HK1170897B - Method and apparatus for determining channel quality index in multiple user-mimo communication networks - Google Patents

Description

Method and apparatus for determining channel quality index in multi-user-MIMO communication network

Cross Reference to Related Applications

Priority of united states provisional application number 61/171292 filed on 21/4/2009, this application claims priority, which is hereby incorporated by reference in its entirety.

Technical Field

The present invention relates to the exchange of channel quality information in a wireless communication system. More particularly, the present invention relates to determining a channel quality index in a communication network, such as an IEEE802.16 based network, supporting multi-user (MU) multiple-input and multiple-output (MIMO) or MU-MIMO operation.

Background

IEEE802.16 is a series of wireless broadband standards developed by IEEE (the institute of electrical and electronics engineers) and is hereby incorporated by reference in its entirety. Ieee802.16e uses OFDMA (orthogonal frequency division multiple access) to carry data. Ieee802.16e supports adaptive modulation and coding such that it is more efficient when channel quality conditions are relatively good, but may use a less robust coding scheme, such as 64QAM (quadrature amplitude modulation), in order to maximize the total amount of channel or data that can be transmitted (transmit) per unit time. However, it is less efficient when the channel quality conditions are relatively poor, but a more robust modulation and coding scheme, e.g., BPSK (binary phase shift keying), may be used in order to improve the channel quality (but at the cost of a lower data transmission rate). Any number of different modulation schemes may be supported depending on the channel conditions. For example, in the best case, a 64QAM modulation scheme may be selected, and in the worst case, a BPSK modulation scheme may be selected. Other intermediate modulation schemes with an intermediate trade-off between efficiency and robustness, such as 16QAM, QPSK, etc., may be selected when the channel conditions are between the two extremes.

IEEE802.16 also supports hybrid automatic repeat request (HARQ) for improved error detection and correction performance.

Another feature of IEEE802.16 is MIMO operation. MIMO is a feature in which transceivers in a network each have multiple transmit and/or receive antennas to allow a directional beam pattern to be achieved by the transmitter in the direction of a desired receiver and/or vice versa, thereby improving the signal strength to the desired receiver without increasing the power at the transmitter. MIMO also tends to reduce interference between channels and improve NLOS (non line of sight) characteristics.

Single-user MIMO or SU-MIMO refers to the use of MIMO in which only one transmitter and receiver communicates on any given communication resource unit at any given time. Multi-user MIMO or MU-MIMO is a technique in which a transmitter (e.g., a base station) uses MIMO to simultaneously transmit two different signals to two different receivers in the same communication resource unit. More specifically, since the transmitter can beamform the transmit beam, it can direct data/signals intended for one receiver (e.g., a mobile station) to that receiver and data/signals intended for another receiver to that other receiver. Two receivers may share the same channel if the two beams are directed in completely different directions and the two receivers are far enough away from the transmitter that the data signal intended for the receiver at each receiver is substantially stronger than the data/signal intended for the other receiver operating on the communication resource unit.

As is well known, in OFDM, each downlink communication channel between a base station and a mobile station actually comprises a particular subset of subcarrier frequencies and a particular subset of time slots (e.g., within a frame) in the overall time and frequency spectrum available to the network. As described above, in MU-MIMO, a base station transmits to two (or more) mobile stations within the same subset of time slots and frequencies. In other words, in MU-MIMO, the base station supports two different communication channels on the same subset of time slots and frequencies. This is why the term "communication resource unit" is used herein to refer to a given subset of time slots and frequencies, and the term "communication channel" is used to refer to a single downlink with a single mobile station, in order to avoid confusion in this specification.

Networks that support MU-MIMO operation will often operate in SU-MIMO until the call load exceeds a certain threshold; some or all of the calls are then switched to MU-MIMO operation in order to support a greater number of simultaneous calls, albeit of lower quality. In some networks, when switching from SU-MIMO operation to MU-MIMO operation, the communication resource units may remain the same size (i.e., occupy the same number of subcarrier frequencies and time slots per unit time) and be shared by exactly two (or more) mobile stations. However, in some networks, the communication resource units for MU-MIMO operation may be of a different size (possibly larger, occupying more subcarrier frequencies and/or time slots per unit time) than the SU-MIMO communication resource units in order to mitigate communication quality degradation for each mobile station sharing the MU-MIMO communication resource units, but at the expense of a smaller increase in overall call capacity.

