FI119014B - Synchronization method and system for real-time wireless transmission of multichannel audio or video data - Google Patents

Description

1,111,014

METHOD AND SYSTEM FOR SYNCHRONIZING WIRELESS TRANSMISSION OF MULTI-CHANNEL AUDIO OR VIDEO DATA

The present invention relates to a wireless synchronization method for audio or video purposes according to the preamble of claim 1.

The invention further relates to a wireless synchronization system for audio or video purposes according to the preamble of claim 3.

The present invention is typically directed to a system for wireless transmission of an isochronous, electromagnetic interference-tolerant, multi-channel digital audio signal of the highest studio quality, including apparatus and methods. This same method can also be used as the basis for 1S multicast transmission of other digital information, such as local delivery of real-time digital video and television signal, when similar real-time and bandwidth requirements are met.

Introduction 20 ·· ·: *. * In current state of the art, studio-quality multi-channel digital • · *. *: The audio signal is first converted to analogue format and then transferred to the speakers • · ·: Channel-specific electrical cables. Digital transmission in electrical cables • · · * · * · · · * or optical fibers is also known. If the speakers are passive, an additional amplifier is required to amplify the signal power of the transmitter and to supply the transmission line * · * · · · · with sufficient power to control the speakers at the appropriate sound level. All of these measures, as well as the physical analog transmission itself, bring with it »» a number of degradation effects on the original signal, such as noise, noise, distortion, group delays, and amplitude and phase errors. Cabling is often cumbersome and may look messy. With careful planning of cables and their placement * ', the above disadvantages can be reduced to some extent, but rarely completely eliminated. The number of cables, their considerable size, and the need for careful planning, combined with laborious installation work 2,119,014, increase costs, plus the skills required and time wastage. Cables and their electromechanical connectors are subject to mechanical faults that are difficult to find and repair. These problems are especially unfortunate in public performances as performers and even the audience move among the cables. Under these circumstances, there are 5 real risks of harness and injury from cables. During art tours, audio equipment is erected and disassembled in a variety of environments, which multiplies these problems, hassles and expenses.

The use of active loudspeakers 10 with integrated and optimized loudspeakers simplifies the situation to some extent. Analog signals can now be of low power and more noise and interference resistant differential signaling is directly available. However, producing multichannel differential signals requires quite expensive high-quality analog electronics as well as Cost-effective differential cables with their connectors.

15

The currently available wireless audio signal distribution systems are non-standardized radio or infrared solutions based on loss-of-audio amplification methods, thus providing poor performance. For this reason, they are mainly used for ancillary purposes, such as in the case of home theater rear speakers 20.

The object of the present invention is to solve, using the techniques described above, ..! I 'problems with isochronous, real-time transmission of the highest studio-quality digital audio • • * * * signal by constructing a new wireless local area network (WLAN) based on international standards * * * 25 25 · · A communication system, transmitter, receiver, and the necessary hardware and operating software for the effective distribution of a limited range of digital audio signals, as well as for testing, configuration, management and control of such a system.

The invention is based on the idea that digital information is transmitted from a central broadcasting *: station to a group of individual loudspeakers in studio digital: \: format via electromagnetic radio waves or through a power distribution network without separate signal cables, typically using international standards 3 mass-produced wireless local area network (WLAN) components. The digital signal is only converted to analog in the active speaker itself and fed locally, through optimized electronic low-pass, band-pass and over-pass filters, to amplifiers and finally to high-quality speaker elements.

5 This guarantees extreme sound quality. Due to the application of mass-produced WLAN technology and its commercially available components, as well as the small number of other integrated circuits needed, the cost of development and the actual system can be considered very reasonable.

10 The method described here replaces wiring with standardized, commercially available wireless LAN technology, as defined in IEEE 802.11 series standards. The special features required for real-time transmission of multi-channel uncompressed studio quality audio have been implemented with the ingenious selection of the WLAN system coordination function, communication modes and control parameters, in combination with a particular top-level firmware control software.

In accordance with a first preferred embodiment of the invention, the receivers are synchronized within a multicast system by a frame termination interrupt formed by a control frame terminating each beacon period at exactly the same 20 times within each beacon period.

• · * • · · * • • • • t • · · *. *: Particularly the synchronization method of the invention is characterized by what φ ·· * ···! is stated in the characterizing part of claim 1.

The synchronization system according to the invention is further characterized by what is stated in the characterizing part of claim 3.

Significant advantages can be achieved with the invention.

• · • · ··· ·· * * · ♦ i. * 30 You can receive studio-quality digital audio reception on a speaker.

· ** 1 • · | * ··· 'avoids all errors associated with the traditional speaker signal path. By converting ···: .t · from digital to analog form only in the active speaker itself, ♦ maximizes sound quality by locating the path of the analog signal to a fixed and 4,119,014 optimized active speaker circuitry according to one embodiment of the invention.

Signal cables, their connectors and differential signal transmitters and receivers, and related materials and installation work can be completely avoided. This eliminates all associated cost, failure, and installation issues. Because mass-produced standard WLAN technology is the basis for the invention, its mass-production cost can be very low according to one embodiment of the invention.

10

Because the multicast mode and frequent multicast synchronization are in use, the inter-channel phase difference can be effectively eliminated in accordance with one embodiment of the invention.

