JP2015154320A - Portable terminal and communication control method thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a mobile terminal capable of improving the throughput of a data flow in which the actual throughput is about the same as the theoretical upper limit of the throughput in data communication. SOLUTION: A mobile terminal has a buffer for temporarily storing communication packets related to each of a plurality of data flows, and is the maximum transmissible data that can be transmitted for one data flow without confirmation of reception from a destination device. Using the information indicating the amount and the RTT (Round Trip Time) in the data flow, the theoretical throughput indicating the theoretical upper limit of the data transfer amount per unit time is calculated, and the actual transmission per unit time is calculated. When the actual throughput indicating the amount of data transfer is compared with the calculated theoretical throughput and it is determined that the throughput has reached the same level, the communication packet related to the data flow is preferentially transmitted from the buffer. [Selection diagram] FIG. 8

Description

  The present invention relates to a network communication technique in a mobile terminal, and more particularly to a technique related to data transfer.

  In recent years, wireless data communication using mobile terminals such as smartphones and tablets has been widely used as wireless communication networks capable of performing high-speed data communication such as LTE (Long Term Evolution) are being constructed. It has become. In addition to connecting to a network using such high-speed data communication and using the Internet or the like with a single mobile terminal, one or more devices such as a PC (Personal Computer) can be used using the tethering function of the mobile terminal. It is also possible to connect to a portable terminal and use the Internet or the like from these devices via the portable terminal.

Incidentally, in general, data communication via wireless communication has a larger data transfer delay than data communication performed only by wired communication. For this reason, various devices have been conventionally used to improve the data transfer amount per unit time (hereinafter referred to as “throughput”).
For example, Patent Document 1 discloses a technique for optimizing a transmission packet size with respect to a TCP window size when communicating with TCP / IP (Transmission Control Protocol / Internet Protocol). In the technique disclosed in Patent Document 1, the transmission packet size is determined so that the accumulated packet size of a plurality of transmission packets is not too much within the TCP window size. As a result, the amount of data that can be transmitted before receiving an acknowledgment ACK (Acknowledgement) from the destination device can be maximized, and the throughput of data transmission can be improved.

JP 2001-237882 A

By the way, the theoretical upper limit value of throughput in TCP / IP (hereinafter referred to as “theoretical throughput”) is determined by the TCP window size and RTT (Round Trip Time), and is proportional to the TCP window size. Inversely proportional.
The actual throughput (hereinafter referred to as “actual throughput”) varies depending on various factors such as the processing capacity of the device, the network bandwidth and the usage state, but the processing capability of these devices and the network bandwidth are sufficiently secured. In spite of this, the actual throughput may not increase. This corresponds to the case where the actual throughput reaches the same throughput as the theoretical throughput (hereinafter, such a state is referred to as a “bottom state”). Since the actual throughput cannot be faster than the theoretical throughput, there is a problem that even if the transmission packet size is optimized in such a case, the actual throughput cannot be improved.

  The present invention has been made in view of such problems, and an object of the present invention is to provide a portable terminal capable of improving the throughput of data communication that has reached its peak.

  In order to solve the above problems, a mobile terminal according to an aspect of the present invention is a mobile terminal that performs data communication including a plurality of data flows with at least one or more devices via a network. A buffer that temporarily stores communication packets related to each of the plurality of data flows, a measurement unit that measures RTT in one data flow of the plurality of data flows, the RTT, and reception from a destination device Using the information indicating the maximum transmittable data amount that can be transmitted without confirmation, the theoretical throughput indicating the theoretical upper limit value of the data transfer amount per unit time when transmitting data via the network is calculated. A calculation unit, and a measurement unit that measures an actual throughput indicating a data transfer amount per unit time actually transmitted in the one data flow; The priority of the one data flow is higher than the priority of the other data flow when the actual throughput satisfies a predetermined standard indicating that the throughput reaches the same level as the theoretical throughput. A priority setting unit that sets the priority, an output unit that generates a communication packet in which the priority is set and outputs the communication packet, and a communication packet having a higher priority among the communication packets stored in the buffer. And a sending unit that sends the data preferentially.

  According to the portable terminal having the above-described configuration, the RTT can be shortened, so that the theoretical throughput can be increased. As a result, the throughput of data communication that has reached its peak can be improved.

The conceptual diagram of the communication system 1 using the portable terminal 10 which concerns on this invention. Schematic of the external appearance of the portable terminal 10. FIG. 3 is a functional block diagram of main parts of the mobile terminal 10. TCP packet data structure. IP packet data structure. FIG. 5 is a sequence diagram showing an outline of data transmission / reception between the PC 20 and the server device 50. The sequence diagram which shows the outline of transmission / reception of the data using TCP window size of PC20 and the server apparatus 50. FIG. 10 is a flowchart of processing in the mobile terminal 10. 6 is a flowchart of throughput calculation and comparison processing in the mobile terminal 10. 10 is a flowchart of processing in the mobile terminal 10 according to the second embodiment. The flowchart of the process in the portable terminal 10 of a modification. An example of UI (user interface) presented when requesting user input for priority setting. Another example of UI presented when requesting user input for priority setting.

<1. Embodiment 1>
<1.1 Overview>
An embodiment of the mobile terminal 10 according to the present invention will be described.
FIG. 1 is a schematic diagram of a communication system 1 including a mobile terminal 10.
The communication system 1 includes a mobile terminal 10, a PC (Personal Computer) 20, a base station 30, a network 40, and a server device 50.

The PC 20 uses the tethering function of the mobile terminal 10 to connect to the mobile terminal 10 and perform data communication with the server device 50 via the mobile terminal 10.
The mobile terminal 10 is connected to the base station 30 via a wireless communication line, and transmits data received from the PC 20 to the server device 50 via the network 40. Data from the server device 50 is transmitted to the PC 20 via the network 40, the base station 30, and the mobile terminal 10. Data communication is performed by a communication protocol based on TCP / IP.

The mobile terminal 10 temporarily stores the communication packet received from the PC 20 in a buffer, and outputs it according to the priority order determined for each communication packet. Further, the mobile terminal 10 outputs the communication packet received from the base station 30 to the PC 20.
By the way, in data communication in TCP / IP, the time from when the communication packet is transmitted from the PC 20 to the server device 50 until the response indicating the reception confirmation is received from the server device 50 is called RTT (Round Trip Time). It becomes an index of network response speed.

  The theoretical maximum value Smax of the throughput can be calculated by Smax = W / R where W is the TCP window size and R is the RTT. The actual data transfer amount per unit time, that is, the actual throughput is the distance between the PC 20 and the server device 50 on the network route, the number of devices relaying / transferring on the route, the processing capability of the server device 50, the network Depending on various factors such as bandwidth and network usage, the actual throughput rarely reaches the same level as the theoretical throughput.

  Conversely, if the actual throughput has reached the same level as the theoretical throughput even though the network bandwidth is sufficiently secured, the actual throughput cannot be faster than the theoretical throughput. The improvement in communication speed cannot be expected, and it will be in a state of peaking. Such a state can be considered because it takes time to transmit a communication packet on the network path, and therefore, the RTT is long. As can be seen from the above formula, when the TCP window size is the same, if the RTT can be shortened, the theoretical throughput increases, the peak state is eliminated, and the actual throughput can be improved.

  As described above, the mobile terminal 10 temporarily stores communication packets in a buffer and outputs them according to a predetermined priority order, but outputs communication packets related to data communication in such a peak state from the buffer with priority. By this process, the time during which the communication packet remains in the buffer of the mobile terminal 10 can be shortened, and as a result, the RTT can be shortened. Therefore, the theoretical throughput is increased, the state of peaking is improved, and the actual throughput is improved.

<1.2 Configuration>
FIG. 2 is a schematic view of the appearance of the mobile terminal 10.
The mobile terminal 10 is a so-called smartphone type mobile communication terminal.
As shown in the figure, the mobile terminal 10 has a receiver hole 151 for transmitting sound output from a receiver (not shown) to the outside of the housing on the main surface of the housing, and a microphone for sound outside the housing. A microphone hole 152 for transmitting to (not shown) is formed, and the main surface of the touch panel 150 is further arranged. Further, on the main surface of the housing, an operation button group 153 for moving a cursor displayed on the touch panel 150 or selecting a display object when the user operates the mobile terminal 10 is arranged. ing.

FIG. 3 is a functional block diagram of the main part of the mobile terminal 10.
As shown in the figure, the mobile terminal 10 includes a control unit 101, an acquisition unit 102, an RTT measurement unit 103, a calculation unit 104, a measurement unit 105, a timer 106, an output unit 107, a buffer 108, a transmission unit 109, and a communication unit 110. , Antenna 111, receiving unit 112, comparing unit 113, priority setting unit 114, tethering unit 115, tethering antenna 116, audio processing unit 117, display unit 118, operation unit 119, storage unit 120, and power supply unit 121. It is the composition which includes.

