Enhanced multi-user transmission

SUMMARY OF THE INVENTION

The object of the present invention is to provide a more flexible transmission system to many users that requires less modification of older devices and procedures.

This object is achieved by a transmitting device according to claim 1, and a receiving device according to claim 12.

Thus, the extended MAC frame, e.g. MU-RTS, is defined. This frame is different from the normal RTS frame because it has many recipients of MAC addresses. This provides a better communication path of the list of identifiers or addresses to other transmission terminals. Although the proposed extended MAC frame has specific fields that are only meaningful / understandable for MU devices, the frame can be broadcast in the older physical layer and has common fields that can be read by all older devices. Therefore, older devices are able to decode bits, interpret common fields and initiate appropriate settings. The interpretation of the extended MAC frame may be a pure MAC process, so that no further information from the physical layer is required. In addition, there is no need to change the interpretation rules for the corresponding existing or older MAC frames. Due to the fact that all other transmission terminals can be at least partially interpreted by all other transmission terminals, their transmission can be considered as broadcast transmission from a physical layer perspective. Consequently, older devices and procedures require little modification.

The proposed transmission devices may be adapted to form the transmission frame only from packets destined for at least one of those other transmission terminals that responded to the transmission request. The receiving device may be adapted to calculate the response time based on its order on the identification list. As a result, other transmission ends (e.g. WLAN stations) can be designed for packets in the frame, so delay and bandwidth can be balanced in various traffic situations.

The identification may for example contain a MAC address, so much longer addresses can be used as compared to the physical layer addresses used in the prior art.

In a particular case, the other ends of the transmission may be candidate transmitters that are addressed to return a request for dispatch to show the intention to send. The proposed device can then be adapted to perform channel estimation for candidate transmitters in accordance with their channel implementation, and to respond to a request to send with an acknowledgment of sending indicating which transmitter can access the channel through which it shapes the direction vector or appropriate channel access information. This is an advantage when the simultaneous reception of multiple packets from different transmitters can be supported and interference between spatial streams can be minimized.

The transmitting device may be configured to use the transmission request as a feedback request for channel status information, to evaluate the channel implementation of said other transmission terminals based on feedback channel status information obtained from said other transmission terminals, and to obtain the appropriate vectors directional for these other transmission terminals, or subcarrier allocation in the case of MU-OFDM. The receiving device is adapted to respond to a transmission request with a channel feedback frame containing channel state information, and optionally also the duration of information indicating the amount of data in its transmission queue. The transmitting device may then be adapted to send frames requesting to be sent including a list of other transmission terminals and corresponding directional vectors for other transmission terminals. The request for transmission may contain information that specifies the format to use for feedback on the status of the channel feedback. The reception of the transmission can be confirmed by the sending device by sending a MAC frame which indicates the other transmission ends from which said transmissions were received. As an example, other transmission ends (i.e. receiving device) from which the transmissions were received, can be indicated by setting the appropriate bits in the bitmap contained in the MAC frame.

In this way, extended MU-DCF can be obtained, which is based on two phases. First, channel status information is obtained from candidate stations. Second, signaling exchange for channel access is taking place. This ensures better coordination of multi-user transmission and reduces possible interference.

In another example, the transmitting device may be adapted to add duration information to the transmission request that indicates the expected duration of all the expected transmission. This measure provides a solution to the hidden problem, when all the receiving terminals allow the calculation of the duration, they must wait before the start of the transmission. More specifically, the receiving device may be adapted to set its network allocation vector according to the content of the duration field contained in the transmission request if the list does not contain any identification of said device. Another list search device may be adapted to calculate the waiting time for their response based on the duration field.

In yet another example, the receiving device may be adapted to respond to the transmission request by combining the MAC data packet unit and the emitting single user feedback mechanism.

In this way, the response can be constructed by combining a conventional response frame (e.g., CTS frame) with a feedback emitting frame, so compatibility with older devices can be improved and the current standard would require only minor modifications.

In another example, the receiving devices may be adapted to use assigned subsets of the multi-carrier transmission signals for the response. In a particular example, the assigned subset may be from the order of said identification in the list. This meter provides an advantage such as the time needed for feedback transmission (e.g., M-CTS and / or M-ACK frames) that can be shortened, as in parallel and not in a dedicated time system. Thus, the signaling load can be reduced and prevents the hidden node problem.

The transmitting or receiving device may include any type of MU transceiver with at least one of a plurality of input strings and a multiple output strand. It is not intended to be limited to the MU MIMO transceiver.

The request for transmission can be transmitted in an older format so that all devices, MUs and older devices can set their NAV accordingly. Older devices can at least read the duration field and therefore you can set their NAV accordingly.