Because an optimized transmission frame size is used, the system-level delay can be minimized to insignificant according to one embodiment of the invention.

According to one embodiment of the invention, the error management method provides a simple and fast best-effort error correction method.

The invention will now be described in more detail with reference to the exemplary embodiments shown in the accompanying drawings, in which * 1 · 1 · 1 '··· 1 shows a general configuration of the ice system according to the invention in the form of a block diagram «· · '··· 1 25.

• · · • φ • 1 · 1 «. Fig. 2 is a block diagram showing an example of a transmission base station according to the invention.

• · 1 * · «• · #« · • · • ·

Figure 3 is a block diagram showing another example of a transmission support station 30 in accordance with the invention.

Fig. 4 is a block diagram showing an example of a receiver according to the invention.

5,119,014

Figure 5 illustrates the structure of audio data representing one multi-channel audio sample according to the invention.

Fig. 6 shows a data structure representing one audio sample eight and a related error management block 5 according to the invention.

Figure 7 illustrates, with the data structure of Figure 6, the error correction principle of the invention.

Fig. 8 is a block diagram illustrating a Medium Access Control (MAC) architecture which may be used in connection with the invention.

Figure 9 illustrates as a data structure a general MAC frame structure that may be used in connection with the invention.

15

Figure 10 illustrates as a data structure a WLAN frame control field that may be used in connection with the invention.

FIG. 11 is a block diagram of various types of MAC addresses of which 20 multicast versions may be used in connection with the invention.

Figure 12 illustrates as a data structure a general beacon frame that may be used in connection with the invention.

• ·· • · • · * · *.:. Fig. 13 shows as a data structure a beacon frame according to the invention.

··· · • · · · · ·

Figure 14 illustrates as a data structure a capability information field that may be used in connection with the invention.

1 * *: 30 Figure 15 shows as a data structure information elements that may be used in connection with the invention *: **:.

• · • · * • ·· • · • 6 119014

Figure 16 illustrates as a data structure the design of a Traffic Indication Map (TIM) element that may be used in connection with the invention.

Fig. 17 illustrates as an information structure an Extended Rate PHY (ERP) data element of a physical layer that may be used in connection with the invention ♦.

Fig. 18 shows an expanded structure of a supported transmission frequency element that can be used in connection with the invention.

10

Figure 19 illustrates as a data structure a non-competitive traffic mode (Contention-Free, CF) parameter set element that may be used in connection with the invention.

Figure 20 illustrates as a data structure a non-competitive traffic mode termination frame 15 (CF-End Frame) which may be used in connection with the invention.

Fig. 21 illustrates as a data structure the frame structure of the Physical Layer Extended Transmission Frequency Orthogonal Frequency Division (ERP-OFDM PHY) which may be used in connection with the invention.

20! Fig. 22 is a graph showing the bandwidth requirement of the invention.

• · • · * * «· • ·

Figure 23 illustrates in tabular form the number of 8 x 32- · · • .., · bit sample records used in connection with the invention in an appropriate arrangement for consecutive data blocks. ·. * For transmitting 25 digital audio signals.

• · · • · • ♦ »>»

Figure 24 is a graph showing the vibration behavior associated with the invention.

# · • · • ♦ * 1 2 3 2 • · 3

Fig. 25 is a block diagram of a general data structure related to the invention with respect to the timing of transmitting the worst case "": 30.

*: * *: Figure 26 is a flowchart illustrating the processing of an audio signal associated with the invention.

♦ ·! 7 119014 In this document, the following terms are used in connection with the invention.

1 wireless local area network (WLAN) 2 audio source 5 3 digital audio transmitter 4 base station 5 remote control 6 receiver 7 loudspeaker, typically an active speaker with built-in amplifiers 10 8 audio system (e.g. Surround Sound 7.1) 9 audio data formatting / sample 10 transmission level audio data formatting 12 sublayer, MAC = Medium Access Control, control of 15 transport means 13 distributed co-ordinate function 14 point co-ordinate function 15 non-competitive transport service 16 competitive transport service 20 17 generic MAC frame structure ·· ·: V 18 frame control j «·! * · “19 Duration / ID • * · ···! Address 20

♦ · * I

• * * ··· * 21 Address 2

# I

* * * I

* ··· * 25 22 address 3 0 9 »• · '' 1 '23 order control,. 24 Address 4 • · · 25 Frame Frame * «I * * 'V 26 Frame Check Sequence (FCS, Frame Control Sequence) e» »V 30 27 MAC Header • · · * ··· * 28 MAC Frame V: 29 WLAN- frame control field 30 traffic policy version 8 119014 31 type 32 subtype 33 toward distribution system 34 from To DS 5 35 additional fragment 36 retry 37 additional data 38 power management 39 Wired Equivalent Privacy 10 40 deflection 41 single address 42 group address 43 multicast address 44 multicast address 15 45 broadcast address 46 general beacon frame 47 frame control 48 duration 49 destination address 20 50 source address · * * • *. * 51 base service set identifier (BSS ID) • ·:. ** j 52 frames · «* ... · 54 Frame Check Sequence (FCS) 25 55 timestamp M» 56 beacon period 57 credentials ·· * · • ** 58 Service Set Identity (SSID) ·· • · 59 optional field ***** 30 60 beacon frame used in this invention ***** 61 frame control: 62 lifetime 9 · * 9 9 · ...: 63 destination address 9 119014 64 source address 65 basic service identifier ( BSSID, Basic Service Set IDentity) 66 Strain Control 67 Frame Frame 5 68 Frame Check Sequence (FCS) 69 Timestamp 70 Lighthouse Period 71 Authority Information

