decoding result, the Viterbi decoding result is again encoded
`and compared with the reception data before Viterbi decod-
`ing so that BER can be estimated.
`
`[0056] Therefore, if frame data having adata length of 10“
`before encoding is encoded at the encoding rate R=V2, BER
`can be estimated in a range of 10’2 to BERs extracted by this
`method are acquired for a plurality of frames, and an average
`of BERs is used for estimating BER. In this manner. BER of
`10"1 or smaller can be estimated and a precision of an
`estimated BER can be improved by absorbing an instanta-
`neous variation in the propagation path.
`
`[005?] Next, description will be made on the operation of
`the packet transfer apparatus constructed as above. Since the
`packet transfer apparatuses PH and P’l‘2 have the same
`structure,
`the packet
`transfer apparatus P'I‘l will be
`described