United States Patent [19]
`Lev et al.
`
`I Ill ~111111111~ Ill Ill II 1111111111111111111111111111111
`
`US005729544A
`[11] Patent Number:
`[451 Date of Patent:
`
`5,729,544
`Mar. 17, 1998
`
`[54] METHOD FOR TRANSMITTING DATA
`PACKETS BASED ON MESSAGE TYPE
`
`(75]
`
`Inventors: Valy Lev; Rod Averbuch; Israel A.
`Cimet, all of Buffalo Grove. ID.
`
`[73] Assignee: Motorola, Inc., Schaumburg, ill.
`
`[21] Appl. No.: 239,656
`
`May 9, 1994
`
`[22] Filed:
`Int. Cl. 6
`[51]
`••••••••••••••••••••••••••••••••••••••••••••••••••••• H04L 12156
`[52] U.S. Cl .............................................. 3701352; 370/353
`[58] Field of Search ........................... 370/60, 60.1, 94.1.
`370/85.7, 95.3. 95.1. 85.2, 351, 352, 353,
`354, 355-357,360,385,387-389.392.
`395, 401. 467, 466, 465, 474; 455/39, 68.
`53.1, 54.1. 54.2, 55.1. 56.1
`
`[56]
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`5/1990 Weik ...................................... 370/60.1
`4,926,416
`5,159,592 10/1992 Perkins ................................... 370/85.7
`5,208,806
`5/1993 Hasegawa .............................. 370/60.1
`8/1994 Tanaka et al .•.........•••.......••••. 370/60.l
`5,339,317
`
`Primary Examiner-Dang Ton
`Attome~ Agent, or Finn-Christopher P. Moreno
`
`[57]
`
`ABSTRACT
`
`Data to be transmitted from an information source (301),
`associated with a local/wide area network (303), to a target
`destination (317) is formatted into a plurality of data pack(cid:173)
`ets. Based on a message type of the data packets, either a
`circuit-switched network (311) or a packet-switched net(cid:173)
`work (309) is selected. When the circuit-switched network is
`chosen, a circuit-switched channel (314) is established to the
`target destination, and the data packets are transmitted to the
`target destination via the circuit-switched channel. When the
`packet-switched network is chosen. a packet-switched chan(cid:173)
`nel (313) is established to the target destination. and the data
`packets are transmitted to the target destination via the
`packet-switched channel
`
`4,698,80'2 10/1987 Goke et al ............................ 370/94.1
`5/1990 Tawara et al .••..•.......•..•••••.•.•. 370/85.2
`4,926,415
`
`6 Claims, 2 Drawing Sheets
`
`ESTABLISH PACKET-swlTCHED CHANNEL
`
`ESTABUSH PAO<Ef.SWITCHED CHAl'l>lEL
`
`405
`
`401
`
`TRANSMT DATA PACKETS
`
`RECEIVE DATA PACKElS VIA
`PACKET-swlT()lEDCHWEL
`
`RECEVE DATA PACKElS VIA
`ORCUIT .sw!TCHED CHAllllEl..
