Audio and multimedia systems often include audio input ports for connecting external audio sources. These audio input ports may include analog input ports with two-wire signal lines, which may include tip, ring, sleeve (TRS) sockets or 3.5 mini jacks, or digital input ports with three or more signal lines.
Integrated Interchip Sound (IIS) is one example of a digital interface standard for connecting audio devices. IIS requires at least three different signal lines. These signal lines include a bit clock line, a left-right clock line to indicate left or right channel audio data, and a multiplexed data line containing the left and right channel audio data. Additional multiplexed signal data lines and a master clock line may also be included in different implementations, further adding to the number of signal lines used to transmit audio. The master clock line may transmit a master clock signal at a higher frequency than the bit clock signal. The master clock signal may be used by a digital signal processor to process the audio data.
As the number of signal lines used to transmit audio increases, production costs and power consumption both increase. Production costs increase because it becomes more expensive to manufacture circuit boards, connectors, and wires to support additional signal lines. For example, not only is there an added materials cost for including the additional pin outs and signal lines on the circuit board, but there is also increased power consumption cost because each added signal line consumes additional power. This increased power consumption may reduce the battery life of portable audio devices between charges, requiring the portable device to be charged more frequently.
Thus, there is a need for a digital audio transmission interface using a minimal number of signal lines to transmit audio.
A digital audio interface in an embodiment of the invention may include two signal inputs to carry audio data. A first signal input may be connected to a signal line carrying digital serial audio data. The second signal input may be connected to a signal line carrying a word clock signal to differentiate the serial audio data transmitted over the first signal line. In the case of a stereo audio data, the word clock signal may differentiate, for example, audio data intended for a right channel from the audio data intended for a left channel. In this case, the word clock signal may correspond to a left-right clock signal. In other embodiments, the audio data may be differentiated differently depending on the configuration, such as in the case that the transmitted audio data include audio for more than two channels.
A bit clock may time the transmission of the digital serial audio data bits so that serial audio data is transmitted at a rate corresponding to the bit clock rate. Since the word clock signal may differentiate a subset of serial audio data bits, such as those bits corresponding to a particular audio channel, the word clock signal frequency may be lower than the bit clock frequency. In an embodiment where the word clock signal differentiates left and right channel audio data in the transmitted serial audio data, the frequency of the word clock signal may range from about 8 kHz to about 192 kHz. The frequency of the bit clock signal may, however, be much higher from about 3 MHz to about 12 MHz.
Since only the word clock signal and serial audio data may be respectively transmitted over each of the two signal lines in an embodiment, the bit clock signal, and other signals such as a master clock signal need not be transmitted. Instead these other clock signals may be derived from the transmitted word clock signal at a receiver to enable the receiver to properly decode the transmitted serial audio data. Because the bit clock signals and other higher frequency clock signals requiring additional power for transmission, such as the master clock signal, may no longer be transmitted, overall power consumption may be reduced.
As discussed previously, the received word clock signal may be scaled at a receiver to regenerate the bit clock signal. Since the word clock signal may associate different audio bits with different channels, the word clock signal may be synchronized to the untransmitted bit clock signal. Accordingly, as long as the phase of the scaled word clock signal is matched to that of the word clock signal, the scaled word clock signal may be synchronized to the original bit clock signal that encoded the transmitted audio bits. One or more phase lock loops (PLL) may adjust the phase of the scaled word clock signal to match the original word clock signal. The PLLs may include an analog PLL or a hybrid digital-analog PLL, which may include a digital PLL coupled to an analog PLL.
The word clock signal input 102 may also be supplied as an input to a signal scaler 120 that scales the word clock signal input 102 to regenerate the bit clock signal 121. The signal scaler 120 may also scale the word clock signal input 102 to one or more additional frequencies depending on the application. For example, the signal scaler 120 may scale the word clock signal 102 to a higher master clock signal frequency 122. This higher master clock signal frequency 122 may be sent to a functional module 130, such as a digital signal processor, to oversample the decoded audio data during signal processing. The signal scaler 120 may also scale the word clock signal input 102 to other frequencies to accommodate other function modules 130.
The signal scaler 120 may include a PLL to adjust the regenerated bit clock signal 121 so that the phase of the regenerated bit clock signal 121 matches the phase of the word clock signal 102. The PLL may include an analog PLL or a hybrid digital-analog PLL having a digital PLL coupled to an analog PLL. The signal scaler 120 may also include one or more frequency multiplier, dividiers, and/or fractional-n synthesizers to scale the word clock signal 102 to regenerate the bit clock signal 121. One or more parameters of the frequency multipliers, dividers, and/or fractional-n synthesizers may be adjustable to accommodate different scaling factors in different applications and embodiments.
