JP5648542B2 - Wireless audio transmission method - Google Patents

Description

  The present invention relates to a wireless audio transmission method for transmitting an audio signal using radio waves.

  In the audio field, it is practical to transmit audio signals by radio waves because it is complicated to connect a plurality of devices with cables or because the cables are not visually beautiful (for example, Patent Documents). 1). The frequency band that can be used relatively freely in consumer devices such as audio signal transmission is limited, and many devices including a wireless LAN use the 2.4 GHz band.

JP 2006-074374 A

  By the way, since the transmission of the audio signal continues for a long time without interruption, if the transmission of the audio signal occupies the communication channel, communication of another wireless device such as a wireless LAN cannot be performed at all. . On the other hand, if the transmission priority of the audio signal is lowered and the wireless LAN communication is prioritized, the transmission of the audio signal tends to be interrupted, and noise is generated in the audio signal reproduced on the receiving side. was there.

  An object of the present invention is to provide a wireless audio transmission method in which both wireless audio signal transmission and communication by other wireless devices are compatible.

The invention of claim 1 is a method for transmitting audio packets between a transmitter and a receiver using a plurality of channels used in combination with other wireless communication,
(Procedure 1) The transmitter and the receiver select the same one channel from the plurality of used channels,
(Procedure 2) After the transmitter performs carrier sense of the selected channel, stores the busy / idle of the channel, and detects that it has become idle,
(Procedure 3) Wait for a waiting time T1 shorter than the shortest packet interval defined by the other wireless communication procedure, and transmit an audio packet
(Procedure 4) After the receiver completes reception of the audio packet, it waits for a waiting time T2 shorter than the shortest packet interval, and returns a response signal.
(Procedure 5) When the transmitter does not receive a response signal from the receiver even after waiting for a waiting time T3 longer than the waiting time T2 and shorter than the shortest packet interval after completing transmission of the audio packet, Resend the audio packet,
(Procedure 6) When the transmitter does not complete transmission / reception of a predetermined number of audio packets after receiving the response signal, the process returns to Procedure 3 to transmit the next audio packet,
(Procedure 7) Upon completion of transmission / reception of the transmission number of audio packets, the transmitter and the receiver switch the selected channel to another channel of the plurality of used channels and return to Procedure 2 (Procedure 8). The number of transmissions of the procedure 6 in each channel is determined based on the stored busy rate of each channel at a predetermined timing while the procedures 2 to 7 are being repeated.

The invention of claim 2 is a method of transmitting an audio packet between a transmitter and a receiver using a plurality of channels used in combination with other wireless communication,
(Procedure 1) The transmitter and the receiver select the same one channel from the plurality of used channels,
(Procedure 2) After the transmitter performs carrier sense of the selected channel, stores the busy / idle of the channel, and detects that it has become idle,
(Procedure 3) Standby is longer than SIFS, which is a packet interval of an acknowledgment response defined by the other wireless communication procedure, and shorter than DIFS, which is the shortest time of a normal packet interval defined by the other wireless communication procedure Wait for time, send audio packet,
(Procedure 4) After the receiver completes reception of the audio packet, it waits for a waiting time T2 shorter than the SIFS, and returns a response signal.
(Procedure 5) When the transmitter does not receive a response signal from the receiver even after waiting for a waiting time T3 longer than the waiting time T2 and shorter than the SIFS after completing the transmission of the audio packet, the audio packet Resubmit
(Procedure 6) When the transmitter has not completed transmission / reception of a predetermined number of audio packets after receiving the response signal, the transmitter waits for a waiting time T1 shorter than the SIFS and transmits the audio packet. Return to step 6
(Procedure 7) Upon completion of transmission / reception of the transmission number of audio packets, the transmitter and the receiver switch the selected channel to another channel among the plurality of used channels and return to Procedure 4 (Procedure 8). The number of transmissions of the procedure 6 in each channel is determined based on the stored busy rate of each channel at a predetermined timing while the procedures 2 to 7 are being repeated.

According to a third aspect of the present invention, in the first and second aspects of the invention, before executing the procedure 1,
(Procedure A) The transmitter scans a plurality of communication channels to detect the usage status of each channel, determines the plurality of usage channels based on the usage status of each channel,
(Procedure B) The transmitter performs a procedure of notifying the receiver of a plurality of determined use channels.