In wireless communication networks with Adaptive Modulation and Coding (AMC), such as IEEE802.16, it is typically the base station that decides which modulation and coding scheme to use. As mentioned above, such a decision is typically based on some measure of the quality of the channel, in particular the downlink channel. Since the base station cannot directly measure the quality of its own downlink channel, the mobile station adapts to determine the downlink channel quality, for example, by observing pilot bits in the downlink channel, bit error correction rates on received data, etc., and then sends channel quality data back to the base station on a control channel. For example, in IEEE802.16, a mobile station transmits a parameter called CQI (channel quality index) to a base station. There are several techniques for generating and transmitting CQI in IEEE802.16, which include codebook-based CQI and sounding-based CQI. In codebook-based CQI, there are a limited number of possible CQI values, each corresponding to a channel condition. Both the mobile station and the base station have codebooks that disclose the meaning of these possible CQI values. The mobile station sends the CQI value to the base station, which requires very little bandwidth, and the base station inserts the CQI value into the codebook to determine what it means. For example, the codebook may reveal that a CQI value of 4 indicates that the base station should select a modulation scheme of 16QAM with a coding rate of 1/2, and reveal others.

However, when in MU-MIMO mode, each mobile station may have to transmit a large amount of CQI data because the mobile stations do not have any information about the other mobile stations with which they share communication resource units.

Disclosure of Invention

To minimize control signaling overhead associated with transmitting CQI data from a mobile station to a base station in a wireless communication network supporting MU-MIMO, CQI during MU-MIMO operation is estimated based on SU-MIMOCQI data, mobile station geometry data, and mobile station PMI (precoding matrix index) data. More specifically, the base station maintains and updates a knowledge pool that correlates the impact of geometry data and learned interfering precoder data with the degradation of CQI values in response to switching from SU-MIMO operation to MU-MIMO operation. Then, when the base station switches from SU-MIMO operation to MU-MIMO operation, it consults the knowledge pool, predicts CQI degradation for each relevant mobile station and subtracts the CQI degradation from the known pre-switched SU-MIMO CQI feedback data to predict a post-switched MU-MIMO CQI for that mobile station.

This saves a very large amount of overhead communication associated with CQI and PMI feedback, since in MU-MIMO the mobile station does not know the identity of other mobile stations with which it shares communication resource units in MU-MIMO operation. Thus, each mobile station will need to transmit a large amount of CQI data, e.g., the best CQI and the worst CQI with respect to different interfering mobile stations, in order for the base station to determine the channel quality in the MU-MIMO channel.

Drawings

Fig. 1 is a block diagram representing an exemplary cellular communication network in which the present invention may be implemented.

Fig. 2 is a block diagram of a logical decomposition of an exemplary OFDM transmitter architecture.

Fig. 3 is a block diagram of a logical decomposition of an exemplary OFDM receiver architecture.

Fig. 4 is a graph illustrating the correlation between geometry and CQI degradation in response to a switch from SU-MIMO to MU-MIMO operation.

Fig. 5 is a flowchart illustrating an operation of a base station according to a specific embodiment of the present invention.

Detailed Description

Fig. 1 illustrates the basic components of an exemplary cellular communication network implementing the present invention. A Base Station Controller (BSC)10 controls wireless communications within a plurality of cells 12, which are served by respective Base Stations (BS) 14. In some configurations, each cell is further divided into a plurality of sectors 13 or regions (not shown). In general, each base station 14 facilitates communication using OFDM with mobile and/or wireless terminals (hereinafter, mobile stations) 16 within the cell 12 associated with the respective base station 14. Movement of the mobile station 16 relative to the base station 14 results in significant fluctuations in channel conditions. As illustrated, the base station 14 and the mobile station 16 each include multiple antennas to provide spatial diversity for communications. In some configurations, a relay station 15 may be provided to assist in communications between the base station 14 and the mobile station 16. A mobile station 16 may be handed off from any cell 12, sector 13, area (not shown), base station 14, or relay 15 to another cell 12, sector 13, area (not shown), base station 14, or relay 15. In some configurations, the base stations 14 communicate with each other and with another network (e.g., a core network or the internet, both not shown) via the backhaul network 11. In some configurations, the base station controller 10 is not required.