72 SSID

10 73 Non-competitive Traffic Parameter Set 74 Traffic Indication Map (TIM) 75 Physical Layer Extended Rate (ERP) 76 Extended Transmission Rates 77 Element Design 15 78 Element ID 79 Length 80 Information 81 TIM Element 82 Element ID 20 83 Length: V 84 DTIM Reading, DTIM = Delivery Traffic Indication Map • · * · * 86 Bitmap Control • · * ···· 87 Partial Virtual Bitmap t * »* ··· * 25 89 ERP Information Element • · • · * ·· 90 Element ID j. . 91 Length *.! 92 non-ERP prevails · · · · 93 use shield · · · * 30 94 Barker primer mode 95 r3-r7 ·· V * 96 advanced supported transfer rates new element design '** · 97 element ID j 10 119014 98 length 99 expanded supported transfer frequencies 100 non-competitive traffic parameter set element design 101 element ID 5 102 length 103 CFP reading 104 CFP, non-competitive sequence 105 CFP, maximum duration 106 CFP, remaining duration 10 107 non-competitive traffic frame (CF-End) 108 MAC header 109 CF-End MAC frame 110 frame control 111 duration 15 112 RA, recipient address

113 BSSDD

114 FCS

115 ERP-OFDM PH Y frame structure,

OFDM = Orthogonal Frequency Division Multiplexing 20 116 Coded / OFDM

·· 1: 1. · 117 Protocol Service Data Unit (PSDU) • · 118 PLCP primer * 119 SIGNAL field ·· · · "··· 1,120 transmission rate 25,121 reserved • ·· • · * · 1 · 1 122 length 123 tail! »V · ', .1 124 parity • · 125 service ·» · · 1 30 127 frame control «1 · * · * ··· 1 128 duration / ID: 129 address 1 130 address 2 11 119014 131 Address 3 132 Frequency Guidance 133 Address 4 134 Frame Frame

5,135 FCS

136 area of interest 137 WLAN repeat / beacon period (N * TU) 138 truncated uncompressed period 139 hub and receiver 10 140 host-to-converter 141 signal processor with buffer 142 USB host controller, USB = Universal Serial Bus 143 USB inputs 144 S / PDIF- inputs, S / PDIF = Sony / Philips Digital InterFace 15 145 analog inputs 146 analog buffers and splitters 147 A / D converters, A / D = Analogue-to-Digital 148 MAC / baseband processor 149 microcontroller 20 150 D / A converter and filter , D / A = Digital-to-Analogue ·· ·: V 151 select analog input • · *. *: 152 A / D conversion • · · ···! 153 select digital input

• M

• · 154 24-bit reformatting • · · **: · * 155 155 select audio input ♦ * · • ***** 156 channels 8 157 no · * 158 yes «· • · 159 format FEC and write to buffer, FEC = Forward Error Correction • · m • · • · * 30 160 calculate missing channels • »· • · '·; * 161 format FEC and write to buffer · 162 samples i-1 **' *: 163 samples ij 12 119014 164 samples i + 1,165 stacked samples i 166 Extended Service Set (ESS)

167 IBSS

5,168 CF polls 169 CF poll request 170 privacy 171 busy 172 antenna structure 10,173 most significant bits 174 audio MAC frame 175 control MAC frame System 15

According to Figure 1, the stiffness system consists of one or more audio sources 2, which can be either digital or analogue sources including direct computer interfaces, for example via a USB interface. Source 2 is coupled to an audio transmitter 3 which is further coupled to a base station 4 which includes an antenna structure 20 for wireless transmission. The transmitter 3 and the base station 4 are typically controlled by ·· ♦ • remote control 5 or a computer. The signal from base station 4 is transmitted wirelessly · ·!. ** | via LAN 1, using isochronous multicast messages to receivers 6 of a multichannel system 8 (e.g., Surround Sound 7.1) including several «·· * · * ... * speakers 7. In other words, audio data from sources 2 is converted to digital data • · · * ·! · * 25 and transmitted to speaker receivers as standard digital wireless • • '· * * * data.

• ♦ • · »*] * Transmitter Base Station • * • ♦ ··· ··· I. * 30 Figure 2 illustrates a simple version of transmitter station 3 which does not include the ··· · · ··· * inputs and figure 3 depicts a device that also has analog inputs 145. Transmitter Base ··· ί.ϊ ί Station 4 is typically an extended IEEE 802.1 lg ♦ * · * '* WLAN ΜΙΜΟ Access Point station operating at 108 Mbit / s, which receives 13,119,014 specified amount of mono, stereo or multi-channel analogue inputs and AES3, S / PDIF, or USB digital audio signals. In practice, 108 Mibt / s is the lowest standardized transfer rate for the rigid systems of this invention. Even higher WLAN transfer rates are expected in the future. They will make it possible to improve error scam techniques by using retransmissions. The analog inputs 145 are electrically buffered 146 and fed to an analog selector that can select one of the stereo inputs for export to a stereo 24 bit 192kbps A / D converter 147 for further Saija parallel converter 140, and a programmable two to eight channel DSP processor.

