A serial transmission protocol and architecture are provided that can be used to forward the raw remote control data in modulated form, between sink (e.g., DTV) and source (e.g., DVD) devices, in addition to other data such as video/audio/control data. The target device need not be in the line of sight of the remote control beam. Daisy-chained devices will pass the remote control signal to the appropriate target device, so that the user can point any remote control at the DTV or other conveniently located device in the system, and still control the actual target device.
General Overview
One embodiment of the present invention is a protocol that enables a very high bandwidth (e.g., 1.5 Gbps (gigabits per second) or higher) in one direction for payload data such as video/audio. The transmission path in this direction is called a forward channel. The protocol also enables a relatively low speed communication link in the opposite direction for carrying processing control data, as well as remote control data. This path is called a backward channel. The protocol can be configured to support all the available formats, and is very flexible, so that adoption of new formats is facilitated. Control and timing signals can also be carried by the forward channel from the source device to the sink device for correct video/audio signal regeneration.
The protocol is herein referred to as the Asymmetrical Serial Multimedia Interface (ASMI) protocol, since it has a high bandwidth requirement in one direction and a low bandwidth requirement in the other. The protocol can be transmitted, for example, via a digital optical fiber link, digital optical wireless link, or a radio frequency (RF) link. Similarly, a combination of such links can be used, such as a high speed optical wireless link for the forward channel and a relatively slower speed RF link for the backward channel. In another particular embodiment, a single optical fiber is used to implement both the forward and backward channels.
The protocol can be used in conjunction with a number of system configurations. In one particular embodiment, a multimedia system configured with a daisy-chain topology is enabled. The system is capable of transmitting uncompressed multimedia data, and includes multiple sources and/or sink devices. The multimedia system can be configured with an automatic and dynamic addressing assignment scheme that allows a user to rewire source and/or sink device cables, wherein all addressing for all networked devices automatically updates to maintain reliable communication. Other system configurations, such as point-to-point and media center configurations, can also be implemented with the protocol.
System Architecture
As can be seen, the system includes a source device 101 and a sink device 109 communicatively coupled via an asymmetric serial multimedia interface (ASMI) link 100. The source device 101 can be, for example, a DVD player, satellite/cable receiver box or set-top box, computer, or other suitable multimedia source. The source device 101 generates multimedia data and sends it via the ASMI link 100, which in this example is an optical fiber. The sink device 109 can be, for example, a digital television, monitor, projector or other suitable display device. The sink device 109 receives multimedia data via the ASMI link 100 and displays that data.
In the example embodiment shown in
Likewise, the sink device 109 includes an ASMI section 111 configured with a remote control process 11a, a video/audio processing section 113, and a control data processing section 115. The video/audio processing section 113 can be in the same device as the ASMI section 111 (as shown in
In operation, the video/audio processing section 103, which can be implemented with conventional technology (e.g., such as that included in a DVD player, satellite/cable receiver box or set-top box, or computer) generates multimedia data (e.g., audio and video), and provides that data to the ASMI section 107. In addition, the control data processing section 105, which can be implemented with conventional technology (e.g., such as that included in a DVD player, satellite/cable receiver box or set-top box, or computer) generates control data, and provides that data to the ASMI section 107. The control data processing section 105 may be further configured to provide additional control functions in accordance with an embodiment of the present invention, as will be apparent in light of this disclosure. These additional control functions (such as those performed by the remote control process 107a) can be implemented within the control data processing section 105, or in a separate control processor module that supplements conventional control functions (as shown in
The ASMI section 111 demultiplexes the received serial stream back into its multimedia (video and audio) and control data components. The video and audio data are provided to the video/audio processing section 113, which can be implemented with conventional technology (e.g., such as that included in a digital TV, monitor, or projector). The video/audio processing section 113 processes that multimedia data so that it can be displayed/sounded to the user. In addition, the received control data is provided to the control data processing section 115, which can be implemented with conventional technology (e.g., such as that included in a digital television, monitor, projector or other suitable display device). The control data processing section 115 may be further configured in accordance to provide additional control functions in accordance with an embodiment of the present invention, as will be apparent in light of this disclosure. These additional control functions (such as those performed by the remote control process 111a) can be implemented within the control data processing section 115 of the sink device 109, or in a separate control processor module that supplements conventional control functions (as shown in
Control data, including remote control data, can also be provided from the sink device 109 to the source device via the backward channel. In particular, control data processing section 115 provides control data to the ASMI section 111, which multiplexes that control data for transmission to the ASMI section 107 (of the source device 101). In addition, the remote control process 111a operates to detect incoming IR remote control data and convert it into electrical signals that can also be multiplexed for transmission to the ASMI section 107. Again, note that remote control process 111a can be integrated into the control data processing 115 (or some other dedicated module) if so desired. In any case, ASMI section 107 then receives and demultiplexes the control data transmitted by the ASMI 111 using the backward channel of the ASMI link 100. That control data is then provided to the control data processing section 105. In addition, the remote control data included in the received control data is regenerated by the remote control process 107a (so that it can be used to control the local source device 101). The ASMI section 107 will be discussed in more detail with reference to
The ASMI link 100 can be implemented using conventional or custom technology (e.g., optical, RF, or combination), and can be wired or wireless or a combination of the two if so desired. In one particular embodiment, the forward and backward channels of the ASMI link 100 are implemented using a pair of optical transceivers communicatively coupled by a single fiber as described in the U.S. patent application Ser. No. 11/173,409. This application Ser. No. 11/173,409, filed on Jun. 30, 2005 and titled, “Bidirectional HDCP Transmission Module Using Single Optical Fiber,” is herein incorporated in its entirety by reference. In such an embodiment, the ASMI section 107 could include or otherwise be coupled with a transceiver module (e.g., VCSEL and photodetector) for implementing the forward channel transmitter and the backward channel receiver. Also, the ASMI section 111 could include or otherwise be coupled with a transceiver module (e.g., LED and PIN detector) for implementing the backward channel transmitter and the forward channel receiver. Numerous wired optical transceiver configurations can be used to effect communication between the devices 101 and 109.