In OFDM modulation, the transmission band for each communication resource unit is divided into multiple orthogonal carriers. Each carrier wave is modulated according to the digital data to be transmitted. Since OFDM divides a transmission band into a plurality of carriers, a bandwidth per carrier is reduced and a modulation time per carrier is increased compared to a single carrier technique. Since multiple carriers are transmitted in parallel, the rate of transmission of digital data or symbols on any given carrier is lower than for single carrier techniques.

OFDM modulation utilizes an Inverse Fast Fourier Transform (IFFT) on the information to be transmitted. For demodulation, performing a Fast Fourier Transform (FFT) on the received signal recovers the transmitted information. In practice, the IFFT and FFT are provided by digital signal processing, respectively, executing suitable algorithms. Accordingly, a characterizing feature of OFDM modulation is the generation of orthogonal carriers for multiple bands within a transmission resource unit. The modulated signals are digital signals having a relatively low transmission rate that can be retained in their respective bands. The individual carriers are not directly modulated by the digital signal. Instead, the carriers are modulated together by the IFFT algorithm.

In operation, OFDM is preferably used at least for downlink transmissions from the base station to the mobile station. Each base station 14 is equipped with n transmit antennas and each mobile station is equipped with m receive antennas. It should be noted that the respective antennas can be used for reception as well as transmission with a suitable duplexer or switch.

Referring to fig. 2, a logical OFDM transmission architecture will be described. Initially, the base station controller 10 will send to the base station 14 data intended to be sent by the base station 14 to the respective mobile station 16, either directly or by means of the relay station 15. The base station 14 may use the CQI associated with the mobile station to schedule data for transmission and select an appropriate modulation and coding scheme for transmitting the scheduled data. The CQI may be received directly from the mobile station 16 or may be determined at the base station 14 based on information provided by the mobile station 16. In either case, the CQI for each mobile station 16 is a function of the degree to which the channel amplitude (or response) varies across the OFDM frequency band.

The scheduled data 44 is scrambled by data scrambling logic 46 in a manner that reduces a peak-to-average power ratio associated with the data, where the scheduled data 44 is a bit stream. A Cyclic Redundancy Check (CRC) for the scrambled data is determined by CRC addition logic 48 and appended to the scrambled data.

Channel coding is performed by the channel encoder logic block 50 to effectively add redundancy to the data to facilitate recovery and error correction at the mobile station 16. Again, the channel coding for a particular mobile station 16 is based on CQI. In some implementations, the channel encoder logic block uses known Turbo coding techniques. The encoded data is then processed by rate matching logic block 52 to compensate for data spreading associated with the encoding.

The bit interleaver logic 54 systematically reorders the bits in the encoded data to minimize the loss of consecutive data bits. The resulting data bits are systematically mapped to corresponding symbols according to the selected baseband modulation by mapping logic 56. Preferably, QAM or QPSK modulation is used. However, if the channel quality is particularly poor, BSK or other highly robust modulation techniques may also be used. The modulation degree is preferably selected based on the CQI determined for a particular mobile station. The symbols may be systematically reordered to further enhance the immunity of the transmitted signal to periodic data loss due to frequency selective fading (fade) by the symbol interleaver logic 58.

At this point, the bit groups have been mapped to symbols representing positions in the amplitude and phase constellation. When spatial diversity is desired, the symbol packets are then processed by a space time block code (STC) encoder logic block 60, which modifies the symbols in a manner that makes the transmitted signal more resistant to interference and easier to decode at the mobile station 16. The STC encoder logic 60 will process the incoming signal and provide n outputs corresponding to the number of transmit antennas of the base station 14. The control system 20 and/or baseband processor 22 will provide a mapping control signal that controls STC encoding. At this time, the symbols for the n outputs represent data to be transmitted and can be recovered by the mobile station 16.

Assuming that the base station has two antennas 28, i.e., n-2, then the STC encoder logic block outputs two symbol streams. Accordingly, each of the symbol streams output by the STC encoder logic 60 is sent to a respective IFFT processor 62, the IFFT processors 62 being shown separately for ease of understanding. Those skilled in the art will appreciate that one or more processors may be used to provide such digital signal processing, either alone or in combination with other processors described herein. The IFFT processors 62 will preferably operate on the respective symbols to provide their inverse fourier transforms. The output of the IFFT processors 62 provides symbols in the time domain. The time domain symbols are grouped into frames associated with prefix-by-prefix insertion logic 64. Each of the resulting signals is up-converted to an intermediate frequency in the digital domain and converted to an analog signal by a corresponding digital up-conversion (DUC) and digital-to-analog (DA) conversion circuit 66. The resulting (analog) signal is then simultaneously modulated at the desired RF frequency, amplified, and transmitted (radiated) via the RF circuitry 68 and antenna 28. It should be noted that the pilot signals known to the intended mobile station 16 are scattered among the subcarriers. The mobile station 16, discussed in detail below, will use the pilot signal for channel estimation.

