JP2000115249A - Data communication terminal and data communication method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide audio data and a moving picture with less delay and intermission with quality in matching with capability of each terminal independently of its processing capability. SOLUTION: The data communication terminal is provided with a main storage device 20, a network interface 40 that receives a data packet from a network 50 and transfers the packet to the main storage device 20, and a central processing unit 10 that processes the data packet transferred to the main storage device 20. The network interface 40 discriminates whether or not the received data packet is a data packet that can be aborted and stores the received data packet that can be aborted to a 1st reception buffer 42a and stores other packets to a 2nd reception buffer 42b. Furthermore, the network interface 40 is provided with a data abort control section 48, which discriminates whether or not the data packet received by the 1st reception buffer 42a satisfies a preset received data packet abort condition and aborts the data packet stored in the 1st reception buffer 42a based on the discrimination result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data communication system for transmitting multimedia data such as audio data and moving image data in real time via a network, such as a desktop conference system and video-on-demand. The present invention relates to a method and a data communication terminal used for the data communication method.

[0002]

2. Description of the Related Art In recent years, a so-called desktop conference system has been used in which communication between remote points is performed by transmitting voice data, moving image data, and the like via a network.

In a data communication terminal used in this desktop conference system, when transmitting audio data or moving image data, an audio signal from a microphone or a moving image signal from a video camera is converted into a digital signal. After encoding, the data is temporarily stored in a main storage device (hereinafter referred to as a main memory).

The audio data and the moving image data stored in the main memory are sent to a central processing unit (CPU).
By UDP / IP (User Datagram Protocol / Inter
(NetProtocol) protocol processing, packetized, and DMA-transferred to the network interface.
In the network interface, a packet header is added to each data packet and transmitted to the network.

At the time of reception by the data communication terminal, the packet data received from the network is decomposed by the network interface, the header is removed, and then the received data such as audio data and moving image data is DMA-transferred to the main memory. The central processing unit reads out the received data stored in the main memory, performs UDP / IP protocol processing, sends the obtained audio data and moving image data to the decoder through the respective interfaces, and decodes the data. , Return to analog signal. And
The sound is output from the speaker, and the moving image is displayed on the display.

FIG. 6 mainly shows an example of the configuration of a network interface portion in a conventional data communication terminal of this type. In FIG. 6, 1 is a central processing unit, 2 is a main memory, 3 is an expansion bus, 4 is a network interface, and 5 is a network.

The network interface 4 comprises a transmission buffer 11 and a reception buffer 12, a network transmission unit 13 and a network reception unit 14, a DMA control unit 15, and an interrupt signal generation unit 16.

At the time of transmission, encoded audio data and moving image data are DMA-transferred to the main memory 2 from an audio interface or graphic interface (not shown) as described above. The central processing unit 1 performs a UDP / IP protocol process on the audio data and the moving image data, and then sends a DMA transfer instruction to the DMA control unit 15. DMA controller 15
Receives the UDP / I from the main memory 2 to the transmission buffer 11 of the network interface 4.
The audio data and the moving image data in packet units subjected to the P protocol processing are DMA-transferred. Then, the data in the transmission buffer 11 is transmitted to the network 5 after the packet header is added in the network transmission unit 13.

At the time of reception, a data packet from the network 5 is received by the network receiving unit 14,
Here, the data is separated into a header part and a data part, and the data in the data part is stored in the reception buffer 12. Receive buffer 1
When the storage of the data packet in the data packet 2 is completed, the interrupt signal generation unit 16 notifies the central processing unit 1 via the bus 3 that the reception of the data packet has been completed as an interrupt signal. At the same time, the DMA control unit 15
The data in the reception buffer is transferred to the main memory 2 through the bus 3
DMA transfer.

Then, the central processing unit 1 reads out the received data stored in the main memory 2 and performs the UDP / I
P protocol processing is performed, and audio data is transferred to an audio interface (not shown), and moving image data is transferred to a graphic interface (not shown).

[0011]

In the desktop conference system, the above processing needs to be performed in real time. Therefore, the quality of transmitted audio and video is
It depends on the bandwidth of the network and the performance of the transmitting terminal and the receiving terminal. In recent years, the bandwidth of a network has been remarkably improved, and networks having a bandwidth of 100 Mbps or more have become widespread, and real-time transmission of high-quality voice and moving images has become possible. The encoding / decoding processing can also be performed at high speed by using dedicated hardware.