`
`410
`
`Pet., Exh. 1015, p. 1
`
`

`
`U.S. Patent
`
`Mar. 17, 1998
`
`Sheet 1of2
`
`5,729,544
`
`-PRIOR ART-
`
`100
`
`-
`
`Wj'
`c111-3 c11s~f119
`
`DATA
`
`TRAFFIC
`
`I
`
`DATA
`
`CHANNEL
`TRAFFIC
`
`FIG.I
`
`200
`
`Wj'
`201203 205 207 209 211 213 215217 219221 223 ~
`I
`
`SOJRCE -
`CHANNEL -
`SOURCE -
`-
`i••r• ' rm1r• {
`
`TI~
`
`-PRIOR ART-
`
`TIME
`
`301
`
`300
`
`311
`
`FIG.2
`
`305
`
`309
`
`PACKET-S'MTCHED TRANSCEIVER 14----'
`~----
`3147
`~313 /
`
`~317
`PACKET-SWITCHED TRANSCEIVER CIRCUIT-SWITa-IED TRANSCEIVER
`
`319
`
`TERMNAL
`
`FIG.3
`
`323
`
`321
`
`Pet., Exh. 1015, p. 2
`
`

`
`U.S. Patent
`
`Mar. 17, 1998
`
`Sheet 2 of 2
`
`5,729,544
`
`BEGIN
`
`PROVIDE INFORMATICN
`
`401
`
`FORMAT INFOOMATICN INTU DATA PACKETS
`
`DETERflttNE t-.ESSAGE TYPE
`
`PACKET
`
`CIRCUIT
`
`ESTABLISH PACKET-SWITCHED CHANNEL
`
`ESTABLISH PACKET-SWITCHED CHAl'f'.JEL
`
`405
`
`406
`
`407
`
`lRANSMT DATA PACKETS
`
`PACKET
`
`ORCUIT
`
`RECEIVE DATA PACKETS VIA
`PACKET-SWITCHED CHANNEL
`
`RECEIVE DATA PACKETS VIA
`CIRa.JIT-sWITCHED CHANNEL
`
`409
`
`410
`
`END
`
`FIG.4
`
`Pet., Exh. 1015, p. 3
`
`

`
`1
`METHOD FOR TRANSMITTING DATA
`PACKETS BASED ON MESSAGE TYPE
`
`5,729,544
`
`FIELD OF THE INVENTION
`The present invention relates generally to communication 5
`systems and, in particular, to the transmission of data
`packets.
`
`BACKGROUND OF THE INVENITON
`
`There exists today a growing need to expand the use of
`traditionally office-based computer applications (e.g., word
`processing programs, electronic mail, etc.) to remote loca(cid:173)
`tions such as the home or car. Data files and/or messages
`generated by such applications, residing on hosts having
`local-area network/wide-area network (LAN/WAN)
`connectivity, are typically transmitted from one location to
`another using high-speed protocols such as the so-called
`Transmission Control Protocol/Internet Protocol (TCP/IP).
`In order to extend the use of these applications to remote
`locations not serviced by a LAN/WAN. it is necessary to
`establish connections between LAN/WANs and other non(cid:173)
`similar communication networks, such as wireless commu(cid:173)
`nication systems.
`A major difficulty in connecting LAN/WANs to wireless
`communication systems is the large disparity in their avail(cid:173)
`able transmission bandwidths and hence, their throughput
`capacities. It is not atypical for a wireless communication
`system to have a transmission bandwidth 10 times less that
`of a LAN/WAN. This disparity also contributes to the widely
`differing protocols used in LAN/WANs and wireless com(cid:173)
`munication systems. The lower throughput capacities asso(cid:173)
`ciated with wireless communication systems has led to the
`use of circuit-switched techniques, whereas the higher
`throughput capacities associated with LAN/WANs has led to
`the use of packet-switched techniques.
`FIG. 1 illustrates an example of a typical, prior art
`circuit-switched communication (100). When initial source
`data (101) becomes available. a channel set-up period (107)
`is required to establish a communication path between the
`source and the destination. For example, in the wireless case,
`the channel set-up period (107) may be the time required to
`request and obtain usage of a particular radio frequency (RF)
`carrier. Regardless of the channel type, the channel. once
`established, remains dedicated for the exclusive use of the
`source and destination.
`Having established the channel. the initial source data is
`transmitted (109) to the destination. As additional source
`data (103. 105) becomes available, it is immediately trans(cid:173)
`mitted (113, 117) through the channel. When necessary.
`usage of the channel is then discontinued during a channel
`tear-down period (119). Advantages of circuit-switched
`techniques are the low overhead requirement (i.e., the
`amount of throughput capacity required for the transmission
`of information other than the source data). as well as the low 55
`delay (i.e., the time difference between the availability of
`source data and its actual transmission). The periods of
`channel inactivity (111, 115) in between periods of data
`availability, however, are a disadvantage of circuit-switched
`techniques. This is a direct result of the dedicated use of the 60
`channel; other sources are unable to utilize the channel.
