JPH05260039A - Frame synchronizing system for codec - Google Patents

Abstract

PURPOSE:To prevent the false acquisition of synchronism caused by a transient state or a real audio code at the time of path connection. CONSTITUTION:Codecs 2-1 to 2-2 are provided with encoding parts 21, frame assembly parts 22, frame aligner parts 23 to exchange received, compressed and encoded data into the clocks of its own codecs by detecting the synchronism of received signals and to output those clocks, and decoding parts 24. An interface part 3-1 provided between the codec 2-1 and a transmission network 1 always monitors the received signals while transmitting an idle symbol and when the codecs are connected by a path through the transmission network 1, the idle symbol from the opposite side codec is detected so that the frame aligner part 23 can acquire frame synchronism. After the frame synchronism is acquired, a synchronizing symbol is transmitted, and the received signal is inputted to the codec 2-1. When the synchronizing symbol from the opposite side codec 2-2 is detected, the interface part 3-1 is operated to start transmitting audio frames.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a frame synchronization system for high-efficiency coded transmission of voice signals between codecs.

[0002]

2. Description of the Related Art With the recent demand for higher efficiency in communication systems, it is required to transmit voice signals in as narrow a band as possible. Therefore, various codecs (Coder-Decoders) for highly efficient encoding and decoding of audio signals are provided. Among these, there is a block coding method that codes a plurality of sample values sampled in time series as one unit, and for example, APC-AB (Adaptive Predi
The ctive Coding with Adaptive Bit allocation method is known.

In this system, a voice signal is divided into a plurality of bands, APC coding is performed for each band, and the number of quantization bits is adaptively assigned in both the frequency domain and the time domain according to the residual power. It is a thing.

As an example of this APC-AB system, the sampling frequency of the audio signal is 8 KHz, the analysis frame length thereof is 32 ms, and the number of band divisions is 3 (0, 1, 1, 2, 2).
.About.4 KHz), the coding frame length is 16 ms (256 bits), and high-efficiency coded data having a transmission rate of 16 Kb / s can be obtained. In this method, samples are not sequentially coded, but a plurality of sample values are coded as a block in block units. The compression encoded data does not correspond in time series to the original time series sampled data. Therefore, when transmitting coded block data consisting of multiple bits that are compression-encoded, a frame bit is added to each block delimiter in order to identify the block, and a frame is constructed and received. By transmitting to the side, the receiving side recognizes the frame delimiter at the position of the frame bit. Further, since the meaning of the coded data in the frame is determined by its bit position, if the frame delimiter is wrong, decoding will not be performed correctly.

[0005]

When a voice signal is encoded with a high efficiency of 16 Kb / s or 8 Kb / s as described above and is transmitted, the encoding is performed so as to have a pattern different from the frame synchronization pattern. However, since the number of frame bits for frame synchronization cannot be increased due to restrictions on transmission efficiency, a specific audio signal may be encoded in the same pattern as the frame synchronization pattern.

In particular, when the same pattern as the frame synchronization pattern as described above occurs in the initial state where the codecs are connected via a switch or the like, the codec considers the pattern as the frame synchronization pattern and synchronizes with it. It will be retracted.

In such a pseudo-synchronized state, the codec of the frame synchronization system cannot correctly decode the high-efficiency coded data, so that the call cannot be made.

The present invention was created in order to solve the above problems, and when a compression-coded voice signal is transmitted in a frame structure between codecs, a transient state at the time of path connection or an actual voice is connected. The purpose is to prevent the pseudo sync pull-in by a code.