However, the processing of the UDP / IP protocol, which is widely used for the transmission of moving image data and audio data, performed by the central processing unit, requires a large amount of processing capability and supports a transmission speed of 100 Mbps or more. To do so, a terminal having a relatively high-performance central processing unit is required.

[0013] Even when a terminal equipped with a high-performance central processing unit is used, if other applications are operated simultaneously while transmitting and receiving audio data and moving image data, the central processing unit temporarily stops. There is a problem that protocol processing is delayed due to lack of processing capability, and phenomena such as delay and interruption of video and audio occur, thereby significantly reducing communication quality.

In a multipoint conference system for holding a conference at a plurality of points, the same voice data and moving image data are broadcast from one terminal to all other terminals in order to use the band effectively. A technique called multicasting is used. In this case, the same quality of voice and moving image is transmitted to each terminal, and communication of different quality cannot be provided for each terminal.

However, in the case of a multipoint conference, there is a high possibility that terminals having various processing capabilities will participate, and the quality required for each terminal may be different.

To solve this problem, Japanese Patent Application Laid-Open No. 9-46338.
In the prior art disclosed in the gazette, data of a plurality of qualities is transmitted by multicast, and information to specify the quality is added to the data to be transmitted to each packet and transmitted.
The receiving terminal employs a method of filtering and using data of desired quality.

However, in the prior art described in this publication, data received from the network and subjected to network protocol processing by the central processing unit is filtered, so that a terminal having a low processing capability performs network protocol processing. However, there is a problem in that the processing ability is exhausted only by using the data and only low-quality audio and moving images cannot be obtained.

As described above, in the conventional transmission and reception of audio data and moving image data, phenomena such as video and audio delays and interruptions occur due to a temporary lack of processing capability in a data communication terminal, and the communication quality is reduced. There is a problem that the temperature is significantly reduced. Furthermore, in the case of multicasting of audio and video, if there is a terminal with low processing capability,
There has been a problem that communication can be performed only with low-quality voice and moving images adapted to the terminal.

The present invention has been made in view of the above-described problems, and has improved the quality of each terminal and reduced delays and interruptions even if the processing capability fluctuates or the processing capability differs between terminals. The purpose is to be able to provide moving images.

[0020]

In order to solve the above problems, a data communication terminal according to the present invention is connected to a network, receives a main storage device, receives a data packet from the network, and transfers the data packet to the main storage device. In a data communication terminal comprising a network interface unit and a central processing unit for processing a data packet transferred to the main storage device, the network interface unit determines whether the received data packet is a discardable data packet. Receiving data determination means for determining whether the received data packet can be discarded, and at least one first reception buffer storing the discardable data packet based on the determination result by the received data determination means; The data packets other than the discardable received data are A second receiving buffer, an interrupt generating means for generating an interrupt to the central processing unit when a data packet is stored in the second receiving buffer, and a second storing buffer stored in the first receiving buffer. It is determined whether the data packet meets the data packet discarding condition, and the first packet is determined based on the determination result.
Data discarding control means for discarding the data packet stored in the receiving buffer of the above. The central processing unit periodically sets a flag indicating that the data packet is stored in the first receiving buffer. To read out the data packets stored in the first reception buffer, and read out the data packets stored in the second reception buffer in response to an interrupt from the interrupt generation means. It is characterized by.

[0021]

According to the present invention having the above-described configuration, the received data determination means determines whether the received data is a data packet that can be discarded. The possible data is stored in a first receive buffer and the other data is stored in a second receive buffer.

The data discarding control means discards the data packet in the first receiving buffer when the data packet in the first receiving buffer satisfies a preset discarding condition.

On the other hand, every time a data packet is stored in the second reception buffer, the data stored in the second reception buffer is notified to the central processing unit by the interrupt generation means and is read out. , All data will be read out without being discarded.

[0024] Therefore, data at a specified data rate or higher can be automatically discarded.
Even if the processing capacity of data communication terminals fluctuates or there is a difference in processing capacity between terminals, it is possible to provide audio and video with little delay and discontinuity with quality that matches the capabilities of each terminal. Become.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of a data communication terminal and a data communication method according to the present invention will be described with reference to the drawings.

[First Embodiment] FIG. 2 shows an example of the overall configuration of a data communication terminal according to this embodiment. That is, the data communication terminal of this embodiment includes a central processing unit (CPU) 10 and a main memory 20.
, The network interface 40, the graphic interface 60, and the audio interface 70 are connected through the extension bus 30.

The graphic interface 60 includes:
A video encoding / decoding unit 61 composed of dedicated hardware for encoding and decoding moving image data is connected. An input side of the video encoding / decoding unit 61 includes a video camera 6 as an example of a moving image information input device.
2 is connected, and a display 63 as an example of a moving image information output device is connected to the output side.

The audio interface 70 is connected to an audio encoding / decoding unit 71 composed of dedicated hardware for encoding and decoding audio data. On the input side of the audio encoding / decoding unit 71, a microphone 7 as an example of an audio information input device is provided.
2 is connected, and a speaker 73 as an example of an audio information output device is connected to the output side.

The network interface 40 sends data stored in the main memory 20 to the network 50, and receives data from the network 50 and transfers it to the main memory 20, as in the above-described conventional example.

This embodiment has a feature in the configuration of the network interface 40. A block diagram of a configuration example of the network interface 40 is shown in FIG.
As shown in FIG.

This network interface 40
Includes a transmission buffer 41 for storing transmission data, a first reception buffer 42a for storing discardable data among data received via the network 50, a second reception buffer 42b for storing other reception data, A network sending unit 43, a network receiving unit 44,
An MA control unit 45, an interrupt signal generation unit 46 as an interrupt generation unit, a reception data determination unit 47, and a data discard control unit 48 are provided.

The transmission buffer 41 and the reception buffers 42a and 42b are constituted by, for example, FIFO memories. In this FIFO memory, data of a plurality of packets is continuously stored. These packets are separated by
A code specified in advance is identified by being inserted into transmission data.

When the storage of the data packet in the second reception buffer 42b is completed, the interrupt signal generation unit 46
As an interrupt signal, it notifies the central processing unit 10 via the bus 30 that the reception of the data packet has been completed. At the same time, the DMA control unit 45 DMA-transfers the data in the reception buffer 42b to the main memory 20 via the bus 30.

Therefore, the reception data stored in the second reception buffer 42b is subjected to the reception processing and the decoding processing in the same manner as in the conventional example shown in FIG. The second reception buffer is selected so as to store, for example, reception data to be decoded in real time without discarding the data. In this example, audio data is selected as this type of data. This is because, in audio, a break in data becomes a break in audio output, so that the quality of reproduced voice is likely to deteriorate.

On the other hand, in this embodiment, the reception buffer 42b stores reception data from which some data can be discarded. In this example, the data which can be partially discarded is moving image data. This is because, even if a moving image is in a so-called frame-dropped state where some data is lost, it is relatively easy to grasp the image content.

The received data determining unit 47 determines whether the received data is data that can be discarded based on the conditions specified for the received data, and based on the determination result, determines whether the received data is the first data.
Receiving buffer 42a or second receiving buffer 42b
To store data. Whether or not the data can be discarded is determined by using information inserted in the packet header, as described later.

The data discarding control unit 48 monitors the data stored in the first receiving buffer 42a, and discards data that satisfies the specified discarding condition from the first receiving buffer 42a. Also, the data discard control unit 4
8 has a flag indicating whether data exists in the first reception buffer 42a.

The DMA controller 45 controls data transfer from the main memory 20 to the transmission buffer 41 and data transfer from the first reception buffer 42a and the second reception buffer 42b to the main memory 20. It also controls the transfer of data from the first reception buffer 42a to the data discard control unit 48. The transfer of data from the first reception buffer 42a to the data discarding control unit 48 is performed in preference to the transfer from the first reception buffer 42a to the main memory 20.

The network sending unit 43 adds a header or the like to the data stored in the transmission buffer 41, generates a network packet such as an Ethernet frame, and sends it to the network 50. The header includes fields such as protocol, priority, transmission address, transmission port, reception port, and data packet type. In this example, whether or not discarding is possible is specified by the transmission address and the reception port among the information of these fields, as described later.

The network receiving unit 44 receives a network packet such as an Ethernet frame from the network 50, and transfers the data with the header removed to the received data determining unit 47.

In the case of this embodiment, the reception data judgment section 47 is configured as shown in FIG. The reception data determination unit 47 will be described.