In another particular embodiment, the forward and backward channels of the ASMI link 100 are implemented using an optical wireless communication channel as described in the U.S. patent application Ser. No. 11/142,882. This application Ser. No. 11/142,882, filed on May 31, 2005 and titled, “High Speed Free Space Optical Detection with Grating Assisted Waveguide,” is herein incorporated in its entirety by reference. In such an embodiment, the ASMI section 107 could include or otherwise be coupled with a transceiver module (e.g., transmitter with grating assisted receiver) for implementing the forward channel transmitter and the backward channel receiver. Also, the ASMI section 111 could include or otherwise be coupled with a transceiver module (e.g., transmitter with grating assisted receiver) for implementing the backward channel transmitter and the forward channel receiver. In such a case, the ASMI link 100 is actually a forward communication link and a backward communication link. Numerous wireless optical transceiver configurations can be used to effect communication between the devices 101 and 109.
In another such embodiment, the forward channel can be implemented using any wireless optical link (such as a VCSEL and PIN detector), and the backward channel can be implemented with a relatively slower RF link (e.g., such as an IEEE 802.11 or other such RF wireless communication link).
ASMI for Source Device
This example ASMI 107 has a downstream port and an optional upstream port. The downstream port includes a high speed forward channel transmission port and a low speed backward channel receiving port. The optional upstream port, which includes a low speed backward channel transmission port and a high speed forward channel receiving port, can receive video/audio signals from an upstream source device in a daisy-chain configuration, as will be discussed in turn.
In operation, the ASMI 107 receives video and audio data from a local video/audio source device (e.g., video/audio processing section 103, such as that included in a DVD player or other such source). The video and audio data is saved into an elastic buffer referred to in
If there is an upstream port, the forward channel demultiplexer (FC-DMX) 201 sends received control data, including any remote control data, to a local control processor 213 included in the remote control process 107a. Recall that processor 213 can be included in the ASMI 107 architecture, or otherwise communicatively coupled with the ASMI 107 (as shown in
The ASMI 107 also has a backward channel demultiplexer (BC-DMX) 209 that demultiplexes control data (including any remote control data) received via the downstream backward channel, so that control data can then be provided to the local control processor 213. The control data from the upstream high speed forward channel (extracted by the FC-DMX 201) and from the downstream low speed backward channel (extracted by the BC-DMX 209) are sent to the local control processor 213 for processing. Any remote control data destined for the local device is converted back into its IR form by operation of the control processor 213 and IR emitter 215. The control processor 213 also transmits the control data to the downstream device via the FC_MUX 205, and to the upstream device (if there is one) via a backward channel multiplexer (BC-MUX) 207, which multiplexes the control data for transmission via the upstream backward channel. The remote control process 107a (including IR emitter 215 and control processor 213) is discussed in more detail with reference to
Communication between the local control processor 213 (and/or the control data processing section 105) and the ASMI 107 can be carried out using custom or conventional technology, such as USB, Ethernet, or universal asynchronous receiver transmitter (UART).
A SERDES (Serializer/Deserializer) block 227 is coupled between the upstream port and the FC-DMX 201 and the BC-MUX 207. Also, SERDES block 229 is coupled between the downstream port and the FC-MUX 205 and the BC-DMX 209. The SERDES blocks 227, 229 are transceiver blocks that convert parallel data to serial data for communication on the ASMI link 100 in serial format, and vice versa. The transmitter section of the SERDES is a serial-to-parallel converter, and the receiver section is a parallel-to-serial converter. Although the SERDES block 227 is shown to be separate from the FC-DMX 201 and BC-MUX 207 in
ASMI for Sink Device
This example ASMI 111 has an upstream port and an optional downstream port. The upstream port includes a high speed forward channel receiving port and a low speed backward channel transmission port. The optional downstream port, which includes a low speed backward channel receiving port and a high speed forward channel transmission port, can provide video/audio signals to a downstream sink device in a daisy-chain configuration, as will be discussed below.
In operation, the ASMI 111 receives video, audio, and control data from the high speed forward channel receiving port. The FC-DMX 201 sends received control data to a local control processor 217. Recall that processor 217 can be included in the ASMI 111 architecture, or otherwise communicatively coupled with the ASMI 111 (as shown in
The control data from the upstream high speed forward channel (extracted by the FC-DMX 201) and from the downstream low speed backward channel (extracted by the BC-DMX 209) are sent to the local control processor 217 for processing. The control processor 217 also transmits the control data to the downstream device (if there is one) via the FC-MUX 205, and to the upstream device via a BC-MUX 207, which multiplexes the control data for transmission via the upstream backward channel. In addition, IR detector 219 is adapted to receive incoming IR remote control signals (provided by a user pointing a remote control at the local sink device), and to convert those IR signals into electrical signals. The control processor 217 then transmits that remote control data (in electrical signal form) to the downstream device (if there is one) via the FC-MUX 205, and to the upstream device via a BC-MUX 207, which multiplexes the control data for transmission via the upstream backward channel. Note that the processing of control information can be implemented the same way as that of source device. The remote control process 111a (including IR detector 219 and control processor 217) is discussed in more detail with reference to
Communication between the local control processor 217 (and/or the control data processing section 115) and the ASMI 111 can be carried out using custom or conventional technology, such as USB, Ethernet, universal asynchronous receiver transmitter (UART), or the like.
