Apparatus and Method for Providing Digital Audio in the FM Broadcast Band

The present invention relates to audio broadcasting and reception, and more particularly to a method and apparatus for providing high quality digital sound signals within the FM broadcast band.

Cable television growth has come from the development of various programming categories and by the technologies which made the delivery of these programs possible. Cable first brought distant TV signals to areas where there was little or no off-air reception. This applies to distant signals and weak signal areas where outdoor antennas are mandatory. The next category to bring major growth to cable was the pay service after the development of reasonable cost satellite delivery systems. After satellite delivery was accepted and less costly, super stations and cable networks formed another category of programming that has become customary and are often termed "extended basic" services. Franchising and local politics has created a generally unprofitable but necessary category called "local origination". Recently, addressable technology and aggressive marketing have caused "pay per view" programming to form another category of programming.

FM (audio) broadcasting over cable has never achieved significant success for two technological reasons: signal quality is poor and there has been no way to collect revenue or control the access to the service.

New digital techniques for the reproduction of sound provide performance that is far superior to analog techniques which have been used in the past. An example of high fidelity sound reproduction using digital techniques can be found in the compact disk technology which has recently enjoyed tremendous success as an alternative to phonograph records and tapes. Digital recording and playback techniques provide reproduction of music that is extremely realistic and free from background noise and distortions which have plagued other high fidelity sound reproduction systems currently in wide scale use.

Commonly-owned, copending U.S. patent application Ser. No. 022,380, U.S. Pat. No. 4,821,097 entitled "Apparatus and Method for Providing Digital Audio on the Sound Carrier of a Standard Television Signal", and incorporated herein by reference, discloses a system wherein the FM audio portion of a standard television signal in the TV band is replaced with digital audio. Three digital audio channels are time division multiplexed on the sound carrier, using combined multiphase and AM modulation. The audio signals are digitized using adaptive delta modulation techniques. Video vertical and horizontal framing, as well as the audio carrier phase reference, audio data bit time and frame reference, and various control data is carried using AM modulation. The digital audio information is carried using multiphase modulation. The composite data stream may be serially encrypted to provide security and prevent unauthorized reproduction of the video and/or audio portions of the television signal.

U.S. Pat. No. 4,684,981, entitled "Digital Terminal Address Transmitting for CATV", discloses producing digital signals of up to four different modes for transmission over an unused television channel in an existing cable television transmission line. High quality audio signals may be transmitted and/or data channels or monaural audio signals, all of which may be transmitted over the single cable television transmission line. Cable television channels have approximately a 6MHz bandwidth, and are transmitted in the TV band which ranges in frequency from 50MHz (channel 2) to 550MHz (channel 50).

Any distribution system which transmits digital audio data (such as a cable television system) must be such that the transmitted audio signal does not interfere with millions of radio sets already in existence which use conventional analog sound circuits. Thus, such things as the channel width of 400 kilohertz (KHz) for each channel within the FM broadcast band should not be changed, subject to narrow tolerances.

It is an object of the present invention to provide a method and apparatus for incorporating digitized audio data within a plurality of channels in the standard FM broadcast band in a manner such that the signal will be recoverable for reproduction of the transmitted audio program on FM radio receivers, with additional digital data receiving circuitry.

In accordance with the present invention, a method and apparatus are provided for transmitting, receiving, and reproducing digital audio signals as discrete carriers frequency allocated as standard FM broadcast signals. An audio signal is digitized using, for example, adaptive delta modulation techniques. Several channels of audio information, such as left and right stereo channels or four voice mono channels can all be digitized and incorporated onto the digital carrier in the FM broadcast band. The digitized audio signal is modulated using multiphase or multilevel amplitude or frequency modulation of the carrier in the FM broadcast band.

The 400KHz spacing of digital carriers in the FM band allows 50 channels of stereo digital quality audio, addressable and encrypted. In a local service area, the FCC spaces FM stations 800KHz apart, meaning that there are a maximum of 25 local FM broadcasters in the most dense markets.

A bandwidth efficient system would use Dolby ADM and would allow data carriers at 400KHz spacing. This spacing is the same as normal broadcast FM. This would allow up to 50 channels in the FM band. The digital channels could be intermixed (staggered) with regular FM channels. There is also the possibility for broadcast (wireless) application.