The header storage buffer 471 stores the header portion of the data sent from the network receiving unit 44. The value stored in the header mask register 475 indicates that all bits of the field to be compared by the header comparator 473 in the header, such as the protocol, priority, transmission address, transmission port, reception port, and data packet type, are “ 1 ", and the bits of the other fields are" 0 ".

The header specification register 474 stores the values of the fields of the discardable data among the fields of the header to be compared, and the values of the other fields are 0.

In the header comparator 473, when the header data is stored in the header storage buffer 471, at the same time, the header storage buffer 471 and the header mask register 475
And the result of the AND operation with the header specification register 4
Exclusive OR with the value of 74. The above operation result is
It becomes "0" only when the values of all fields specified in the header of the received data match the specified values.

The data separating section 472 includes a header comparator 47
3 is “0”, the header storage buffer 47
1 data to the first reception buffer 42a,
The data continuously transmitted from the network receiving unit 44 is transferred to the first receiving buffer 42a until the end of the packet is detected.

On the other hand, if the operation result of the header comparator 473 is other than "0", the data in the header storage buffer 471 is transferred to the second reception buffer 42b, and the data continuously transmitted from the network Is transferred to the second reception buffer 42b until the end of the data is detected.

Next, in the case of this embodiment, the data discarding control section 48 is configured as shown in FIG. That is, the data discarding control unit 48 of this embodiment includes a data discarding unit 481 and a maximum retained data amount designation register 48.
2 and a held data amount counter 483.

The maximum holding data amount specifying register 482 is
This register specifies the maximum data amount to be stored in the first reception buffer 42a. Retained data amount counter 483
Is a counter indicating the amount of data stored in the first reception buffer 42a, which is counted up by a write signal WE to the first reception buffer 42 from the reception data determination unit 47, The countdown is performed by the read signal RE for one reception buffer 42a.

The data discarding means 481 stores the value of the held data amount counter 483 in the maximum held data amount designating register 4.
82, and if the former is greater than the latter,
It issues a data read request REQ from the first reception buffer 42a to the control unit 45. Reading based on the data read request REQ is continued until a code indicating the end of the packet is read, and the read data is discarded.

Next, the operation of transmitting voice data and moving image data using the data communication terminal of this embodiment will be described with reference to FIGS.

The audio signal of the conference participant input from the microphone 72 is A / D-converted and encoded by the audio encoding / decoding unit 71, and then the audio signal is transferred to the main memory 20 by the audio interface 70.
Will be transferred.

The video signal of the conference participant input from the video camera 62 is also A / D-converted and encoded by the video encoder / decoder 8, and transmitted to the main memory 20 by the graphic interface 60. D
MA transfer is performed.

The audio data and the moving image data stored in the main memory 20 are subjected to UDP / IP protocol processing by the central processing unit 10 and then DMA-transferred to the network interface 40. In the network interface 40, data from the main memory 20 is stored in the transmission buffer 41. After the header of the data in the transmission buffer 41 is added to the data by the network
Sent to 0. As described above, the header includes the IP address of the sender, the reception port number of each data, the data packet type, and the like. In the case of this example, a different number for the receiving port number for each data such as audio data and moving image data is given in advance in a data communication system including the network 50.

Next, the operation of receiving voice data and moving image data using the data communication terminal of this embodiment will be described with reference to FIGS.

Before starting the data reception, the central processing unit 10 specifies the header mask register 475 and the header specification register 474 of the reception data judgment unit 47 of the network interface 40 and the maximum holding data amount specification of the data discard control unit 48. The value of the register 482 is set.

In the case of this example, in order to designate moving image data from a certain sender as data that can be discarded,
The IP address of the sender and the port number for receiving the moving image data are designated. That is, the header mask register 4
The bit corresponding to the sender IP address of 75 and the reception port number of the moving image data is set to “1”, and the other bits are set to “0”. The header specification register 474 contains:
The value of the sender's IP address and the value of the receiving port number of the moving image data are set.

Then, the maximum holding data amount specifying register 4
In 82, for example, a data amount for three frames is specified.
By specifying the number of frames, a temporary delay in the reception processing is caused by a time corresponding to the specified data amount, in this case,
If the time is within three frames, a moving image can be reproduced without dropping frames.