Fig. 3 illustrates the process of receiving the transmitted signal at the mobile station 16 either directly from the base station 14 or by means of the relay 15. Once the transmitted signals reach each of the antennas 40 at the mobile station 16, the respective signals are demodulated and amplified by respective RF circuitry 70. For the sake of brevity and clarity, only one of the two receiver halves (receiverhalf) is described and illustrated in detail. Analog-to-digital (a/D) converter and down-conversion circuitry 72 digitizes and down-converts the analog signals for digital processing. The resulting digitized signal may be used by an automatic gain control circuit (AGC)74 to control the gain of the amplifiers in the RF circuitry 70 based on the received signal level.

Initially, the digitized signal is provided to synchronization logic 76, the synchronization logic 76 including coarse synchronization logic 78, the coarse synchronization logic 78 buffering a number of OFDM symbols and calculating an autocorrelation between two consecutive OFDM symbols. The resulting time index corresponding to the maximum value of the correlation result determines a fine synchronization search window that is used by fine synchronization logic block 80 to determine the precise frame start position based on the header. The output of the fine synchronization logic 80 facilitates frame acquisition by the frame location logic 84. Proper frame positioning (alignment) is important so that the subsequent FFT processing provides an accurate conversion from the time domain to the frequency domain. The fine synchronization algorithm is based on the correlation between the received pilot signal carried by the header and the local copy of the known pilot data. Once frame alignment acquisition occurs, the prefix of the OFDM symbol is removed by prefix removal logic block 86 and the resulting samples are sent to frequency offset correction logic block 88, and frequency offset correction logic block 88 compensates for the system frequency offset caused by the unmatched local oscillators in the transmitter and receiver. Preferably, the synchronization logic 76 includes frequency offset and clock estimation logic 82, the frequency offset and clock estimation logic 82 being header based to help estimate this effect on the transmitted signal and provide these estimates to the correction logic 88 to properly process the OFDM symbols.

At this point, the OFDM symbol in the time domain is ready for conversion to the frequency domain using FFT processing logic block 90. The result is a frequency domain symbol that is sensed to processing logic 92. The processing logic block 92 extracts the scattered pilot signals using scattered pilot extraction logic block 94, determines a channel estimate based on the extracted pilot signals using channel estimation logic block 96, and provides channel responses for all subcarriers using channel reconstruction logic block 98. To determine the channel response for each subcarrier, the pilot signal is essentially a plurality of pilot signals dispersed in a known pattern in the time and frequency domains in the data symbols throughout the OFDM subcarriers.

The processing logic block 92 compares the received pilot symbols with the pilot symbols expected in a particular subcarrier at a particular time slot to determine the channel response for the subcarrier that sent the pilot symbol. The results are interpolated to estimate the channel response for most, if not all, of the remaining subcarriers within which pilot symbols are not provided. The actual and interpolated channel responses are used to estimate the overall channel response, which includes the channel responses for most, if not all, of the subcarriers in the OFDM channel.

The frequency domain symbols and the channel reconstruction estimates (derived from the channel response from each receive path) are provided to the STC decoder 100. The STC decoder 100 provides STC decoding on both receive paths to recover the transmitted symbols. The channel reconstruction information provides equalization information to the STC decoder 100 sufficient to remove the effects of the transmission channel when processing the respective frequency domain symbols.

The recovered symbols are put back in order using a symbol deinterleaver logic block 102, the symbol deinterleaver logic block 102 corresponding to the symbol interleaver logic block 58 of the transmitter. The de-interleaved symbols are then demodulated or de-mapped into a corresponding bit stream using de-mapping logic block 104. The bits are then deinterleaved using a deinterleaver logic block 106, the deinterleaver logic block 106 corresponding to the bit interleaver logic block 54 of the transmitter architecture. The deinterleaved bits are then processed by a de-rate matching logic block 108 and presented to a channel decoder logic block 110 to recover the original scrambled data and CRC checksum. Accordingly, the CRC logic block 112 removes the CRC checksum, checks the scrambled data in a conventional manner, and provides it to descrambling logic block 114 for descrambling using a known base station descrambling code to recover the originally transmitted data 116.