[0009]

FIG. 1 is a block diagram of a frame synchronization system of a codec according to the present invention. As shown in FIG. 1, the above problems are caused by a coding unit 21 for compressing and coding a voice signal, a frame assembling unit 22 for adding a frame bit to the coded data to create a voice frame, and synchronizing a received signal. A codec 2-2 having a frame aligner unit 23 for detecting and receiving the received compressed coded data by switching to its own clock and a decoding unit 24 for decoding the compressed coded data output by the frame aligner unit 23, In the method of transmitting and receiving audio signals with frame synchronization between 2 and 2,
Received signals are constantly monitored while transmitting idle symbols at frame intervals. When the codecs are connected by a path via the transmission network 1, the idle symbols from the other codec are detected and the frame aligner unit 23 is detected. Synchronous pull-in is performed, when the frame synchronization is established, the sync symbol is sent out, the received signal is input to the codec 2-1 and when the sync symbol from the other codec 2-2 is detected, the audio frame is started to be sent out. The interface section 3-1 is provided between the codec 2-1 and the transmission network 1 and is solved by the frame synchronization system of the codec of the present invention.

[0010]

The idle symbol consisting of a specific bit pattern with a large number of bits is transmitted prior to the start of the transmission of a voice frame carrying actual voice data. Idle symbols can be configured with a sufficient number of bits for synchronization pull-in, and at this time, voice coded data is not received and frame synchronization pull-in is performed only with idle symbols, so pseudo-sync pull-in can be prevented. .. Then, when frame synchronization using the idle symbol is established, a synchronization symbol is sent to notify the other party of the establishment of synchronization,
Input the received signal to the codec. And
When the sync symbol from the partner codec is detected, the transmission of the voice frame carrying the high-efficiency coded data of the real voice signal is started. Once the frame synchronization is established, even if the pseudo frame synchronization pattern is transmitted, the frame aligner protects the synchronization so that it is not drawn into the pseudo frame synchronization pattern. It will be possible to prevent it from becoming.

[0011]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a configuration diagram of a frame synchronization system of a codec of the present invention, FIG. 2 is a diagram showing a frame format, and FIG. 2 is a diagram showing an operation sequence.

In FIG. 1, reference numeral 1 is a transmission network including an exchange, 2-1 and 2-2 are codecs, and 3-1 and 3-2 are interface sections provided between the transmission network and the codec. The codecs 2-1 and 2-2 are an encoding unit 21 that compresses and encodes a 64 Kb / s audio PCM signal into, for example, a block unit of 16 Kb / s, a frame that adds a frame bit to the compression code to create an audio frame. The assembling unit 22 detects the frame synchronization pattern of the received signal, establishes frame synchronization, outputs the synchronization detection signal when the frame synchronization is established, and outputs the compressed encoded data by changing the clock of the received compressed coded data to the clock of its own codec and outputting the same. , A decoding unit for decoding the received compressed code into 64 Kb / s PCM audio data
Having 24.

The interface units 3-1 and 3-2 are transmission selection units.
31, a symbol detection unit 32, and a gate unit 33. The transmission selection unit 31 is controlled by the symbol detection unit 32 as shown in FIG.
One of the idle symbol frame, the synchronization symbol frame, and the voice frame shown in (1) is selected and transmitted to the transmission transmission path.

FIG. 2 shows an example of the frame format. The idle / synchronization symbol frame has an idle symbol of a specific pattern (eg, 10101010) in a plurality of bits (eg, 8 bits) at the head of the frame, or a specific pattern different from the idle symbol (eg, 0010).
0100) sync symbol, and the subsequent bits are all 1's frame. In the audio frame, the first two bits are frame bits in which "0" and "1" are alternated for each frame, and the rest are compression-encoded actual audio data.

The symbol detector 32 constantly monitors the received signal from the reception transmission line, and controls the transmission selector 31 to select an idle symbol frame when no signal is received. When an idle symbol is detected from the received signal, a sync signal is created and supplied to the frame aligner unit 23.
Upon receiving this synchronization signal, the frame aligner unit 23 performs the synchronization pull-in of the built-in frame synchronization circuit, and when the synchronization is established, notifies the symbol detection unit 32 of the synchronization detection signal to that effect. When the synchronization detection is notified, the symbol detection unit 32 controls the transmission selection unit 31 to selectively transmit the synchronization symbol frame and opens the gate unit 33 to input the reception frame to the codec.