In the environment specified as described above, the data from the network 50 is received by the network receiving unit 44 in FIG. 1, and after being separated from the header and data, is sent to the received data determining unit 47.

The header comparator 47 of the reception data judgment section 44
At 3, the sender IP address and port number in the header are compared with the values specified in the header specification register 474, and if they match, the data is stored in the first reception buffer 42a and does not match If the data is
The data is stored in the second reception buffer 42b. In the case of this example, the moving image data is stored in the first reception buffer 42a, and the audio data is stored in the second reception buffer 42b.

The audio data stored in the second reception buffer 42b is immediately DMA-transferred to the main memory 20 by the DMA controller 45. At the same time, the completion of data reception is notified to the central processing unit 10 by an interrupt signal generated by the interrupt signal generation unit 46.

The central processing unit 10 receiving this notification
Reads the audio data stored in the main memory 20, performs the UDP / IP protocol processing, and transfers the data to the audio interface 70. The audio data transferred to the audio interface 70 is sent to the audio encoding / decoding unit 71, where the audio data is decoded, and the D / A
It is converted and output from the speaker 73.

The moving image data stored in the first receiving buffer 42a is read periodically according to an instruction from the central processing unit 10. The procedure is described below.

The central processing unit 10 periodically reads the value of the held data amount counter 483 of the data discarding control unit 48 of the network interface 40, and if the value is not 0, all data in the first reception buffer 42a is read. DMA transfer to the main memory 20 is performed.

The DMA transfer is performed by the DMA controller 45. The period in which the central processing unit 10 reads the value of the held data amount counter 483 is specified to be equal to, for example, the time interval of a moving image frame. However, when the CPU processing capacity is temporarily reduced due to the use of another application or the like, the read cycle may be long.

A part of the moving image data stored in the first reception buffer 42a may be discarded without being transferred to the main memory 20. In a moving image in which one frame is transmitted as one data packet, a part of data is discarded, so that a frame rate is reduced.

As described above, the central processing unit 10
Polling the value of the held data amount counter 483 of the data discarding control unit 48 as a flag indicating whether or not the data packet is stored in the first reception buffer 42a, and determining whether to discard the data according to the flag. I do.

For the algorithm for discarding data,
This will be described with reference to the configuration diagram of FIG.

When the data packet is stored in the first reception buffer 42a, the value of the held data amount counter 483 increases by the size of the data packet. At this time, if the value of the held data amount counter 483 is equal to or smaller than the value held in the maximum held data amount designation register 482, the data in the first reception buffer 42a is held as it is.

When the value of the held data amount counter 483 exceeds the value of the maximum held data amount designation register 29, first, the DMA from the first reception buffer 42a is read.
If a transfer is occurring, stop it. Next, data is read one byte at a time from the first reception buffer 42a, and the value of the held data amount counter 483 is reduced by one. This operation is continued until a code indicating the end of the packet is read. The read data is discarded.

Then, again, the held data amount counter 48
3 is compared with the value of the maximum held data amount specifying register 482, and if the value of the held data amount counter 483 is still equal to or larger than the value of the maximum held data amount specifying register 482,
Data destruction continues.

The moving image data DMA-transferred from the first reception buffer 42a to the main memory 20 is subjected to UDP / IP protocol processing by the central processing unit 10 and then transferred to the graphic interface 60 for video encoding. After being decoded by the decoding unit 61, it is D / A converted and displayed on the display 63.

According to the above embodiment, the audio data is transferred to the main memory 20 immediately upon receipt, so that the audio data is processed with higher priority than the moving image and is not discarded. The quality is guaranteed.

On the other hand, in the case of a moving image, the maximum data rate of data transferred to the main memory 20 can be set based on the maximum amount of retained data and the read time interval. If the number exceeds the threshold, a part of the moving image data is discarded. Therefore, by setting the maximum data rate in accordance with the processing capability of the terminal, a moving image having a quality suitable for the processing capability of the terminal can be displayed without stopping the moving image.

When the processing capacity of the terminal is temporarily insufficient due to the use of another application or the like, the data rate transferred to the main memory 20 is automatically reduced by increasing the read time interval. However, consumption of the processing capacity of the terminal is prevented. This temporarily reduces the quality of the moving image, but can prevent the moving image from being delayed or stopped.