In parallel with recovering the data 116, a CQI, or at least information sufficient to allow the base station 14 to create a CQI, is determined and transmitted to the base station 14. As described above, the CQI may be a function of the carrier-to-interference ratio (CR) and the degree to which the channel response varies across subcarriers in the OFDM band. For this embodiment, the channel gains for each subcarrier in the OFDM band will be used to transmit information that is compared against each other to determine the degree to which the channel gains vary across the OFDM band. While a variety of techniques may be used to measure the degree of variation, one technique is to calculate the standard deviation of the channel gain for each subcarrier over the entire OFDM band used to transmit the data.

Fig. 1-3 provide one specific example of a communication system that may be used to implement embodiments of the present invention. It should be understood that embodiments of the present invention can be implemented with communication systems having architectures that differ from the specific example, but that operate in a manner compatible with the implementation of the embodiments described herein.

Many wireless telecommunication systems and standards, including IEEE802.16, 3GGP Long Term Evolution (LTE), and UMB, have adopted or at least supported codebook-based closed-loop MIMO, which, as described above, allows mobile units to transmit very small segments of CQI data that the base station can use to convert to robust information about channel quality.

Also as mentioned earlier, IEEE802.16, LTE, and UMB also support multi-user (MU) MIMO. MU-MIMO allows a network to increase the number of communication channels that can be simultaneously supported by sharing a single OFDM communication resource unit between two or more mobile stations using essentially Spatial Division Multiple Access (SDMA).

In codebook-based MU-MIMO, mobile stations feed their CQI and Precoding Matrix Index (PMI) values back to the base station for use by the base station, for determining what modulation and coding scheme to use to communicate with the respective mobile station, and so on.

Precoding refers to the process of beamforming with multiple transmit antennas by appropriately weighting the signals provided to each transmit antenna so that the signal strength is maximized at the receiver. Thus, the PMI is a data set that the mobile station sends to the base station to inform the base station which specific precoding scheme the mobile station wishes the base station to use for downlink communication from the base station to the mobile station. It should be noted, however, that many communication networks also allow a base station to ignore (override) a mobile station's precoding scheme request and select its own precoding scheme and inform the mobile station of the precoding scheme so that the mobile station can correctly receive data from the base station. In any case, the CQI and PMI transmitted from the mobile station to the base station indicate channel quality.

In MU-MIMO, the mobile station is unaware of the identity of the other mobile station or stations with which it shares the OFDM communication resource unit. Therefore, a large amount of CQI feedback must be sent to provide the base station with appropriately enough information to make informed (wellformed) decisions about which modulation and coding schemes to utilize with these mobile stations. For example, the information may include a plurality of CQIs for SU-MIMO and MU-MIMO, respectively, e.g., SU-CQI, best MU-CQI, worst MU-CQI, possibly a plurality of best MU-CQI and worst MU-CQI for different possible precoding schemes for communicating with other mobile stations. (hereinafter, the other mobile station or stations with which a given mobile station shares an OFDM communication resource element will sometimes be referred to as an interference precoder or multiple interference precoders.) this increased control signaling overhead for transmitting CQI and PMI control data is undesirable.

According to the present invention, the mobile station is not required to transmit large amounts of CQI feedback data when entering and being in MU-MIMO mode, but rather the base station incorporates knowledge pool circuitry that maintains and updates a knowledge pool that correlates mobile station geometry information and/or learned information about the impact of interfering precoders with expected CQI degradation. The expected CQI degradation is then combined with CQI feedback data from the mobile station while in SU-MIMO operation to predict its MU-MIMO CQI.

More specifically, each base station maintains a knowledge pool that includes a mapping from interfering precoder information and user geometry information to SU-mimo cqi feedback degradation. Information about the impact of interfering precoders may be obtained from HARQ (hybrid automatic repeat request) statistics received during SU-MIMO and MU-MIMO operation, as will be described in more detail below.