Each of FC-MUX 205 and BC-MUX 207 can be programmed or otherwise configured to provide simple multiplexing. Likewise, each of FC-DMX 201 and BC-DMX 209 can be programmed or otherwise configured to provide complementary demultiplexing. In one particular embodiment, the multiplexers (FC-MUX 205 and BC-MUX 207) are implemented as a multiplex state machine, and the demultiplexers (FC-DMX 201 and BC-DMX 209) are implemented as a demultiplex state machine. Each of these MUX/DMX state machines will be discussed in turn. Other techniques for serializing data for transmission, and then deserializing that data for receiver processing can be used here, as will be apparent in light of this disclosure.
SERDES (Serializer/Deserializer) block 227 is coupled between the upstream port and the FC-DMX 201 and the BC-MUX 207. Also, SERDES block 229 is coupled between the downstream port and the FC-MUX 205 and the BC-DMX 209. The SERDES blocks 227, 229 are transceiver blocks that convert parallel data to serial data for communication on the ASMI link 100 in serial format, and vice versa. The transmitter section of the SERDES is a serial-to-parallel converter, and the receiver section is a parallel-to-serial converter. Although the SERDES block 227 is shown to be separate from the FC-DMX 201 and BC-MUX 207 in
Automatic/Dynamic Addressing Assignment
As previously discussed, the multimedia system can be configured with an automatic and dynamic addressing assignment scheme that allows a user to rewire source and/or sink device cables, wherein all addressing for all networked devices automatically updates to maintain reliable communication. In more detail, each source and sink device is assigned with an address. This will allow control information (including remote control data) to be passed among all source devices and sink devices.
The address assignment mechanism in accordance with one embodiment of the present invention is designed to satisfy the following requirements: (1) the assignment is fully automatic, so no switch and other user interface is needed; (2) the assignment is dynamic, such that the address is updated to reflect any network cabling change by the user; and (3) the address and its assignment scheme are transparent to the user. The address is based on the position and ordering of each device in the daisy-chain connection.
In one particular embodiment, the address assignment mechanism is implemented using the forward channel data format. In particular, the forward channel format includes a field (e.g., one byte) called upstream device address. In one such case, the address has the following 8 bit format:
This upstream device address field makes it possible for the downstream device to update its address whenever the user changes the cable configuration. For example, if the upstream port address received by a downstream device has changed (e.g., because the user has added a new piece of equipment, such as a second DVD player), then the local address assignment for that downstream device is updated so that its address is one plus the address of the received upstream port address. Detecting a change in upstream port address can be carried out, for example, by the local control processor 213 or 217 (or other local computing function). The control processor 213 or 217 can then compute the new address for the device.
With the type of conventional IR detector in
As shown in
Note that the bandwidth of the amplifiers 274, 268 should be at least twice the maximum frequency of the IR signal 250. However, since noise generally increases with electrical bandwidth, the IR detector 219 preferably can have just enough bandwidth to receive the highest desired signal frequency to obtain maximum signal reception sensitivity. Optical remote controls send digital data by amplitude modulation of a digital carrier signal, typically in the range of 20-56 KHz with some newer ones proposing use up to 100 KHz. Thus, in one embodiment, the amplifiers 274, 268 have bandwidths of 200 KHz. The AC coupling capacitors 269, 271 ensure that the amplified offset voltages do not saturate the output voltage of the final stage, and their capacitance values are chosen to ensure that the lowest desired frequency of 20 KHz is transmitted.
With the IR detector 219 in
In order to regenerate the IR signal 250 based on the samples of the IR detector output 254 received at the source device 101, clock synchronization between the source device 101 and the sink device 109 is needed to prevent IR data packet FIFO over-run or under-run. Clock synchronization is provided by conventional SERDES blocks 227, 229 in the ASMI 107 and ASMI 111. SERDES blocks are typically capable of providing the backward channel speed necessary for the sampling rate while synchronizing the clocks between the source device 101 and the sink device 109. Thus, the sampling by the control processor 217 can be configured to sample the IR signal 250 based on the back channel SERDES 227 clock of the ASMI 111 (sink device) for transmission to the source device 101 over the ASMI link 100, and the IR signal 250 can be regenerated on the source device 101 based on the recovered DESERDES 229 clock of the ASMI 107 (source device). This is possible because SERDES designs typically include clock and data recovery circuits allowing a receiver to extract embedded clock information and re-timed data from an incoming encoded data stream. The extracted DESERDES clock is synchronized with the transmitter clock, making the IR regeneration operate at the same rate as that of the IR transmitters and preventing over-run or under-run problems.