In the above-described embodiment, a case has been described in which the number of the first reception buffers 42a is one. However, a plurality of first reception buffers may be provided. In this case, the maximum data rate or the read time interval is changed for each of the plurality of first reception buffers, so that the maximum data rate of the transfer to the main memory 20 is changed for each of the plurality of first reception buffers 42a. For example, the image quality can be changed for each sender of the moving image, or the priority of the processing can be set.

In this case, in order to change the read time interval from the plurality of first receiving buffers, a flag indicating that data packets of the receiving buffer are stored is set to a period at which the central processing unit polls, and What is necessary is just to change for each of several 1st reception buffer. Also, by setting the maximum held data amount to a different value for each of the plurality of first reception buffers, the reading speed can be changed for each type of data packet.

In the above embodiment, the moving image data from one sender is designated as discardable data. However, the protocol, priority, transmission address, transmission port, reception port, and It is needless to say that moving image data from a plurality of senders can be specified as discardable data or data other than moving image data can be specified as discardable data by an arbitrary combination such as a data packet type. .

[Second Embodiment] Next, a data communication terminal according to a second embodiment of the present invention will be described. The configuration of the data communication terminal according to the second embodiment is as follows.
As compared with the data communication terminal according to the above-described first embodiment, only the configuration of the data discarding control unit 48 in the network interface 40 is different. Therefore,
In the following description, only the components of the data discard control unit 48 according to the second embodiment will be described, and the description of the same components as those of the first embodiment will be omitted. FIG. 5 shows a block diagram of a configuration example of the data discarding control unit 48 in the case of the second embodiment.

The data discarding control unit 48 according to the second embodiment includes a data discarding unit 481, a minimum holding time designation register 484, and a packet time register 485.

The minimum holding time designation register 484 stores the first
Is a register for designating the minimum time for which the data stored in the reception buffer 42a of FIG. The packet time register 485 is a register that records, for each data packet, the time when the data packet was stored in the first reception buffer 42a or the generation time of the data packet written in the header of the data packet.

The data discarding means 481 calculates the current time,
The current time is compared with the value of the time stored at the beginning of the packet time register 485, and the current time is
5 is greater than or equal to the value stored in the minimum holding time designation register 484 from the time stored at the beginning of
A request REQ for reading data from the first reception buffer 42a is issued to the MA control unit 45. Reading is continued until a code indicating the end of the packet is read, and the read data is discarded.

Next, an operation when transmitting voice data and moving image data using the data communication terminal of the second embodiment will be described.

The operation of the data communication terminal according to the second embodiment differs from the operation of the data communication terminal according to the first embodiment only in the operation of the data discard control unit 48. Therefore, only the operation of the data discard control unit 48 will be described below.
A description of the same parts as in the first embodiment will be omitted.

An algorithm for discarding data is as follows.
This will be described with reference to the configuration diagram of FIG.

When the data packet is stored in the first reception buffer 42a, the last entry of the packet time register 485 is the time ts when the data packet was stored in the first reception buffer 42a or the header of the data packet. The generation time tg of the data packet written therein is written.

At this time, at the same time, the value of the first entry of the packet time register 485 is read and compared with the current time tp. When the difference Δt between the current time and the time ts or tg read from the packet time register 485 is equal to or smaller than the value stored in the minimum holding time designation register 484, the data in the reception buffer 42a is held as it is.

If Δt is equal to or greater than the value of the minimum holding time designation register 484, first, the first reception buffer 4
When the DMA transfer from 2a is being performed, this is stopped. Next, data is read one byte at a time until a code indicating the end of the packet is read from the first reception buffer 42a. The read data is discarded.

When the reading until the code indicating the end of the packet is read from the first reception buffer 42a is completed, the first entry of the packet time register 485 is discarded. Then, the value of the time at which the packet becomes newly the first entry of the packet time register 485 is read out again, and the read value is compared with the current time.
If the value is 4 or more, the data of the packet is discarded in the same manner as described above, and the first entry of the packet time register 485 is discarded. Thereafter, the above-described data discard operation is performed until the time difference Δt becomes equal to or smaller than the value of the minimum holding time designation register 484.

According to the above-described second embodiment, the audio data is transferred to the main memory 20 immediately upon receipt, so that the audio data is processed with priority over the moving image and is not discarded. , Voice quality is guaranteed.