The geometry information may be any available information indicative of the channel quality, such as SNR (signal to noise ratio) or BER (bit error rate). In Kumar, s.; monthal, g.; nin, j.; ordas, I.; pedersen, k.i.; in one embodiment disclosed in Mogensen, P.E, AutonomousIntercellInterferencAvoidedanceundersiderFractionLoadynaudLockLongTermEvention, VehicularTechnologyConference, 2009, VTCSpring2009, IEEE69th, published12June2009(ISSN 1550-2252; PrintISBN: 978-1-4244-_S) Total inter-cell interference (P) averaged with fast fading_I) And noise (P)_N) The ratio of the sums, i.e.:

geometric (or G-factor) P_S/(P_I+P_N)

Fig. 4 is a graph illustrating an exemplary correlation between geometry and CQI degradation when switching from SU-MIMO to MU-MIMO. For example, when the geometry is low (e.g., -20dB), the average CQI degradation when switching from SU-MIMO to MU-MIMO operation is 3 dB. When the geometry is 0dB, switching from SU-MIMO to MU-MIMO results in 4.5dB of average CQI degradation. With a geometry of 10dB, the handover results in an average CQI degradation of almost 6 dB. Thus, it can be seen that the geometry of the mobile station has an effect on the amount of expected CQI degradation when the system switches from SU-MIMO to MU-MIMO.

The impact of different interfering precoders on CQI can be gathered by having the base station compare the HARQ statistics in the first OFDM frame after it switches from SU-MIMO to MU-MIMO operation with the HARQ statistics of the last frame before switching from SU-MIMO to MU-MIMO operation. The increase in automatic repeat request (ARQ) between these two frames is typically strongly correlated with CQI degradation due to interfering precoders.

In operation, each mobile station feeds back its preferred single-user PMI data and its CQI data to the base station during SU-MIMO. When the base station decides to schedule MU-MIMO operation, it pairs two or more mobile stations to share MU-MIMO resource units and adjusts the CQIs of the two mobile stations as indicated by the knowledge pool mapping. A resource unit is a particular combination of OFDM subcarriers and time slots. May be the same unit as used for SU-MIMO or may be a different unit.

More specifically, circuitry in the base station receives and stores the SU-mimo cqi received from each of the mobile stations in its cell. The circuitry also receives and stores preferred single-user PMIs reported by each of the mobile stations. Finally, the mobile stations can also be configured to transmit geometry information (e.g., the aforementioned G-factor) to the base station, such that the base station also receives and stores the geometry information for each mobile station. The geometry information may be determined, for example, at the time of network entry and at other suitable times after network entry (e.g., at fixed intervals and/or upon the occurrence of certain defined events).

When the base station decides to initiate MU-MIMO operation, it knows the PMIs and CQIs for all mobile stations in its cell, and can use this information to determine which mobile stations to combine in the MU-MIMO resource unit. At this point, for each mobile station to be placed in MU-MIMO mode with one or more other mobile stations, prediction circuitry at the base station is capable of inserting into the knowledge pool the most recent geometry information for that mobile station and the most recent preferred PMIs received from other mobile stations with which that mobile station will share OFDM communication resource units, and determining the predicted CQI degradation as a function of the interference precoding information and the geometry information.

The base station then takes the last reported SU-mimo CQI value for the mobile station and subtracts the expected CQI degradation from it to get the expected MU-mimo CQI for the mobile station. This is done for each mobile station sharing a communication resource unit. A precoding selection circuit in the base station will then select a modulation and coding scheme from the CQI codebook for downlink transmission to the mobile station based on the expected CQI. Optionally, the expected CQI may apply some further offset or correction factor.

In one embodiment, a knowledge pool can be maintained separately for each mobile station. Alternatively, data for multiple mobile stations can be commonly maintained in a single knowledge base, and knowledge pool information for multiple mobile stations can be used to predict CQI degradation for individual mobile stations when switching from SU-MIMO mode to MU-MIMO mode. In addition, each mobile station can maintain its own knowledge pool or the entire network or a portion thereof (e.g., a BSC), and can collect information for each mobile station across multiple base stations/cells or the entire network.

The characteristic of the knowledge pool that correlates interfering precoder information and/or geometry information with CQI degradation can take a variety of forms, in accordance with the principles of the present invention. In one simple example, the expected SU-MIMO to MU-MIMO CQI degradation as a function of the interfering precoder modulation scheme may be maintained in one table, the expected CQI degradation as a function of the geometry factor may be maintained in another table, and the two CQI degradation numbers from the tables can simply be added to obtain an overall predicted CQI degradation value.

In more complex implementations, algorithms may be utilized for factors that would interfere with any correlation of the precoder modulation scheme and the geometry factor to each other in affecting CQI.