Hereinafter, these components constituting the mobile terminal 10 will be described.
(Control unit 101, storage unit 120)
The control unit 101 includes an acquisition unit 102, an RTT measurement unit 103, a calculation unit 104, a measurement unit 105, a timer 106, a comparison unit 113, a priority setting unit 114, an output unit 107, a transmission unit 109, a communication unit 110, and a reception unit 112. , The tethering unit 115, the sound processing unit 117, the display unit 118, the operation unit 119, and the power supply unit 121 to control the mobile terminal 10 mainly with a mobile terminal control function, a tethering function, a throughput calculation function, a throughput comparison function, A function for realizing a priority setting function and a transmission packet transmission function;

The mobile terminal control function realizes a function equivalent to a general function of a conventional mobile terminal, for example, a call function, an Internet site browsing function, a mail transmission / reception function, a standby function, a user interface control function, an application execution function, etc. It is a function.
The tethering function performs wireless data communication with the PC 20, transmits data from the PC 20 to the server device 50 on the network using the wireless communication function of the mobile terminal 10, and transmits data from the server device 50. This is a function for transmitting to the PC 20. The tethering function controls the tethering unit 115, the sending unit 109, the receiving unit 112, and the communication unit 110, and the data received from the PC 20 by the tethering unit 115 is transferred to the sending unit 109 via the communication unit 110 to the base station 30. This is a function of transmitting data from the server device 50 received by the receiving unit 112 via the communication unit 110 to the tethering unit 115 and transmitting it to the PC 20.

  The throughput calculation function includes a theoretical throughput calculation function and an actual throughput measurement function. The theoretical throughput calculation function controls the acquisition unit 102, the RTT measurement unit 103, the calculation unit 104, and the timer 106, and causes the calculation unit 104 to acquire the TCP window size acquired by the acquisition unit 102 and the RTT measured by the RTT measurement unit 103. This is a function for calculating theoretical throughput using The actual throughput measurement function is a function that controls the measurement unit 105 and the timer 106 to cause the measurement unit 105 to measure the actual throughput of data transmitted from the PC 20 to the server device 50.

The throughput comparison function controls the comparison unit 113 and causes the comparison unit 113 to compare the theoretical throughput calculated by the calculation unit 104 with the actual throughput measured by the measurement unit 105 to determine whether or not the actual throughput has reached its peak. This is a function for determining
The priority order setting function controls the priority setting unit 114 and causes the priority setting unit 114 to determine the priority of the communication packet transmitted from the PC 20 to the server device 50 based on the comparison result in the comparison unit 113. This is a function for causing the output unit 107 to output.

The transmission packet transmission function is a function for controlling the transmission unit 109 to cause the transmission unit 109 to output the communication packets stored in the buffer 108 to the communication unit 110 in descending order of priority of the communication packet.
Note that the function of the control unit 101 is realized by, for example, a CPU (Central Processing Unit) of the mobile terminal 10 executing a program stored in the storage unit 120.

The storage unit 120 includes, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, and the like. The storage unit 120 stores a program that defines the operation of the control unit 101 and a threshold value used to determine whether or not the actual throughput of the communication packet has reached its peak.
(Tethering unit 115, tethering antenna 116)
The tethering unit 115 performs wireless data communication with the PC 20, transmits data from the PC 20 to the server device 50 on the network using the wireless communication function of the mobile terminal 10, and receives data from the server device 50. Is transmitted to the PC 20.

The tethering antenna 116 is, for example, a metal monopole antenna, and is used for the mobile terminal 10 to perform wireless communication with the PC 20 using the tethering function.
Note that the function of the tethering unit 115 is realized by, for example, a CPU (Central Processing Unit) of the mobile terminal 10 executing a program stored in the storage unit 120.
(Output unit 107, buffer 108, sending unit 109)
The output unit 107 has a function of giving the priority set by the priority setting unit 114 to the communication packet transmitted to the server device 50 output from the PC 20 and outputting it to the buffer 108.

The buffer 108 is a memory such as a RAM, and has a function of temporarily storing communication packets transmitted from the PC 20 to the mobile terminal 10.
The sending unit 109 has a function of outputting communication packets stored in the buffer 108 to the communication unit 110 in descending order of priority according to the priority set in the communication packet.

Note that the functions of the output unit 107 and the sending unit 109 are realized by executing a program stored in the storage unit 120 by the CPU of the mobile terminal 10.
(Communication unit 110, antenna 111)
The communication unit 110 includes a communication LSI (Large Scale Integration) and is controlled by the control unit 101. The communication LSI has a function of transmitting / receiving data to / from an external server using the Internet network using the TCP / IP protocol, for example.

The communication unit 110 has a function of connecting to the base station 30 via the antenna 111 and transmitting the communication packet output from the transmission unit 109 to the server device 50. Further, the communication packet transmitted from the PC 20 has a function of outputting a communication packet indicating reception confirmation transmitted from the server device 50 to the receiving unit 112.
Note that the function of the communication unit 110 is realized by the CPU of the mobile terminal 10 executing a program stored in the storage unit 120.

The antenna 111 is, for example, a metal monopole antenna, is connected to a communication LSI, and is used for communication performed by the communication LSI.
(Receiver 112)
The receiving unit 112 has a function of receiving a communication packet from the server device 50.
Specifically, the receiving unit 112 receives, for example, a communication packet sent from the communication unit 110 as a response including a reception confirmation for the communication packet addressed to the server device 50 from the PC 20, and receives the communication packet from the acquisition unit 102 and the RTT. Output to the measurement unit 103.

Note that the function of the receiving unit 112 is realized by the CPU of the mobile terminal 10 executing a program stored in the storage unit 120.
(Acquisition unit 102, RTT measurement unit 103, calculation unit 104, timer 106)
The acquisition unit 102 has a function of extracting the TCP window size from the information of the header part of the communication packet from the server device 50 received by the reception unit 112 and outputting the TCP window size to the control unit 101 and the calculation unit 104.

The RTT measuring unit 103 controls the timer 106 to transmit a communication packet related to data transferred by the PC 20 to the server device 50, and to determine the time until the receiving unit 112 receives the communication packet for confirming the receipt of the communication packet. Has the function to measure. Then, it has a function of outputting the RTT obtained from the measured time to the calculation unit 104.
The calculation unit 104 has a function of calculating the theoretical throughput using the TCP window size value from the acquisition unit 102 and the RTT value from the RTT measurement unit. Specifically, when the TCP window size value is W and the RTT value is T _RTT , the theoretical throughput Smax can be calculated by Smax = W / T _RTT .

Note that the functions of the acquisition unit 102, the RTT measurement unit 103, and the calculation unit 104 are realized by executing a program stored in the storage unit 120 by the CPU of the mobile terminal 10.
Upon receiving an instruction from the RTT measurement unit 103, the timer 106 transmits a communication packet to the server device 50, and the reception unit 112 receives a reception confirmation communication packet for the communication packet from the server device 50. It has a function to measure the time until. In addition, the timer 106 has a function of receiving the instruction from the measurement unit 105, measuring the time taken for the output unit 107 to output a certain amount of communication packets, and outputting the measured time to the measurement unit 105.
(Measurement unit 105, comparison unit 113, priority setting unit 114)
The measurement unit 105 controls the timer 106 to measure the time, and the function of measuring the data amount of the communication packet transmitted from the output unit 107 to the server device 50 within the time measured by the timer 106. Have The measuring unit 105 has a function of calculating actual throughput from the time output by the timer 106 and the amount of data transmitted during that time. Specifically, assuming that the time obtained from the timer 106 is Td and the amount of data transmitted within the time Td is D, the actual throughput St can be calculated as St = D / Td.

  The comparison unit 113 compares the theoretical throughput calculated by the calculation unit 104 with the actual throughput measured by the measurement unit 105, determines whether or not the actual throughput has reached its peak, and sends the determination result to the priority setting unit 114. Output. Specifically, the comparison unit 113 obtains a ratio of the actual throughput to the theoretical throughput, and whether or not the actual throughput has reached a peak depending on whether or not the ratio exceeds a predetermined threshold (for example, 0.95). Determine whether. When the ratio exceeds the threshold, the comparison unit 113 determines that the actual throughput has reached its peak.

  The priority setting unit 114 has a function of determining a priority to be set for the communication packet output from the PC 20 based on the comparison result output from the comparison unit 113 and outputting information on the determined priority to the output unit 107. Have. Specifically, the priority setting unit 114 sets the priority of the communication packet addressed to the server device 50 to another communication packet when the comparison result in the comparison unit 113 indicates that the actual throughput has reached the peak. Information indicating that the priority is higher than the priority is output to the output unit 107. Otherwise, information indicating that the priority of the communication packet is not changed is output to the output unit 107.