The use of the DSP allows any channel matrixing to generate derived channel signals from the right and left channel stereo signals. Alternatively, if the AES3 or S / PDIF digital input 144 was selected, it is converted to a parallel format by shift register 140 and then fed into a 32-bit signal processor 141 memory buffer, which transforms the signal into WLAN-15 for 8-channel, 24-bit parallelism. Using the USB 2.0 interface 143 allows direct 8-channel 24-bit 192kbps audio to be used as input, which is directly stored in 32-bit parallel storage. Resampling from lower sampling frequencies to 192 kilo samples / s can be done by DSP based on, for example, curve fitting of a third-order polynomial fitted through four sample points 20: *, · combined with an increase in resolution to 24 bits from possibly smaller • t • * «*. *: for sample lengths. After these processing steps, a prior "· · ··.! error scrambling code octet 11 (Figures 6 and 7) in front of each 24-bit audio sample 9 ·· • ** .. *. The resulting 32-bit samples are buffered to memory for transmission to · · * · ** j · '25 as eight sample records 9 representing current samples of eight * · * ··· * audio channels.

• Transmitter station 3 has a 16 kilobyte ring buffer memory area 141 or FIFO type • ♦ • · buffer for intermediate storage of incoming data awaiting transmission. After initialization • I «• ♦ · i. * 30, transmitter station 3 uses a TU unit to set the value of the non-competitive beacon period, and the CFPMaxDuration parameter sets to 512 ps. This ··· V adjustment is renewed when the incoming audio stream ends, which minimizes the reaction time at the beginning of the next · · ** · audio stream and thus the size of the memory buffer. Alternatively, it is possible to 14 119014 live fill the audio stream by filling the buffer with zero data as long as incoming audio is missing. Once the audio data stream has begun and there is at least 6.6 kilobytes of free space in the ring buffer or FIFO buffer, the uncompetitive sequence will be reprogrammed to three TUs to set the CFPMaxDurati-5 on parameter to 2,782 ps. The digital audio transmission over the wireless LAN will also begin at this stage. Depending on the traffic situation of the WLAN, it may take 459 ... 1448 ps before the first audio data is output, during which 2.8 ... .9.2 kilobytes of additional audio data has arrived in the buffer. Therefore, the space requirement limits for the input buffer are 9.4 ... 15.8 kilobytes, depending on the timing ratio between the audio source 10 and the WLAN. For this reason, a 16 kilobyte ring buffer or FIFO buffer is sufficient in all situations. The signal processor 141 fills its buffer with the incoming audio data and the audio data is transferred from the buffer either by a timer-controlled DMA controller or another processor so that the correct WLAN data rate is continuously maintained. There is a fast standardized parallel-to-serial or serial interface between the processor and the WLAN circuits and the software application program interface (API) of the WLAN circuit is used to control communication as described above. The transmission delay is up to about 1500 ps. In addition, reception buffering causes an additional 500 ps delay. Thus, the total delay is less than 2000 ps. This corresponds to less than one meter of sound propagation, making the delay insignificant for the human listening experience.

* * · F · *** • WLAN portion of base station (and receivers) complies with IEEE 802.1 lg standard ··· * ”* Range and transfer rates published by Atheros Inc. and Airgo Inc. with extension • · * «nuts. In addition, an MIMO-type antenna structure 172 is usually used.

25 The nominal bit rate is 108 Mbit / s. These advanced IEEE 802.1 lg WLAN implementations also include an efficient • transmission error repository mechanism that distributes,. ·, Possible burst path bursts into single bit reception errors at t · · • ·, * ··, reception and able to stack them all at the octet level. This feature is utilized in the · • · · · 30 forward-facing error correction method we define.

The competition-based, individually addressed message exchange between base station 4 and receiver stations is used to configure, monitor, and control the receivers and associated audio devices. The transmitter base station 4 further has a handheld infrared remote control receiver, a USB 2.0 computer traffic receiver / transmitter, and a universal USB 2.0 transmitter / receiver for Bluetooth and WLAN handheld remote adapters.

5 j System Configuration, Status Monitor, and Control The system configuration, status monitor, and control are performed using the handheld remote control (s) or (personal) PC application program (-10 programs) as described above.

i

Receivers; i i

As shown in Figure 4, receiver 6 typically consists of a MIMO antenna subsystem 172 and an IEEE 802g compatible WLAN circuit with Atheros or Airgo range and zoom rates. Typically, software-controlled multicolor LEDs are included to assist in the identification and status monitoring of individual speakers 7 in configuration, status monitoring, and control operations. WLAN is used at a nominal speed of 108 Mbit / s. The received audio data stream is buffered into a 20 16KB ring input or FIFO memory buffer and audio output from the buffer ·· t • is triggered when the buffer capacity reaches 6.6KB.

• ·:. ** i The 32-bit processor 149 selects the currently selected channel data of this receiver and is fed to a 24-bit D / A converter 150 followed by a high-quality ··· volume control circuit and a buffer amplifier to produce a studio-level analog audio signal corresponding to the selected audio channel of this various audio receiver unit. The channel selection is made by the configuration and control software using WLAN's competitive traffic services.