On the other hand, in the case of moving image data, the maximum data rate of data transferred to the main memory 20 can be set based on the minimum holding time and the read time interval, and the data reception speed exceeds the maximum data rate of this data. In this case, part of the moving image data is discarded.

Therefore, by setting the maximum data rate in accordance with the processing capability of the terminal, a moving image of a quality suitable for the processing capability of the terminal can be displayed without stopping the moving image. Further, when the processing capacity of the terminal is temporarily insufficient due to use of another application or the like, the data rate transferred to the main memory 20 is automatically reduced due to an increase in the reading time interval, and the processing of the terminal is reduced. Capacity consumption is prevented. This temporarily reduces the quality of the moving image, but can prevent the moving image from being delayed or stopped.

Also in the second embodiment, a plurality of first reception buffers are provided for each type of data packet,
By setting the minimum time at which a discardable data packet should be held to a different value for each of the plurality of first reception buffers, the reading speed can be changed for each type of data packet.

[Third Embodiment] Next, a data communication terminal according to a third embodiment of the present invention will be described. The data communication terminal according to the third embodiment has the same hardware configuration as the data communication terminal according to the first embodiment or the data communication terminal according to the second embodiment.

The third embodiment is applied to the following.
This is a case in which moving image data is broadcast from one terminal to a plurality of terminals using a multicast technique. The moving image data transmitted in the third embodiment is hierarchically encoded data. The hierarchically encoded data in this example is composed of the lowest hierarchical data obtained by encoding the lowest resolution moving image and the higher hierarchical data obtained by encoding the difference between the higher resolution image and the lowest resolution image. Become. The data of a higher hierarchy can be prepared not for one data but for a plurality of hierarchies.

The transmitting terminal assigns an identification number to the data of each layer in order from the lowest resolution. When transmitting the data, the identification number is added as a part of the header, and then the data is multicast. Send by

On the other hand, each receiving terminal is configured so that some of the identification numbers can be designated as data that can be discarded according to the processing capability of the terminal itself. As a result, in the third embodiment, it is possible to specify the minimum resolution at which the display of all frames is guaranteed and to freely switch the resolution to be displayed for each receiving terminal. Become.

In order to specify the minimum resolution at which the display of all frames is guaranteed, data having an identification number larger than the identification number of the resolution to be guaranteed is specified as discardable data. The data of the resolution to be guaranteed is temporarily stored in the reception buffer 42b, and all data is stored in the main memory 2b.
0, the display of all frames is guaranteed.

When the resolution to be displayed is freely switched, the first reception buffer 4 for holding the discardable data
2a are prepared for the identification numbers, and data having the respective identification numbers are held in separate first reception buffers 42a. Then, the central processing unit 10 can display the image of the resolution to be displayed by reading out only the data of the buffer in which the data having the identification number or less of the resolution to be displayed is stored. In this case, all other data is discarded.

According to the third embodiment described above,
Even when the transmitting terminal broadcasts to multiple terminals by multicasting, each receiving terminal can display a moving image with a resolution according to the processing capability of its own terminal and freely switch the resolution to be displayed It is possible to do.

[0100]

As described above, according to the present invention, since a part of transmitted data can be automatically discarded at a data rate higher than a certain data rate, the processing capability of the data communication terminal varies. In addition, even when there is a difference in processing capability between terminals, a flexible real-time communication system capable of providing audio and moving images with little delay and no interruption with quality adapted to the capability of each terminal can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram of a main part of an embodiment of a data communication terminal according to the present invention.

FIG. 2 is an overall block diagram of an embodiment of a data communication terminal according to the present invention.

FIG. 3 is a block diagram showing a configuration example of some blocks of the block diagram of FIG. 1;

FIG. 4 is a block diagram showing a configuration example of some blocks of the block diagram of FIG. 1;

FIG. 5 is a block diagram of a main part of another embodiment of the data communication terminal according to the present invention.

FIG. 6 is a block diagram of a main part of a conventional data communication terminal.