Note that the functions of the measurement unit 105, the comparison unit 113, and the priority setting unit 114 are realized by executing a program stored in the storage unit 120 by the CPU of the mobile terminal 10.
(Audio processor 117)
The audio processing unit 117 is configured by, for example, a receiver and a microphone. The voice processing unit 117 is controlled by the control unit 101 and has a function of performing voice input / output processing in a call performed by the mobile terminal 10.

The receiver is controlled by the control unit 101 and has a function of converting an electrical signal sent from the control unit 101 into sound and outputting the sound.
The microphone has a function of converting sound into an electric signal and sending the converted electric signal to the control unit 101.
(Display unit 118, operation unit 119)
The display unit 118 is a liquid crystal display included in the touch panel 150, for example.

The display unit 118 is controlled by the control unit 101 and has a function of generating and displaying an image to be presented to a user who uses the mobile terminal 10.
The operation unit 119 is, for example, a touch pad and operation button group 153 included in the touch panel 150.
The operation unit 119 is controlled by the control unit 101 and has a function of accepting a contact operation or a button pressing operation from a user using the mobile terminal 10.

Specifically, the touch pad of the operation unit 119 has a function of converting a contact operation from the user to a display object such as an icon displayed on the liquid crystal display of the touch panel into an electrical signal and outputting the electrical signal. Further, the operation unit 119 has a function of converting a user's button pressing operation into an electric signal, moving a cursor displayed on the liquid crystal display, and selecting a display object such as an icon.
(Power supply unit 121)
The power supply unit 121 is, for example, a secondary battery pack that can be repeatedly charged.

The power supply unit 121 has a function of supplying power for driving each functional unit of the mobile terminal 10 according to an instruction from the control unit 101.
<1.3 Data structure>
(TCP packet)
FIG. 4 is a diagram illustrating a data structure of a TCP packet used for TCP / IP communication. In FIG. 4, the data structure of the TCP packet is illustrated by a plurality of lines every 32 bits, but in actuality, it is data of a sequential bit string.

The TCP packet is composed of a header part and a data part 312.
The header part includes a source port number 301, a destination port number 302, a sequence number 303, an ACK number 304, a data offset 305, a reserved area 306, a control bit 307, a window size 308, a checksum 309, an urgent pointer 310, And option 311. There may be a case where the option 311 is not provided (the size is 0). The TCP packet has a structure in which a data portion 312 follows this header portion.

The transmission source port number 301 has a data length of 16 bits and is a number for identifying the port number used for transmission by the transmission source of the TCP packet.
The destination port number 302 has a data length of 16 bits and represents a standby port number used for receiving a communication packet in the destination device.
The sequence number 303 has a 32-bit data length and is a field for designating a sequence number for ordering the data to be transmitted. For each byte of data to be transmitted, a sequence number is assigned in ascending order one by one to specify how far the data has been transmitted. Specifically, the sequence number is managed on the transmission side of the communication data, and each time data is transmitted, the sequence number is added by the number of bytes of the transmitted data, and transmitted in a TCP packet.

  The ACK number 304 has a data length of 32 bits, and is a field that indicates to which byte position the received data has been received. The ACK number 304 is set in a response TCP packet and transmitted to indicate how far the data receiving side has received. The value of sequence number +1 of the data position where reception has been completed is set. Note that this ACK number field is valid only when an ACK flag described later is on.

  The data offset 305 has a data length of 4 bits and represents a position where the data portion 312 starts. Since the data part immediately follows the header part, the value of the data offset 305 represents the size of the header. The value of the data offset 305 is set with 32 bits (4 bytes) as one unit. That is, if the option 311 is not provided, the header portion is 20 bytes, so the value of the data offset 305 is set to “5 (“ 0101 ”in binary number)”.

The 6-bit data length following the data offset 305 is reserved as a reserved area 306, and currently “0” is set.
The control bit 307 has a data length of 6 bits, each of which is a 1-bit URG (urgent) flag, ACK flag, PSH (push) flag, RST (reset) flag, SYN (synchronize) flag, and FIN (finish) flag. Consists of.

From the URG flag to the FIN flag, the initial value is “0” by default. Set to "1" to enable the corresponding flag.
Hereinafter, transmission or reception of a TCP packet in which the ACK flag is set to be valid may be expressed as “transmit ACK” and “receive ACK”, respectively. The same expression may be used for other flags.

The URG flag indicates that “emergency data” is included in the TCP packet.
If the ACK flag is “1”, it indicates that a valid ACK number is included in the TCP header. Actually, “1” is set in the ACK flag of all TCP packets except the first TCP packet transmitted at the time of establishing the TCP connection. On the other hand, since the contents of the ACK number 304 have no meaning in the first TCP packet transmitted, “0” is set in the ACK flag.

The PSH flag is a flag for requesting that received data be immediately delivered to a higher-level device.
The RST flag is set when it is desired to interrupt or reject the TCP connection. If this RST flag is set, the receiving side considers that the connection request has been rejected and must discard or forcibly terminate the current TCP connection.

  When the TCP connection is interrupted for a long time due to an error or the like and the consistency between the sequence number and the ACK number cannot be obtained, this RST is transmitted instead of ACK, and the current TCP connection is forcibly terminated. In addition, if an RST is transmitted instead of an ACK in response to a TCP connection request, this indicates that the connection is rejected. For example, RST is transmitted when resources on the server side are insufficient to respond to an open request or when it is desired to reject a connection request from an unauthorized IP address.

  When a TCP connection is requested, a TCP packet in which “1” is set in the SYN flag is transmitted. Thereby, the TCP connection open process is started. Since the TCP connection is a two-way communication path, the SYN flag is set to be valid for the first connection request packets transmitted from both sides, but the SYN flag of the second and subsequent TCP packets is “0”. Is set. When receiving SYN, the receiving side synchronizes its own ACK number with the received sequence number. This prepares for subsequent communications.

The FIN flag is used for terminating the TCP connection. A TCP packet in which the FIN flag is set to be effective means that it is not necessary to receive any more data. When the FIN is received, a termination process is started. When FIN is sent from both sides, the TCP connection is terminated, and resources such as an internal buffer prepared for the TCP connection are released.
The window size 308 has a data length of 16 bits and is used to transmit the TCP window size on the receiving side to the transmitting side. The transmitting side determines the maximum amount of data that can be transmitted before receiving an ACK from the receiving side, according to the TCP window size notified from the receiving side. When the value of the window size 308 is “0”, it means that data cannot be received, and is used to request the transmission side to temporarily stop data transmission. Since the window size 308 is 16 bits long, a TCP window size of up to 65,535 bytes can be set.

The acquisition unit 102 of the mobile terminal 10 according to the present embodiment reads the window size 308 included in the communication packet sent from the server device 50 and acquires the TCP window size.
The checksum 309 has a 16-bit data length and is a field for storing inspection data for checking the consistency of the TCP packet. The checksum value is calculated using “1's complement arithmetic”.

The urgent pointer 310 has a data length of 16 bits, and when urgent data exists in the TCP packet, the position of the urgent data is set. At this time, the set value is a value represented by an offset from the sequence number 303. If urgent data exists in the TCP packet, “1” is set in the URG flag.
The option 311 has a variable data length of 32 bits and is used for setting various characteristics in the TCP connection. For example, MSS and window size scaling options are used to set each TCP connection.

  The data part 312 has a variable data length and follows immediately after the header part. If urgent data is present, the urgent data follows immediately, followed by normal data. There is also a packet having only a TCP header in which no data exists in the data portion. For example, a TCP packet transmitted to simply notify the TCP window size or notify keep-alive does not require a data part. Keep alive is a communication function for notifying and confirming that a TCP connection is active by sending and receiving empty TCP packets at regular time intervals even when no data communication is performed. . If no communication is performed, it is determined that there is no communication, and the communication line is disconnected or the TCP connection is timed out and disconnected. This is used to prevent this.

(IP packet)
Next, the data structure of the IP packet will be described.
FIG. 5 is a diagram illustrating a data structure of an IP packet used for TCP / IP communication. In FIG. 5, the data structure of the IP packet is illustrated by a plurality of lines every 32 bits, but in actuality, it is data of a sequential bit string.

The IP packet is composed of a header part and a data part 414.
The header part includes a version 401, a header length (IHL: Internet Header Length) 402, a service type (TOS: Type Of Service) 403, a datagram length 404, an ID 405, a flag 406, a fragment offset 407, and a TTL (Time To Live) 408. , Protocol number 409, header checksum 410, source IP address 411, destination IP address 412, and option 413. Note that there may be no option 413 (size is 0). The IP packet has a structure in which a data portion 414 follows this header portion.