Each of the eight channel receivers 6 has its own internal crystalline clock, of which it produces a sampling clock signal of 192,000 samples per second. These clocks are restarted with CF-End control message end frame interrupts on each of the 3 ··· · (at the end of the reception gap of ί ί 072 ps.) The differential analog output of the D / A converter is the differential input of the active speaker of the channel configured for it.

16 119014

In addition, eight-channel digital audio is produced for use with standardized AES3 (S / PDIF) coaxial and opto output. This enables the implementation of wireless studio-quality 8-channel digital audio link. Another pair of AES3 (S / PDIF) coaxial and opto outputs are manufactured for digital input active speakers.

5

Remote Control Terminals There are two methods for controlling the ice system, a battery-powered handheld terminal 5 and application software that can be implemented on a variety of platforms including Linux, MS Windows, Apple, and Symbian operating systems.

Handheld remote control Handheld remote control 5 consists of a keypad, a small display, a processor and a communications kennel link to the access point. The keyboard functions allow selection of the audio source 2, configuration and control of the audio speaker group 8 and the individual speaker 7.

The speaker group 8 as well as the individual speaker 7 can be softly activated and deactivated and their overall and individual sound levels can be adjusted. The handheld remote control j communicates with the transmitter base station 4 via infrared, Blootooth or WLAN link 20. Base station 4 transmits controls to the speakers via individual receivers ·· »• V using a competitive mode of communication and either group or individual addressing.

• ·!, ** i The remote control 5 has a panic button and a function that causes instant soft mute to all speakers 7.

·· 1 • · • * * · · · · · · · · · · · · · · · · · · · · · · · · · ·

The system described above can be completely controlled by a computer running • · · * jj · 1 configuration, status monitoring and control application software. Commands and Responses • 1 j * "'communicates with the sending base station using Bluetooth, IrDA, LAN, WLAN, or j: 1.1 30 USB 2.0 links.

··· • 1 • · # ·· ··· • · · * · · · π 119014

Method

5 and 6, the apparatus of the invention transmits isochronically, in real time, eight completely independent and uniform canonical uncompressed channels with 24 to 5 bit 192,000 samples per second digital audio streaming with an effective 8 bit forward error correction code (FEC) supporting audio transmission. 11 for each individual sample. Groups of eight discrete 24-bit samples with their associated 8-bit FEC octets 11, combined with 32 octets or 256 bits, are referred to herein as the transmission-level audio data-10 format. Thus, the continuous audio application level digital audio data bandwidth requirement is 36.864 Mbit / s. Together with the FEC octet 11, the data bandwidth requirement of the application becomes 49.152 Mbit / s. In addition, there are MAC,

additions caused by framing, WEP encapsulation and physical layer framing, and the effect of time spent on competitive traffic. This results in even WLAN I

A 15 bit rate of 54 Mbit / s is insufficient for this application. Required performance cannot be achieved with today's standard WLAN technologies. The new transfer method described below is based on the innovative use of non - competitive multicast

With Point Coordination Function (PCF) as defined by IEEE 802.11 standards.

20 bandwidth can be optimally divided into non-competitive PCF medium access mode I

** * j V and the conventional Decentralized Control Function (DCF) •! ** l between the competitive access mode so that multichannel isochronous 4 ... i digital audio transmission is possible. With 108 ··· • · * ··· * Mbit / s extensions of the IEEE 802.1 lg + WLAN network and using ERP-OFDM PHY layer framing, it is possible to • • * * ·! · * 25 move the target eight (8) independent 24 + 8-bit, 192 * ** # * ··· * kilobase / s digital audio stream isochronically coupled with normal competition based 4 WLAN data traffic. The large number of channels, high resolution and high sampling frequency * * »• · · provide today's best commercially available wireless sound quality distribution.

44 4 • 4 4 1:. · 30 a ·· I 4 I '···' Data Structure 4 4 4 • · · 4 • 4 18 119014

According to Fig. 25, the object of the invention is to transmit a sufficient number of audio blocks (transmission-level audio data formatting) 10 in order to produce high-quality audio. First, the beacon period 137, which is set by software settings, must be properly selected to achieve the goal. The beacon signal, which defines the length 137 of the beacon period 5, is transmitted at intervals determined by an integer as defined in the IEEE 802.11 g WLAN standard. This integer receives values between 1 and N. In other words, the beacon period 137 is the product of the beacon number and the time unit (TU). One time unit TU in the IEEE 802.1 lg WLAN standard is 1024 ps and therefore the beacon 137 is a multiple of TU (1024 ps). However, the standard specifies that for each beacon 10 beacon 137, sufficient time must be reserved for racing traffic, more specifically sufficient time for the maximum frame size, the ACK frame, the two basic time frames, and the two SIFS times. In accordance with the present invention, a sufficient number of TU time units of lighthouse 137 has been found to be three (3). This gives enough time to transmit 12 audio MAC frames 174 and 2 control MAC frames 175 15 within one beacon period 137. Each audio MAC frame 174 includes 49 to 51 transmission level audio data format blocks 10, the number of blocks being determined according to the table of FIG. In this figure, a single row represents the contents of audio MAC frame 174 during one uncompetitive period 138 in beacon period 137. As shown in Figure 23, the predefined sequence is repeated every 125 beacon periods. With this well-defined sequence, the average flow rate of the audio input and> · · • * t:. * WLAN output is matched and their vibration between them is minimized, as shown in Figure 24. This results in · · * •• J at the same time minimizes the buffer memory requirement for both transmitter 3 and * ·· • * ··· * receivers 6.