Another alternative would be to transmit 44KHz sampled 16 bit linear PCM (Compact Disc Format) spaced at 1.2MHz between channels, allowing 16 channels in the FM band.

The Dolby system could be built at low cost. From a marketing standpoint, it is recommended that discrete carriers be used rather than full video channels with time division multiplexing (TDM). This results in lower cost, more acceptable use of spectrum to the cable operator and more robust performance.

When the present invention is used in conjunction with a cable television system, three primary components are used. These are the addressable controller (also referred to as "headend controller"), the headend encoder, and the subscriber converter (also known as the "subscriber terminal"). Both the addressable controller and encoder are present at the headend from which the cable television signals are sent by the cable system operator. The addressable controller controls all subscriber terminals in the cable television system, controls the encoders/decoders associated with the system, configures scrambling modes, service codes, and encryption keys, and orchestrates the dissemination of all decryption keys. The encoder of the present invention is a headend device consisting of a number of subcomponents including an audio digitizer, audio scrambler, tag insertion logic, addressable controller interface logic, and modulator circuitry.

The subscriber converter is a device located at each subscriber's residence and contains an RF converter module, demodulator, addressable control interface logic, subscriber interface logic, audio decryptor and digital to analog ("D/A") converter.

Control data communicated over the FM path time division multiplexed with the digital audio data between the headend controller and the encoder typically includes a signature used to protect sensitive information communicated over the path, tag and audio encryption keys, key usage identifiers, sampling mode data, audio service code, and price and morality rating data. Data which pertains to the digital audio service and is sent to the subscriber terminal over the FM path includes a signature used to protect sensitive information communicated over the path, tag and audio decryption keys, and authorization information.

The following commonly-used abbreviations may be used throughout this application: kilohertz (KHz), megahertz (MHz), frequency modulation (FM), television (TV), adaptive delta modulation (ADM), amplitude modulation (AM), cable television (CATV), pulse code modulation (PCM), time division multiplex (TDM), pulse modulation (PM), pulse amplitude modulation (PAM), pulse width modulation (PWM), frequency division multiplexing (FDM), Quadrature Phase Shift Keying (QPSK), radio frequency (RF), audio frequency (AF), direct current (DC), and Federal Communications Commission (FCC).

Other objects, features and advantages of the invention will become apparent in light of the following description thereof.

FIG. 1 is a schematic block diagram of the digital audio system of this invention.

FIG. 2 is a schematic block diagram of a digital FM broadcast and cable interconnection system, according to the invention.

FIGS. 3A and 3B, combined, are a schematic block diagram of a digital FM receiver, according to the invention.

FIG. 4 is a schematic of a multiphase modulator suitable for use in the digital audio system of this invention.

FIG. 5 is a polar diagram showing phase relationships for the multiphase modulator of FIG. 4.

FIG. 6 is a schematic of a Costas loop QPSK detector suitable for use in the digital receiver of this invention.

FIG. 7 is a schematic block diagram of a digital FM broadcast and cable interconnection system, according to the invention.

FIG. 8 is a schematic block diagram of a digital FM receiver, according to the invention.

Digital information such as digitized audio, addressing data, and auxiliary data may be combined together to form a composite digital data stream. This digital signal may then be modulated on to a carrier for transmission. The modulation may cause amplitude, phase, or frequency variation of the carrier. In order to maintain channel spacing similar to previously established analog transmission standards, multilevel (AM), multiphase (PM, i.e., QPSK) or multi frequency (FM) must be used. QPSK or eight frequency FM are approaches, when combined with an efficient digital audio sampling system such as Dolby ADM, that can allow coexistent digital and analog modulated carriers in the FM broadcast band at the normal frequency allocations. QPSK is the preferred transmission modulation means as its signal to interference ratio required for operation without data errors is less than with eight level FM.

FIG. 1 shows the major elements of the digital audio system of this invention, portions of which will be described in greater detail hereinafter. As described herein, the digital audio system is applied to a cable television network, but it will be evident that the techniques described are applicable to wireless broadcast of digital audio.