[Explanation of symbols]

 Reference Signs List 10 central processing unit 20 main memory 30 expansion bus 40 network interface 41 transmission buffer 42a, 42b reception buffer 43 network transmission unit 44 network reception unit 45 DMA control unit 46 interrupt signal generation unit 47 reception data determination unit 48 data discard control unit 50 Network 60 Graphics interface 70 Audio interface 471 Header storage buffer 472 Data separation unit 473 Header comparator 474 Header designation register 475 Header mask register 481 Data discarding means 482 Maximum retained data amount designation register 483 Retained data amount counter 484 Minimum retention time designation register 485 packet time register

 ──────────────────────────────────────────────────続 き Continuing on the front page F term (reference)

Claims (11)

[Claims] 1. A main storage device connected to a network,
A data communication terminal comprising: a network interface unit that receives a data packet from the network and transfers the data packet to the main storage device; and a central processing unit that processes the data packet transferred to the main storage device. Means for determining whether a received data packet is a discardable data packet; and at least one of the at least one discardable data packet being stored based on a result of the determination by the received data determination means. A first reception buffer, a second reception buffer in which data packets other than the discardable reception data are stored based on a result of the determination by the reception data determination means, and a data packet stored in the second reception buffer. When stored, it is assigned to the central processing unit. Interrupt generation means for generating only the data packet stored in the first reception buffer; determining whether the data packet stored in the first reception buffer satisfies a data packet discarding condition; and storing the data packet in the first reception buffer based on the determination result. And a data discard control unit for discarding the data packet, wherein the central processing unit periodically polls a flag indicating that the data packet is stored in the first reception buffer, A data communication method comprising: reading data packets stored in a first reception buffer; and reading data packets stored in the second reception buffer in response to an interrupt from the interrupt generation means. Terminal. 2. The data communication terminal according to claim 1, wherein the received data determination unit discards the received data packet by comparing a value of at least one field in a header of the received data packet with a designated value. A data communication terminal for determining whether or not a data packet is possible. 3. The data communication terminal according to claim 1, wherein said data discarding control means is configured to store the data amount in said first reception buffer in a preset first reception buffer. A data communication terminal for discarding an oldest data packet among data packets stored in the first reception buffer when the amount is equal to or more than an upper limit value of the amount. 4. The data communication terminal according to claim 1, wherein said data discarding control means includes a storage unit for storing information of a minimum time at which said discardable data packet is to be held, and said discardable data is stored. Recording the time at which the packet was received, and discarding the data packet from the first reception buffer if the difference between the recorded time and the current time is greater than or equal to the minimum time. Data communication terminal. 5. The data communication terminal according to claim 1, wherein the data discarding control means includes a storage unit for storing information on a minimum time at which the discardable data packet is to be held, Wherein the data packet is discarded from the first reception buffer when the difference between the data packet generation time recorded in the first reception buffer and the current time is equal to or longer than the minimum time. 6. The data communication terminal according to claim 1, wherein a plurality of said first reception buffers are provided for each type of data packet, and wherein said central processing unit indicates that a data packet is stored. A data communication terminal characterized in that a readout speed is changed for each type of data packet by setting a different polling period for each of the plurality of first reception buffers. 7. The data communication terminal according to claim 3, wherein a plurality of said first reception buffers are provided for each type of data packet, and an upper limit value of a data amount stored in said first reception buffer is set to a plurality. The data communication terminal according to claim 1, wherein the reading speed is changed for each type of the data packet by setting a different value for each of the first reception buffers. 8. The data communication terminal according to claim 4, wherein a plurality of said first reception buffers are provided for each type of data packet, and a minimum time during which said discardable data packet is to be held is set. A data communication terminal that sets a different value for each of the plurality of first reception buffers, thereby changing a reading speed for each type of the data packet. 9. A data communication method for transmitting and receiving multimedia data using the data communication terminal according to any one of claims 1 to 8, wherein the transmitted and received data is classified into a plurality of arbitrary types, An identification number is assigned to each of the classifications. The transmitting terminal inserts the identification number into a header of a data packet to be transmitted and transmits the data packet. The receiving terminal can discard a data packet having a specific identification number. A data communication method characterized in that a data packet is specified. 10. The data communication method according to claim 9, wherein the data to be transmitted and received is moving image data hierarchically coded to a plurality of resolutions, and the transmitting terminal determines at least a data packet to be transmitted according to the resolution. A data communication method characterized by being classified into two or more types. 11. The data communication method according to claim 9, wherein the transmitting terminal broadcasts the same data to a plurality of terminals, and the receiving terminal comprises: A data communication method, wherein a different data packet is designated as the discardable data packet in accordance with the processing capability of each terminal or a request of a user.