  According to the present invention, audio packets are transmitted and received with priority over other wireless communication such as a wireless LAN while moving through a plurality of channels, so that audio packets can be transmitted without interruption and each channel can be transmitted. It is possible to minimize the influence on the wireless communication system being used.

Configuration diagram of a wireless audio system according to an embodiment of the present invention A diagram for explaining a plurality of communication channels defined in IEEE802.11b and channels used by the wireless audio system The figure explaining the transmission procedure of the audio packet in the same wireless audio system Flow chart showing operation of transmitter of wireless audio system The figure explaining the schedule decision system of a wireless audio system

  FIG. 1 is a configuration diagram of a wireless audio system according to an embodiment of the present invention. FIG. 2 is a diagram for explaining an audio packet transmission procedure of the wireless audio system.

  This wireless audio system includes a transmitter 1, a receiver 2, a portable audio player 3 connected to the transmitter 1, and a stationary speaker 4 with an amplifier connected to the receiver 2. The transmitter 1 digitizes the audio signal reproduced / input from the portable audio player 3 and transmits it on the 2.4 GHz band quasi-microwave. The receiver 2 receives the high-frequency signal transmitted from the transmitter 1, D / A converts the audio signal superimposed on the high-frequency signal, and inputs the audio signal to the speaker 4 with amplifier. The digitized audio signal is transmitted from the transmitter 1 to the receiver 2 in one direction at a high speed, and the control signal such as a command is bidirectionally communicated between the transmitter 1 and the receiver 2 at a low speed.

  Communication between the transmitter 1 and the receiver 2 is performed while moving on 13 channels defined in IEEE802.11b. When transmission / reception (transmission) of a predetermined number of audio packets is completed on any channel, the channel is moved, and the next predetermined number of audio packets are transmitted on the moved channel. By repeating this, the audio packet is transmitted without interruption, and communication in a specific channel is completely impossible even if the wireless LAN system of IEEE802.11 (b) exists in the radio wave reachable range. There is no such thing.

  Here, with reference to FIG. 2, 13 channels defined in IEEE 802.11b will be described. Thirteen channels are set at intervals of 5 MHz from 2412 MHz to 2472 MHz. When the communication bandwidth is 20 MHz, four channels Ch1, Ch5, Ch9, and Ch13 can be used simultaneously as shown by the solid line in FIG. Further, when the transmission bandwidth is 22 MHz, three channels can be used simultaneously. In FIG. 2B, a combination of three channels Ch1, Ch7, and Ch13 is shown by a solid line, but there are various combinations such as Ch1, Ch6, and Ch11. In the following embodiment, a procedure for transmitting an audio packet while sequentially moving three channels (Ch. A, B, C) will be described. However, in the present invention, the number of channels used is not limited to three.

  With reference to FIG. 3, an audio packet transmission procedure in the wireless audio system will be described. The transmitter 1 and the receiver 2 start transmission of audio packets on the first channel (Ch.A). Thereafter, the transmitter 1 and the receiver 2 transmit a predetermined number of audio packets for each channel while sequentially moving a plurality of channels (used channels Ch. A, B, C) as shown in FIG. .

  An audio packet transmission procedure in each used channel will be described with reference to FIG. When the transmitter 1 starts transmission of an audio packet on a used channel (for example, Ch. A), first, carrier sense of that channel is performed. If another communication device is transmitting a signal on this channel (busy: BUSY), that fact is stored and the state is monitored to change to idle. When the channel state changes to idle, or when the channel is idle as a result of carrier sense, transmission of an audio packet is started after waiting for a waiting time T1 (for example, 3 μs). The reason that the busy state is stored at the start of channel use is to determine the number of transmission packets for each channel based on the busy rate of each channel. Details thereof will be described later.

  If the other communication device is a wireless LAN terminal compliant with IEEE802.11b, after the channel state changes to idle, it waits for at least 10 μs (SIFS: Short Inter Frame Space (see FIG. 3D)). Usually, after waiting for 50 μs or more (DIFS: DCF Inter Frame Space + backoff time), the transmission of the packet is started. For this reason, the transmitter 1 can always transmit an audio packet without causing the wireless LAN terminal to start communication. This audio packet is received by the receiver 2.

  When the transmission of the audio packet is completed, an acknowledgment ACK is transmitted from the receiver 2. The waiting time T2 from the completion of packet reception at the receiver 2 to the transmission of the acknowledgment ACK is, for example, 2 μs. Also in this case, it is possible to transmit an acknowledgment response ACK prior to the wireless LAN terminal in the same manner as described above.

  In addition, as shown in FIG. 3C, the transmitter 1 retransmits the same audio packet when the acknowledgment response ACK is not received even after waiting for T3 (for example, 5 μs) after transmitting the audio packet. Since this waiting time T3 is also shorter than the shortest packet interval (SIFS) of the wireless LAN, the audio packet can be retransmitted without being interrupted by the terminal of the wireless LAN. The size of one audio packet is about 500 μs, and audio data for about 1 ms can be transmitted with this size. Therefore, even if retransmission occurs once every few times, the buffer does not underflow on the receiver 2 side.

  The wireless audio system is not intended for long-distance communication such as a wireless LAN, and is intended for packet transmission over a short distance of about 10 meters. Because it is a system. Of course, if the standby time is lengthened, the distance margin between devices of the wireless audio system and the throughput margin of the high-frequency circuit and the processing unit can be increased.

  In the description of this figure, the waiting times T1, T2, and T3 are all shorter than SIFS (10 μs). However, even if the waiting times T1, T2, and T3 are longer than SIFS, the shortest DIFS time (50 μs). If shorter, communication is possible without being interrupted by other wireless LAN terminals. In this way, if the wireless LAN (IEEE802.11b) acknowledgment ACK transmission (10 μs standby) is given priority over the audio packet transmission start (T1> 10 μs) of the wireless audio system, the wireless audio system Waiting for completion of a series of packet transmission procedures for wireless LAN packet transmission and acknowledgment reply, audio packet transmission is started, and transmission start of the audio packet is slightly delayed. In view of the efficiency, it can be said that it is more efficient to give priority to the acknowledgment ACK of other wireless LAN terminals and complete a series of packet transmission procedures. Further, only T1 immediately after Busy in FIG. 3B is made longer than SIFS (10 μs) to give priority to the transmission of the wireless LAN confirmation response, and the subsequent waiting times T1, T2, T3 are made shorter than SIFS and transmitted. The efficiency may be improved. As the shortest packet interval of the present invention, when “another wireless communication procedure” is IEEE802.11b, the shortest time of SIFS or DIFS can be applied.

  In this procedure, when the transmitter 1 and the receiver 2 transmit the audio packet, the audio packet is always transmitted with priority over the packet of the wireless LAN and the sound is not cut off. However, if this is continued for a long time, this channel (for example, Ch. In A), the wireless LAN system cannot communicate at all. Therefore, when transmission of a predetermined number of packets (for example, 3 to 6) of audio packets is completed, it moves to the next channel (for example, Ch. B) and continues transmission of audio packets in the same procedure. Similarly, when transmission of a predetermined number of audio packets is completed on this channel, the transmission is further moved to the next channel (for example, Ch. C), and transmission of audio packets is continued in a similar procedure. This minimizes the influence on the wireless LAN while continuing the transmission of audio packets without interruption.

  Here, a method for determining the predetermined number of packets, that is, the number of transmitted packets at each use channel will be described. As described above, at the start of the transmission procedure on each used channel, whether or not the channel is busy (that is, busy) is stored. At an appropriate timing, the stored busy frequencies are integrated to calculate a busy rate. That is, the ratio between the number of times each channel is used and the number of times of busy is the busy rate. Based on the busy rate of each channel, the number of packets is allocated so that the number of transmission packets of a channel with a low busy rate is increased and the number of transmission packets of a channel with a high busy rate is reduced.

  Next, a method for determining the number of transmission packets for each used channel will be described. In this embodiment, the number of packets is assigned to each used channel so that 15 packets can be transmitted by making a round of three used channels. The number of transmitted packets p (1), p (2), p () of the used channels C (1) (= Ch.A), C (2) (= Ch.B), C (3) (= Ch.C) 3) For example,

It is calculated by the following formula. here,
r1 = [idle rate of channel C (1)] − 50
r2 = [idle rate of channel C (2)] − 50
r3 = [idle rate of channel C (3)] − 50
And The idle rate is calculated as “100−busy rate”. When r1, r2 and r3 become negative, it may be set to 0. Further, depending on the rounding processing of int, the sum of p (1), p (2), and p (3) may be larger or smaller than 15. In this case, the maximum value p (n) It may be adjusted by increasing or decreasing.

Through the above processing, assignment of the number of transmission packets according to the congestion of the channel bandwidth of each used channel is determined. The method for determining the number of transmission packets is not limited to this method.
Since the busy frequency is not accumulated at the beginning of transmission of audio packets, the number of transmission packets may be allocated equally to all used channels.
p (1) = 7, p (2) = 5, p (3) = 3
As described above, the number of packets may be fixedly assigned in order of use channel.

  This transmission packet number determination process may be performed at an appropriate interval. In order to cope with instantaneous channel use, it may be performed at short intervals of about 1 to several seconds. In addition, when dealing with steady channel use, it may be performed at a long interval of about 1 to several minutes.

  Returning to FIG. 1, the configuration of the transmitter 1 and the receiver 2 will be described. The transmitter 1 includes an audio input unit 10 to which the portable audio player 3 is connected, an A / D converter 11 that converts an audio signal input from the portable audio player 3 into a digital signal, a controller 12 that controls the operation of the transmitter 1, A communication circuit 13 and an antenna 14 for performing communication through a communication channel are provided. The communication circuit 13 transmits an audio signal and receives an acknowledgment ACK from the receiver 2. The communication circuit 13 performs mutual communication of control signals such as transmission and reception of schedules with the receiver 2. Further, the communication circuit 13 checks the busy / idle by watching the communication channel. Channel switching of the communication circuit 13 is performed under the control of the controller 12.

  The receiver 2 includes a built-in antenna 24, a communication circuit 20 that performs communication on each communication channel, a controller 21 that controls the operation of the receiver 2, a D / A converter 22 that converts a received digital audio signal into an analog signal, and An audio output unit 23 is provided for outputting an audio signal to the speaker 4 with amplifier. The communication circuit 20 receives the high frequency signal transmitted from the transmitter 1 and demodulates the audio signal and the control signal, and transmits a signal such as an acknowledgment ACK input from the controller 21 to the transmitter 1. The audio signal demodulated into the baseband digital signal by the communication circuit 20 is input to the D / A converter 22. The control signal demodulated by the communication circuit 20 is input to the controller 21. The D / A converter 22 converts the input digital audio signal into an analog audio signal and outputs the analog audio signal from the audio output unit 23 to the speaker with amplifier 4.

  FIG. 4 is a flowchart showing an audio packet transmission processing operation. This processing operation is executed by the controller 12 of the transmitter 1. First, a schedule is determined (S1). The schedule consists of the used channel C (1-3) and the number of transmission packets p (1-3) for each channel. As the channels to be used, three channels with appropriate intervals may be selected as shown in FIG. When the schedule is determined, the channel is scanned to search for a receiver, and communication with the receiver 2 is established (S2). When communication with the receiver 2 is established, the schedule determined using the control channel is transmitted (S3). Thereafter, the packet is transmitted by the following processing.

  1 is set to the counter n of the used channel (S4). It moves to this selected use channel C (n) (S5). Next, a counter i for counting the number of packets transmitted on the selected use channel is set to 1 (S6). Then, it is determined whether the channel is busy based on carrier sense (S7). If the channel is busy (YES in S7), the busy counts are accumulated (S8), and the process waits in S7 until the busy state ends. Note that the busy count in S8 is not accumulated during this standby. When the channel state shifts to idle (NO in S7), it waits for the waiting time T1 (S9) and transmits the i-th audio packet (S10). It is assumed that the carrier sense of the communication channel is also performed during the standby time T1.

  When the transmission of the audio packet is completed, it waits for T3 until an acknowledgment ACK is received from the receiver 2 (S11, S12). If an acknowledgment ACK is received before T3 elapses (YES in S11), it is determined whether the packet number counter i is the number of packets p (n) to be transmitted on this use channel (S13). If the packet number counter i is less than p (n) (NO in S13), 1 is added to i (S14), and the process returns to S9 to transmit the next audio packet.

  On the other hand, in S11 and S12, if no acknowledgment response ACK is received from the receiver 2 even after the standby time T3 has elapsed (YES in S12), the process returns to S10 to retransmit the i-th audio packet.

  If the packet number counter i is p (n) in S13 (YES in S13), it is assumed that the transmission of the audio packet in this use channel is completed, and after switching the use channel in synchronization with the receiver 2, S5 Resume transmission of the following audio packets.

  Here, the use channel is switched in the following procedure. First, it is determined whether the counter n of the used channel is 3 (S15). If n is less than 3 (NO in S15), 1 is added to n (S16), and the process returns to S5 and moves to the use channel C (n) designated by n (S5). On the other hand, if n is 3 (YES in S15), the process returns to S4 to reset n to 1 (S4), and moves to the use channel C (1) designated by n (= 1) (S5). .

  FIG. 4B is a flowchart showing transmission packet number assignment processing. This process is executed at an appropriate timing (for example, every 1 to several seconds or every 1 to several minutes) while the processes of S4 to S16 in FIG. This process is preferably executed at the timing indicated by B in FIG. 4A while returning from S15 to S4.

  This process is a process of reallocating the number of transmission packets of each used channel based on the busy rate of each used channel so far. First, the busy frequency of each used channel is totaled to calculate the busy rate (S30). And based on this total result, the transmission packet number p (1-3) of each use channel is calculated (S31). The busy rate calculation method and the transmission packet number calculation method are as described above. Then, the calculated transmission packet number p (1-3) is transmitted to the receiver 2 (S32). Subsequent audio packet transmission / reception is performed for each packet. By this processing, it becomes possible to assign the number of transmission packets according to the real-time congestion status of each used channel. Each time this processing is performed, the integrated value of the busy frequency of each used channel may be reset, or the integration may be continued as it is.

  In the above description, the channel to be used is selected by appropriately selecting three channels that are appropriately separated as shown in FIG. 2 (B). Check the channel, select a channel that has little effect even if packet transmission is performed preferentially, and use this channel to transmit audio packets to minimize the impact on other wireless communications. You may do it.

  Hereinafter, a method for determining the use channel and the number of transmission packets (schedule) for each use channel will be described with reference to FIG. In the following embodiment, a case will be described in which three used channels are selected with a communication bandwidth of 20 MHz.

  FIG. 5A is a flowchart showing the schedule determination processing operation. This process is executed by the controller 12 described later. In this process, first, all 13 channels in the 2.4 GHz band are scanned (S51), and the usage status of each channel is tabulated from the scan results (S52). Then, based on the result of the aggregation, the channel used for packet transmission (used channel) and the number of packets transmitted once in each used channel are determined (S53).

  With reference to FIGS. 5B and 5C, channel scanning and usage status totalization methods will be described. Channel 1 to channel 13 are scanned and carrier sense is performed to detect busy (in use) / idle (empty) of each channel. This is repeated a plurality of times (for example, 100 times) at regular intervals (for example, every 10 ms). The time required for scanning is 1 second when it is repeated 100 times every 10 ms. In this way, by repeatedly performing busy / idle scans for all channels, it is possible to detect an average usage situation that is not instantaneous for each channel. Note that most of the time required for channel scanning is the time required for channel switching (reception frequency switching) of the receiver of the communication circuit 13. If a receiver capable of high-speed channel switching can be realized, the channel scanning time can be greatly reduced.

  After these multiple scans, the number of busy times (busy frequency) is counted for each channel. FIG. 5C is the busy frequency distribution graph. When the scan is repeated 100 times, the busy frequency (frequency) becomes the busy rate (%). In the example of the figure, communication is performed using channel 1 and channel 6, and the busy rate of channel 6 has a high probability of about 60%. The communication channel band extends from channel 5 to channel 7. Further, channel 1 has a busy rate of about 20%, and the channel band of this communication extends to channels 1 and 2.

  Then, the channel band of packet transmission is applied on the frequency distribution graph, and three channels having less influence are determined as used channels. In the example of FIG. 5C, the channel 13 and the channel 9 that do not substantially affect the currently used channel are determined as the first and second used channels C (1) and C (2), and then The channel 1 that has no influence on the channel 6 that is used at a high busy rate and has an influence on the channel 1 that is used at a low busy rate is determined as the third used channel C (3).

  When the used channel is determined in this manner, the number of transmission packets may be determined in advance based on the busy rate of each used channel. In this case, the calculation method of the number of transmitted packets p (1), p (2), and p (3) may be the same as S31 in FIG.

The above description is a method of determining three used channels by performing a channel scan at the start of communication. However, the used channels may be reviewed during communication. As described above, in the transmission method of the present embodiment, an audio signal of about 1 ms can be transmitted using a packet of about 500 μs. Therefore, if transmission is performed without error, there is a margin of audio data on the receiver 2 side. Therefore, the schedule may be recreated by performing the channel scan shown in FIG. 5 again at an appropriate timing during transmission execution. The appropriate timing is, for example, when a predetermined number of audio packets can be transmitted without retransmission at regular intervals, or when a message indicating that the buffer is full is received from the receiver 2 side 2. In the case where the channel scan is executed halfway, the time may be shortened by reducing the number of repetitions (100 in the example of FIG. 5).

  In this embodiment, the device that reproduces the audio signal is the portable audio player 1, but the reproducing device is not limited to a portable device.

1 Transmitter 12 Controller 13 Communication Circuit 2 Receiver 21 Controller

Claims (3)

A method of transmitting audio packets between a transmitter and a receiver using a plurality of channels used in combination with other wireless communication,
(Procedure 1) The transmitter and the receiver select the same one channel from the plurality of used channels,
(Procedure 2) After the transmitter performs carrier sense of the selected channel, stores the busy / idle of the channel, and detects that it has become idle,
(Procedure 3) Wait for a waiting time T1 shorter than the shortest packet interval defined by the other wireless communication procedure, and transmit an audio packet
(Procedure 4) After the receiver completes reception of the audio packet, it waits for a waiting time T2 shorter than the shortest packet interval, and returns a response signal.
(Procedure 5) When the transmitter does not receive a response signal from the receiver even after waiting for a waiting time T3 longer than the waiting time T2 and shorter than the shortest packet interval after completing transmission of the audio packet, Resend the audio packet,
(Procedure 6) When the transmitter does not complete transmission / reception of a predetermined number of audio packets after receiving the response signal, the process returns to Procedure 3 to transmit the next audio packet,
(Procedure 7) Upon completion of transmission / reception of the transmission number of audio packets, the transmitter and the receiver switch the selected channel to another channel of the plurality of used channels and return to Procedure 2 (Procedure 8). The wireless audio transmission characterized in that the number of transmissions of the procedure 6 in each channel is determined based on the stored busy rate of each channel at a predetermined timing while the procedures 2 to 7 are being repeated. Method. A method of transmitting audio packets between a transmitter and a receiver using a plurality of channels used in combination with other wireless communication,
(Procedure 1) The transmitter and the receiver select the same one channel from the plurality of used channels,
(Procedure 2) After the transmitter performs carrier sense of the selected channel, stores the busy / idle of the channel, and detects that it has become idle,
(Procedure 3) Standby is longer than SIFS, which is a packet interval of an acknowledgment response defined by the other wireless communication procedure, and shorter than DIFS, which is the shortest time of a normal packet interval defined by the other wireless communication procedure Wait for time, send audio packet,
(Procedure 4) After the receiver completes reception of the audio packet, it waits for a waiting time T2 shorter than the SIFS, and returns a response signal.
(Procedure 5) When the transmitter does not receive a response signal from the receiver even after waiting for a waiting time T3 longer than the waiting time T2 and shorter than the SIFS after completing the transmission of the audio packet, the audio packet Resubmit
(Procedure 6) When the transmitter has not completed transmission / reception of a predetermined number of audio packets after receiving the response signal, the transmitter waits for a waiting time T1 shorter than the SIFS and transmits the audio packet. Return to step 6
(Procedure 7) Upon completion of transmission / reception of the transmission number of audio packets, the transmitter and the receiver switch the selected channel to another channel among the plurality of used channels and return to Procedure 4 (Procedure 8). The wireless audio transmission characterized in that the number of transmissions of the procedure 6 in each channel is determined based on the stored busy rate of each channel at a predetermined timing while the procedures 2 to 7 are being repeated. Method. Before performing step 1 above,
(Procedure A) The transmitter scans a plurality of communication channels to detect the usage status of each channel, determines the plurality of usage channels based on the usage status of each channel,
(Procedure B) The wireless audio transmission method according to claim 1 or 2, wherein the transmitter performs a procedure of notifying a receiver of a plurality of determined use channels.

Images