At the headend, or cable transmission center 10, a plurality of television channel transmissions 12 are provided for distribution over a cable transmission line 14, according to known techniques. Further, according to known techniques, a plurality of subscribers 16 (one shown) are connected, each by a cable drop 18, to a distribution network 20. Each subscriber is provided with either a cable-ready television, capable of tuning in excess of 100 TV channels, or is provided with a converter 22 which converts a selected one of the hundred or so TV channels to a preselected channel, such as channel 3, which may be received by an ordinary, non-cable-ready television set 24. It is further known to provide "special" channels within the band of television channels that may only be viewed by a subscriber on a special basis. To this end, digital address signals 26 are provided on the cable 14, and the converter 22 either permits or restricts viewing the special TV channels in response to the digital address signals, again according to known techniques.

According to the invention, audio-only source material is also provided over the cable 14 in the following manner. A channel 30 of audio source material 32 is provided to a digitizer 34 which converts the source material into digital format. Such conversion of the audio source material to digitized audio may be performed according to a variety of known techniques. The digitized audio source material is provided to an FM band exciter 36 which provides the digitized audio source material as a radio frequency (RF) signal to the cable 14. One channel 30 is illustrated, but several channels may be provided. Each channel may contain stereo program material. The RF output of each channel 30 occupies up to 400 kilohertz (KHz), preferably in the standard FM broadcast band which ranges typically from 88-108 megahertz (MHz). There are fifty 400KHz channels available in the standard FM band. Thus, although up to fifty distinct audio channels 30 could be provided, it is preferable to provide only up to 25 channels of digitized audio material in alternate (every other) channels in the standard FM band.

The digitized audio signals from the audio channels are provided from the subscriber cable drop 18 to a digital FM band receiver 38, which is described in greater detail hereinafter with respect to FIGS. 3A and 3B. As with the "special" TV channels, access to some or all of the digitized audio channels may be restricted by the use of address signals 26 which are imbedded in the multiplexed digital channel.

It is also possible to provide nondigitized audio source material over the cable 14. A channel 40 of audio source material 42 is provided to a conditioning circuit 44 to adjust the signal level of the source material. The conditioned audio source material is provided to an FM exciter 46 which provides the conditioned audio source material as a radio frequency signal to the cable 14. One channel 40 is illustrated, but several channels may be provided. As with the digitized channels 30, the RF output of each nondigitized channel 40 occupies 400KHz in the 20MHz FM band, and can be provided as a nonspecial (non-address restricted) channel to the subscriber.

Advantageously, the nondigitized channels 40 can be interspersed between the digitized channels 30. Alternatively, but less effectively, the digitized channels 30 can be assigned to a portion, such as the upper 10MHz of the standard FM band while the nondigitized channels 40 reside in the lower 10MHz of the standard FM band.

A combiner 48 combines the signal outputs of the TV channels 12, the address information 26, the digitized audio channels 30 and the nondigitized audio channels 40 onto the cable 14.

FIG. 2 shows an FM broadcast station 50 and cable interconnection system. A studio 52 provides audio source material (similar to elements 32, 42 of FIG. 1) as unprocessed audio signals in stereo along "left" and "right" signal paths.

In one instance, the audio signals are provided to an FM stereo encoder and loudness processor 54 (similar to element 44 of FIG. 1), and from there are provided onto an FM exciter 56 (similar to element 46 of FIG. 1). The output of the exciter 56 is amplified by a high power amplifier 58 and broadcast over the airwaves by an antenna 60 as stereo multiplex (MPX) FM in the FM broadcast band.

In another instance, the audio signals are provided to a Dolby adaptive delta modulation (ADM) encoder 62, and from there are provided as a data stream to a digital processor and combiner 64. The digital processor and combiner 64 operates under the control of a text and control computer 66.

The output of the digital processor and combiner 64 is provided in one instance to an FM exciter 68, similar to the FM exciter 56. The output of the exciter 68 is amplified by a high power amplifier 70, similar to the amplifier 58, and broadcast over the airwaves by an antenna 72, similar or unitary with the antenna 60, as digital FM in the FM broadcast band for reception by a digital FM receiver, such as is disclosed in FIGS. 3A, 3B.