`These advantages and disadvantages make the use of circuit(cid:173)
`switched techniques most efficient for longer
`communications, such as file transfers or fax transmissions.
`FIG. 2 illustrates an example of a typical, prior art 65
`packet-switched communication (200). A key difference
`between packet-switched and circuit-switched methods is
`
`10
`
`2
`that the channel, when used in a packet-switched manner. is
`not dedicated and is available for use by multiple sources
`and destinations. Once available, the initial source data
`(101) is partitioned into data packets (203, 207. 211) for
`transmission. The data packets (203, 207, 211) occupy
`available time-slots that include capacity for overhead data
`(201, 205, 209). Due to the commonality of the channel. the
`overhead data (201, 205, 209) typically comprises target
`destination identification information so that data intended
`for a particular destination may be properly routed.
`As additional source data (103, 105) becomes available, it
`is again formatted into data packets (217, 221, 227) and
`placed into available time-slots with their associated over(cid:173)
`head data (215, 219, 225). Advantages of packet-switched
`methods are that set-up/tear-down periods are not required.
`15 Also, multiple communications may be intermingled on the
`channel. Assuming the use of channels having equivalent
`bandwidths, packet-switched methods are less efficient rela(cid:173)
`tive to circuit-switched methods due to the additional
`overhead, typically leading to larger throughput delays.
`20 Delays are further lengthened when time-slot availability is
`reduced due to heavy use of the channel. These disadvan(cid:173)
`tages can be overcome by increasing the packet-switched
`channel's transmission bandwidth, if possible. to accommo(cid:173)
`date the larger overhead and need for additional time-slots.
`25 Typically, packet-switched techniques are most efficient in
`the transmission of shorter communications. such as elec(cid:173)
`tronic mail messages or paging services.
`In order to establish connectivity between LAN/WANs
`and wireless communication systems, the incompatibilities
`30 of their respective packet-switched and circuit-switched
`protocols need to be resolved. One solution to this problem
`is to directly transmit the packet-switched data, including
`the overhead data for each packet. over a circuit-switched
`(i.e., wireless) channel. This is inadequate. however. because
`35 the differences in throughput capacities would require an
`inordinate amount of packet-switched information to be
`buffered prior to transmission over the circuit-switched
`channel. Even if the circuit-switched channel has sufficient
`bandwidth. this solution becomes inefficient due to channel
`40 inactivity during periods of low packet volumes.
`Another solution is to establish a circuit-switched trans(cid:173)
`mission for each data packet or group of data packets. While
`this solution might be acceptable for low volumes of
`packets, it becomes severely inefficient for increasing packet
`45 volumes due to the set-up and tear-down overhead.
`As the previous examples indicate. the efficiency with
`which packet-switched data is transmitted depends on the
`type of message being transmitted. That is, shorter packet(cid:173)
`switched messages may be efficiently transmitted over
`so packet-switched channels, whereas longer packet-switched
`messages may be efficiently transmitted over circuit(cid:173)
`switched channels. Therefore. a need exists for a method that
`allows packet-switched data to be transmitted in the most
`efficient manner.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`FIG. 1 illustrates a circuit-switched communication in
`accordance with prior art.
`FIG. 2 illustrates a packet-switched communication in
`accordance with prior art.
`FIG. 3 illustrates a data network that includes a local/
`wide-area network and a wireless communication system in
`accordance with the present invention.
`FIG. 4 illustrates a logic diagram which may be used by
`a data network when transmitting packet-switched and/or
`circuit switched information in accordance with the present
`invention.
