JP2804385B2 - Voice packet communication device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice packet communication apparatus in a multimedia packet communication technique for transferring real-time information such as voice information in packets.

[0002]

2. Description of the Related Art FIGS. 5 and 6 show a "voice packet system".
(Saito et al., Nippon Telegraph and Telephone Corporation, Tsuken Report, 38, p.
231 to 237, 1989). In FIG. 5, a voice terminal 1 such as a telephone and a facsimile is a PBX 2
The PBX 2 is connected to the voice packet communication device 4 via a TTC standard 2 Mb / s I / F 3 in which 30 channels are time-division multiplexed. The voice packet communication device 4 is connected to the packet multiplexing device 6 via the X21 I / F 5, and the packet multiplexing device 6 is connected to another packet multiplexing device via the high-speed digital line 7.

FIG. 6 is a configuration diagram of a conventional voice packet communication device 4. In the figure, the voice packet communication device 4 includes:
A PBX interface 8 connected to the I / F 3;
A CODEC unit (encoder / decoder) 9 connected to the X interface 8 and provided for each channel and the audio PAD 1
The voice PAD 10 is connected to the CODEC 9 and performs a layer 2 protocol between the packet assembling / disassembling unit 21 for assembling and disassembling the voice packet, the data memory 12, the CPU 13 and the packet multiplexer 6. Line control unit 14 that terminates and transmits and receives voice packets
have.

FIG. 7 shows a format of a voice packet when assembled by the voice PAD 10. In FIG. 7, 16 is a header of the voice packet, 17 is a data portion of the voice packet, and a plurality of samples of the embedded ADPCM encoded voice data. 17a is an M1 block that collects MSBs, 17b is an M2 block that collects MSB-1, 17c is an M3 block, and 17d is an M4 block.

Next, the operation will be described. In FIG. 5, the voice signal from the PBX 2 is
And transmitted to the other party via the packet multiplexing device 6. Further, the packet multiplexing device 6 that has received the voice packet from another packet multiplexing device via the high-speed digital line 7 transfers the voice packet to the voice packet communication device 4 via the X21 I / F 5.

The sound PAD operation in transmission and reception will be described with reference to FIG. In the audio PAD 10 on the transmission side, only the audio part of the audio signal is embedded ADPCM.
Encoded and according to the packet format shown in FIG.
The encoded audio data is assembled into a single audio packet by combining a block in which the upper bits are collected and a block in which the lower bits are collected, and further multiplexes a plurality of channels to transmit to the packet multiplexer 6.

[0007] In the case of reception, a voice PA which has received a voice packet from the packet multiplexer 6 via the X21 I / F 5
D10 decomposes a packet composed of a block in which the upper bits are collected and a block in which the lower bits are collected for each channel, restores the encoded audio data, and decodes the encoded audio data by the CODEC unit 9; F3 and PBX
2 to the voice terminal 1.

[0008]

Since the conventional packet communication terminal is configured as described above, when a voice packet is transmitted in a voice packet communication device accommodated in a time division multiplexing device, it is blocked when a line is congested. Congestion is avoided by discarding.
4, the block discarding control is realized based on the queue length of the transmission buffer. However, if the block discarding control is accommodated in a time-division multiplexing device on a low-speed line, the delay time increases if the block discarding control is performed. There was a problem. In other words, when the queue length of the block discard buffer is monitored during line congestion, a considerable load is placed on the CPU during the discard processing time, and when the line speed is low because the buffer queue is controlled to some extent, the delay time increases. There was a problem of becoming.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and an audio packet communication apparatus capable of shortening the discard processing time and minimizing the delay time even at a low line speed. The purpose is to gain.

[0010]

A voice packet communication device according to the present invention detects a sound state of each channel and calculates a transmission amount on a line based on the detected sound state. A block discarding control unit including a discarding unit that performs discarding and discarding based on a calculation result based on a discarding algorithm is newly provided.

[0011]

The traffic calculator and the discarder according to the present invention calculate the amount of information to be transmitted to the line and perform block dropping before the information is transmitted to the line control LSI. A voice packet communication device that can be guaranteed is realized.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a voice packet communication device according to the present embodiment. In the figure, the same or corresponding parts as in the conventional example shown in FIG. In the figure, as a new configuration, in addition to the configuration shown in FIG. 6, the voice PAD 10 has a traffic calculation function of calculating a sound state of each channel and calculating a transmission amount on a line based on the calculated state. A block discarding control unit 20 comprising a discarding function for discarding and discarding by a discarding algorithm based on the calculation result, and a packet transmitting / receiving unit for discarding a voice packet block when transmission is congested based on the processing result of the block discarding control unit 20 15 is provided. In addition, the line control unit 14 21 I / F is accommodated.