In another instance, the output of the digital processor and combiner 64 is provided as 8-level data to an FM modulator 74 which inserts the digitized audio signals onto a cable television transmission line 76 via a directional splitter 78 to a cable headend 80 (corresponding to element 10 of FIG. 1). Preferably, the digital audio signals are transmitted over the cable 76 at 5-30MHz, which is reserved for upstream (to the headend) communication over the cable 76, and which is outside of the spectrums of both the audio channels (30 and 40 of FIG. 1) and television channels (12 of FIG. 1) that are provided by the headend 80 to subscribers 82.

In FIG. 2, the exciter 68 could as well be a QPSK modulator, and the modulator 74 could as well be a QPSK modulator.

In the arrangement shown in FIG. 2, the headend 80 is provided with a digital demodulator and remodulator 84 for receiving and demodulating the digitized audio signals from the off-site modulator 74 in the 5-30MHz band, and for remodulating and transmitting digitized audio signals on the transmission line 76 in the FM band (88-108MHz) to subscribers 82. The techniques of providing "special" audio channels, as well as the spacings of digitized and nondigitized audio channels within the FM band, discussed with respect to FIG. 1, are equally applicable in the system of FIG. 2.

FIG. 2 contemplates that several stations 50 will provide digital audio channels, generally one channel per station, to the cable system operator (CSO) 80. As noted therein, this is readily accomplished over existing cable transmission lines 76 in a band (such as 5-30MHz) reserved for upstream communication to the CSO. More details of the system are shown with respect to FIG. 7.

FIGS. 3A and 3B show a digital FM band receiver. Generally, FIG. 3A shows a tuner section 100, and FIG. 3B shows a decoder section 101. The input 102 to the receiver is either a cable transmission line (14 of FIG. 1; 76 of FIG. 2), or a suitable antenna and preamplifier (not shown). It is contemplated in this example that the receiver will tune from 72-120MHz, in an "expanded" FM band, to provide a large number of audio channels while avoiding the used TV channels, and gaps therebetween, such as the 4MHz gap between TV channels 4 and 5.

The signals from the input 102 are provided to a double tuned tracking filter (DTTF) 104, from there to an amplifier 106, on to a single tuned tracking filter (STTF), and to a mixer 110, according to known techniques. The mixer 110 receives a second input from an oscillator 112, so that the output of the mixer 110 is at an intermediate frequency (IF) of 10.7MHz for a selected channel. The channel selection process is under control of a tuning synthesizer 114, integrating amplifier 116, STTF 118 and amplifier 120, interconnected as shown, and impressing an appropriate signal on a line 124 to the DTTF 104, STTF 108 and oscillator 112 to effect channel selection, according to known techniques.

The selected audio channel is provided at the intermediate frequency (IF) to a filter network comprising a bandpass filter 126, amplifier 128 and bandpass filter 130, as shown, according to known techniques, and is ready for detection.

In one embodiment of the invention, a detector 132 is provided which comprises an FM detector 134, such as a Sanyo LA1150, which provides an 8 level data output to an analog-to-digital (A/D) converter 136, such as a 4-bit CMOS device. The detector 132 is suitable for digital audio received in 8-frequency modulated FM format.

The output of the A/D device 136 is provided as a data stream over a bus to a demultiplexer and decryption logic circuit 138 which separates the data stream into control bits and channel indication (tag bits) and encrypted digital audio data bits (demultiplexing functions) and decrypts the digitized audio data into a suitable form for a Dolby decoder 140. The audio data is decrypted into three serial streams per audio channel consisting of basic delta modulation parameters for "left" and "right" channels and companding data streams for "left" and "right" channels.

The demultiplexed control and channel data separated out from the data stream by the element 138 are provided to a microprocessor (uP) 142 which controls the overall operation of the receiver. Channel selection is provided by an infrared (IR) receiver and/or a keyboard 144, which information is passed on by the microprocessor 142 to the tuning synthesizer 114. A unique address, or serial number for the receiver is stored in a nonvolatile memory (NVM) 145, for addressing by the CSO, as discussed with respect to FIG. 1.