Point-to-Point Configuration
As can be seen, the sink ASMI 111 is connected to the source ASMI 107 via the ASMI link (optical fiber) 100. The ASMI 107 receives multimedia data (video/audio) and control data from a source device (e.g., DVD player that includes video/audio processing section 103 and control data processing section 105 with integrated control processor 213), multiplexes that multimedia data and control data into a serial data stream, and transmits that data stream to the sink ASMI 111 via its downstream port and the ASMI link 100. The sink ASMI 111 receives the serial data stream at its upstream port, and demultiplexes it into separate video/audio streams for display/sounding and control data processing by the sink device (e.g., digital TV that includes video/audio processing section 113 and control data processing section 115 with integrated control processor 217). The previous discussions with reference to
In this example configuration, external video circuitry is used. However, it will be appreciated in light of this disclosure that the ASMI 107 can be integrated into a source device and ASMI 111 can be integrated into a sink device (as shown in
Daisy-Chain Configuration
As can be seen, each ASMI 107 and 111 in the daisy-chain receives control data from at least one of two sources: from the upstream device via its high speed forward channel receive port, and from the downstream device via its low speed backward channel receive port. The source ASMI #1 does not have an upstream device, and the sink ASMI #M does not have a downstream device. All the control data received at anyone ASMI is sent to the local processor (e.g., control processor 213/217) for processing, and is not relayed to the downstream device directly. If the control data traffic is not targeted to the local device, the local control processor 213/217 will transmit that traffic to other devices via either the backward channel or the control data field in the forward channel data stream, which will be discussed in turn.
In addition, each ASMI 107 and 111 transmits control data to at least one of two places: to the downstream device via its high speed forward channel transmit port, and to the upstream device via its low speed backward channel transmit port. The addressing and processing of the control data can be carried out, for instance, by the local control processor 213/217 and its associated protocols (which may be proprietary or conventional). The ASMI mechanism described in this example system design passes the control data to and from the local processor, and transmits it via the high speed forward channel and low speed backward channel.
In one embodiment, IR remote control data captured at one of the sink devices and converted into electrical signal data is communicated to one of the source devices in sampled, modulated form, using the backward channel of the ASMI link 100. In a similar fashion, remote control data captured at one of the source devices can be communicated to one of the sink devices using the forward channel of the ASMI link 100 in sampled, modulated form (although this would not be typical, given user propensity to aim the remote control at a sink device such as a DTV, as opposed to a source device, particularly if that source device is out of the user's forward field of view). In any case, the IR remote signal can then be regenerated and provided to control the target device accordingly.
In a multi-source, multi-sink configuration, there is typically more than one multimedia stream. In order to support this operation, a stream number is used. In more detail, and in accordance with one particular embodiment, each multimedia data stream in the forward channel of the ASMI link 100 has a stream number field. The local control processor 213 of each source device programs or otherwise assigns the stream number. The ASMI 107 of the source device transmits its local multimedia data with this assigned stream number in its header, as will be explained in turn. Similarly, the local control processor 217 of each sink device also programs a stream number. Processor 217 sends any received multimedia data stream including a stream number that matches this local stream number to the local device for display/sounding by operation of the ASMI 111. If the stream number does not match, then processor 217 relays the data stream to the downstream device.
The stream number mechanism facilitates user interface for a stream selection process. In particular, and in accordance with an embodiment of the present invention, each source ASMI 107 transmits with its own address as the stream number. In other words, ASMI #1 will transmit multimedia stream number one, and ASMI #2 will transmit multimedia stream number two, and so on. This can be a default stream number assignment scheme if so desired. Each sink ASMI 111 (e.g., in conjunction with its local control processor 217) will detect the total number of active streams and display them to the user (e.g., via a pull down list or other suitable user interface mechanism). The user can then select the desired video stream by selecting one stream from the displayed list. The user selected stream number can be programmed into one or more ASMIs 111, so that each programmed ASMI 111 will know to deliver stream data having that stream number to the local video/audio circuitry. Any number of known user interface techniques and menu schemes can be used to allow the user to access stream information available from the local control processor 217 of the sink device.
In one particular embodiment, for a source ASMI 107, the local data is first stored in the local buffer 211. If there is an upstream port, the data received from that port can be saved in the relay buffer 203. The merge of the local multimedia data and the data from the upstream device can be implemented, for example, with a simple multiplexing (e.g., FC-MUX 205 and BC-MUX 207), as previously explained with reference to
Just as with the example configuration shown in
The transmitted multimedia data (e.g., uncompressed video/audio or any other data communicated by a remote controlled multi-component system) will be associated with a bit rate (e.g., constant bit rate associated with uncompressed multimedia, or variable bit rate data). In any such cases, assume the available bandwidth for the ASMI link 100 and architecture is greater than or equal to the desired highest multimedia data rate to which the system will employ.
Media Center Configuration
The source ports of the media center 305 receive multimedia data from a corresponding source device 101 via a forward channel of an ASMI link 100. The ASMI 111 for that port of the media center 310 effectively operates as a forwarding state machine, by routing the received data to the proper sink port. Likewise, the sink ports of the media center 305 receive control data from a corresponding sink device 109 via a backward channel of an ASMI link 100. As will be appreciated in light of this disclosure, the control data may include a representation of IR remote control data captured at the sink device 109. The ASMI 107 for that port of the media center 310 effectively operates as a forwarding state machine, by routing the received control data to the proper source port. In one particular embodiment, each ASMI 111 and ASMI 107 operates under the control of one or more control processors (e.g., processors 213 and 217, generally referred to a “control processor” for this example embodiment) of the media center 305. As discussed herein, the control processor can evaluate address information and/or stream information included in received data packets to effect proper routing of data to its intended destination.
Data Format
In one example embodiment, the multimedia and control data in the ASMI link 100 is transmitted in packets. Each packet covers a fixed number of video pixel clocks. The packet rate on the serial interface is: Pixel clock rate/Packet size (in pixels). Based on this fixed packet size scheme, the sink-side can readily regenerate all the multimedia data and control signals.
Serial Clock Interface
The clock rate for the high speed forward channel of the ASMI link 100 is normally independent of the video clock rate. There may be more than one video/audio source, each with different clocks. Even for one video stream, the video clock can still be different for different video modes (e.g., 480p, 720p, 1080i, etc). In one particular embodiment, the high speed forward channel of the ASMI link 100 runs at a constant clock. The source device (or first source device in the daisy-chain configuration) generates this clock. All downstream devices recover this clock and use it for ASMI operation, including traffic receiving and relaying. In one embodiment, the control processor 217 (or a dedicated module included in ASMI circuit 111) is programmed or otherwise configured to implement known clock recovery techniques.
Multiple Streams
As previously discussed, a multimedia system can be designed to support multiple video/audio streams, in accordance with an embodiment of the present invention. This is useful, for example, for multi-source and/or multi-sink configurations, and for picture in picture applications. In one particular embodiment, there is an 8 bit stream number in the header of each multimedia (e.g., video/audio) data packet to indicate to which multimedia stream this packet belongs. Based on a user's stream selection, the control processor 217 programs a target stream number in the ASMI 111 of each sink device. In addition, each ASMI 107 will transmit all multimedia data from its source device with its corresponding stream number. The downstream sink device will compare the received stream number with its programmed stream number. If the two match, the stream is delivered to the local video circuit (e.g., video/audio processing section 113, such as that in a digital TV) for display/sounding. Otherwise, the packet is relayed to the next down stream port for like processing, until it is received at its intended destination.
Time Stamp Scheme
Videos are displayed in pixels, lines, and frames. Uncompressed video/audio data must be exactly synchronized with timing control signals like HSYNC (Horizontal Synchronization) and VSYNC (Vertical synchronization). In an HDMI/DVI system, there are also CTL control signals to synchronize HDCP and digital audio application. These timing control signals change their state at a specific pixel clock. The status of the signals and the time they are changing are carried over the high speed ASMI link 100 for the sink device to recreate the original video/audio.
In one particular embodiment of the present invention, each of the source devices includes or is otherwise operatively coupled to a circuit configured to record all the control signal changes (e.g., such as a dedicated circuit, control processor 213, or a microcontroller with I/O ports for receiving control signals and port routines to log signal changes). After each signal change, the new control signal values and the relative time at which the change takes place are recorded. These recorded timing data are generally referred to as time stamps. In one such case, all the time stamps are relative to the first pixel time of the packet (or some other consistent time stamp point). In addition, all the time stamps and timing control signal values are transmitted to the sink device together with the multimedia data.
At the sink-side, the video regeneration circuit (e.g., video/audio processing section 113, such as that in a digital TV) starts pixel counting from the first pixel for each packet data. When the count reaches each time stamp, the control signals are generated according to the value in the packet. In this way, all the control timing signals are regenerated at exactly the same time as they are recorded at the source-side. Note that the number of time stamps per packet time is not constant. Thus, a counter can be used to count the number of time stamps for each packet. In one such embodiment, this counter is transmitted from the source device ASMI 107 to the sink device ASMI 111, together with all the time stamps.
Blank Suppression Scheme
Typical video data includes active video pixels and about 15%-20% blank pixels. These blank pixels do not carry useful information and consume valuable bandwidth if being carried over a constrained link. As described in U.S. patent application Ser. No. 10/864,755, these blank pixels can be suppressed to save serial link bandwidth. In such an embodiment, the ASMI link will transmit active pixel data only. Application Ser. No. 10/864,755, filed on Jun. 8, 2004, and titled “Scheme For Transmitting Video and Audio Data of Variable Formats Over a Serial Link of a Fixed Data Rate,” is herein incorporated in its entirety by reference. In one such embodiment, the user can select to enable/disable blank pixel suppression for their particular multimedia communication system.
Forward Channel Data Format
In one embodiment of the present invention, the forward channel data transmitted on the ASMI link 100 is in the form of packets. One packet carries the active video data for the pre-specified number of video clocks. All fields in this example format are in the byte boundary, except the number of audio words header field and the number of audio channels header field, which can be combined into one byte. This facilitates transmitting the data using a standard serialize/deserialize (SerDes) interface and with a standard 8B/10B encoding scheme.
Each forward channel packet has the following format shown in Table 1, in accordance with one particular embodiment:
The Packet Sequence Number is used for each device to synchronize with each other. It is incremented by one for each packet transmitted at the source. The number is wrapped back to zero when its maximum value is reached. Each downstream device should see a constantly incrementing number for this field if no packet is lost.
Video Stream Number is used for multi-video stream modes. It indicates to which video stream this packet belongs. Each device is programmed with a local steam number by the local control processor 213/217. The default value of this field is the same as the device address. The source device will use this number to transmit its multimedia data streams. The sink device will compare this field with its locally programmed number.
Number of Audio Words indicates the number of audio words transmitted in this packet. Since the audio circuit and video circuit may not share the same clock, the number of audio words generated during one video packet time is not constant. Source devices set this field and the sink devices use this field to demultiplex the correct number of audio words from the stream.
Number of Audio Channels indicates the number of audio channels transmitted in this packet. Data for each audio channel is transmitted sequentially until all the channels are transmitted.
Video Mode indicates the video mode being transmitted by this packet (e.g., 480p, 720p, 1080i, etc).
Number of Active Video Words indicates the number of active video words transmitted by this packet. Since only the active video data is being transmitted (in one particular embodiment), the number of active video data determines the packet size. The difference between Number of Video Words and Number of Active Video Words is the number of video clocks in the blank period.