`
`Pet., Exh. 1015, p. 4
`
`

`
`5,729,544
`
`
`
`25
`
`3
`4
`321) transparently provide data to, and receive data from,
`DESCRIPTION OF A PREFERRED
`the data-capable terminal (323). That is, data packets are
`EMBODIMENT
`exchanged with the data-capable terminal (323) as though it
`Generally, the present invention provides a method for
`were directly coupled to the LAN/WAN (303). The data-
`transmitting information of a given message type to a target
`destination. This is accomplished by formatting the data to 5 capable terminal (323), which may be a PC-based facsimile
`be transmitted into a plurality of data packets. Based on a
`unit for example, controls the reception and transmission of
`corresponding message type of the data packets, either a
`information based on the information type.
`circuit-switched network or a packet-switched network is
`So configured, the information source (301) can send
`selected. When the circuit-switched network is chosen, a
`packet-switched data intended for the target destination
`circuit-switched channel is established to the target 10 (317). via the LAN/WAN (303), to the network interfacer
`(305). The network interfacer (305). in turn, can select either
`destination, and the data packets are transmitted to the target
`destination via the circuit-switched channel. When the
`the packet-switched transceiver (309) or the circuit-switched
`packet-switched network is chosen. a packet-switched chan-
`transceiver (311) to transmit the data to the target destination
`nel is established to the target destination, and the data
`(317).
`packets are transmitted to the target destination via the 15
`FIG. 4 illustrates a logic diagram which may be used
`packet-switched channel. Such a method allows packet-
`when transmitting information between the information
`switched information to be transmitted in the most efficient
`source (301) and the target destination (317) in accordance
`with the present invention. The information transfer illus-
`manner to a target destination.
`The present invention can be more fully described with
`trated in FIG. 4 is hereinafter described in the context of an
`reference to FIGS. 3 and 4. FIG. 3 illustrates a data network 0 application using the TCP/IP protocol in a simplex manner 2
`(300) that includes an information source (301), a local/wide
`(i.e., in one direction only). However. it is important to note
`area network (LAN/WAN) (303). a network interfacer (305),
`that the process illustrated in FIG. 4 may also work for other
`a packet-switched transceiver. (309), a circuit-switched
`protocols, such as the AppleTalk Transaction protocol, oper-
`transceiver (311), a packet-switched channel (313), a circuit-
`ating in a duplex manner, i.e., from the information source
`switched channel (314), and a target destination (317). The
`(301) to the target destination (317), and vice versa.
`information source (301) may comprise an electronic mail or
`At step 401, it is assumed that information. intended for
`word processing application running on a personal computer
`transmission to the target destination (323), is provided by
`(PC) or workstation. The LAN/WAN (303) may comprise a
`an application operating at the information source (301).
`public or private data network such as Integrated Services 30 This information is in a digital form and may comprise a
`short electronic mail message or a long document file. The
`Digital Network (ISDN), Frame Relay, or Asynchronous
`Transfer Mode (ATM) network.
`information source (301) formats the information into data
`The network interfacer (305) provides a link between the
`packets ( 402). Each data packet includes a header that
`LAN/WAN (303) and an RF communication system com-
`comprises a first target destination identifier, such as the 1P
`prising the packet-switched transceiver (309), the circuit- 35 address, for uniquely identifying the data-capable terminal
`switched transceiver (311), the packet-switched channel
`(327). A TCP session header, including data fields for
`(313). the circuit-switched channel (314), and the target
`indicating a message type, is provided in addition to the data
`destination (317). The network interfacer. (305) may be a
`packets.
`group of modems providing multiple access paths to the
`When initiating the TCP session, the message type is
`public switched telephone network (PSTN). or it may be a 40 determined by the application sourcing the information and
`is indicated in the TCP session header. By examining the
`packet gateway/router that incoiporates many different pro-
`tocols to communicate with public or private data networks.