* · · * ·· 25 · * · »♦ *** Bandwidth Allocation 9 9 9 9 9 9 9 9 M As shown in Figure 25, to ensure the timely transport of audio data, it is necessary to: maximum repetition rate of non-competitive episodes 138. At the same time ·· * * ·: 30 maximum network capacity must be reserved for audio traffic. The IEEE802.il standard requires that • i * V: During each repeated non-competitive period, there must be sufficient competition traffic time to send at least one maximum-size data frame ***** along with its acknowledgment frame, two SEFS- episode and two basic time slots.

is 119014 Using 108 Mbit / s bit rate and ERP-OFDM PHY frame, this time is 212 + 40 + 2x 10 + 2x9 = 290 ps. As described in the IEEE 802.11 standard, racing traffic may advance to shorten the non-competitive period 138 at the beginning with a maximum value of one RTS control frame, one CTS control frame,.

i 5 the sum of one maximum data frame, one ACK control frame, and four 1 SIFS time durations. At 108Mbps and ERP-OFDM PHY framing, this requirement is 40+ 40+ 212 +40+ 4x 10 = 372 ps. The non-competitive section begins with a beacon frame 67 (Figure 13) and is followed by one SIFS. At 108 Mbit / s and using ERP-OFDM PHY, this requirement is 76 + 10 10 = 86 ps. The non-competitive section is terminated by CF-End frame 109 (Figure 20). 108

At Mbit / s and using ERP-OFDM PHY, this requirement is 40 ps. The remaining time within the repeated uncompetitive playback period is over for uncompetitive data traffic. Since the granularity of the non-competitive period is one 1024 ps units of time (TU), the time remaining for non-competitive traffic when the 15 non-competitive periods are set to 1 TU is 1024 - 290 - 372 - 86 - 40 = 236 ps. Considering the maximum data frame size defined by IEEE 802.11, the additional load caused by MAC, WEP, and PHY encapsulation, and the PIFS times required between consecutive data frames, only 72 samples of 8-channel 24 + 8 bit, 192kbps data can be transmitted during this time ..

... 20 Due to the sampling frequency of audio, the real-time requirement is slightly over 196 samples • · \ during each TU. Therefore, this is not a viable solution. Setting the duration of the uncompetitive period to · ·· * / 2 TU, for uncompetitive data traffic, or for the duration ····. ···. 1 260 ps, which allows transmission of up to five largest samples of 72 • • · · · · · frames equivalent to 1 155 ps, plus one 27-sample frame counter ·· * · ”** 25 for 104 ps. This will take a total of 387 samples over two TUs with a minimum real-time requirement of slightly more than 2 x 196 = 392 samples. Thus, this is a bit too little capacity for real-time audio transmission. Thus, it is evident that the smallest and ··· ϊφ> 4ϊ most desirable non-competitive traffic repetition intervals are then exactly three TUs, 3; V. x 1024 = 3,072 ps

! "*: 30 ···. ···. In order to optimize the smooth flow of data and minimize buffering needs, the average number of samples per unit • * · TU per unit should be kept close to 1 024 / 1000 x 192 = 196.61 as much as possible by varying the size of the data frames in an appropriate manner as shown in Figure 23. The frame size algorithm, one of the key innovations of the invention, is presented as follows: Uncompetitive time is initially decomposed into 12 variable size buffers. The size of the buffer 5 ranges from 49 to 51 sample records each, according to the following size determination rules: The first, second, fourth, fifth and sixth buffers of successive 12 buffer sets contain 51 samples each, while the first buffer of the third set contains 50 sample records. buffers contain 49 samples. This sequence is repeated every se for the 10th buffer set up to the 125th buffer set, whereupon the whole cycle restarts.

The only exception to the above rule is the first buffer of the 60th buffer set, which contains 50 samples instead of 51 as would otherwise be required by the internal repeat rule.

As shown in Figure 24, this algorithm ensures that the buffering color remains less than +/- in all buffer sets and returns zero at every 125 samples at the end of the control set. This adjustment algorithm achieves the worst-case 180 ps margin within the uncompetitive data transfer time. This time is sufficient to transfer one additional 1,632 data octet frames, or alternatively two 450 octets frames of user data, which can be used for desired other real-time control purposes. This arrangement also enables efficient, competitive user-based data transfer, which, along with real-time audio transmission, is approximately 12 Mbit / s. Loops- • · · • · ·; · License-based traffic is available for rigid system configuration and control, as well as for other independent data traffic.

As noted above, the duration of at least three TUs is required for the beacon * * *; “* 25 repetition intervals to reserve enough bandwidth for uncompetitive

• · I

• · ** 'for isochronous audio traffic. By selecting a minimum value of three TU time units,. and minimizes both the lag time and the need for buffering. The choice of a minimum value of three · · · TU time units will still allow competitive traffic the maximum bandwidth • • · with uncompetitive isochronous audio traffic.