Fig. 5 is a diagram conceptually illustrating an operation of a base station according to the principle of one embodiment of the present invention. It should be understood that the flow chart is conceptual in order to illustrate the main processes of the present invention, and does not represent the actual operation of the base station process, which would include many other processes and steps. Further, some "steps" in the flowcharts represent actions that occur continuously or are driven with interruptions such that they do not occur at a particular time. In fact, the overall sequence of steps in the flow chart is exemplary only.

The base station continuously maintains and updates a knowledge pool associating interfering PMIs and geometry information with CQI degradation, as shown in step 501.

In step 503, the base station receives and stores the preferred PMIs for SU-MIMO operation requested by each mobile station within the corresponding cell. Under normal operating conditions, the base station performs this step virtually continuously.

Next, in step 505, the base station receives and stores CQI from each mobile station in a corresponding cell. Again, this is an operation that may occur normally and continuously in the base station as part of the normal operating protocol.

Next, in step 507, the base station reads the geometry factor received from each mobile station. This may also be performed continuously by the base station under normal operating conditions.

Next, in step 509, when the base station decides to switch from SU-MIMO mode to MU-MIMO mode with respect to one or more OFDM communication resource units, the flow will continue until steps 511, 513, 515 and 517. On the other hand, if the base station does not decide to switch to the MU-MIMO mode, step 511 and 517 will not be performed. Instead, the base station will simply continue to operate in normal SU-MIMO mode, which is simply represented in the figure as a flow branch back to first step 501.

When the call load is below a predetermined threshold, the base station will typically operate in SU-MIMO mode in order to provide the highest quality reception in each channel. However, if the load exceeds a certain threshold (e.g., the number of available channels), it will switch to MU-MIMO mode in order to serve the increased call load, although this is likely to have a lower channel quality. In this example, we will assume that in MU-MIMO mode, only two mobile stations share OFDM communication resource units in MU-MIMO.

Thus, in step 511, the base station checks the CQI, PMI and geometry of the two mobile stations that are to share a given OFDM communication resource unit. Next, in step 513, for each of these mobile stations, the base station consults a knowledge pool to determine a predicted CQI degradation according to the following function: (1) the geometry factor of the mobile station and (2) the interfering PMI of another mobile station. Next, in step 515, the base station subtracts the predicted CQI degradation from the last recorded SU-mimo CQI for the mobile station to determine a new MU-mimo CQI for the mobile station. The base station then selects a modulation scheme and a coding scheme for each of these mobile stations as a function of the predicted MU-mimo cqi, and so on.

Finally, in step 517, for each mobile station sharing the communication resource unit, the base station compares the HARQ information in the first frame after switching to MU-MIMO mode with the HARQ information for the most recent SU-MIMO frame, estimates the effect of the interfering precoders of the other mobile stations on the CQI of each of these mobile stations, and updates the interfering precoder information in the knowledge pool accordingly.

It will be appreciated that the present invention may be implemented within a base station, relay station or other node of a network. The processing disclosed herein may be performed entirely within a single node of the network or may be distributed among a number of different nodes, such as one or more base stations and Base Station Controllers (BSCs). Further, the processes may be implemented by any number of circuit types, including but not limited to programmed general-purpose computers, digital signal processors, combinational logic circuits, analog circuits, Application Specific Integrated Circuits (ASICs), firmware, hardware, software, and various combinations of any of the foregoing. Further, it will be understood that many of the steps disclosed herein may be performed by devices that conventionally exist in base stations or other nodes of a network. As just one example, the transmission and reception of the various data discussed herein may be performed using existing transmitters, receivers, data processors, antennas, and other devices already present in the nodes of the wireless network for performing all other data and control signal communications in the network.

The present invention does not necessarily require any modification to the mobile station to operate in accordance with the present invention. However, other mobile stations can be readily modified to operate in accordance with the principles of the present invention.

Having thus described some particular embodiments of the invention, various alterations, modifications and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements as are made obvious by this disclosure are intended to be part of this description though not expressly stated herein, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description is by way of example only and is not limiting. The invention is limited only as defined in the following claims and the equivalents thereto.