Version 401 has a data length of 4 bits and indicates the version of the IP protocol.
In the communication system 1 of the present embodiment, since communication using IPv4 (Internet Protocol version 4) is used, the value is always “4 (“ 0100 ”in binary number)”.
The header length 402 has a 4-bit data length and indicates the size of the header part.

The value of the header length 402 is set with 32 bits (4 bytes) as one unit. That is, if the option 413 is not provided, the header portion is 20 bytes, so the value of the header length 402 is set to “5 (“ 0101 ”in binary number)”.
The service type 403 has a data length of 8 bits and is used to specify the priority of the IP packet.

  Of the 8 bits of the service type 403, 3 bits from the head are called IP Precedence and are used to set priority. That is, IP priority can be set to eight levels using IP precedence, and the higher the value, the higher the priority. For example, “0” is generally set in IP precedence for IP packets transmitted from a PC, and “5” is generally set for IP packets from voice devices such as IP phones.

When the mobile terminal 10 according to the present embodiment determines that the actual throughput at the time of transmission from the PC 20 to the server device 50 has reached its peak, the IP Precedence value in the communication packet related to the transmission is a value greater than “0”. By setting to, the priority is set high.
In the IP Precedence setting, the values “6” and “7” are used for network control such as routing, and the range of values for which the user can set the priority is “0” to “5”. is there.

  The datagram length 404 has a data length of 16 bits and indicates a value representing the size of the entire IP packet in bytes. Since the data length is 16 bits, the maximum amount of data that can be transmitted in one IP packet is 65,535 bytes (= 64 Kbytes). Further, since the minimum size of the header portion is 20 bytes, the minimum value of the datagram length 404 is 20 bytes + the byte size of the data portion 414.

When an IP packet is fragmented, the datagram length indicates the size of each IP fragment, not the size of the entire original IP packet.
The ID 405 has a data length of 16 bits and is used to identify a fragmented IP packet. Fragmentation of an IP packet (hereinafter referred to as “fragmentation”) is a function of transmitting an IP packet by dividing it into several parts. As described above, the maximum size of an IP packet is 64 Kbytes, but a network medium that can transmit data of this size at one time is rare. Therefore, transmission is performed after the IP packet is divided to a size that can be transmitted. Then, the IP packets received after being divided by the destination device are combined and reconfigured into the original size IP packet. The transmission source device sets a random 16-bit numerical value in the ID 405 every time an IP packet is transmitted. Since all packets that are fragmented from one IP packet are set to the same ID, they serve as identifiers when the destination device reconfigures them into one IP packet.

The flag 406 is special flag information having a data length of 3 bits and used for fragmentation. Of the 3 bits, the first 1 bit is reserved and is always set to “0”. The remaining two bits are an MF (More Fragment) bit and a DF (Don't Fragment) bit, respectively.
The MF bit is a flag indicating whether or not a fragmented IP packet continues. When one IP packet is divided into a plurality of packets, the MF bit is set to “0” in the last packet, and the MF bit is set to “1” in the other packets. While the MF bit is “1”, it indicates that a fragmented IP packet follows further.

The DF bit is a flag indicating whether or not fragmentation of the IP packet is prohibited. A router or the like on the communication path performs fragmentation of the IP packet if necessary, but does not perform fragmentation when the DF bit is set to “1”.
The fragment offset 407 has a data length of 13 bits, and indicates an offset value for indicating which part in the original IP packet corresponds to the fragmented IP packet. Although the fragment offset 407 is only 13 bits long, fragmentation is performed in units of 8 bytes. Therefore, by setting the value obtained by dividing the position of the original IP packet by 8 to the fragment offset 407, 64 Kbytes. A range can be represented.

The TTL 408 has a data length of 8 bits and represents the time when the IP packet is valid on the network. The value of TTL 408 is updated when header processing is performed in a device that relays IP packets on a communication path. When the value of TTL 408 is “0”, the IP packet is discarded. The value set here is a value in seconds.
The protocol number 409 has a data length of 8 bits and represents the type of network protocol corresponding to a higher rank than the IP protocol. Specific protocols set by the protocol number 409 include ICMP, TCP, UDP, GRE, and the like. When indicating each protocol of ICMP, TCP, UDP, and GRE, “1”, “6”, “17”, and “47” are set in the protocol number 409, respectively.

  The header checksum 410 has a data length of 16 bits, and the checksum of the header part is set. The data portion 414 is not a checksum target. This is because if the IP packet is fragmented, the checksum cannot be calculated because there is no data in the data part. The checksum value is calculated using “1's complement arithmetic”.

  The transmission source IP address 411 has a data length of 32 bits and indicates the IP address of the transmission source device. If there is an IP address of the transmission destination, the IP packet can be delivered to the transmission destination. However, when a response packet is returned from the transmission destination to the transmission source, the IP address indicated by the transmission source IP address 411 is changed. An IP packet is transmitted with the destination IP address 412 set.

The destination IP address has a data length of 32 bits and indicates the IP address of the destination device. In routing processing on the communication network, routing processing is performed based on the destination IP address 412 and an IP packet is delivered to the target device.
In the present embodiment, when the PC 20 transmits data to the server device 50, the IP packet in which the IP address of the own device is set as the transmission source IP address 411 and the IP address of the server device 50 is set as the destination IP address. Generate and send.

The option 413 has a variable data length of 32 bits, and is used to realize an additional function along with transmission of an IP packet. Note that the data length of the option 413 may be zero.
The option 413 is not normally used, but can be used for, for example, an IP packet passage log (time recording) or a routing process. Specifically, it can be used to forcibly specify a routing route or to record a route (IP address of a router, etc.) through which an IP packet has passed.

The data part 414 has a variable data length and immediately follows the header part. Usually, packets such as TCP, UDP, and ICMP are included. When the data portion 414 includes a TCP packet, the data portion 414 of the IP packet includes at least 20 bytes of data.
As described above, the data flow can be distinguished by four sets of the source and destination port numbers obtained from the TCP packet and the source and destination IP addresses obtained from the IP packet.

<1.4 Operation>
FIG. 6 is a sequence diagram illustrating an outline of data transmission / reception between the PC 20 and the server device 50 in the communication system 1.
In the sequence diagram of FIG. 6, it is assumed that TCP / IP communication between the PC 20 and the server device 50 is established, and a packet transmission / reception procedure for establishing communication is omitted.

When transmitting data to the server device 50, the PC 20 generates a communication packet related to the data to be transmitted in accordance with the TCP / IP protocol, and starts transmitting the communication packet (step S101).
The communication packet of TCP data # 1 output from the PC 20 is transmitted to the base station 30 via the portable terminal 10 and delivered to the server device 50 connected to the network 40 (step S102).

When receiving the communication packet of TCP data # 1, the server device 50 generates TCP ACK # 1 as a response to the reception confirmation and transmits it to the PC 20 (step S104).
The TCP ACK # 1 communication packet output by the server device 50 is sent to the mobile terminal 10 via the base station 30, and is delivered to the PC 20 connected to the mobile terminal 10 by tethering (step S105).

When receiving the TCP ACK # 1 (step S106), the PC 20 determines that the TCP data # 1 has been transmitted normally, and completes the transmission of the TCP data # 1 (step S107).
In order to continue transmitting data, the PC 20 generates TCP data # 2 and transmits a communication packet (step S108). Then, TCP data # 2 is transmitted as in step S102 (step S109).

  The server device 50 receives the TCP data # 2 (step S110), and returns a TCP ACK # 2 as a response to the reception confirmation of the TCP data # 2 (steps S111 and S112). Note that the processes of step S110, step S111, and step S112 are the same as the processes of step S103, step S104, and step S105, respectively.

When the PC 20 receives the TCP ACK # 2 (step S113), the PC 20 determines that the TCP ACK # 2 has been transmitted normally and completes the transmission of the TCP data # 2 (step S114).
Thereafter, the PC 20 repeats the process of transmitting the communication packet and receiving the reception confirmation until all the desired data is transmitted. Similarly, the server device 50 transmits ACK every time a communication packet is received. In this manner, communication packets related to data communication are transmitted / received between the PC 20 and the server device 50 via the portable terminal 10.

By the way, each time one TCP packet is sent, waiting for reception of an ACK has a large overhead related to data transmission and reception and is inefficient. Therefore, actually, data transmission / reception as shown in FIG. 7 is performed based on the TCP window size.
In the sequence diagram of FIG. 7, it is assumed that TCP / IP communication between the PC 20 and the server device 50 is established as in FIG. 6, and a packet transmission / reception procedure for establishing communication is omitted. . Further, it is assumed that the PC 20 and the mobile terminal 10 have acquired the TCP window size by the communication packet related to the data flow sent from the server device 50 when establishing the data communication.