`appropriate portion of the TCP session header (i.e .• the TCP
`The packet-switched transceiver (309) comprises a
`port number), the network interfacer (305) determines the
`transmitter/receiver pair capable of communicating in a
`message type (403). The preferred classification of all pos-
`packet-switched manner, as described previously. Likewise, 45 sible message types can be predetermined and stored in the
`network interfacer (305). If the message is a short type, such
`the circuit-switched transceiver (311) comprises a
`transmitter/receiver pair capable of communicating in a
`as electronic mail. the packet-switched network (i.e .• the
`circuit-switched manner. as described previously. Manage-
`packet-switched transceiver) is selected (404). The packet-
`ment of each transceiver's (309. 311) respective channel
`switched channel (313) is established (405) in the sense that
`(313. 314) (i.e .• handoffs. channel assignmen~ etc.) can be 50 the data packets are placed into the shared, packet-switched
`done by the transceivers (309. 311) themselves, or jointly
`channel (313) as time slots become available.
`Conversely. if the message is a long type. such as a file
`with the network interfacer (305). Additional data
`manipulation. such as error correction. may also be per-
`transfer. the circuit-switched network (i.e.. the circuit-
`formed by the transceivers (309. 311) and/or the network
`switched transceiver) is selected (404). Using conventional
`interfacer (305).
`55 call set-up techniques, the network interfacer (305) estab-
`The target destination (317) further comprises a mobile
`lishes (406) the circuit-switched channel (314) to the target
`destination (317). For example, an identification number of
`packet-switched transceiver (319) and a mobile circuit-
`switched transceiver (321) coupled to a data-capable terrni-
`the mobile circuit-switched transceiver (321), uniquely asso-
`nal (323). The mobile circuit-switched transceiver (321)
`ciated with the IP address of the data-capable terminal (323),
`may comprise a Groupe Speciale Mobile (GSM) digital 60 may be used to establish the exclusive, circuit-switched
`channel (314). It is understood that availability of both the
`cellular telephone or a IDENTM mobile radio. both by
`Motorola. Inc. The packet-switched transceiver (319) may
`packet-switched and circuit-switched channels (313, 314)
`comprise an ARDIS® mobile radio by Motorola. Inc. The
`may also be considered when selecting the proper path.
`packet-switched transceiver (319) and the mobile circuit-
`Having established either a packet-switched or circuit-
`switched transceiver (321) transceive information via the 65 switched channel (313, 314). the data packets are transmit-
`packet-switched channel (313) and the circuit-switched
`ted to the target destination at step 407. When the packet-
`channel (314). respectively. The mobile transceivers (319,
`switched channel (313) is used, the data packets may be
`
`Pet., Exh. 1015, p. 5
`
`

`
`5,729,544
`
`5
`transmitted in their entirety. i.e., the headers for each data
`packet are transmitted. In a preferred embodiment. when the
`circuit-switched channel (314) is used, the header portion of
`each data packet is not transmitted, with the exception of the
`first data packet. (fhe interested reader will find additional
`details regarding a way to transmit packets through a circuit(cid:173)
`switched channel in U.S. Pat. No. 5.483,524 entitled
`METHOD FOR TRANSMITTING DATA PACKEI'S VIA
`A CIRCUIT-SWITCHED CHANNFL, filed on even date
`herewith and being commonly assigned, which patent is
`incorporated herein by this reference.) Other methods of
`transmitting packet-switched data via a circuit-switched
`channel may be used without departing from the spirit of the
`present invention.
`Upon transmission of the data ( 407), it is determined at
`the target destination (317) whether the data is being trans(cid:173)
`mitted via the packet-switched or circuit-switched channel
`(408). When the packet-switched channel (313) is being
`used, this determination is done by examining the header
`portion of incoming time slots to detect the presence of data
`packets addressed to the target destination (317). When 20
`transmitted via the circuit-switched channel (314), the actual
`establishment of the circuit-switched channel (314) can be
`used to indicate the impending arrival of data.
`When transmitted via the packet-switched channel (313),
`the data packets, including their respective TCP/IP headers, 25
`are received ( 409) and forwarded to the data-capable termi(cid:173)
`nal (323). When transmitted via the circuit-switched channel
`(314), the data packets are received (410) and forwarded to
`the data-capable terminal (323) after having appropriate
`TCP/IP headers added to each data packet In this manner. 30
`the method of actual transmission of the data packets is
`transparent to the data-capable terminal.
`The present invention provides a method for transmitting
`information of a given message type to a target destination.
`With such a method, the inefficiencies inherent to the 35
`transmission of variable-length, packet-switched data are
`overcome. This is achieved through the provision of both
`packet-switched and circuit-switched communication paths.
`Based on the message type, the optimal path can be selected,
`thereby improving efficiency.
`We claim:
`1. A method comprising the steps of:
`providing information to be transmitted, which informa(cid:173)
`tion comprises one message type of a plurality of
`message types as indicated by a TCP session header 45
`included in the information;
`formatting the information into a plurality of data packets;
`automatically selecting as between a circuit switched
`network and a packet switched network based on the
`one message type indicated by the TCP session header, 50
`wherein the circuit switched network is selected when
`a long type is indicated by the one message type, and
`wherein the packet switched network is selected when
`a short type is indicated by the one message type;
`when the circuit switched network is selected, establish- 55
`ing a circuit switched channel and transmitting the
`plurality of data packets via the circuit switched chan(cid:173)
`nel;
`
`40
`
`6
`when the packet switched network is selected, establish(cid:173)
`ing a packet switched channel and transmitting the
`plurality of data packets via the packet switched chan(cid:173)
`nel.
`2. The method of claim 1 wherein the step of establishing
`the circuit switched channel includes the step of establishing
`an RF-based communication channel as the circuit switched
`channel.
`3. The method of claim 1 wherein the step of establishing
`10 the packet switched channel includes the step of establishing
`an RF-based communication channel as the packet switched
`channel.
`4. A method for communicating information between a
`source and at least one destination, the method comprising
`15 the steps of:
`at the source;
`providing information to be transmitted, which informa(cid:173)
`tion comprises one message type of a plurality of
`message types as indicated by a TOP session header
`included in the information;
`formatting the information into a plurality of data packets;
`automatically selecting as between a circuit switched
`network and a packet switched network based on the
`one message type indicated by the TCP session header,
`wherein the circuit switched network is selected when
`a long type is indicated by the one message type. and
`wherein the packet switched network is selected when
`a short type is indicated by the one message type;
`when the circuit switched network is selected, establish(cid:173)
`ing a circuit switched channel and transmitting the
`plurality of data packets via the circuit switched chan(cid:173)
`nel;
`when the packet switched network is selected, establish(cid:173)
`ing a packet switched channel and transmitting the
`plurality of data packets via the packet switched chan(cid:173)
`nel;
`at the at least one destination;
`determining that the plurality of data packets is being
`transmitted as between the circuit switched network
`and the packet switched network;
`when the plurality of data packets is transmitted over the
`circuit switched network, receiving the plurality of data
`packets via the circuit switched channel; and
`when the plurality of data packets is transmitted over the
`packet switched network. receiving the plurality of data
`packets via the packet switched channel.
`S. The method of claim 4 wherein the step of establishing
`the circuit switched channel includes the step of establishing
`an RF-based communication channel as the circuit switched
`channel.
`6. The method of claim 4 wherein the step of establishing
`the packet switched channel includes the step of establishing
`an RF-based communication channel as the packet switched
`channel.
`
`* * *
`
`*
`
`Pet., Exh. 1015, p. 6
`
`

`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`CERTIFICATE OF CORRECTION
`5,729,544
`March 17, 1998
`Lev et al.
`
`PATENT NO.
`
`DATED
`
`INVENTOR(S) :
`
`It is certified that error appears in the above-indentified patent and that said Letters Patent is hereby
`corrected as shown below:
`
`Column 6, line 19 reads "TOP" should be --TCP--.
`
`Signed and Sealed this
`
`Eleventh Day of August 1998
`
`€1.a~
`
`BRUCE LEHMAN
`
`Attest:
`
`Attesting Officer
`
`Commissioner of Patents and Trademark.\
`
`Pet., Exh. 1015, p. 7