As described previously, the Upstream Device Address field specifies the address of the upstream device and can be used for automatic address learning and multi video stream application.
Control Word is used to carry general purpose control data, including remote control data. Other control data can be, for example, HDCP related control information, home entertainment control information, Internet related information, and other such control information, as needed. Control words are used to exchange data among the control processors 213/217 in each device. In the forward channel of the ASMI link 100, and according to one particular embodiment of the present invention, a control word has the following format shown in Table 2:
In one particular embodiment, the control processor 213/217 in each source or sink device performs segmentation and re-assembly functions and transmits data by this eight byte payload field. Either in point-to-point or daisy-chain configuration, the whole received control word is passed to the local control processor 213/217 for interrogation. In the same way, the local control processor 213/217 will generate all the control data traffic for the downstream device.
Audio Word has variable length and carries all the audio data. The length of this field is determined by the number of audio channels and the number of audio words fields. At the beginning of the audio word field, a four bit audio mode is first transmitted.
1: I2S
2: SPDIF
Others: Reserved for future expansion
The audio clock count field is used for the sink device's audio PLL to regenerate the original audio clock. This field indicates the number of audio clocks counted during the past packet time. In one particular embodiment of the present invention, the audio words are transmitted in the following order shown in Table 3:
Video Word has variable length and carries all the video data. The length of this field is determined by the number of active video words. Although the number of video pixel clocks during one packet time is constant, the number of active pixels is variable since there is blank pixel time. Note that the ASMI does not interpret or alter the video data content. It simply passes the data in the same order from source to sink devices. The video word field sequentially carries all the video data from the source device to the sink device. For a typical 24 bit video (standard HDMI/DVI interface and most video devices today), every pixel of video is carried by three bytes of high speed serial data.
For higher resolution (e.g., 30 bit) video, the number of bits for each video pixel may not be multiples of bytes. The video word field still sequentially carries all the video data from the source device to the sink device. The following scheme can be used, in accordance with one embodiment of the present invention for 30 bit video transmission: Each pixel of video shall be carried by four bytes of high speed serial data. Although this transmission scheme can cause the waste of 2/32=6% of bandwidth, it simplifies the data alignment and circuit design. Thus, a trade-off between wasted bandwidth and simplified alignment/circuit design can be considered when implementing a given application.
Backward Channel Data Format
The data format for the backward channel will now be described, in accordance with one embodiment of the present invention. In one particular embodiment of the present invention, the backward channel data has a similar format as the control data in the forward channel.
In one such case, the backward channel packet has a Preamble field and a Start Frame Delimiter (SFD), as is typical in communication networks. A Packet Size field can also be provided, so as to enable a variable packet size, if so desired. In one particular implementation, the source address and destination address have the same meaning as that of the forward channel control data. The control data can also have variable size (up to 256 bytes in this example). As will be apparent in light of this disclosure, the control data can include a representation of the samples of the modulated IR remote control data 254 captured at a device included in the system. In one particular embodiment, a CRC field having a two byte length is also provided. An example backward channel packet is shown in Table 4:
In one particular embodiment of the present invention, each sink ASMI has a state machine for demultiplexing (e.g., FC-DMX 201 and BC-DMX 209) incoming data, and a state machine for multiplexing (e.g., FC-MUX 205 and BC-MUX 207) outgoing data. In one such embodiment, the demultiplexing state machine receives data from the upstream ASMI and separates the data into three data streams (video, audio, and control). The multimedia data (video and audio) with a video stream number that matches or otherwise corresponds to the sink device is stored in the local buffer 211 for display/sounding. The multimedia data with a non-matching video address is stored in relay buffer 203 for relaying to the downstream port. The control data is transferred to the local control processor 213/217, as previously explained. All these separations can be performed by the demultiplexing state machine based on the predefined packet format discussed herein. The multiplexing state machine of each sink ASMI receives remote control data (e.g., captured by IR detector 219 and processed by control processor 217) from the user's remote control, and multiplexes that control data onto the backward channel and/or forward channel for transmission via the ASMI link 100.
Each source ASMI has a state machine for demultiplexing (e.g., FC-DMX 201 and BC-DMX 209) incoming control data (including remote control data), and a state machine for multiplexing (e.g., FC-MUX 205 and BC-MUX 207) outgoing data. In one such embodiment, the demultiplexing state machine receives remote control data from the downstream ASMI, and regenerates the IR remote control signal (e.g., with processor 213 and IR emitter 215). The source multiplexing state machine receives multimedia data (video and audio) from the local video/audio circuit (e.g., DVD player etc.) via the local buffer 211, and takes relayed upstream multimedia data from the relay buffer 203 if there is any, as well as the control data from the local control processor 213. The multiplexing state machine then multiplexes all the data and transmits that data to the downstream ASMI via the ASMI link 100.
Multiplexing State Machine
The following pseudo code illustrates the multiplexing state machine, in accordance with one embodiment of the present invention:
Demultiplexing State Machine
The following pseudo code illustrates the demultiplexing state machine, in accordance with one embodiment of the present invention:
HDCP over ASMI
HDCP (High Definition Content Protection) is a protocol developed for the HDMI and DVI interface to carry protected video/audio signals. One embodiment of the present invention can be used to implement the HDCP protocol over the ASMI link 100. The employed communication schemes are based on HDCP specification. All the key vectors and key exchange schemes are the same as HDCP. The generation and application of the encryption polynomials are the same as that in the standard HDCP protocol.