The output of the Dolby decoder 140 is provided as "left"and "right" audio channels to a stereo amplifier 146, and to stereo outputs 148 for use with standard audio components. It is anticipated that a relay will be provided at the output 148 to switch between other source material (not shown) and the digital audio output of the receiver, to cover instances where a user's amplifier component has limited inputs available.

In an alternate embodiment, the detector 132 is a quadrature phase shift key (QPSK) detector. This, of course, presupposes that the digital modulation of audio data signals occurring, for instance at element 34 of FIG. 1 and element 64 of FIG. 2, occurs in the QPSK mode. It is apparent that reception of multilevel AM or FM modulated digital signals can suffer from multipath reception problems (reflections) when transmitted over the airwaves (see 72, FIG. 2) especially with respect to stereo transmissions. QPSK displays greater immunity to this problem.

QPSK techniques are well known, and are disclosed for instance in the aforementioned commonly-owned, copending U.S. patent application Ser. No. 022,380, which is incorporated herein by reference.

FIG. 4 shows a multiphase modulator 200 suitable for use as the FM band exciter 36 of FIG. 1. Serial data input is provided to a serial/parallel converter 202, filtered by two digital filters 204 and 205 and provided to two digital-to-analog converters 206 and 207, as shown. The output of each digital-to-analog converter 206 and 207 is provided to a balanced mixer 208 and 209, respectively. The output of a carrier oscillator 210, operating in the FM band, is split by a splitter 211 and provided, in one instance, to one of the mixers 209, and in another instance is phase shifted 90 degrees by a phase shifter 212 prior to being provided to the other mixer 208. The outputs of the two mixers are combined at a combiner 213, the output of which is digitized audio RF output in the FM band. Multiphase modulation technique is described in greater detail in the aforementioned U.S. patent application Ser. No. 022,380, and is incorporated by reference herein.

The multiphase modulator 200 is used to modulate the digital audio data.

As shown in the polar diagram 220 in FIG. 5, the audio data is modulated such that each two bit symbol appears 90 degrees apart on the axes of the polar diagram. The rightmost bit in each of the two bit symbols is shifted out of the transmitting shift register first, and into the receiving shift register first. There are four data points 222, 224, 226, 228 shown in polar diagram 220 on the circle 221 which represent the normal amplitude of the carrier signal.

FIG. 6 shows a known Costas loop carrier recovery system 250, which is suitable for decoding a QPSK signal according to known techniques. Such an arrangement could be advantageously employed as the detector 132 of FIG. 3B.

FIG. 7 shows a digital broadcast system 300. A playlist computer 302, for instance at a station controls the selection of audio source material from a disc player 304. The output of the disc player is digitized by a Dolby digitizer 306, and passed on as one of 16 inputs (channels) 308 to a formatter/encryptor/tag inserter/EDC inserter ("inserter") 310. The inserter 310 combines the digital audio output of the digitizer 306 with other source material, which may or may not be digital audio. The inserter 310 formats and encrypts the source material on each channel 308, tags it to identify a program access level, and provides error detection and correction (EDC) functions. The output of the inserter 310 is multiplexed by a multiplexer 312, modulated by a modulator 314 and transmitted over a single video satellite uplink 316.

At the receiving end, a satellite dish 320 receives the multiplexed output of the inserter 310, demodulates it at a demodulator 322, demultiplexes it at a demultiplexer 324 and provides it as a data stream to an EDC correct/control data insertion device 326. Each of the 16 demultiplexed data streams is error corrected by the device 326 and provided to a QPSK broadcast modulator 328, such as been hereinbefore described. A computerized billing system 330 exercises control over a radio controller 332, which is comparable to the address module 26 of FIG. 1 for permitting/restricting access to program material by subscribers. The address information from the radio controller 332 is inserted by the device 326 into the data streams.

As shown in FIG. 7, 16 individual outputs 334 of the device 326, each corresponding to a channel of source material, is provided to a QPSK modulator (one, 328 shown), and combined by a combiner 336 onto a transmission line 338 for distribution to subscribers (one shown) having an appropriate terminal 340.