FIG. 2 is a diagram showing the block discarding control timing between the packet assembling / disassembling unit 21, the block discarding control unit 20, and the packet transmitting / receiving unit 15 in FIG. The timing at which audio data is assembled in the packet assembling / disassembling unit 21 and discarded by the packet transmitting / receiving unit 15 is shown. The packetization trigger timing Tp is CODEC
The period of the audio data from the unit 9 for one packet is shown, and generally includes 16 ms and 32 ms. In the figure, Ts is the time during which the start sample timing of one packet is shifted from the previous channel for load distribution. The block discarding process in the block discarding control unit 20 is as follows.
According to the procedure shown in FIG. 3, the detection is performed after detecting the sound / non-speech state of the channel n at the packetization trigger timing Tp. The voice packet communication device 4 having the above configuration is an element of the voice packet communication system shown in FIG.

Next, the operation will be described. 1 and 2
In the above, the voice-encoded data embedding ADPCM by the CODEC unit 9 is divided into blocks by the packet assembling / disassembling unit 21, then packetized by the packet transmitting / receiving unit 15, and transmitted from the line control unit 14 to the line. In this case, the voiced state information generated in the CODEC unit 9 is sampled from the packet assembling / disassembling unit 21 by the block discarding control unit 20 at the cycle of the packetization trigger timing Tp, and is transmitted to the line at the timing of the packetization trigger timing Tp. The traffic amount is calculated, and to what extent block discarding is appropriate is calculated by a block discarding control algorithm described later, and block discarding control is performed on the packet transmitting / receiving unit 20.

That is, in the discard algorithm for discarding blocks shown in FIG. 3, first, at step S1, N
Is calculated. N is a number indicating the number of channels of voice data that can be passed through the voice band obtained by subtracting the non-discardable modem and signaling signal bands from the line speed, and the calculation formula is as follows. N = V _V / V _EA = (V _T −n × V _D ) / V _EA where V _V : band assigned to voice data,
V _EA : M devet ADPCMM encoding speed, V _T : line speed, V _D : band of modem signal, etc.

Next, in step S2, a process of reading the current number T of channels in the sound state from the CODEC 9 is performed. Thereafter, the T value and the N value are compared in step S3, and if T <N, it is determined that no processing has been performed. That is a state that can you to sending voice packets without blocking discarded. If T> N, it is necessary to discard blocks, so the number K of overflowing blocks is calculated in step S4. The formula for calculating K is that a voice packet has four blocks (M1,
In the case of the (M2, M3, M4) configuration, K = (T−N) × 4.

If it is determined in step S5 that the number of overflow blocks K is in the range of K.ltoreq.T, the process proceeds to step S10, where M4 blocks are discarded for each channel. Also, K>
If T, M3 and M4 blocks are discarded for each channel in step S6, and M4 blocks are discarded for each channel from the point of K> K (steps S7 to S).
9). It should be noted that the K value is implemented by the transmission restriction process before starting communication so that the K value falls within the range of 2T> K.

Therefore, in the above embodiment, by employing the above-described block discarding processing means and algorithm, it becomes possible to use the line efficiently at the time of line congestion.

[0019]

As described above, according to the present invention, by calculating the amount of information transmitted to the line, the line control L
Since block dropping is performed before transmission to the SI, uniform voice quality and a small delay time can be guaranteed for each channel.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a voice packet communication device according to one embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a processing relationship between a CODEC unit 9 and a block discard control unit 20 in FIG.

FIG. 3 is a diagram showing a procedure of block discarding in a block discarding control unit 20 in FIG. 1;

4 is a schematic configuration diagram of a voice packet communication system to which the voice packet communication device 4 shown in FIG. 1 is connected.

FIG. 5 is a schematic configuration diagram of a voice packet communication system to which a conventional voice packet communication device 4 is connected.

FIG. 6 is a block diagram showing a conventional voice packet communication device.

FIG. 7 is a format diagram of an audio packet when assembled by the audio PAD 10;

[Explanation of symbols]

 Reference Signs List 1 voice terminal 2 PBX 4 voice packet communication device 6 packet multiplexer 9 CODEC unit (encoder / decoder) 10 voice PAD 12 data memory 13 CPU 14 line control unit 15 packet transmission / reception unit 18 time division multiplexer 20 block discard control Part 21 Packet assembling / disassembling part

Continuation of the front page (72) Inventor Hikaru Kawawada 5-1-1, Ofuna, Kamakura-shi, Kanagawa Prefecture, Communication Systems Laboratory, Mitsubishi Electric Corporation (56) References 1-135154 (JP, A) JP-A-2-202250 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H04L 12/56

Claims (1)

(57) [Claims] An encoding unit accommodated in a time division multiplexing device and encoding input information such as an audio signal from a PBX,
A voice packet composed of a packet assembling part which is composed of a block composed of a block in which only the discardable information is collected from the encoded data encoded by the encoding part and a block in which only the non-discardable information is collected, and a line control part In the communication device, a traffic calculating unit that detects a sound state of each channel and calculates a transmission amount on a line based on the state, and discards and discards in accordance with a discard algorithm determined based on the traffic calculation result. A voice packet communication device comprising a block discarding control unit comprising: