Summary of the invention
In order to solve frequency-shift keying synchronized broadcasting system with the audio frequency delay problem, the 1PPS pps pulse per second signal that GPS is provided is as timing signal, mode frequency division multiplexing by the FSK modulation transmits in channel, the invention provides an adding and an extracting method with the road voice-frequency signaling.Described technical scheme is as follows:
A kind of method that solves synchronous frequency-modulating broadcast system synchronization audio frequency time-delay calculation and adjustment is characterized in that, the audio front end server of described system is carried out following steps:
Steps A: communicate with human-machine interface module, obtain frequency-shift keying synchronized broadcasting system applies layer user data;
Step B: adopt the mode of poll to select the data picture format that will send, synchronization frame and Non-alignment Frame be transmission in turn, and per second only sends a synchronization frame;
Step C: from the data framing of application layer until data link layer;
Step D: the data of data link layer are carried out the FSK modulation;
Step e: receive digital audio and video signals from audio input end, after low-pass filtering, and with the FSK waveform adder after the modulation of road voice-frequency signaling, after CODEC D/A conversion, export to back level equipment by numeral or analog audio interface.
Preferably, the data picture format that described system audio front-end server sends comprises synchronization frame, site information frame and command frame, synchronization frame comprises the time delay reference point as timing signal, the site information frame provides parameter control for each exciter, and command frame then can be concentrated setting to the parameter of all exciters in the whole network.
Preferably, the frame head of described synchronization frame aligns with the pps pulse per second signal that GPS provides.
Preferably, the frame head of described synchronization frame and Frame carry out CRC check respectively.
Preferably, for the FSK modulation of data in link layer, adopting tables look-up adds the method for keying.
A kind of method that solves synchronous frequency-modulating broadcast system synchronization audio frequency time-delay calculation and adjustment is characterized in that, the Digit FM Exciter of described system is carried out following steps:
Steps A: the digital audio and video signals that receives is carried out low-pass filtering, use for the exciter coded modulation; The digital audio and video signals that receives is carried out high-pass filtering, obtain the FSK modulated wave of signaling;
The coherent detection algorithm of step B:FSK is to carry out demodulation by the square-law maximum of relatively correlator output.The FSK modulated wave that obtains and the in-phase component and the quadrature component of frame head data are carried out related calculation respectively, obtain two groups of correlation computations results, square operation is done in two class value pointwises, and the result is carried out addition, preserve operation result.Simultaneously ask for maximum in the result, the maximum corresponding points that find in a frame period are the frame synchronization point, thereby recover the accurate sampled point relative position relation of whole transmission frame, realize sampled point synchronously, bit synchronization and frame synchronization.
Step C: the signaling data that obtains is carried out CRC check and frame parsing, obtain correct network management information, and calculate the audio frequency time delay in conjunction with the 1PPS standard timing signal of GPS.
Preferably, among the step B, utilize modulated wave and frame head to carry out related operation and realize that frame head is synchronous, and realize that sampled point is synchronous.
Preferably, the system clock of transmitting-receiving two-end is all phase-locked with the 10MHz benchmark frequency marking that GPS provides; At set intervals, the original position of demodulation is finely tuned.
The beneficial effect that technical scheme of the present invention is brought is:
Transmit in the frequency division multiplexing mode with the road voice-frequency signaling by provided by the invention, receiving terminal only needs high-pass filtering and carries out the FSK demodulation that just can to obtain transmitting terminal added with the road voice-frequency signaling, this adding method with the road voice-frequency signaling is applicable to that any transmission bandwidth is the audio transmission system of 20KHZ, has extensive applicability, and this method need not to add any optional equipment, brings great convenience for the user in the actual sets network process.
Embodiment
The invention will be further described below in conjunction with the drawings and specific embodiments, but not as a limitation of the invention.
In the synchronous frequency-modulating broadcast of the present invention, the 1PPS pps pulse per second signal that GPS is provided is as timing signal, transmits in channel with audio stream is multiplexing together.As shown in Figure 2, at transmitting terminal, the pps pulse per second signal that the audio front end server provides GPS is as joining in the audio stream with the road voice-frequency signaling.At receiving terminal, exciter will extract with the road voice-frequency signaling, and timing signal that will be wherein compares with local GPS pulse per second (PPS), just can obtain audio frequency time delay that transmission course causes.In addition,, can realize Long-distance Control like this, for whole Network Management provides very big convenience to each exciter with the down management information that can also comprise network in the voice-frequency signaling of road.
From broadcasting the transmission of control centre to each emission station FM exciter, the ad-hoc networks that can be utilized as synchronous frequency-modulating broadcast provides realizes for audio signal noted earlier.These ad-hoc networks can be the PDH/SDH networks of fiber optic network, microwave network or telecommunications.It is exactly the construction cost costliness that but this special network has a general shortcoming, is difficult to promote.
So wish to use very ripe DVB transmission link at present.The satellite transmits link of DVB at first, at present, the Digital Television that each overwhelming majority of central, and province level TV station has had the DVB-S program of broadcasting TV programs by satellite.And one road satellite television programming can also attach six road audio frequency except the television field frame that comprises program itself.Can be the audio signal that will transmit as the transmission that is combined with each other of one tunnel supplemental audio and TV programme.Owing to need not to ward off in addition separately channel, so there is not extra channel occupancy expense.In addition, the satellite broadcasting wide coverage is suitable for the establishment of large tracts of land synchronous frequency-modulating broadcast network.And satellite broadcasting is not limited by geographical conditions, and some geographical conditions is abominable, is unsuitable for the place that Wireline is laid, and can receive the signal of satellite transmits yet.The popularity rate of cable TV network in the city of DVB system is quite high at present, and the audio transmission scheme of link realization via satellite almost can be transplanted to cable TV network fully.These all make the DVB transmission link become the first-selection of setting up synchronous frequency-modulating broadcast network.But, no matter be dedicated network, or the DVB network, all may bring the uncertain variation of audio frequency time delay, so no matter adopt any transmission means, all need to carry out the time delay adjustment.
At DVB transmission link the most frequently used in the synchronous frequency-modulating broadcast network, adding method with the road voice-frequency signaling has been done following consideration: some the MPEG-2 encoder that uses in the DVB system is only supported the analogue audio frequency input at present, and the network provider does not generally allow existing equipment is changed, so need to have joining in the analog audio stream of timing signal, in transmission course, remain identical time delay to guarantee both with the road voice-frequency signaling.The method that adds is exactly earlier to the modulating with the road voice-frequency signaling of digital form, and is again that modulation signal and analogue audio frequency is together multiplexing.Since lower with road voice-frequency signaling data transfer rate, and consider that the modulation and demodulation of FSK are simpler, so present embodiment has adopted the modulation system of FSK.
With the multiplex mode of the modulation waveform of road voice-frequency signaling and analogue audio frequency two kinds of time division multiplexing and frequency division multiplexings can be arranged.
Time division multiplexing at first is shown in accompanying drawing 3 and 4.At first, the audio stream of importing is passed through sample rate conversion, it is compressed, come command transmitting to reserve time slot in time domain.During reception, with higher sample rate composite signal is received earlier, again it is separated.In the separating process, the rotary switch of transmitting-receiving two-end must be opened simultaneously or be closed, if error appears in this synchronizing process, signaling will be sneaked in the audio signal, and the quality of the audio signal broadcasted is produced serious disturbance.In the implementation procedure of reality, because the pseudo-signaling of very possible appearance in the signal, so this strict synchronism is difficult to guarantee that this allows to occur anything but, so do not adopt this multiplex mode in this preferred embodiment in the equipment of broadcast level.
Then see frequency division multiplexing again, as shown in Figure 5.Because the frequency spectrum of most of audio signal all concentrates on below the 15KHz, so the thought of frequency division multiplexing is exactly will be high-end with the frequency spectrum that the road voice-frequency signaling is modulated to more than the audio signal 15KHz by the FSK modulation.Like this, at receiving terminal, only need a simple low pass filter just can the filtering signaling to the influence of audio signal, can not influence the audio signal quality of broadcast.And this multiplex mode is applicable to that any audio transmission bandwidth is the system of 20KHz, has versatility widely.Therefore, present embodiment has selected for use the mode of frequency division multiplexing to realize multiplexing with road voice-frequency signaling and analogue audio frequency.
According to the analysis of front as can be known in the synchronous frequency-modulating broadcast the most frequently used audio transmission mode be the transmission link of DVB.In this transmission link, need to use the MPEG-2 encoder.The use meeting of this MPEG-2 encoder exerts an influence to the transmission with the road voice-frequency signaling of frequency division multiplexing.Because the coding standard of MPEG-2 has made full use of the psychologic acoustics characteristic of people's ear audio signal has been carried out lossy compression method.And people's ear is insensitive to the radio-frequency component in the audio frequency, so if high frequency range component is too little or by temporal masking or masking spectrum, encoder just can not be to its transmission of encoding.Because when adopting frequency division multiplexing, the frequency spectrum that has been modulated onto audio signal with the road voice-frequency signaling is high-end, so, occur the phenomenon that certain period do not encode and transmit signaling probably through behind the MPEG-2 encoder.
At the characteristic of MPEG-2 encoder, present embodiment has carried out prudent selection to the transmission parameter with the road voice-frequency signaling.Here, the modulating frequency of FSK is selected lower frequency as far as possible, as 17KHz and 18KHz.The modulation amplitude of FSK requires moderate, and what present embodiment was selected is-24dBFS.The duration of FSK modulation symbol has been selected 1ms.Duration is shorter, like this, when owing to the MPEG-2 coding a certain section FSK modulation waveform being lost, just can find a new frame complete signal as early as possible, carries out correct demodulation again.Because the data transfer rate of signaling is lower, so the data of losing can be remedied by the retransmission data of back, the audio frequency time delay does not need per second all to go to adjust yet, so can not influence the performance of signaling.In addition,, also audio signal has been carried out the low-pass filtering of 15KHz, to reduce the interference of audio signal to signaling at transmitting terminal.In the frame structure of design, also added CRC check, detect the demodulation error code that may exist.
The application has designed the architecture with road voice-frequency signaling transmission system layering of frequency division multiplexing.Mainly comprising four layers, is respectively physical layer, data link layer, transport layer and application layer.The corresponding relation of the architecture of native system and ISO/OSI and TCP/IP as shown in Figure 6.
System adopts the architecture of layering, can bring two benefits: the one, and, the layering abstract according to different qualities realizes a clear and definite function, can avoid each layer function confusion for every layer.The 2nd,, each layer is relatively independent, thereby makes the task of being assigned to each layer independently to realize.Like this, when certain solution that provides when one deck is wherein upgraded, can not influence other layer.
The concrete formation of each layer as shown in Figure 7.
Three kinds of function frames have been defined with the road voice-frequency signaling: synchronization frame, site information frame and command frame.Synchronization frame comprises the time delay reference point as timing signal, and the site information frame provides parameter control for each exciter, and command frame then can be concentrated setting to the parameter of all exciters in the whole network.During transmission, synchronization frame and Non-alignment Frame send in turn, and per second only sends a synchronization frame.The frame head of synchronization frame aligns with the pps pulse per second signal that GPS provides, as shown in Figure 8.
Signaling transmission system is to realize on the common treatment hardware platform of a base-band audio signal.This hardware platform comprises seven modules as shown in Figure 9: human-machine interface module, digital audio transceiver module, analogue audio frequency CODEC module, FPGA and CPLD module, DSP module, power module and clock module.
Wherein, analogue audio frequency CODEC module and digital audio transceiver module are the input/output modules of this platform.The simulation or the audio signal of digital form are sent in the level module of back by these two modules and to be handled; Voice data output after perhaps system can will handle by these two modules is used for next stage equipment.
The DSP module is whole core of hardware platform, and nearly all signal processing function is all finished in this module.
FPGA and CPLD module are used for providing the transmission of signal and the buffer memory of data for above-mentioned several modules.The part signal processing capacity also realizes in this module.
Human-machine interface module provides the interface of an interchange for human-computer dialogue.Power supply and clock module then are the essential conditions of the normal operation of entire circuit.
Transmitting terminal, audio front end server just, the function that realize has the following aspects:
(1) with the single chip communication of controlling liquid crystal panel and serial ports of computers, obtains system applies layer user data;
(2) adopt the mode of poll to select the data picture format that will send;
(3) from the data framing of application layer until data link layer;
(4) data of data link layer are carried out the FSK modulation;
(5) receive digital audio and video signals from input, after the 15KHZ low-pass filtering, and with the FSK waveform adder after the modulation of road voice-frequency signaling, after CODEC D/A conversion, export to back level equipment by analog audio interface.
The software function module schematic diagram of transmitting terminal as shown in figure 10.
For CRC check, DSP has adopted a kind of based on the quick CRC check algorithm by byte calculating of tabling look-up.The thinking of this algorithm of tabling look-up is the residue coding schedule of a byte information of first off-line structure, tables look-up and XOR can be tried to achieve the CRC residue of multibyte information according to this coding schedule.Because byte information is totally 8 binary elements, so always have 256 kinds of various combinations.Generator polynomial G (x)=x with most widely used CCITT suggestion
16+ x
12+ x
5+ 1 is example, and every kind of combination removes through generator polynomial G (x), just produces the check value of two bytes, so byte residue coding schedule accounts for 256 words (512 bytes) altogether.
CRC check fast algorithm flow chart as shown in figure 11.
For the FSK modulation of data in link layer, adopting tables look-up adds the method for keying.This method has three advantages: (1) speed fast (2) can guarantee phase place, and (3) realization is simple continuously.Its algorithm schematic diagram as shown in figure 12.
The receiving terminal of system, Digit FM Exciter end just, the function that finish mainly contains the following aspects:
(1) digital audio and video signals that receives is carried out the 15KHZ low-pass filtering, use for the exciter coded modulation.
(2) digital audio and video signals that receives is carried out the 15KHZ high-pass filtering, obtain the FSK modulated wave of signaling.
(3) FSK modulated wave and the frame head that obtains carried out related operation, to simplify the demodulation of FSK.
(4) signaling data that obtains is carried out CRC check and frame parsing, obtain correct network management information, and calculate the audio frequency time delay in conjunction with the 1PPS standard timing signal of GPS.
The software function module schematic diagram of receiving terminal as shown in figure 13.
Filter Design
In native system, all adopt digital filter in DSP, to realize filtering to digital signal.Digital filter mainly is divided into two classes: finite impulse response (FIR) filter and infinite impulse response (IIR) filter.The FIR filter has the impulse response of a finite length, is the filter with ideal linearity phase place and constant group delay.Now the summary of benefits with the FIR filter is as follows:
● desirable linear phase response can be provided.
● can adopt foolproof algorithm to realize.
● reduced algorithm to the limited arithmetic precision and the sensitivity of rounding error.
● guaranteed stability.
The ideal linearity phase place of FIR filter is meant that the phase response of filter is with frequency linearity decay (speed that is phase change is constant).The constant group delay is used the value that particular importance is arranged to the application of ask for something preservation signal waveform shape as the audio or video of high speed data transfer or high-fidelity.We are based upon on this point basis especially to the reckoning of audio frequency time delay, so the Design of Filter here all adopts Finite Impulse Response filter to realize.
The THE DESIGN OF WINDOW FUNCTION method is the basic design method of FIR filter, has obtained using widely.In the present embodiment, with THE DESIGN OF WINDOW FUNCTION method designing filter the time, can use triumphant damp window.Because triumphant damp window is one of useful and the most optimum window construction.For given stopband attenuation, it provides minimum main lobe width, the transition band of steepest just, and just in this point, it is optimum.During specific implementation, the empirical equation according to triumphant damp window calculates filter parameter N and β earlier, and the firl function that provides by MATLAB obtains desired filter coefficient vector then, calls in the DSP program with the form of coefficient table again.
Sampled point is synchronous
Sampled point is a synchronization concept that proposes at native system synchronously.Sampled point requires synchronously to separate the bit synchronization timing error that timing recovers and is no more than an audio sample point interval, and the synchronization timing of ascending the throne can be as accurate as sampled point.Only in this way, could satisfy the required precision that the audio transmission time-delay is calculated.The synchronous realization of sampled point is particularly important, and it is the basis of follow-up series of computation.
The FSK demodulation
The FSK demodulation is adopted relevant and the square-law decision device finish sampled point synchronously, frame synchronization, method is as shown in figure 14.
The coherent detection algorithm of FSK is to carry out demodulation by the square-law maximum of relatively correlator output.The FSK modulated wave that obtains and the in-phase component and the quadrature component of frame head data are carried out related calculation respectively, obtain two groups of correlation computations results, square operation is done in two class value pointwises, and the result is carried out addition, preserve operation result.Simultaneously ask for maximum in the result, the maximum corresponding points that find in a frame period are the frame synchronization point, thereby recover the accurate sampled point relative position relation of whole transmission frame, realize sampled point synchronously, bit synchronization and frame synchronization.Calculate maximum outgoing position homogeneous phase simultaneously at continuous 3 times, we think that system enters sampled point precise synchronization state.Adopt above-mentioned demodulating algorithm, can realize simultaneously that bit synchronization, frame synchronization and sampled point are synchronous.
After realizing synchronously, the transmission frame data are carried out demodulation, separate timing employing method as shown in figure 15.Demodulation is carried out CRC check after obtaining whole transmission frame data, if the CRC check result is correct, then preserves demodulating data, resolves back output.If CRC check result is incorrect, then abandon data.
The application has proposed in a kind of digital synchronous frequency modulation broadcast system adding and the extracting method with the road voice-frequency signaling.By the analysis of the audio transmission mode that may use synchronous frequency-modulating broadcast network, proposed a kind ofly will add universal method in the simulated audio signal with the road voice-frequency signaling.This method will be modulated to the high-end frequency of audio frequency with the road voice-frequency signaling by the FSK modulation earlier, be sent to transmission link with the audio signal frequency division multiplexing more together.Receiving terminal only needs high-pass filtering and carries out the FSK demodulation that just can to obtain transmitting terminal added with the road voice-frequency signaling.This adding method with the road voice-frequency signaling is applicable to that any transmission bandwidth is the audio transmission system of 20KHz, has extensive applicability.And this method need not to add any optional equipment, brings great convenience for the user in the actual sets network process.
The architecture Design with the road audio transmission system of frequency division multiplexing has taken into full account MPEG-2 encoder in the present the most frequently used DVB transmission system to the audio frequency radio-frequency component, also promptly with the influence of road voice-frequency signaling.And the parameters that adopted in the signaling transmission course and strategy have been made detailed regulation at these influences.
Above-described embodiment is a more preferably embodiment of the present invention, and common variation that those skilled in the art carries out in the technical solution of the present invention scope and replacement all should be included in protection scope of the present invention.