Once the frequency of the word clock signal 102 has been multiplied at multiplier 211, the multiplied frequency may be sent to a coupled analog PLL 220. The analog PLL may include a frequency divider 221, frequency multiplier (not shown), and/or a fractional-n synthesizer 222. The parameters of these frequency dividers, multipliers, and/or fractional-n synthesizers may be programmable to accommodate different desired scaled output frequencies. For example, a divisor of the frequency divider 221, and the parameters R, M, N, resulting in an average frequency multiplier of (R+N/M) for the fractional-n synthesizer 222, may be programmable.
The frequency dividers, multipliers, and/or fractional-n synthesizers may generate one or more scaled frequencies depending on the particular application. For example, as shown in
In box 301, the transmitted word clock signal may be scaled to regenerate a clock signal that encoded the serial audio data. The scaling factor that regenerates the clock signal may be preprogrammed if the encoding clock signal is known and fixed. If the encoding clock signal varies, the parameters of the scaler may be reprogrammed to regenerate a matching clock signal. The reprogramming may occur by providing a clock signal identifier to the scaler to enable the scaler to set its parameters to generate the matching clock signal.
As discussed previously, a phase of the word clock signal may be matched to a phase of the clock signal that encoded the serial audio data. This is because the word clock signal may differentiate different audio bits in the serial audio data stream, so the start of a word clock signal cycle may coincide with the start of a bit cycle representing a first audio bit to be differentiated. Similarly, the end of a word clock signal cycle may coincide with the end of a bit cycle representing a last audio bit to be differentiated.
Thus, as long as the phase of the scaled word clock signal matches a phase of the original word clock signal, the regenerated clock signal may be synchronized to the phase of the original clock signal. A phase lock loop may adjust phase synchronization between the scaled word clock signal and the original word clock signal to maintain synchronization.
In box 302, the scaled word clock signal may assist in decode the serial audio data. Since the word clock signal may be scaled to regenerate the encoding clock signal, the regenerated clock signal may also assist in decoding the serial audio data by identifying audio bits and bit transitions.
In box 303, the decoding of the serial audio data may also include using the original word clock signal to identified the set of the digital audio bits to be differentiated. For example, if the word clock signal differentiates audio bits intended for a left and right audio channels, then the word clock signal may also be used during the decoding to identify those decoded audio bits intended for the left channel and those intended for the right channel.
In box 401, the serial audio data may be encoded and transmitted over the first signal line using a clock signal to encode the audio data. The encoding clock signal need not be transmitted. The serial audio data may include different set of audio data, such as multiple channels of audio data. For example, the serial audio data may include left channel audio data and right channel audio data.
A word clock signal may identify different audio data bits, such as audio data intended for the left and right channels, encoded in the serial audio data. Because the word clock signal may identify particular audio bits, a phase of the word clock signal may match a phase of the audio bit encoding clock signal. This phase matching may enable the word clock signal to identify the start and end of the encoded bits representing a particular set of audio data, such as left or right channel data.
In box 402, the word clock signal may be transmitted over the second signal line. In box 403, the word clock signal may be scaled to regenerate the higher frequency encoding clock signal. As discussed previously, the scaling parameters regenerating the encoding clock signal may be fixed in those instances where the encoding clock signal is fixed and may be variable in those instances where the encoding clock signal frequency may vary. An encoding signal identifier, which may identify the encoding clock signal, or may include one or more scaler parameters, may select the particular encoding clock signal used.
Then, at each rising edge of the recalculated and phase matched bit clock signal 530, the serial audio data signal 520 may be sampled to identify and decode the audio bits transmitted over the serial audio data signal 520. In this case, the decoded audio bits 540 may be 110100 (0 for low state and 1 for high state). The word clock signal 510 may also be sample at each rising edge of the recalculated bit clock signal 530 to identify a channel associated with the respective decoded audio bit 540, such as a first channel corresponding to a low signal state and a second channel corresponding to a high signal state. Other variations may be used in other embodiments.
The foregoing description has been presented for purposes of illustration and description. It is not exhaustive and does not limit embodiments of the invention to the precise forms disclosed. Modifications and variations are possible in light of the above teachings or may be acquired from the practicing embodiments consistent with the invention. For example, some of the described embodiments may refer to scalers having digital and analog phase lock loops, but other scalers may include only analog phase lock loops. Similarly, different scaler configurations may include different combinations of one or more frequency multipliers, frequency dividers, fractional-n synthesizers, delta-sigma synthesizers, and/or other scaling circuits.
This application relates to co-pending application “Hybrid Digital-analog Phase Locked Loops,” filed as U.S. patent application Ser. No. 13/155,561, also filed on Jun. 8, 2011, which is incorporated herein by reference in its entirety.