Data transmitted over the forward channel serial link are encrypted. In the source device, the data is encrypted at the parallel bus before the SerDes. On the sink-side, the data is decrypted at the parallel bus of the SerDes output. In a typical HDMI or DVI based HDCP system, the source device periodically (about every two seconds) reads link verification value Ri′ from the sink device. The link verification fails if the Ri′ value is wrong.
If the data is carried over an optical wireless link for the forward channel, the backward channel can be implemented with an RF wireless link (e.g., 802.11 or other suitable RF wireless link technology). A typical wireless link has more noise and more transmission errors. However, there are techniques that can be used to improve the reliability of the HDCP link verification in the noise environment. One technique is more retransmission. In particular, the link verification value Ri′ can be sent from sink to source every millisecond (or so), continuously. A few corrupted values will be detected and not cause any problem. Another technique is where the source-side can compare the received value with its newly generated value and previous value. Either match indicates a valid HDCP verification. With this technique, any delay of the new Ri′ caused by network latency or re-try latency will not cause HDCP verification failure.
For the sink-side to decrypt the data correctly, the sink and source device must be in synchronization for the packets and the data words. Packet data format for the ASMI as described herein provides an easy way for this synchronization. In particular, each packet header contains a packet sequence number. Sink and source devices can use this sequence number to get synchronized with each other. All the key exchange, key update, polynomial generation, data encryption and decryption are all synchronized with this sequence number.
Remote Control Signal Processing
The forward channel of this example embodiment provides high bandwidth for video, audio and/or other types of user data, such as IR data, to the video sink-side (e.g., DTV), while the backward channel with relatively low data bandwidth is utilized to carry user data for link integrity checking and other purposes (including transmission of remote control data) to the video source-side (e.g., DVD player). IR control signal data from the user's remote control in modulated form is captured and sampled by the remote control process 111a and converted into electrical signals that are multiplexed on the backward channel by operation of ASMI 111. Those remote control signals are communicated over the ASMI link 100 and received at the video source device ASMI 107, which demultiplexes the remote control signals and provides them to the remote control process 107a. The remote control process 107a can then convert those control signals into a form usable by the video source device to activate the requested control on the source device. For instance, the remote control process 107a can convert the control signals back into IR form to illuminate an IR sensor of the video source device control circuit. Other embodiments may provide the control signals in another form (e.g., electrical signal) suitable to activate the video source device control circuit.
In more detail, and in accordance with one embodiment of the present invention, the output 254 of the IR detector 219 replicates the shape of the input IR signal 250. When the IR detector 219 detects IR control data (e.g., after a user engages the remote control of the video source device to advance through chapters of a DVD or other such source device control), the IR detector 219 outputs a stream sequence of modulated, high and low pulses 254. The stream starts with a period of low signal, and ends with a period of high signal. As explained above, this IR output 254 is sampled by the control processor 217 and represented as digital “0” and “1” stream levels, where “0” is for the low part of the IR signal 254, and “1” for the high part of the IR signal 254.
Two factors are considered when the sampling period is selected: signal distortion and the amount of data stream to represent the detected IR control data signal. In general, the shorter the sampling period, the lower the IR signal distortion level when it is restored at the video source-side. However, the shorter the sampling period, the more data that is needed to characterize the signal.
In one particular case, eight samples are grouped as a byte. In such an embodiment, the ASMI 111 is configured to place a fixed length of bytes (e.g., 128 bytes) into a FIFO (First-In First-Out) buffer (included in the control processor 217 or in a dedicated buffer, not shown). For example, with a 4 Mbps sampling rate, 128 bytes of data could hold 256 microseconds of IR signal (128×8 bits/4 Mbps=256 microseconds). The FIFO buffer contents are processed into a packet by operation of the control processor 217. The packet is then multiplexed onto the backward channel of the ASMI link 100 by operation of BC-MUX 207 for transmission over the ASMI link 100 to the ASMI 107 of the video source device.
After the ASMI 107 of the video source-side receives the IR control data packet, it sends an acknowledgement (ACK) packet to the ASMI 111 by the forward channel to inform the ASMI 111 of the IR control data packet reception. If the received IR control data packet is corrupted on the back channel transmission, the ASMI 107 will not send the ACK packet to the ASMI 111 of the video sink-side. If the ASMI 111 does not receive the ACK packet in a given period of time (e.g., 50 microseconds), the ASMI 111 will resend 425 the IR control data packet, until it receives an ACK packet or until a timeout occurs. The timeout mechanism is used to prevent resending of IR data packet in perpetuity (e.g., in the event of a disabled link 100 or other communication failure). If a 4 Mbps sampling rate is used for the IR signal, then the backward channel bandwidth is at least 16 Mbps for this retransmission scheme 425. After timeout, the ASMI 111 discards the IR control data, and prepares to send the next IR control data packet if any IR packet is pending to be sent.
The remote control process 107a of ASMI 107 can regenerate the received remote control signal data in various ways. In one particular embodiment, the direct output of digital signal is used (i.e., no conversion to IR). Such a direct digital output can readily interface with other digital remote control circuitry included in the video source device. Alternatively, the digital remote control data signal (including the modulation information) can be converted to a modulated IR signal. For instance, the direct digital output can be used to drive an IR LED to emit a corresponding IR signal. This type of IR signal can also be readily received by a video source device, which typically has built-in IR detector circuitry.