  When transmitting data to the server device 50, the PC 20 generates a communication packet having a size not exceeding the data amount indicated by the TCP window size in accordance with the TCP / IP protocol, and starts transmitting the communication packet (step S201). ). In the example of FIG. 7, four communication packets of TCP data # 1 to # 4 are generated and transmitted (steps S202 to S205).

When the server device 50 receives the TCP data # 1 to # 4 (steps S206 to S209), it returns TCP ACKs # 1 to # 4 to the TCP data # 1 to # 4, respectively (steps S202a to S205a). . As shown in the figure, the server device 50 can transmit the TCP data # 2 to # 4 before confirming the reception of the TCP ACKs # 1 to # 3.
In FIG. 7, the TCP data # 1 to # 4 are received in numerical order, but the server device 50 may receive the TCP data # 1 to # 4 in any order. This is because the arrival order may change depending on the network route that these communication packets follow before reaching the server device 50 from the mobile terminal 10. Further, the order of TCP ACKs # 1 to # 4 received by the PC 20 may be any order.

The TCP ACK # 1 to # 4 communication packets output by the server device 50 are sent to the mobile terminal 10 via the base station 30 and delivered to the PC 20 connected to the mobile terminal 10 by tethering.
When receiving the TCP ACKs # 1 to # 4, the PC 20 determines that the TCP data # 1 to # 4 can be normally transmitted, and completes the transmission of the TCP data # 1 to # 4 (step S210).
The time from transmission of TCP data to reception of TCP ACK is RTT. For example, the time from the transmission of TCP data # 1 in step S202 to the reception of TCP ACK # 1 in step S202a is RTT # 1. Similarly, the time until the reception of TCP ACKs # 2 to # 4 for the transmissions of TCP data # 2 to # 4 is RTT # 2 to # 4, respectively. In this embodiment, the theoretical throughput is calculated using the average value of RTT within a certain period.

  As described above, the TCP window size can be used to continuously transmit communication packets as long as the TCP window size is not exceeded, and the TCP window size can be transmitted without waiting for the reception of an ACK for each communication packet from the server device 50. It is possible to transmit communication packets up to a data amount not exceeding. As a result, the overhead time related to the receipt confirmation of the transmission data can be shortened.

Next, a description will be given of processing in the mobile terminal 10 in the case where data flow processing related to data communication is performed in the sequence shown in FIG.
FIG. 8 shows a flowchart of data flow processing related to data communication in the mobile terminal 10.
The mobile terminal 10 not only relays a data flow related to data communication between the PC 20 and the server device 50, but also performs data communication between the PC 20 and another server device, or a device other than the PC 20 and a device on the network 40. It is also assumed that the data flow related to data communication with is relayed.

The control unit 101 controls the timer 106 to initialize the value of the timer Td used for calculating the actual throughput to 0 (Td = 0) (step S11).
Next, the control unit 101 receives, via the tethering unit 115, the SYN output from the PC 20 for requesting connection to the server device 50 in order to establish communication between the PC 20 and the server device 50, and sends it to the output unit 107. Output. The sending unit 109 takes out a communication packet indicating SYN from the buffer 108 and outputs it to the communication unit 110. The communication unit 110 transmits SYN to the base station 30 via the antenna 111 (step S12). The SYN received by the base station 30 is delivered to the server device 50 through the network 40. The server device 50 that has received SYN transmits an ACK that permits connection to the PC 20. The receiving unit 112 of the mobile terminal 10 receives the ACK addressed to the PC 20 (step S13), and outputs the ACK to the PC 20 via the tethering unit 115. By transmitting SYN from the PC 20 and receiving ACK as a response from the server device 50, communication between the PC 20 and the server device 50 via the mobile terminal 10 is established.

In addition, the control unit 101 controls the measurement unit 105 to measure the total data amount D of the communication packet output from the output unit 107 after initializing the timer Td in the communication packet addressed to the server device 50 from the PC 20. .
The acquisition unit 102 of the mobile terminal 10 acquires the TCP window size W from the window size 308 of the TCP packet 300 included in the communication packet from the server device 50 received by the reception unit 112 (step S14).

Next, the control unit 101 controls the timer 106 to initialize the value of the timer T for measuring the elapsed time and the value of the variable i (T = 0, i = 0) (step S15).
Further, the control unit 101 controls the timer 106 to initialize the value of the timer T _RTT (i) for measuring _RTT (T _RTT (i) = 0) (step S16).
The tethering unit 115 receives the communication packet Wp (i) transmitted from the PC 20 to the server device 50, and outputs it to the control unit 101 (step S17).

The control unit 101 controls the output unit 107 to output the communication packet Wp (i) received from the PC 20 to the buffer 108.
During the data transmission / reception with the server apparatus 50, the control unit 101 transmits the communication packet from the server apparatus 50 to the previous window size W and then transmits the communication packet to the accumulated packet amount transmitted from the time when the ACK is received. Whether or not the packet amount (cumulative packet amount + Wp (i), hereinafter referred to as “estimated cumulative packet amount”) obtained by adding the packet amount of the communication packet Wp (i) to be exceeded exceeds the TCP window size W. Determination is made (step S18).

  If the estimated cumulative packet amount does not exceed the TCP window size W (step S18: NO), the control unit 101 instructs the sending unit 109 to send the communication packet Wp from the buffer 108 to the server device 50 to the sending unit 109. (I) is taken out and output to the communication unit 110. Communication unit 110 transmits communication packet Wp (i) to base station 30 (step S19).

Next, the control unit 101 determines whether or not the receiving unit 112 has received an ACK for any of the communication packets Wp (x) (x = 0 to i) transmitted so far from the server device 50. (Step S20).
When the receiving unit 112 receives an ACK for the transmitted communication packet Wp (x) from the server device 50 (step S20: YES), the control unit 101 controls the timer 106 and stops the timer T _RTT (x). And hold the value at that time (step S21). Next, the control unit 101 increments the variable i by 1 (step S22). Then, it is determined whether or not the timer T is equal to or greater than a predetermined elapsed time threshold Tth (for example, 1 second) (step S23).

On the other hand, when the receiving unit 112 has not received an ACK for the transmitted communication packet Wp (x) from the server device 50 (step S20: NO), the timer T _RTT (x) is not stopped and the process of step S22 is performed. Process. When receiving an ACK for the transmitted communication packet Wp (x), the control unit 101 stops the timer T _RTT (x) and holds the value at that time.

If the timer T does not exceed the threshold value Tth, the control unit 101 repeats the processing from step S16. Otherwise, the control unit 101 returns ACK for all the transmitted communication packets Wp (x) (x = 0 to i). It is determined whether or not it has been received, and when the ACK is received, the timer T _RTT (x) corresponding to the received ACK is stopped and the value at that time is held (step S24).
When ACK is received for all the transmitted communication packets (step S24: YES), the control unit 101 calculates the average of T _RTT (i) measured so far and sets it as T _RTT (step S25). In this embodiment, it is assumed that all ACKs for transmitted communication packets are normally received.

Next, the control unit 101 resumes transmission of communication packets addressed to the server device 50 accumulated in the buffer, and performs calculation processing and comparison processing between the actual throughput and the theoretical throughput (step S26). Details of this process will be described later.
On the other hand, when the assumed accumulated packet amount exceeds the TCP window size (step S18: YES), the control unit 101 instructs the sending unit 109 to repeat the processing without sending the communication packet to the server device 50. The process proceeds to step S24 (step S18: YES).

  The control unit 101 determines whether or not the ratio (St / Smax) of the actual throughput value St calculated in step S25 to the theoretical throughput value Smax is larger than a threshold value Rth (for example, Rth = 0.95) (step step). S27). When St / Smax is larger than the threshold value Rth, that is, when the actual throughput is about the same speed as the theoretical throughput (step S27: YES), the control unit 101 determines that the actual throughput has reached its peak. The priority of the communication packet addressed to the server device 50 is set high (step S28). Specifically, for example, the value of IP Precedence included in the service type 403 of the IP packet 400 addressed to the server device 50 is set to a value larger than “0” (for example, “5”). It is assumed that “0” is set as the IP Precedence value of communication packets of other data flows. Furthermore, the communication packet generation process in the mobile terminal 10 of the present embodiment includes a process of rewriting a part of the header part of the communication packet from the PC 20 as described above.

Next, the control unit 101 controls the timer 106 to initialize the value of the timer Td used for calculating the actual throughput (Td = 0) (step S29).
The control unit 101 determines whether data transmission from the PC 20 to the server device 50 is completed (step S30). Specifically, a communication packet in which the FIN flag of the control bit 307 of the TCP packet is valid, that is, a communication packet in which “1” is set in the FIN flag is transmitted from the PC 20 to the server device 50. As a response, when a communication packet in which the FIN flag of the control bit 307 of the TCP packet is valid is received, it is determined that the data transfer is completed.

If it is determined that the data transfer has not been completed (step S30: NO), the process returns to step S14 to continue the data transfer process. If the data transfer is complete (step S30: YES), the process is terminated.
(Calculation and comparison of actual and theoretical throughput)
Next, calculation and comparison processing of actual throughput and theoretical throughput in the mobile terminal 10 will be described. This process is performed in step S26 in FIG. 8 and step S56 in FIG.

FIG. 9 shows a flowchart of calculation and comparison processing of the actual throughput and the theoretical throughput in the mobile terminal 10.
The calculation unit 104 calculates the theoretical throughput Smax using the TCP packet size W and _TRTT acquired by the acquisition unit 102 (step S31). Specifically, the calculation unit 104 calculates the theoretical throughput Smax by Smax = W / T _RTT .

Next, the control unit 101 controls the measurement unit 105, and in the communication packet addressed to the server device 50 received from the PC 20, the total data amount D output by the output unit 107 from the initialization of the timer Td to the present time. Obtain (step S32).
Then, the control unit 101 controls the measurement unit 105 to calculate the actual throughput St (step S33). Specifically, the measurement unit 105 calculates the actual throughput St by St = D / Td using the total data amount D and the value of the timer Td indicating the time taken to output the total data amount D. To do.

Further, the measuring unit 105 calculates a ratio Rt of the actual throughput to the theoretical throughput (step S34). Specifically, the measurement unit 105 calculates the ratio Rt by Rt = St / Smax. Note that St does not have a value greater than Smax, so Rt is a value of 1 or less.
<1.5 Discussion>
In the mobile terminal 10 according to the present embodiment, in data communication between devices relayed by the own device, the actual throughput is the same as the theoretical throughput based on the theoretical throughput calculated from the TCP window size and the RTT. It is determined whether or not communication is being performed, that is, whether or not the communication speed has reached its peak. Then, by setting the priority of the communication packet of the data flow at which the communication speed has reached its peak, the communication packet is preferentially transmitted from the mobile terminal 10 and the time during which the communication packet remains in the mobile terminal 10 is set. Shorten. As a result, the RTT of the data flow whose communication speed has reached its peak can be shortened, and the theoretical throughput is increased. Therefore, it is possible to eliminate the state where the actual throughput has reached the peak, and to improve the communication speed of the data flow whose communication speed has reached the peak.
<2. Second Embodiment>
<2.1 Overview>
In the mobile terminal 10 according to the first embodiment, the priority of communication packets related to the data flow is set higher based on whether or not the actual throughput has reached the peak for a predetermined data flow between specific devices. This shows an example of improving throughput.

In the second embodiment, for each data flow relayed by the mobile terminal 10, it is determined whether or not the actual throughput has reached a peak, and one data flow among the data flows that have been determined to have peaked is determined. The portable terminal 10 that improves the actual throughput will be described.
<2.2 Configuration>
Since the hardware configuration of the mobile terminal 10 of the present embodiment is basically the same as that of the mobile terminal 10 of the first embodiment, the same reference numerals as those of the mobile terminal 10 of the first embodiment are used.

  The control unit 101 of the mobile terminal according to the first embodiment determines whether or not the actual throughput has reached the peak for the communication packet related to data communication between the PC 20 and the server device 50, and determines the priority of the communication packet. The setting process is performed, but the control unit 101 of the mobile terminal 10 according to the present embodiment is different in that it determines whether or not all data communication to be relayed has reached a peak. When there is one data flow that has reached its peak, the priority for the data flow is increased by the same processing as described in the first embodiment. Further, when there are a plurality of data flows that have reached the peak, one of the data flows is selected to increase the priority.

<2.3 Operation>
FIG. 10 is a flowchart of processing in the mobile terminal 10 according to the second embodiment.
Note that the mobile terminal 10 of the present embodiment not only relays the data flow related to data communication between the PC 20 and the server device 50, but also performs data communication between the PC 20 and other server devices, as in the first embodiment. Alternatively, it is assumed that a data flow related to data communication between a device other than the PC 20 and a device on the network 40 is also relayed.

First, the control unit 101 initializes a counter count for counting the number of data flows that have reached the peak (count = 0) (step S41).
Next, the control unit 101 controls the output unit 107, analyzes the communication packet output to the buffer 108, and selects one data flow (step S42). Each data flow has at least one value of the source port number 301 and destination port number 302 of the TCP packet 300 and the source IP address 411 and destination IP address 412 of the IP packet 400 related to the data flow. Be different.

The control unit 101 instructs the acquisition unit 102 to cause the acquisition unit 102 to acquire the TCP window size W from the communication packet for the selected data flow (step S43).
Next, the control unit 101 controls the timer 106 to initialize the value of the timer Td (Td = 0) for use in calculating the actual throughput (step S44).
Control unit performs steps S45~ step S55, performs processing for T _RTT calculated while transmitting a communication packet. Steps S45 to S54 are the same as steps S15 to S25 in FIG. 8 described above.

Next, the control unit 101 calculates and compares the actual and theoretical throughputs already described with reference to FIG. 9 (step S56).
The control unit 101 determines whether or not the ratio (St / Smax) of the actual throughput value St calculated in step S56 to the theoretical throughput value Smax is larger than a threshold value Rth (eg, Rth = 0.95) (step step). S57). When St / Smax is larger than the threshold value Rth, that is, when the actual throughput is about the same speed as the theoretical throughput (step S56: YES), the control unit 101 determines that the actual throughput has reached its peak. Then, the value of the counter count is incremented by 1 (step S58).

  If the control unit 101 has not reached its peak (step S57: NO), the control unit 101 does not increment the counter count by 1 and determines whether there is a different unprocessed data flow (step S59). In some cases, the processing from step S42 to step S59 is performed until there is no unprocessed data flow. Note that the processing from step S42 to step S59 may be performed in parallel for different data flows.

When the determination of whether or not the actual throughput has reached the end for all the different data flows is completed, the control unit 101 ends the process when the value of the counter count is “0” (step S60: “0”). To do.
When the value of the counter count is “1” (step S60: “1”), the control unit 101 sets a high priority for the data flow that has reached its actual throughput (step S61). A specific setting method is the same as step S28 in FIG.

  When the value of the counter count is “2” or more (step S60: “2”), the control unit 101 selects a data flow that increases the priority according to a predetermined priority order (step S61). Specifically, the order in which data transfer to the device specified by which IP address should be prioritized is determined in advance according to the IP address of the transmission destination, and the IP address addressed to the highest order is determined according to the order. Select a data flow.

Next, the control unit 101 sets the priority of the selected data flow high (step S62).
<2.4 Discussion>
In the mobile terminal 10 according to the present embodiment, in data communication between devices relayed by the own device, the actual throughput is the same as the theoretical throughput based on the theoretical throughput calculated from the TCP window size and the RTT. When there are a plurality of data flows that are communicating, the priority order of one data flow that has the highest priority order determined in advance by the user among the corresponding data flows can be increased. Thereby, the user can improve the communication speed about the data flow which wants to improve the communication speed most.

<2.5 Modification>
<2.5.1 Overview>
In the second embodiment, when there are a plurality of data flows that are peaked, one data flow is selected according to a predetermined priority order, and the priority order of communication packets related to the data flow is set high. However, if one of the peaked data flows is selected and the priority of communication packets related to the data flow is set high, the priority of communication packets related to other data flows is relatively low. And other communication speeds may be reduced. In addition, when the communication partner device and data communication contents are known in advance, it is easy to set the priority of each data communication, but the communication partner device and data communication contents are various, In some cases, it is difficult to preset the priority order of data communication. Therefore, in the present modification, a description will be given of the mobile terminal 10 that allows the user to select a data flow that sets a higher priority when there are a plurality of data flows that are peaked.

Since the hardware configuration of the mobile terminal 10 of this modification is basically the same as that of the mobile terminal 10 of the second embodiment, the same reference numerals as those of the mobile terminal 10 of the second embodiment are used.
In the second embodiment, when there are a plurality of data flows that are peaked, the control unit 101 selects one data flow according to a predetermined priority order. On the other hand, in this modification, when there are a plurality of data flows in a peaked state, it is not always necessary to set the priority order for each data communication in advance, and each time the user is instructed to select a data flow to increase the priority. Different points are required. For this reason, the mobile terminal 10 according to the present modification displays a list of data flows that are peaked on the display unit 118, and accepts a data flow selection instruction from the user.
<2.5.2 Operation>
Hereinafter, the processing operation of the mobile terminal 10 in the present modification will be described.

FIG. 11 is a flowchart of processing in the mobile terminal 10 according to the modification.
The operation of the process before step S60 in the figure is the same as the operation of steps S41 to S59 in FIG.
When the determination of whether or not the actual throughput has reached the end for all the different data flows is completed, the control unit 101 ends the process when the value of the counter count is “0” (step S60: “0”). To do.

In addition, when the value of the counter count is “1” (step S60: “1”), the control unit 101 sets a high priority for the data flow in which the actual throughput reaches its peak (step S75). A specific setting method is the same as step S27 in FIG.
When the value of the counter count is “2” or more (step S60: “2”), the control unit 101 causes the display unit 118 to display a list of corresponding data flows (step S71).

  FIG. 12 is an example of a data flow selection menu displayed on the display unit 118. A UI 500 in FIG. 12 is an example of a UI displayed on the display surface of the display unit 118. In the list display area 501 of the UI 500, a list of data flows that have reached the peak is displayed. In the example of the figure, a list of data flows for each server of the communication partner is displayed. The user can select a data flow whose priority is set higher by selecting a data flow whose priority is set higher from the list of data flows displayed on the display unit 118 and touching an OK button 502. . When the user touches the cancel button 503, it indicates that the priority order of any data flow is not set high. In addition, when “any data communication is as it is” is selected and the OK button 502 is tapped, the priority order of any data flow is not set high.

  By such a user operation, the operation unit 119 accepts an instruction operation by the user (step S72). When the operation unit 119 receives that one data flow has been selected in accordance with an instruction from the user (step S73: YES), the control unit 101 acquires information indicating the selected data flow (step S74). The priority of the communication packet related to the data flow is set high (step S75).

On the other hand, when the data flow is not selected (step S74: NO), the process ends without performing the process of setting the priority of the communication packet related to the data flow high.
<2.5.3 Summary>
As described above, when there are a plurality of data flows in a peaked state, the mobile terminal 10 of the present modification displays a list of the data flows in a peaked state to the user. And an instruction as to which data flow priority is set higher. Therefore, the user can appropriately select the data flow that increases the priority order each time according to the status of data communication.
<3. Other variations>
The mobile terminal according to the present invention has been described above based on the embodiment. However, the mobile terminal can be modified as follows, and the present invention is limited to the mobile terminal as described in the above-described embodiment. Of course not.

(1) In the first and second embodiments and the modified examples, the theoretical throughput and the actual throughput are compared using the ratio. However, the comparison may be made based on the difference between the theoretical throughput and the actual throughput. For example, when the actual throughput is within a predetermined difference with respect to the theoretical throughput, it may be determined that the peak has been reached.
(2) In the first and second embodiments and the modified examples, the TCP window size, the RTT, and the data transfer amount are periodically measured, and the comparison is performed based on the measurement results. However, the present invention is not limited to this example. You may perform based on the result of having measured TCP window size, RTT, and the amount of data transfers in the fixed period of the start of data transfer. Thus, the processing can be simplified by making a comparison based on the result within a certain period of data transfer start.

  (3) In the first and second embodiments and the modification, the RTT value used when calculating the theoretical throughput is the average value of the RTT in the communication packet within a certain time, but is used when calculating the theoretical throughput. The value of RTT is not limited to this. The RTT may be calculated by any method using the communication packet related to the data flow. For example, an RTT value for each communication packet may be used, or an RTT average value for a certain number of communication packets may be used instead of an RTT for a communication packet within a certain time.

(4) In Embodiments 1 and 2, it has been described that all ACKs for transmitted communication packets can be normally received. However, in actual communication, ACK may be lost on the network path or time may be exceeded, and ACK may not be received normally. In such a case, error processing such as retransmission of a communication packet is performed according to the TCP / IP communication protocol. Then, if it can not normally receive the ACK may be determined the T _RTT without considering the RTT according to the communication packet.

  (5) In the second embodiment, an example is shown in which one data flow for increasing the priority is selected. However, a plurality of data flows for increasing the priority may be selected instead of one. If there is a margin in the total data transfer amount on the communication path, even if the priority of multiple data flows is increased in the communication packet related to the peaked data flow, the communication speed of other data flows will be increased. May not have an effect. Therefore, in such a case, the priority of communication packets related to a plurality of data flows may be set high instead of one of the data flows that has reached the peak.

  Further, when it can be determined that the effect of improving the throughput by increasing the priority is small, it is not necessary to perform the process of increasing the priority of any data flow. For example, if all the data flows are capped, the total data transfer amount in the original communication path may be small. In such a case, the communication related to any data flow If the packet priority is set high, the communication speed of other data flows may be further deteriorated. In such a case, the priority of any data flow may not be changed.

On the other hand, if only some data flows out of the plurality of data flows are considered to have a sufficient processing capacity, network bandwidth, and the like. In such a case, it may be determined that there is an effect of improving the throughput by raising the priority, and the priority of the data flow that has reached the peak may be set higher.
(6) In the first and second embodiments and the modification, the example has been described in which the priority of the communication packet related to the peaked data flow is set higher than that of other data flows. Not limited to. It is only necessary that the communication packet related to the data flow that has reached the peak can be set to be transmitted earlier than the communication packet related to the other data flow. For example, the priority of the communication packet related to the data flow other than the data flow that has reached the peak may be set lower than the priority of the communication packet related to the data flow that has reached the peak.

Note that the generation of the communication packet related to the data flow in the mobile terminal according to the present invention includes a process of rewriting a part of the header portion of the communication packet received from the PC 20 as described in the embodiment.
Further, in the first and second embodiments and the modification, the example in which the priority of the communication packet related to the data flow in the peaked state is set to “5” which is the highest priority that can be set by the user is shown. The priority may not be set to the highest priority. For example, the priority set for communication packets related to other data flows by the user may be examined, and a priority higher than those priorities may be set. If the highest priority among the priorities set for communication packets related to other data flows is “1”, the priority of the communication packet related to the data flow that has reached the peak state is set to “2”. Or more.

(7) In the first and second embodiments, the example in which the priority is set by changing the value of IP Precedence has been shown, but the setting of the priority is not limited to this. Any priority may be set as long as the priority of the communication packet output from the buffer can be set. For example, priority control defined by a wireless protocol such as LTE may be used.
(8) In the first and second embodiments and the modification, the case where there is one buffer used for temporarily storing communication packets is shown, but a plurality of buffers may be used. For example, a buffer for each priority may be prepared, and communication packets may be distributed and stored in buffers corresponding to the respective priorities. In this case, it is not always necessary to set a value for IP Precedence. In this way, each time the transmission unit 109 outputs a communication packet to the communication unit 110, the transmission unit 109 stores the communication packet in the high priority buffer without checking the priority of the communication packet stored in the buffer. Communication packets accumulated in the communication unit 110 can be output preferentially from communication packets.

  (9) In the first and second embodiments and the modification, the example in which the communication speed is improved for the data flow from the PC 20 connected to the mobile terminal 10 by the tethering function has been described. It is not limited to the data flow from the PC 20 connected to the mobile terminal 10. Any data flow may be used as long as the mobile terminal 10 outputs to an external network. For example, it may be a data flow in data communication between the mobile terminal 10 itself and the server device 50.

(10) The plurality of data flows are data flows that are distinguished by the IP address of the communication destination, but may be a series of data flows that are distinguished by a certain standard. For example, each data flow may be a data flow that is distinguished for each port number and each application. Alternatively, the data flow may be different for each data to be transmitted.
FIG. 13 is an example of a UI that presents a list of data flows that have reached the peak as a list of data flows for each application.

  The UI 510, the list display area 511, the OK button 512, and the cancel button 513 in the figure correspond to the UI 500, the list display area 501, the OK button 512, and the cancel button 513 in FIG. In the list display area 501 of FIG. 12, an example of displaying a list of data flows that are peaked for each communication partner server has been shown, but in the example of FIG. 13, the data flow for each application that performs data communication Is displayed. The user presented with this list can select an application whose communication speed is most desired to be improved from the list.

  (11) In the second embodiment, when there are a plurality of data flows in a peaked state, the data flow for increasing the priority is determined in advance according to the priority order determined for each data flow type. The determination of the data flow to be increased is not limited to this method. Any method may be used as long as one data flow for increasing the priority can be determined. For example, the data flow with the largest ratio of the theoretical throughput to the actual throughput, that is, the data flow with the highest actual throughput may be selected.