Further, the symbol detection unit 32 controls the transmission selection unit 31 so as to selectively transmit a voice frame when a synchronization symbol is detected from the received signal. Next, the frame synchronization pull-in operation at the time of connecting the transmission paths between the codecs in the system having the above configuration will be described with reference to the diagram showing the operation sequence of FIG.

When the path is not connected, the symbol detector 32 does not receive the symbol from the reception transmission path, so the symbol selector 31 sends out the idle symbol frame.

When the path is connected to the other codec via the transmission network 1 including the exchange, the idle symbol frame is first received via the reception transmission line. Then, the symbol detection unit 32 detects an idle symbol, creates a synchronization signal indicating the phase of the received frame from the reception timing of the idle symbol, and the frame aligner unit 23 of the codec.
Supply to the synchronous circuit of. When the synchronization signal is received over a predetermined number of protection stages, frame synchronization in the codec is established, so the frame aligner unit 23 notifies the symbol detection unit 32 of the synchronization detection signal. The symbol detection unit 32 notified of the synchronization detection signal controls the transmission selection unit 31 to selectively transmit the synchronization symbol frame. Similarly, since the synchronization symbol frame is also transmitted from the partner codec, the symbol selecting unit 32 that receives this and detects the synchronization symbol recognizes the establishment of frame synchronization of the partner codec and transmits the voice frame from the own codec. The transmission selection unit 31 is controlled. At this time, in the codec, the PLL of the synchronizing circuit in the frame aligner unit 23 is locked to the phase of the received frame, so that it can be correctly synchronized with the voice frame sent from the other side. After that, the synchronization circuit in the frame aligner unit 23 performs synchronization detection based on the first frame bit of the audio frame, and the decoding unit 24 decodes the received audio data in a predetermined manner so that the audio signal is correctly transmitted and received.

As described above, since the voice frame is not transmitted until the synchronization is established, the synchronization is established by the idle symbol frame and the synchronization symbol frame, and the voice data in the transient state at the time of path connection is not received, the voice data is not received. Correct synchronization is performed without the occurrence of pseudo-synchronization pull-in due to the same pattern as the internal synchronization pattern.

When the synchronization detection disappears due to a path disconnection or the like, the symbol selection unit is notified of that fact, and the symbol selection unit returns to the initial state and starts transmitting an idle frame.

[0021]

As described above, according to the present invention, even if the switching connection by the transmission network is repeated, the synchronization pull-in can be surely performed at the correct position at the time of the path connection between the codecs of the frame synchronization system, and the transition can be performed. It has an effect of preventing the pseudo synchronization due to the state and the synchronization pull-in due to the voice actual code, and largely contributes to the improvement of the reliability of the voice transmission by the compression encoding.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a frame synchronization system of a codec of the present invention.

FIG. 2 is a diagram showing a frame format.

FIG. 3 is a diagram showing an operation sequence.

[Explanation of symbols]

1 ... Transmission network, 2-1, 2-2 ... Codec, 21 ... Encoding section, 22
... frame assembling section, 23 ... frame aligner section, 24 ... decoding section, 3-1, 3-2 ... interface section, 31 ... transmission selecting section, 32 ... symbol detecting section, 33 ... gate section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04L 29/08

Claims (1)

[Claims] 1. A coding unit (21) for compressing and coding a voice signal.
And a frame assembling unit (22) that adds a frame bit to the encoded data to create a voice frame, and a frame aligner unit that detects the synchronization of the received signal and transfers the received compressed and encoded data to its own clock and outputs it. (23) and a codec (2-1, 2-2) having a decoding unit (24) that decodes the compression coded data output by the frame aligner unit (23), and an audio signal is obtained by frame synchronization. In the transmission / reception method, the received signal is constantly monitored while transmitting idle symbols in the frame cycle, and the transmission network (1) is provided between the codecs.
When a connection is established via the path via the other side, the idle code from the other codec is detected and the frame aligner unit (23) performs synchronization pull-in. -1)
To the codec (2-1) and the transmission network (1) when the sync symbol from the other codec (2-2) is detected. ) Is a frame synchronization method of the codec, which is provided between