Methodology
The method begins with sampling 405 the modulated IR remote control signal in modulated form (without demodulation), and grouping 407 the IR remote control signal samples into a data packet (one or more data packets can be used, depending on the packet format used). The method proceeds with sending 409 the remote control data packet to the source device via a serial communication link. As previously explained, one embodiment implements the communication link with a single optical fiber 100 having a forward channel and a backward channel. The remote control data packet can be communicated using the backward channel of the serial communication link. In other embodiments, the remote control data packet is communicated wirelessly using an optical beam (e.g., IR transceivers configured with grating assisted waveguides). Other embodiments can use an RF link (e.g., 802.11) to transmit the remote control data packet to the source-side.
At the source-side, the method includes receiving 411 the remote control data packet at the source-side, and then determining 413 if the received remote control data packet is corrupted. If not, then the method continues with sending 415 an ACK packet to the sink device, and regenerating 417 the remote control signal. Note that the IR signal is in modulated form, including the carrier frequency. Thus, this regeneration may include providing an electrical output signal (which may or may not have the same format as the received signal), or an IR signal generated by driving an IR light-emitting diode with the received remote control signal. The method continues with providing 419 the regenerated remote control signal to the source device, thereby controlling that device based on the user's remote control action provided at the sink device. If the received packet is corrupted, the method does not send 421 an acknowledgement (ACK) packet to the sink device.
At the sink-side, the method includes determining 423 whether an ACK packet has been received within the pre-set time limit (e.g., 50 microseconds). If not, the method continues with re-sending 425 the data packet to the source device, and determining 427 if a timeout has occurred. In one such embodiment, a timeout can be signaled if the remote control data packet has been re-sent more than three times, although numerous other timeout schemes can be used. If a timeout has occurred, then the method continues with discarding 429 the remote control data packet (note that this situation will rarely occur). If no timeout has occurred, then the method continues by repeating steps 423 and 425 until either an ACK packet is timely received or a timeout occurs. If an ACK packet is received in step 423 or a timeout has occurred 427 and the packet is discarded in step 429, the method will go back to step 405 and repeat steps 405 through 429 as needed, notwithstanding timeouts and other such desired flow control mechanisms.
Remote Control Signal Format
The scheme described herein to sample an IR remote control signal at one device and send it to another device via an ASMI is a universal and integrated IR repeating solution. As is known, IR remote control signals have different standards among manufactures. For a specific manufacture, the format of IR remote control signals is typically consistent due to the compatibility consideration of the legacy product. In this case, a method can be employed to recognize the IR remote control signal from a number of remote control manufactures, and to abstract that signal to a few bytes of data that describe the IR remote control signal. To send these fewer bytes of data requires lower bandwidth.
Generally stated, the ASMI scheme described herein need not be aware of the meaning of this vendor-specific data, because the modulated IR signal is detected as is in modulated form and is regenerated as is in modulated form. There is no need for the source device to understand the meanings of the particular codes used in the IR signal to regenerate the IR signal, since the source device received the IR signal over the ASMI link in modulated form itself.
By utilizing two-way communication as described by one embodiment herein, an IR remote control signal can be securely duplicated on the source-side when the user provides the remote control signal to the sink-side. The IR repeating is “universal” because it does not require knowledge of the modulation frequency of the IR signal. The IR repeating scheme in this disclosure will work with any modulation frequency, because the modulation frequency is embedded in the sampled IR control signal itself that is transmitted from the sink device to the source device over the ASMI link. The integrated solution makes controlling a video source device easy for a user, in that the user can intuitively point the IR remote control at the video sink device (e.g., TV) to control a source device that is not in the consumer's forward field of view. No separate cabling (other than the optical fiber) for carrying the remote control signals is needed. Rather, the remote control signaling data is carried, for example, in the same optical fiber as the payload data (e.g., video and/or audio), or via a wireless transmission (e.g., RF or optical). In either such cases, a forward channel can simultaneously carry payload data from the source device to the sink device.
The foregoing description of the embodiments of the invention has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed. Many modifications and variations are possible in light of this disclosure. For example, multimedia typically includes video and audio data. However, embodiments of the present invention can operate with any type of data, such as presentation slides, study materials, graphics (e.g., digital art or slide shows, with or without audio), audio only (e.g., music or audio books), and video (e.g., movies with or without audio), or other types of data that can be presented in a system having one or more sink devices and one or more source devices that operate under remote control. It is intended that the scope of the invention be limited not by this detailed description, but rather by the claims appended hereto.
This application is a continuation-in-part application of, and claims priority under 35 U.S.C. § 120 from, U.S. patent application Ser. No. 11/423,381 filed on Jun. 9, 2006, entitled “Integrated Remote Control Signaling,” which is incorporated by reference herein in its entirety. This application is also related to (i) U.S. patent application Ser. No. 11/406,558 filed on Apr. 18, 2006, entitled “Protocol for Uncompressed Multimedia Data Transmission,” to (ii) U.S. patent application Ser. No. 11/406,875, filed on Apr. 18, 2006, entitled “EDID Pass Through via Serial Channel,” to (iii) U.S. patent application Ser. No. 11/173,409, filed on Jun. 30, 2005, entitled “Bidirectional HDCP Transmission Module Using Single Optical Fiber,” to (iv) U.S. patent application Ser. No. 11/142,882, filed on May 31, 2005, entitled “High Speed Free Space Optical Detection with Grating Assisted Waveguide,” and to (v) U.S. patent application Ser. No. 10/864,755, filed on Jun. 8, 2004, entitled “Scheme For Transmitting Video and Audio Data of Variable Formats Over a Serial Link of a Fixed Data Rate,” all of which are incorporated by reference herein in their entirety.
Number | Date | Country | |
---|---|---|---|
Parent | 11423381 | Jun 2006 | US |
Child | 11747071 | US |