By using this method, it is possible to select a data flow that can expect the most improvement in throughput.
(12) In the second embodiment and the modification, the example in which the UI for data flow selection is displayed on the display unit 118 of the mobile terminal 10 has been described. However, the UI display is the display of the mobile terminal 10. The display is not limited to the unit 118. It may be displayed so that the user can perform an operation for selecting a data flow. For example, you may display on the display of PC20 which is communicating with the portable terminal 10 by the tethering function. In this case, the user can select a data flow whose priority is set high from the data flows displayed in a list by operating a keyboard, a mouse, or the like connected to the PC 20.

  (13) All or part of the components of the mobile terminal 10 described in the embodiment may be realized as a circuit that executes the function, or by executing a program by one or a plurality of processors. It may be realized. Moreover, it is good also as what is comprised as a package of IC, LSI, and other integrated circuits. This package is incorporated into various devices for use, whereby the various devices realize each function as shown in the embodiment.

  (14) Machine language for causing the processor of the mobile terminal 10 and the various circuits connected to the processor to execute the process of the mobile terminal 10 shown in the embodiment (the process shown in FIGS. 8 to 11, etc.) A control program composed of high-level language program codes can be recorded on a recording medium, or can be distributed and distributed via various communication paths. Such a recording medium includes an IC card, a hard disk, an optical disk, a flexible disk, a ROM, a flash memory, and the like. The distributed and distributed control program is used by being stored in a memory or the like that can be read by the processor, and the processor executes the control program to realize each function as shown in each embodiment. Will come to be. In addition to directly executing the control program, the processor may be compiled and executed or executed by an interpreter.

(15) The communication control method according to the present invention (the processes shown in FIGS. 8 to 11 and the like) may be realized by a computer using a computer program, or may be transmitted as a digital signal composed of the computer program. .
The present invention also provides a computer-readable recording medium such as a flexible disk, hard disk, CD-ROM, MO, DVD, DVD-ROM, DVD-RAM, BD (Blu-ray ( (Registered trademark) Disc), or recorded in a semiconductor memory or the like. Further, the present invention may be the computer program or the digital signal recorded on these recording media.

The computer program or digital signal according to the present invention may be transmitted via an electric communication line, a wireless or wired communication line, a network represented by the Internet, data broadcasting, or the like.
(16) The above embodiment and the above modifications may be combined as appropriate.

  The portable terminal according to the present invention is useful for improving the throughput of data transfer in network communication.

10 Mobile terminal 20 PC
30 base station 40 network 50 server apparatus 101 control unit 102 acquisition unit 103 RTT measurement unit 104 calculation unit 105 measurement unit 106 timer 107 output unit 108 buffer 109 transmission unit 110 communication unit 111 antenna 112 reception unit 113 comparison unit 114 priority setting unit 115 Tethering section 116 Tethering antenna 118 Display section 119 Operation section 150 Touch panel 301 Source port number 302 Destination port number 308 Window size 403 Service type 411 Source IP address 412 Destination IP address

Claims (13)

A mobile terminal that performs data communication including a plurality of data flows with at least one device via a network,
A buffer for temporarily storing communication packets related to each of the plurality of data flows;
A measuring unit for measuring RTT (Round Trip Time) in one data flow among a plurality of data flows;
The theory of data transfer amount per unit time when data is transmitted through the network using the RTT and information indicating the maximum transmittable data amount that can be transmitted without confirmation of reception from the destination device A calculation unit for calculating a theoretical throughput indicating a typical upper limit value;
In the one data flow, a measurement unit that measures an actual throughput indicating a data transfer amount per unit time actually transmitted;
The priority of one data flow is higher than the priority of another data flow when the actual throughput meets a predetermined criterion indicating that the actual throughput has reached the same throughput as the theoretical throughput. A priority setting section for setting
An output unit for generating a communication packet in which the priority is set and outputting the communication packet to the buffer;
A mobile terminal comprising: a transmission unit that preferentially transmits a communication packet having a high priority among communication packets stored in the buffer. The mobile terminal according to claim 1, wherein the priority setting unit determines that the criterion is satisfied when a ratio of the actual throughput to the theoretical throughput is equal to or greater than a predetermined value. The priority setting unit determines that the criterion is satisfied when an absolute value of a difference between the theoretical throughput and the actual throughput is equal to or less than a predetermined value. Mobile device. When the actual throughput of two or more data flows among the plurality of data flows satisfies the criterion, the priority setting unit satisfies the criterion based on a predetermined importance for each data flow. The data flow having the highest importance is selected from the data flows, and the priority in the selected data flow is set to be higher than the priority of the other data flows. The portable terminal described. When the actual throughput of two or more data flows among the plurality of data flows satisfies the criterion, the priority setting unit is exceptionally different from any other data flow for any data flow satisfying the criterion. The mobile terminal according to claim 1, wherein the process of setting the priority to be higher than the priority is not performed. Furthermore, a presentation unit that presents information related to the data flow to the user;
A reception unit that receives input from the user,
The presenting unit, when an actual throughput of two or more data flows among the plurality of data flows satisfies the criterion,
Presenting the user with information identifying each data flow meeting the criteria,
The accepting unit accepts user input related to the information presented by the presenting unit;
The priority setting unit identifies a data flow whose priority is to be increased based on the user input, and sets the priority in the data flow higher than the priority of other data flows. Item 2. A portable terminal according to Item 1. The data communication is data communication based on TCP (Transmission Control Protocol),
The mobile terminal according to claim 1, wherein the information indicating the transmittable data amount is a window size related to the one data flow. The mobile terminal according to claim 7, wherein each of the plurality of data flows is a data flow identified by an IP address of a transmission destination device. The mobile terminal according to claim 7, wherein each of the plurality of data flows is a data flow identified by a port number related to the data communication. The mobile terminal according to claim 7, wherein each of the plurality of data flows is a data flow identified by an application that outputs data included in the data flow. A method for controlling a mobile terminal that performs data communication including a plurality of data flows with at least one device via a network,
A buffer for temporarily storing communication packets related to each of the plurality of data flows;
A step of measuring an RTT (Round Trip Time) in one of the plurality of data flows;
The theory of data transfer amount per unit time when data is transmitted through the network using the RTT and information indicating the maximum transmittable data amount that can be transmitted without confirmation of reception from the destination device Calculating a theoretical throughput indicating a typical upper limit;
Measuring the actual throughput indicating the amount of data transferred per unit time actually transmitted in the one data flow;
The priority of one data flow is higher than the priority of another data flow when the actual throughput meets a predetermined criterion indicating that the actual throughput has reached the same throughput as the theoretical throughput. The steps to set
An output step of generating a communication packet in which the priority is set and outputting the communication packet to the buffer;
A transmission step of preferentially transmitting a communication packet having a high priority among the communication packets stored in the buffer. A control program for a mobile terminal that performs data communication including a plurality of data flows with at least one device via a network,
A buffer for temporarily storing communication packets related to each of the plurality of data flows;
A measurement step of measuring RTT (Round Trip Time) in one data flow among a plurality of data flows;
The theory of data transfer amount per unit time when data is transmitted through the network using the RTT and information indicating the maximum transmittable data amount that can be transmitted without confirmation of reception from the destination device A calculation step for calculating a theoretical throughput indicating a typical upper limit value;
In the one data flow, a measurement step of measuring an actual throughput indicating a data transfer amount per unit time actually transmitted;
The priority of one data flow is higher than the priority of another data flow when the actual throughput meets a predetermined criterion indicating that the actual throughput has reached the same throughput as the theoretical throughput. A priority setting step to set,
An output step of generating a communication packet in which the priority is set and outputting the communication packet to the buffer;
A control program for causing the portable terminal to execute a sending step for preferentially sending a communication packet having a high priority among the communication packets stored in the buffer. A computer-readable recording medium that records a control program for a portable terminal that performs data communication including a plurality of data flows with at least one device via a network,
A buffer for temporarily storing communication packets related to each of the plurality of data flows;
A measurement step of measuring RTT (Round Trip Time) in one data flow among a plurality of data flows;
The theory of data transfer amount per unit time when data is transmitted through the network using the RTT and information indicating the maximum transmittable data amount that can be transmitted without confirmation of reception from the destination device A calculation step for calculating a theoretical throughput indicating a typical upper limit value;
In the one data flow, a measurement step of measuring an actual throughput indicating a data transfer amount per unit time actually transmitted;
The priority of one data flow is higher than the priority of another data flow when the actual throughput meets a predetermined criterion indicating that the actual throughput has reached the same throughput as the theoretical throughput. A priority setting step to set,
An output step of generating a communication packet in which the priority is set and outputting the communication packet to the buffer;
A recording medium recording a control program for causing the portable terminal to execute a transmission step of preferentially transmitting a communication packet having a high priority among communication packets stored in the buffer.

Images