Further advantageous changes are set out in the dependent claims.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will now be described based on various examples in accordance with the accompanying drawings in which:

Fig. 1 shows a block diagram of a multi-user MIMO transmission system in accordance with various examples;

Fig. 2 shows a four-way compliance procedure according to the first example;

Fig. 3 shows the MU-RTS frame structure with multiple receiver field addresses in accordance with the first example;

Fig. 4 shows the M-ACK frame structure with instructions for using the antenna according to the first example;

Fig. 5 shows a two phase channel access procedure according to the second example;

Fig. 6 shows the MU-CFR frame structure according to a second example;

Fig. 7 shows the MU-RTS frame structure with the Tx direction vectors according to the second example;

Fig. 8 shows the MU-CTS frame structure according to a second example;

Fig. 9 shows a diagram illustrating a hidden node problem;

Fig. 10 shows a more general MU-RTS frame structure according to the fourth example;

Fig. 11 shows the MU MIMO uplink mechanism according to the fifth example;

Fig. 12 shows the structure of a C4T frame according to the fifth example;

Fig. 13 shows the structure of the RTS frame according to the fifth example;

Fig. 14 shows the MU-CTS frame structure with the Tx direction vectors according to the fifth example; and

Fig. 15 shows the MU-ACK frame structure according to the fifth example.

DETAILED DESCRIPTION OF EXEMPLARY EXAMPLES

In the following example, preferred embodiments are described on the example of the MU MIMO system as shown in Fig. 1.

Referring to Fig. 1, the MU MIMO access point (AP) provides WLAN access for an exemplary number of four stations from 21 to 24 as shown in Fig. 1 and has multiple antennas. AP 10 includes N different M1 to MN processing steps to provide different coding and / or modulation schemes to which input signals can be used and which can be selectively connected to at least one of a plurality of antennas.

Typically, two types of MIMO techniques can be used in two directions between AP 10 and each of the stations from 21 to 24 based on channel spreading properties, i.e. the structure of the spatial correlation matrix in the receiver's antenna pattern. For high correlations of the received signal, various directional algorithms can be used, while for low correlations of the received signal - diversity (DIV) and multiplexing (MUX) approaches can give better performance. In MUX schemes, multiple streams are transmitted simultaneously, each using one dedicated antenna. This increases throughput with a factor equal to the number of streams transmitted. In DIV schemes, many antennas are used differently. The transmitter uses only one antenna for the basic DIV scheme. A receiver with multiple antennas receives a lot of copies of the transmitted signal so that, using an appropriate signal processing algorithm, it obtains a significantly higher signal-to-noise ratio (SNRs). In the scheme combining MUX and DIV, more transmit antennas are active, but the receiver, as in all DIV schemes, may still have more antennas than the number of streams. Multiplexing is present, but the receiver receives more information about the transmitted signal than in the case of pure MUX.

These examples improve the multi-user support for an IEEE 802.11-based network by using MU-RTS and M-CTS frames for channel access, and M-ACK when confirming correctly received packets. Optionally, the adaptive transmission channel can be selected by selecting only a subset of stations that responded together with M-CTS to create another MIMO frame. The decision can be made on the basis of information obtained from M-CTS. In this way, MU diversity can be used. In addition, channel quality information can be feedback via M-CTS and M-ACK frames, which indicate a subset of the antennas accepted for transmission. Depending on the complexity and accepted load, the decoded information can be more accurate, such as the received SNR from each antenna, for example. Inter-frame spaces between adjacent packets can be provided to ensure coexistence with older IEEE 802.11 stations.

Fig. 2 shows the four-way compliance procedure according to the first example. The proposed MU-DCF is based on conventional M-DCF, where a four-way compliance procedure is proposed to facilitate channel access with multiple users prior to data transmission.

The following additional MAC protocol functionalities are proposed during the transmission cycle in MIMO scenarios, compared to the conventional M-DCF protocol.

Referring to Fig. 2, transmission is initiated, e.g., by AP 10 and transmitting the MU-RTS frame as shown in Fig. 3, which is a MAC frame containing multiple receiver addresses used for addressing e.g. three (R # 1 to R # 3) of the four exemplary stations from 21 to 24 shown in Fig. 1. The MU-RTS frame may be preceded by a training sequence allowing channel estimation on the receiver side, e.g. at one selected station from 21 to 24. As an alternative option, the channel can also be estimated in parallel with the MU-RTS frame transmission, e.g. by preamble transmission. It should be noted that the transmission of preambles for channel estimation may result in the MU-RTS frame becoming not able to be decoded at the physical layer.

After receiving the MU-RTS frame, the station selection (R # 1 to R # 3) that are present on the receiver list will be answered using the M-CTS frame. The order of responses is by default determined by the respective subsequent receivers in the list. The first M-CTS is transmitted after a short interframe delay (SIFS), and the next are transmitted after a reduced interframe delay (RIFS).