Another beneficial feature of this system 300 is that locally (versus remotely, by satellite) originated audio source material 342, such as for simulcast with a television program, may be combined by a combiner 344 onto the transmission line. This would be achieved by digitizing the source material 342 with a digitizer 346, for each of a plurality of channels 348, providing the digitized source material to an inserter similar to the inserter 310 (but not requiring the EDC insertion function), and QPSK modulating the combined output thereof with a modulator 350 for broadcast on the transmission line 338. Although not shown, video channels could also be combined for broadcast over the line 338.

FIG. 8 shows a digital FM receiver 400, similar in many respects to that shown and described with respect to FIGS. 3A and 3B. As will become evident, however, a notable difference is that the receiver 400 of FIG. 8 is suitable for receiving both QPSK digitized and nondigitized FM signals.

Signals received on an antenna 402 are provided to a tunable RF bandpass filter 404, to a variable gain amplifier 406, and to a tunable RF bandpass filter 408. The output of the RF bandpass filter 408 is provided to a mixer 410, which receives a second input from a variable frequency oscillator 412, so that the output of the mixer 410 is at an intermediate frequency (IF) for a selected channel. The channel selection process is under control of a tuning synthesizer 414 which receives the output of the variable frequency oscillator 412, and provides a signal based on the output of the oscillator 412 to effect channels selection by the RF bandpass filters 404 and 408. The output of the mixer 410 is provided to an intermediate frequency (IF) bandpass filter 416, tuned to 10.7MHz, to an amplifier 418, and to a second IF bandpass filter 420 tuned to 10.7MHz. The IF bandpass filters 416 and 420 are preferably wide type ceramic filters. The output of the second IF filter is the signals received on the antenna 402 corresponding to a selected channel in the FM band. A dotted line 422 encloses the elements 404, 406, 408, 410, 412, 416, 418 and 420, such as would be found in a standard FM tuner.

The output of the second IF bandpass filter 420 is provided to both a QPSK demodulator 430 and to an FM detector 432. In one signal path, the FM detector detects the audio component of the incoming signal and provides such as an AF signal to a stereo demultiplexer 434 (for stereo broadcasts), the output of which is provided to a digital or analog switch 436 as left and right audio channel signals. The FM detector 432 also provides a signal to the variable gain amplifier 406 to automatically control the gain thereof in accordance with known automatic gain control (AGC) techniques.

In another signal path, the QPSK demodulator provides a bit stream to a logic array 438, when there is a digital signal received in the selected channel. An indication of the existence of such a digital signal, indicative of a digitized audio broadcast being received, is provided as a logic signal to the digital analog switch 436. The output of the logic array 438 is provided to a Dolby ADM decoder 440, the output of which is provided to the digital analog switch 436 as left and right audio channel signals (for stereo broadcasts).

The analog/digital switch 436 selects between the outputs of the Dolby ADM decoder 440 and the stereo demultiplexer 434, under control of the logic array 438, and provides left and right audio signals from one or the other to audio amplifiers 442 and 444.

The advantages of the invention are multifaceted. As mentioned hereinbefore, the digitized audio channels can be interspersed between nondigitized audio channels, each occupying 400KHz in the FM band. The Federal Communications Commission (FCC) requires at least 800KHz between standard FM channels in a market, which translates into only 25 stations in the 88-108MHz band. However, since digitized audio channels may be interspersed between standard (nondigitized) channels, up to 50 channels (stations) could be provided in the standard FM, 20MHz wide band. This is highly pertinent to both airwave and cable transmission. Due to the availability of, in essence, twice the number of stations, there is plenty of room created for original local stations, basic premium stations and "pay-per-listen" stations, all, in marked contrast to the aforementioned U.S. Pat. No. 4,684,981, without usurping a TV video channel.

Another advantage is that most of the degradation in a standard FM signal occurs within the cable network (transmission line) itself. This problem is overcome by the use of digitized audio channels as one-to-one replacements for the standard audio channels. The ultimate result is that listeners will be able to receive audio broadcasts that are more in line with digital disc (CD) recordings which are becoming so popular. Further, the possibility of providing high quality audio via cable may add a new impetus to the radio industry.