• · ·: · * 30 • · · • m *; * 'The combination of these three limiting factors is called the optimum * * · for this application.

* «21 119014

error Management

According to Figure 7, the error management method is optimized for simplicity and speed, assuming that the rigid system is used by people for listening to multi-channel studio-5 speech and music. This means a simple and fast best-company error-fooling procedure that either completely eliminates the remaining false-frame application-level transmission errors or reduces their audio penalties to an unobservable level. The method utilizes a long 24-bit audio data sample and a high sampling rate of 192kbps, as well as the ability inherent in 10 extended IEEE 802.llg + implementations to convert transmission burst errors into single bit errors at reception. However, this error scam method is not valid for applications where errors cannot be tolerated at all.

As shown in Fig. 7, before the transmission step, the eight-bit Forward Error Control (FEC) 11 octet is coupled before the most significant octet 173 in each individual sample 162-165. This FEC octet 11 is formed by duplicating the most significant octet 173 of the data sample. Due to the WLAN's internal transmission error scrambling method, almost all of its remaining reception errors are single-bit errors. Therefore, it is sufficient to offset the effect of one bit 20 errors. This is done by comparing the most significant data · · · · · * · · · · · · · · · · · · · · · · · octet to the corresponding received · FEC-octet • and * · *. * *: The most significant sample octet 173 is replaced by the average of the most significant octet 162 of the previous sample and the next most significant octave of the following flawless sample 164. This effectively eliminates almost all remaining errors in the 25 · '···' 25 most significant data octets. Because the lower importance bits represent less than 4% * * * · · · * of the dynamic range of the audio, their significance is lost. Due to the high sampling rate, the residual errors are not audible to the human ear.

It is important to remember that multicast simultaneously transmits the specified multicast * * to all recipients of all * channel * audio data. This may: (! List various intelligent error correction methods based on data from adjacent audio channels if such methods appear attractive. If special error correction hardware is still available or a lot of computing power is available at both the transmitter and the receivers, effective known 8/24 error correction methods to provide a more comprehensive error correction.

5 Pacing i

As shown in Figure 20, the system's internal synchronization is based on periodically repeated frame termination interrupts generated from CF-End frames 109 at exactly 2,782 ps from the beginning of each regularly repeated, uncompetitive 3,072 ps 10 episodes. The frame end interrupt of this control message 109 synchronizes all the receivers 6 with an inaccuracy in the difference in latency between the interrupts of the various receivers. Because all receivers are programmed to wait for this particular interrupt, the system-level synchronization vibration caused by the interrupt latency is of the order of one machine instruction sequence (plus a very small inter-processor 15 crystal clock phase difference random vibration). In practice, this total vibration is of the order of 100 ns and is in no way perceptible to the human ear. By comparison, the duration of the 192kbps audio sampling period is 5.21 ps.

Detailed Description of the WLAN Transmission Round 20 • t • t *, As shown in Figure 25, when idle, when no audio signal is present, the transmitter is programmed to * * ·· operate at a beacon period value of one time unit (1 TU). In this way, the response delay of the system * "!! i when a new incoming audio stream starts and thereby minimizes the size of the transmitter buffer. When the beginning of the audio stream is detected, the transmitter enters the active transmission mode.

• · ·

, ···, 25 During the active transmission phase, the beacon period is reprogrammed to three TUs

* * I

• · · I

time unit, and as a result, each WLAN audio transmission time slot is programmed to last three 1,024 ps TU time units each, i.e., a total of 3,072 ps.

• · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · Point Point Co Co Co Co Co Point Point Co Co Point Co Point Co Co Point Point Co Point Co Point Point. Lighthouse Repeat • ♦ * * i. ···. 30 and thus the recurrence interval for non-competitive traffic is also set to three TU time units, and each such period includes both the non-competitive and the non-competitive basis. The duration of the reserved non-competitive period is set to 2 782 ps ♦ ♦ using the CFPMaxDuration parameter in the beacon frame 67, and this setting leaves 23 1 1 901 4 guaranteed 290 ps decentralized control function (DCF) for racing traffic. This time is long enough to transmit a maximum length data frame during the competition period, with associated acknowledgments and associated frame spacers, as required by IEEE 802.11. It also means that at least 6.16 Mbit / s of bandwidth (when using the largest data frame size) is always available for competitive traffic. The uncompetitive period reserved for heavy traffic on large frames will sometimes be shortened at the front end when the transmission of the frame occurs during the start of the intended uncompetitive period. Because this competition-based information exchange may also include RTS, CTS, and ACK control-10 frames with associated inter-frame time elements in addition to said largest data frame, up to a maximum of 372 ps may be consumed with media occupied from the beginning of the uncompetitive period.