The above procedure can be programmed as a subprogram based on the following pseudo structure:

n is the station position in the list of the receiver in the MU-RTS frame after receiving the MU-RTS, it is waiting for SIFS during (the station has not yet started its M-CTS transmission) if (n = 1) transmits M-CTS otherwise if (the channel is busy) {waiting for the channel to be free n = n-1 waiting for RIFS}

otherwise {n = n-1 is waiting for RIFS}

After this, the MU-RTS frame will be sent, the transmitter (e.g. AP 10) will start searching for the channel, and may continue after the following pseudo code structure (if the subsystem has been implemented):

after SIFS, the first M-CTS is expected during (waiting for more M-CTS frames) if M-CTS does not arrive when it is expected, after RIFS the next M-CTS is expected;

otherwise the next M-CTS is expected after + RIFS transmission;

The transmitter (e.g. AP10) does not receive any, receives pair or all of the M-CTS frames from the addressed subset of stations. From the received, it can read the information contained in the CAB field (or any other channel status information that could be contained in the M-CTS frame), and can create a MIMO frame from those packets intended only for stations that responded. This can be expressed by the following pseudo code structure:

if (# M-CTS received> 0) creating and transmitting a MIMO frame from packets for stations that responded, optionally using some scheduling strategies; otherwise, it starts channel access for the next transmission;

When using the scheduling strategy, MIMO frames may not contain packets for some stations that responded with an M-RTS frame. Classification can include adaptive switching between SU ​​and MU operating modes.

Stations receive the MIMO frame, and generate the M-ACK frame, e.g. on the same principles as in M-DCF. The order and separation of M-ACK frames can be equal to the M-CTS frame.

Finally, when the transmitter (e.g. AP 10) receives M-ACK frames, it removes confirmed packets from the queue and initiates subsequent transmissions. Unconfirmed packets may be resent.

As can be deduced from Fig. 2, successive M-CTS and M-ACK frames are separated by RIFS. The number of "m" users that are served in this way can be determined by the following relationship:

SIFS + (m-2) RIFS <DIFS (1)

This condition ensures that even if all M-CTS / M-ACK packets except the last are lost, the last station will still have a free channel to broadcast its own M-CTS / M-ACK. If the condition is not met, the channel may be free for a longer duration than the DIFS of the ongoing transmission, which will allow starting a new one.

Fig. 4 shows the M-ACK frame structure according to the first example with instructions for using the antenna for the next transmission. As already mentioned, M-CTS frames according to the MDCF procedure contain a bitmap for antenna feedback (CAB field). This information can be used in the transmitter to use the adaptive channel, where only packets for stations with good channel condition can contribute to the MIMO frame. By including the CAB field in the M-ACK frame as shown in Fig. 4, channel feedback can be obtained without requiring the M (U) -RTS - M-CTS compliance procedure.

In addition, other IEEE 802.11e options such as transmission capabilities (TxOP), block acknowledgments (BA), or no acknowledgments can also be combined with the above procedures to further increase performance.

In addition, an estimated load can be carried out. The induced load will now be shown based on the example in Fig. 1 with a single transmitter (AP 10) that transmits sixteen packets, four to each of the four stations from 21 to 24. If all stations from 21 to 24 will have four antennas (every is not necessary, but only an example contrary to the example in Fig. 1), the 4x4 multiplexing scheme can be used as follows: SU case:

x (M-RTS + M-CTS + MIMO frame + M-ACK) (2)

MU case:

x (MU-RTS + 4M-CTS packages + MIMO frame + 4M-ACK) (3)

This means that the transmission procedure clearly takes longer in the following MU case, and the average delay per packet is the same in both cases. However, the average delay per station for a SU varies considerably depending on whether it is the first or last station to receive its packets. For some stations, such long delays may not be acceptable.

So far, in the example above, it was assumed that at the beginning all four packets for all four stations from 21 to 24 were already generated. But there are many applications offering low load, on the other hand with very stringent delay requirements that do not allow too long to wait for further packets for the current station to create a full MIMO frame. Otherwise, sending immediately what is present in the queue will often mean the transmission of a single spatial stream, which actually means load multiplication. If there are no restrictions on the MAC packets for spatial stream mapping, the MAC packet may, however, be transmitted using multiple spatial streams.

In the event of heavy loads, the proposed MU approach will be very beneficial by reducing fluctuations.

Before the MU RTS / CTS compatibility mentioned above in connection with the first example, the transmitter (e.g. AP 10) may send MU MIMO a channel feedback packet (MU - CFR) request preceded by training sequences that request channel state feedback (CSI) from candidate receiving stations. The transmitter therefore has the CSI of all candidate receivers and can evaluate them all for possible MU MIMO transmission and choose the right, transmitting directional vectors to reduce interference between spatial streams.

The MU RTS packet can be modified in such a way that instead of communication with the proposed bitmap of the antenna, the proposed directional vector for each of the streams is forwarded to stations from 21 to 24. This mechanism provides the possibility of using directional transmission as the MU MIMO transmission mechanism and checks whether stations from 21 to 24 can be administered simultaneously with MU MIMO transmission using the proposed direction vectors selected by the transmitter. In general, the proposed improvement ensures better coordination of MU MIMO transmission and reduces possible interference among streams. Thus, it improves the performance of the entire network.

In the second example, a new MU-DCF with two phase channel access procedure is shown. MU MIMO transmission is carried out with directional support of the IEEE 802.11 based network using MAC frames, such as MU-CFR, MIMO channel feedback (M-CF), MU-RTS and M-CTS frames, for channel access, and MAC M-ACK frame to confirm correctly received packets. Optionally, MU MIMO adaptive transmission can be used by modifying the direction vectors for MO MIMO transmission only with a subset of stations that responded with M-CTS to create another MIMO frame. This decision can be made based on information collected from the M-CTS frame. If the channels already have this information, MU-CFR and M-CF frames may not be needed for the channel access mechanism and MU MIMO transmission.

MU MIMO transmissions increase spectral efficiency and ensure better use of resources. However, MIMO systems provide high bandwidth for many stations and at the same time can become critical when the wireless network is compact.

Fig. 5 shows a two phase channel access procedure according to the second example. The proposed MU-DCF procedure is based on the first phase of obtaining channel status information from candidate stations / receivers, and the second phase of MU-RTS and M-CTS exchange for channel access similar to the first example.

In the first phase, the transmission is initiated by a multi-access channel (MU-CFR) feedback request frame in which the transmitter (e.g. AP 10 in Fig. 1) basically requests channel status information from multiple receivers (e.g. a subset of stations from 21 to 24 in Fig. 1).

Fig. 6 shows the MU-CFR frame structure according to a second example, which may be preceded by a training sequence for channel estimation at the receiver. The proposed MU-CFR frame is a MAC frame with multiple receiver address fields.

After receiving the MU-CFR frame, the stations present on the receiver list correspond to the MIMO channel feedback frame (M-CF). The M-CF frame is a MAC frame and can have the same MIMO CSI feedback frame format in the SU MIMO system in accordance with IEEE 802.11n. The order of responses can be determined by default by corresponding receivers in the list. The first M-CF frame is transmitted after the SIFS break, and the next after the corresponding RIFS breaks.

In the second phase, after the transmitter sent the MU-CFR frame and received the corresponding MIMO information about the channel status of each candidate receiver, it evaluates the station channel realization for possible MU MIMO transmissions and determines the appropriate forwarding vector for each station / spatial stream. Then, it proceeds in the channel reservation phase as in the first example, but with a different MU-RTS and M-CTS frame format.

Fig. 7 shows the MU-RTS frame structure according to a second example with a variable number of receiver addresses and which is amplified by a variable number of additional Tx direction vectors.

However, there is an alternative frame structure to MU-RTS. MU-RTS can be a generic MU-RTS without Tx direction vectors. The training sequence that immediately follows MURTS can be directional according to TX direction vectors. The receivers then estimate the direction channel and feed the CSI back to the receiver. This information serves the same purpose as confirmed direction vectors forwarded by stations 21 to 24 to AP 10. This approach is more advantageous because it reduces the number of bits listed. However, the format of the expected feedback from each station must be transmitted in the MU-RTS frame, so not only stations can format the feedback, but also other stations can predict the duration of the M-CTS frame sent by the station.

Fig. 8 shows the proposed new MU-CTS frame structure according to the second example.

If the transmitter intends to send more than a spatial stream to a single station, it simply repeats the receiver address consistently as many times as required for the spatial streams of that receiver. In this case, the proposed Tx direction vector for each spatial stream will be different than expected. In this way, the transmitter can easily change the operating mode from MU MIMO transmission to SU MIMO transmission.

The proposed new M-CTS frames with reference to the second example contain a variable number of confirmed Tx direction vectors, so the variable number of streams allocated to stations can be evaluated independently. This information can also be used in the transmitter to adaptively change the direction vectors to make better use of the channel. Thus, with the above-mentioned improvements provided by the second example, the transmitter can use directional transmission to reduce interference along the spatial streams it sends.

The proposed compliance procedure protocol, e.g. by referring to the first and second examples above, may suffer from the so-called "hidden node problem". The third example shows the modification that solves this previously mentioned node problem.

Fig. 9 illustrates the hidden node problem. It is assumed that AP 10 in Fig. 1 with the CR10 communication range will want to establish MU MIMO transmission to stations 21 and 22 with the respective communication ranges CR21 and CR22 in a situation where AP 10 is placed between stations 21 and 22. Due to the location of the station 21 and 22, first station 21 may search for transmission from AP 10, but may not seek transmission from second station 22. Similarly, the second station 22 may search for transmission from AP 10, but may not search for transmission from the first station 21. This means that the M-CTS frame from station 21 cannot be received by the second station 22. If the second station 22 detects an empty channel ( i.e. there is no other transmission or interference), based on the algorithm explained in the first example, the second station 22 will send the M-CTS frame after waiting for the SIFS plus RIFS period. As a result, the M-CTS frame sent by the second station 22 will collide with the M-CTS frame sent by the first station 21 on AP 10.

Due to the possibility of some stations seeking MU MIMO reception that are not in mutual communication range, it is proposed in the third example that all stations looking for MU MIMO reception watch the end of the MU-RTS frame and calculate the duration they must wait before they start transmitting their M frames -CTS, taking into account the order of M-CTS transmission, the separation between each transmission, and the duration of each M-CTS transmission. The searching station then sends the M-CTS frame when the specified duration (starting from the time when the last transmission of the MU-RTS frame was received by each station) has expired. While waiting for the transmission of the M-CTS frame, the station does not need to read the channel, because the correct reception of the MU-RTS frame implies a reserve channel between the station and AP 10.

Referring to the third example, this improved procedure is proposed for setting the MU MIMO subordinate transmission.

As in the first example, the AP 10 initiates MU MIMO transmission by transmitting the MU-RTS frame, which is a MAC frame containing multiple receiver addresses and which now also includes field duration. The proposed field duration may include the predicted duration "d" of the entire expected transmission, from the end of the MU-RTS frame to the last MACK frame. If the number of receivers seeking MU MIMO reception is "N", T (x) is the duration of "x", where "x" is data or control frame transmitted by wireless means, then the expected duration "d" can be obtained as follows:

d = 3 * aSIFStime + N * T (M-CTS) + T (MU MIMO frames) + N * T (M-ACK) + 2 * (N-1) * aRIFStime (4) where aSIFStime is the duration of SIFS and aRIFStime is the duration of the RIFS. In addition, T (MU MIMO frame) is the estimated duration of the MU MIMO frame calculated by scheduling the MU MMO transmitter. The scheduling operation of such scheduling depends on each individual implementation.

The M-CTS frame has a fixed length. In addition, all M-CTS frames can be transmitted using the same modulation and coding scheme. Thus, the duration T (M-CTS) of the M-CTS frame is known through AP 10. Similarly, the M-ACK frame has a fixed length, and is also required to send using the same modulation, coding scheme and hence duration T (M-ACK) of the M-ACK frame is known in AP 10.

By setting the suggested field duration, all stations that do not seek MU MIMO reception but which receive the MU-RTS frame can set their network allocation vector (NAV) in relation to the field duration in the MU-RTS frame, and therefore do not transmit during reserved duration.

After receiving the MU-RTS frame, the seekers respond with the M-CTS frame, structurally e.g. as proposed in conventional M-DCF. The order of responses is determined by the order of the receivers in the MU-RTS frame address list. The first M-CTS frame is transmitted after the aSIFStime duration, and the next after the M-CTS plus aRIFStime duration. Since the transmission medium from stations seeking to AP 10 is currently reserved through MU-RTS transmission, there is no longer a need for stations 21 to 24 to read the medium before transmission.

Also, due to the possible problem of hidden nodes, the reception of preceding M-CTS frames cannot be used to signal the transmission of another M-CTS. Therefore, it is proposed to control M-CTS transmissions as follows.

After receiving the MU-RTS frame, the "n" position of the waiting station in the order of receivers is determined and the time of the "t" station must wait before transmitting its M-CTS frame which is calculated by the following equation:

t = aSIFStime + (n-1) * (T (M-CTS) + aRIFStime) (5)

The field duration in M-CTS can be obtained from the field duration of the MU-RTS frame subtracted from the value (t + T (M-CTS). The procedure in AP 10 after transmission of the MU-RTS frame may refer to the first example. MU-MIMO frames, stations can follow the same procedure in transmitting their M-ACK frames.

After receiving the MU MIMO frame, the time "t" STA must wait before transmitting its M-ACK frame and can be calculated based on the following equation: t '= aSIFStime + (n-1) * (T (M-ACK) + aRIFStime ) (6)

The duration of the field in the M-ACK frame can thus be obtained from the duration of the MU MIMO frame field minus the value (t '+ T (M-ACK)).

The proposed mechanism can also be used in the first phase of two-phase channel access in the case of MU MIMO according to the second example, namely the control of M-CF frame transmission by search stations.

In addition, the proposed mechanism can be used in the second phase of two-phase channel access for the MU MIMO described in the second example, namely to control the transmission of M-CTS frames. It should be noted that the duration of each M-CTS frame may be different due to the many spatial streams intended for a single receiver. However, this information is provided in the MU-RTS frame, and therefore the AP 10 and all seek stations may take this into account when calculating the "t" parameter.

As explained above, the MU-RTS and M-CTS frame structures defined in the first and second examples require a new MAC frame format.

In this fourth example, it is proposed to combine the MAC aggregate data unit (A-MPDU) and the emitting feedback mechanism to allow subordinate MU MIMO transmission with minimal addition and change to the current specification such as IEEE 802.11n. This is advantageous because the proposed subordinate MU MIMO is compatible with earlier standard versions.

By using the proposed combination, only the MU-RTS frame needs to be defined, while the M-CTS frame can be constructed based on the sum of conventional CTS frames and the emitting feedback frame. In addition, the more general MURTS frame can be used with some additional information required in the carried MURTS in the additional frame connected to the MPDU.

The IEEE 802.11n specification defines an A-MPDU frame, where two or more MAC MPDU or MAC frames can be transmitted in one possible transmission. To locate MPDU within A-MPDU, the delimiter precedes MPDU. In fact, this connection is intended for attaching unacceptable frame management actions (e.g. feedback frame) to the data frame or control response transmission frame (e.g. CTS).

There are several emitting mechanisms defined in IEEE 802.11n, some of which have dedicated emission emitting result frames. These frames may be CSI frames, uncompressed control frame, compressed control frame, feedback frame indicating antenna selection. The CSI frame contains channel state information for each subcarrier in each spatial stream and SNR for each stream. The uncompressed control frame includes a transfer control matrix for each subcarrier in each spatial stream, such as SNR for each spatial stream. The compressed control matrix contains the control contains the compressed control matrix format. The feedback frame indicating the antenna selection contains information for the receiver regarding the selected antennas for the next transmission.

Using the proposed cumulative MPDU and the directly specified feedback frame, the M-CTS frame proposed in the first example can be redefined as a combination of a conventional CTS frame and a feedback frame indicating the antenna selection. The proposed antenna bitmap field (PAB) can be carried in a feedback frame indicating the choice of antenna. The proposed M-CTS frame in the second preferred example can be redefined as a combination of a conventional CTS frame and a compressed or uncompressed control frame. The confirmed fields of the Tx direction vectors can be transferred in the control frame. The MIMO control field in the control frame can be set, so the "No." parameter sets the number of spatial streams intended for CTS-transmitting, the "Nc" parameter sets the number of spatial streams (or space-time streams) of the RTS-transmitting (e.g. AP 10 in Fig. 1) sends (this information can be obtained from the number of receiver addresses, or the number of long high-bandwidth training fields (HT-LTF) set on the packet preamble), the "Nb" parameter, the number of bits used to present the coefficient, is set in the same way as the number of coefficients used by the AP 10 in the MU-RTS frame, so the parameter, Ng "determines the number of carriers grouped into one.

Since the additional fields in the proposed MU-RTS in the first and second examples are the same as the additional fields in the M-CTS frame, it is also proposed to redefine the MU-RTS frame as a combination of a more general MU-RTS frame and a feedback frame indicating the choice of antenna or uncompressed feedback frame.

Fig. 10 shows a more general MU-RTS frame structure according to the fourth example. This general MU-RTS frame can also be used for other multi-available transmissions, e.g. OFDM multi-available transmission as initially mentioned.

Similar to the redefined M-CTS frame, additional fields can be carried in a feedback frame indicating the choice of antenna or control frame. It should be noted that the AP 10 can determine the feedback format through the appropriate settings in the MIMO control field.

The MU-RTS frame as proposed in the first and second examples requires confirmation of sending directional vectors or antenna bitmaps from the addressed stations. This means that the addressed stations must perform a channel measurement. Therefore, AP 10 in Fig. 1 can send training sequences, for example HT-LTF in preamble. The number of HT-LTF to send must be equal to or higher than the number of spatial streams (or space-time streams) for the AP 10 to intend to send the actual MU MIMO transmission.

In addition, it is proposed in the fourth example to use the MU-RTS / CTS message exchange as a carrier for the emitting mechanism. This emitting mechanism carried by the MU-RTS / CTS message can be used when the scheduling can process channel state information (CSI) transmitted in the M-CTS frame so quickly that the MU MIMO frame is scheduled in the selected aSIFStime after receiving the last M-CTS . If this is not relevant, a two-phase approach can be used. It should be noted that the distance between phase one and phase two has not been determined. Hence, in the second phase, confirmation of the TX area can check if the CSI in AP 10 is outdated.

Instead of sending the MU-RTS frame with the proposed directional vectors or antenna bitmaps, the AP 10 can indicate to all addressed stations on the list the channel measurement performance by setting the "not sounding" bit in the HT_SIGNAL field to 0. The combination of this bit and the MU-RTS frame may result in measuring channel through the addressed station and provide channel status information (CSI) in conjunction with the CTS frame. The format of the CSI frame sent by each station, namely the above bits Nb and Ng, can be the same for each station. These parameters can be set by AP 10 and passed through the high-bandwidth control field (HTC) contained in the MU-RTS frame by using the format control function of the frame, or specified in terms of fixed values. If the format is transferred, new fields can be entered in the HTC field, which can take over the currently reserved fields, or the existing field can be redefined. The "Nc" parameter is determined by the number of spatial dimensions AP 10 would like to emit, and this is due to the number of HT-LTF. In addition, "No." values ​​for each station can be forwarded by AP 10 so that each station can predict the duration of the CTS aggregate frame. These values ​​can also be passed through additional fields defined in the HTC field.

In this fifth example, an improved MAC protocol is proposed that supports MU MIMO transmission with the direction in the upstream direction of MU transmission systems, such as IEEE 802.11-based WLANs. The fifth example thus includes the first example for uplink scenarios where a common receiver is able to handle multiple transmitters simultaneously. The fifth example is based on a MAC mechanism in which a common receiver (e.g. AP 10 in Fig. 1) initiates transmission by transmitting a transmission call frame (C4T) to candidate transmitters (e.g., stations 21 to 23 in Fig. 1). The addressed candidate transmitters respond by sending M-RTS frames to show their transmission intentions to a common receiver with sequences preceded by training to estimate the channel at the receiver. As an alternative, training sequences can be provided in a particular preamble. The receiver estimates the channels from each transmitter and assesses the candidate transmitters in relation to their channel implementation. The receiver can also find suitable transmitting directional vectors for each transmitter and can respond to M-RTS frames with the MU-CTS frame, which can indicate which transmitters can have access to the channel through which it transmits directional vectors. Then, MU MIMO transmission can be started.

The proposed new mechanism referring to the fifth example thus provides a superior channel access mechanism for MU MIMO transmission supporting directional transmissions, where a common receiver simultaneously supports the reception of many packets from different transmitters. Thus, the spectral efficiency of the system can be increased. The embedded directional transmission mechanism ensures good coordination of MU MIMO transmission among many transmitters so that interference among spatial streams is minimized. In addition to the new mechanism, it allows switching between SU ​​and MU MIMO mode transmissions if channel implementation is not appropriate for MU MIMO transmission.

In particular, in the proposed MU MIMO mechanism, the MAC for the parent scenario of C4T, M-RTS and MU-CTS frames can be used to access the channel, and the MU-ACK frame can be used to confirm correctly received packets. Optionally, adaptive MU MIMO transmission can be provided by modifying transmit direction vectors for MU MIMO transmission only with a subset of stations whose spatial streams are correctly received to build the next MIMO frame. The decision may be based on information collected from e.g. an incorrect correction code check (such as cyclic code redundancy (CRC)) of received packets.

Next, the proposed MAC master procedure is described in greater detail based on five steps as shown in Fig. 11.

In a first step, a common receiver (e.g. AP 10 in Fig. 1) transmits a connection to the transmission frame (C4T) for initiating MU MIMO transmission on the uplink. In the C4T frame, it indicates the addresses of all master supported MU MIMO stations, which is a variable number. Alternatively, the AP 10 may decide to only search for a subset of the master, supported MU MIMO stations.

The C4T frame can also carry a request for emitted and an indication of the number of spatial dimensions to be emitted. Alternatively, the number of spatial dimensions to be emitted can be normalized for the possibility of estimating the channel to AP 10, which can be obtained from HT AP 10, which can be transmitted in beacon frames bound by corresponding frames, etc. The field duration can be set to cover the duration of the transmission up to the beginning of the MU MIMO transmission. As explained above, this duration can be obtained from the sum of the duration of the response from the station, the duration of the MU-CTS frame and the intervals separating the SIFS / RIFS frames. Because the number of stations assigned to MU MIMO transmission is not known when the C4T frame is executed, the duration of the MU-CTS frame is also not known. For the calculation of the duration of the C4T field, it is assumed that the MU-CTS frame contains the maximum number of fields and the MCS used, is the same as that used for C4T frame transmission. It should be noted that in the same frame, the CSI report format (e.g. parameters Nb, Ng, Nc, No.) for each station must be transferred. Me mechanism carrying this information may be the same as that used in the MU MIMO subordinate transmission. The duration also takes into account CSI reports. The C4T frame can be transmitted regularly. The repetition rate may then depend on the number of parent possible access points (APs) MU MIMO. The frequency may be forwarded to other APs in the beacon frame.

Fig. 12 shows an example of a C4T frame structure with multiple transmitter address fields according to the fifth example.

After each station receives a call, e.g. a C4T frame, it responds in the second step of the procedure by sending an M-RTS frame to indicate the intention to transmit to the indicated receiver. Here, the "M-RTS frame" means the M-DCF RTS frame and includes RTS frame fields and additional fields like CSI field. The M-RTS frame can be replaced by a combination of an RTS frame and a CSI feedback frame.

The order of M-RTS frames is by default determined by the order of the transmitters in the C4T frame list. The first M-RTS frame is broadcast after the SIFS period, and subsequent ones are broadcast after the corresponding RIFS periods. The physical protocol data unit (PPDU) carrying the M-RTS frame may be broadcast PPDU. The duration can be the sum of two durations at which the first duration begins the SIFS period after the end of the M-RTS transmission to the beginning of the MU MIMO transmission, and the second duration is the duration of the data frame transmission if the MCS used for the M-RTS frame could be used to forward pending data. From this field duration, AP 10 can learn about the amount of data to be sent by the station, and thus can set the field duration in the MU-CTS frame accordingly.

Fig. 13 shows the M-RTS frame structure as used in the fifth example.

After the AP 10 receives M-RTS frames from candidate broadcast stations and estimates the station channel implementation, it performs the third step of the station channel implementation procedure for the possible MU MIMO transmission and finds the appropriate broadcast direction vector for each station or spatial stream.

Then, he proceeds to book the channel by sending a MU-CTS frame in which he indicates which transmitters can have access to the channel by which they transmit directional vectors.

Fig. 14 shows an example of a proposed enhanced MU-CTS frame structure with Tx direction vectors according to the fifth example.

Alternatively, a more general MU-CTS frame can be used that does not carry the Tx direction vectors. Then, the cumulative control frame (compressed or uncompressed) can be used to transfer the Tx direction vectors. MCS used for transmission via assigned stations can also be transferred in these frames, e.g. in HTC fields. The duration field can be set to the duration of the longest spatial stream plus the SUFS period and the time needed to transmit the M-ACK frame.

In the fourth step of the procedure, stations can access the channel by using the Tx direction vectors indicated in the MU-CTS frame.

Finally, in the fifth step, after the completion of the MU MIMO master transmission, the AP 10 can send a MU-ACK frame, which confirms the successful receipt of packets transmitted simultaneously by the assigned stations.

Fig. 15 shows the corresponding MU-ACK frame structure in combination with the fifth example. This confirmation can be transferred in the field of confirmed bitmap packets (APB) whose length is equal to the number of Rx addresses in the MU-CTS frame. Successful packet reception can be confirmed, e.g. by setting the bit corresponding to the "1" broadcasting station.

With the proposed mechanism of the fifth example, the receiver can initiate and coordinate MU MIMO transmissions on the master link by finding the appropriate transmit direction vectors and providing this information to the transmitters, thus, it provides an effective channel access mechanism and interference avoidance technique for the master MU MIMO transmission .

In the sixth example, the mechanism is proposed to reduce the load in MUDCF. Most MU-DCF loads are generated by multiple corresponding M-CTS and M-ACK frames with their SIFS periods and preamble before each frame. Using a multi-channel access scheme other than time division multiple access (TDMA) significantly improves MU-DCF network performance.

In MIMO systems, spatial multiplication of frames is possible, but channel knowledge in the transmitter cannot be assumed. In OFDM systems such as IEEE 802.11a, the use of OFDMA transmission leads to the least hardware complexity. However, other regimens like MC-CDMA or CDMA may have a similar effect.

In the case of OFMDA, using e.g. subcarrier quadrants, short packets such as the MCTS and M-ACK frame, are not four times longer - because much of the frame is a preamble. Depending on the physical layer, the M-CTS and M-ACK frames are longer by a few symbols. Assuming that the packet size is 1024 bytes, the physical layer mode for 54Mb / s data packets, the 36Mb / s physical layer mode (and other important parameters as in the IEEE 802.11a standard), the transmission window has a duration of 338μs in SU mode, 578μs in MU (TDMA) mode, and 362μs in MU (OFDMA) mode.

Hence, it is proposed in the sixth example to reduce the time needed to transmit M-CTS and M-ACK frames in MU operation mode in MU-DCF. This reduces the MU load almost to that of the SU system, while maintaining the above mentioned benefits of MU-MIMO transmission.

Instead of transmitting M-CTS and M-ACK frames in TDMA mode, all subcarriers are divided into subsets and each subset is assigned to one station that must send the M-CTS and M-ACK frame.

Information regarding the mapping of the subcarrier files to the station can be determined from the order of the receivers in the address list sent in the MU-RTS frame. Hence, M-CTS and MACK frames are transmitted simultaneously, so the SIFS and preambles that precede each frame in IEEE 802.11 networks and which are load-bearing in MIMO systems are parallel. Depending on the characteristics of the physical layer, M-CTS and M-ACK frames can be only a few symbols longer.

Briefly, the transmitting device, receiving device, system and method of performing multi-access transmission to multiple transmission terminals have been described, where the request for transmission, e.g. MU-RTS frame, MU-CFR frame, or C4T frame, are transmitted to multiple transmission terminals, and where the request is provided with a MAC frame that contains a list of at least two receiving terminal identifications that are requested to respond to the request. The various benefits of further refinement and strengthening of this underlying overall concept are illustrated in the above embodiments.

It should be mentioned that the present invention is now limited to the above embodiments and can be used for any multi-user scheme, not just MU MIMO. More specifically, the invention applies to all types of MIMOs based on WLAN, especially DCF systems. The protocol works in both single-access (SU) and multi-access (MU) modes. Performance improvements over M-DCF can be expected in strongly interconnected systems and slave APs where many connections are present. In addition, the invention applies to all wireless multi-access systems with a random MAC access mechanism. You should expect an increase in the spectrum performance of a wireless network where traffic is asymmetrical and most of the traffic comes from a single transmitter like an AP in a sub-scenario or a server that distributes data to multiple stations, and / or where most of the traffic is intended for a single receiver such as an AP in the scenario slave or gateway that provides access to the external network.

The final but even more important is that it should be noted that the term "includes" or "including" when used in a specification containing claims is intended to determine the presence of identified features, meaning stages or components, but does not exclude the presence or addition of one or more other characteristics, stage, components or their group. Further, references indicating the plural of the previous element in the claims do not exclude the presence of a plurality of such elements. In addition, any reference marks do not limit the scope of the claims.