The worst case scenario for broadcast data for audio data is as follows. The anticipated start of the race-15 episode is coming, but broadcasting the longest possible competition-based sequence has just begun. It causes a 372 ps shortening of the uncompetitive period. Only after this shortening delay can a 40 ps beacon message be sent which sets the NAV status. The transmission of the first audio data block begins when the additional 10 ps SIFS time has elapsed. This means a total of 459 ps be- ... 20 after the start of the non-competitive period. In smaller frontal shortening situations, • the broadcasting software adds an appropriately sized silent fill time to reach 459 • · · ps. This arrangement ensures that the first audio bit is always transmitted. within the same relative momentum of 3 072 ps within an uncompetitive replay period. On offer • · · * . . *. therefore, the transmission time for uncompetitive audio data is 3 072 - 458 - 290 - 40 - 10 = 2 • · · 25 274 ps. In the worst-case scenario, the first audio buffer contains • • ♦ ten-one 24 + 8-bit sample records. The following eleven audio data blocks contain: Y each 49 sample records. When the MAC and ERP-OFDM PHY Frame Fields are attached, · · ·, the corresponding frame transfer times for the sample records 49, 50, and 51 become · * · *; 156 ps, 160 ps, and 160 ps, respectively. The largest size of 12 doubles is 51 + 11 x 49 • *: ***; 30 audio records. These twelve blocks are moved sequentially so that there are 19 ps PIFS sequences in each of their slots. Thus, the worst-case total time from the first ····· to the first audio bit is 160 + -11 x 156 + 11 x 19 - 2,085 ps. The effective time-compression of 24 1 1 901 4 in WLAN transmission is thus 2,085 / 3,072 = 68 percent. The twelve audio blocks are followed by an additional 19 ps PIFS period followed by a 180 ps programmed idle delay, followed by a 40 ps CF-End Broadcast Frame 109 to terminate the non-compete period and remove the NAV state previously set to 5 by the start of the beacon frame. This happens at exactly the same time that the uncompetitive period would have ended also based on the timer set by the CFPMaxDuration parameter of the beacon. The resulting 180 ps non-competitive time margin of the total available time of 2,284 ps represents only eight percent of the total non-competitive time margin. This is the beginning of a competition-based 10 episode that allows the transmission of one maximum-size data frame, the ACK1 response and associated two-frame SIFS times, and two basic time slots, as required by the IEEE 802.11 standard.

Transmitter and Base Station Functions 15

As shown in Figure 26, the system selects n analog inputs 145, or m AES (S / PDIF) digital inputs 144/143, between steps 151 or 153. If an analog input was selected, the buffered analog stereo signal would be converted in the next step 152 to 2 x 24 bit s with two-channel A / D converter. If a digital ... 20 input was selected, the incoming digital bitstream would also be reformatted in step * # 1 • · j *. 1. 154 to 24-bit 192 kS / s format. In step 156, the number of audio channels is checked; • 1 · ♦ · · * and if some channels are missing 157, they are counted in step 160.

write the FEC and write it to the buffer together with the audio data in step 161. I

. 1]! ' If the number of channels is eight in step 156, then the data and FEC channels • · · j. ·. ”25 can be directly into the buffer in steps 158 and 159.

• · »··. ·, ·. Multicasting, in this application, refers to the procedure whereby all receivers receive the same data packet from which they extract suitable data.

· »·

In other words, all eight loudspeaker receivers receive the data of all • · · · · 30 audio channels, but only extract from that data the data that is allocated to each of the channels represented by each of them.

··· • · · * · · 25 119,014

The invention is applicable to a wide variety of isochronous communication systems, but as described herein, it is particularly well suited for multichannel audio applications.

Consequently, video applications are also suitable for some embodiments of the present invention.

In addition to WLAN broadcast media, the present invention is also applicable to UltraWideband (UWB) radio transmission technology or HomePlug AV type transmission technology where power network cabling is also used for data transmission. In the latter case, the transmission system 10 is not drone wireless, but since active speakers always require external power supply via cable, no additional cables are required for data transmission.

j ♦ ♦ • · 1 1 1 1 * 1 * 1 * * 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ··· • »• 1 · • · f • · · · 1 • • · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · t · • ·!

Claims (4)

26 1 1 901 4 A method for synchronizing a multicast digital audio or video data in your wireless real-time applications (6,7) with a streaming multicast method, wherein: - the data is organized into frames (17,174,175) consisting of control frames (175) and audio or video frames (174); or video data (9,10) is transmitted in multicast to a plurality of 10 receivers (6) within successive beacon periods (137), characterized in that: - audio or video data is synchronized between the plurality of receivers (6) using 15 -End the frame interrupt formed by the control frames. A method according to claim 1, characterized in that the interrupt command is a frame end interrupt command. 20 M I; ', · 3. Digital hijacked audio or video data wirelessly in real-time. 1 ·: In applications (6,7), a streaming multicast system consisting of ··· ··· «··· •» * ··· 1 - means for organizing data into beacon slots (137) including • · 1 · 25 and audio or video data frames (174) and control frames (175), and • · "· '- means for transmitting thus organized frames (174,175) .. as a multicast to multiple recipients (6) within successive beacon periods *» (137), järjestelmä * 1 30 characterized in that the system comprises: • 1 m 27 1 1 901 4 - means for synchronizing audio or video data (9,10,174) between a plurality of receivers (6) in each beacon period (137). ) with the interrupt command placed inside. A rigid system according to claim 3, characterized in that the frame interrupt interruption of the CF-End control frame, which is transmitted at a precisely timed interval within the beacon period (137), is used to provide low-vibration re-synchronization. ** · • t · • • • • • • 1 • ·· • 1 ··· ···· * ·· * · I • · ··· | • · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · F «i 119014