Claims (18)

1. A base station in a wireless communication network for wirelessly communicating with a plurality of mobile stations, the base station supporting multi-user multiple-input and multiple-output MU-MIMO communication, comprising:

a transmitter;

a receiver;

a plurality of antennas;

a knowledge base circuit that maintains a knowledge base of data indicative of channel quality for each mobile station and a correlation of a precoding scheme for each other mobile station and a channel quality index, CQI, between the base station and each mobile station for differences in single user MIMO operation versus multi-user MIMO operation;

determining circuitry to determine a precoding scheme selected by each mobile station for single-user MIMO operation, a CQI for single-user MIMO for each mobile station, and data indicative of channel quality for each mobile station, wherein the determining circuitry is further to determine one or more mobile stations to share a first communication resource unit comprising orthogonal frequency division multiplexing, OFDM, subcarriers and slot data when the one or more mobile stations are operating in MU-MIMO operation; and

prediction circuitry, responsive to switching a first mobile station of the pair of mobile stations from single-user MIMO operation to multi-user MIMO operation, to use the knowledge base to predict CQI degradation for each of the first and second mobile stations based on precoding scheme data and data indicative of channel quality corresponding to each of the mobile stations sharing the first communication resource unit.

2. The base station of claim 1, wherein for each mobile station sharing the first communication resource unit in MU-MIMO operation, the prediction circuitry uses a knowledge base to determine an effect on the prediction of CQI of: (a) data indicative of channel quality for that mobile station and (b) interference effects of the selected single-user precoding scheme for each other mobile station sharing the first communications resource element.

3. The base station of claim 2, further comprising:

precoder selection circuitry to select a modulation scheme for downlink communication for each of the mobile stations sharing the first communication resource unit in MU-MIMO operation as a function of the respective predicted CQI degradation for that mobile station and the CQI for single-user MIMO.

4. The base station of claim 1, wherein the base station uses orthogonal frequency division multiple access in a downlink communication channel.

5. The base station of claim 1 wherein the knowledge base circuit maintains a separate knowledge base for each mobile station.

6. The base station of claim 1 wherein the knowledge base circuit maintains a common knowledge base for a plurality of mobile stations.

7. The base station of claim 1, wherein the data indicative of channel quality for each mobile station comprises a geometry factor received from the mobile station via the antenna, the geometry factor being a ratio of an expected received signal power received by one mobile station to a sum of an average total inter-cell interference and noise.

8. The base station of claim 1, wherein the data indicative of channel quality for each mobile station is obtained upon entry into the wireless communication network.

9. The base station of claim 1, further comprising:

circuitry for receiving a hybrid automatic repeat request, HARQ, from a mobile station in communication therewith; and

circuitry for:

(a) comparing HARQ statistics for communications with the first mobile station under single-user MIMO operation with HARQ statistics for communications with the first mobile station under multi-user MIMO operation sharing communication resource elements with at least one second mobile station, the at least one second mobile station using a particular precoding scheme to generate comparison data; and

(b) estimating an effect of a particular precoding scheme of the at least one second mobile station on the CQI of the first mobile station using comparison data, wherein the data in the knowledge base regarding the correlation of the CQI with the precoding scheme of the at least one second mobile station with which the first mobile station shares the communication resource unit is generated at least in part from the comparison data.

10. The base station of claim 1, further comprising:

circuitry for receiving a hybrid automatic repeat request, HARQ, from a mobile station in communication therewith; and

circuitry for:

(a) comparing HARQ statistics for communications with a first mobile station under single-user MIMO operation with HARQ statistics for communications with the first mobile station in a first frame after entering multi-user MIMO operation sharing communication resource units with at least one second mobile station, the at least one second mobile station using a particular precoding scheme to generate comparison data; and

(b) estimating an effect of a particular precoding scheme of the at least one second mobile station on the CQI of the first mobile station using the comparison data, wherein the data in the knowledge base regarding the correlation of the CQI with the precoding scheme of the at least one second mobile station with which the first mobile station shares the communication resource unit is generated at least in part from the HARQ statistical comparison data.

11. A method of predicting downlink channel quality between a base station and a mobile station in a wireless communication network in multi-user multiple-input multiple-output, MU-MIMO, operation, the method comprising:

maintaining a knowledge base of data indicative of channel quality for each mobile station and at least a correlation of a precoding scheme for each other mobile station and a difference in channel quality index, CQI, between the base station and each mobile station for single user MIMO operation versus multi-user MIMO operation;

determining a precoding scheme selected by each mobile station for multi-user MIMO operation, CQI for single-user MIMO for each mobile station, and data indicative of channel quality for each mobile station;

determining one or more mobile stations to share a first communication resource unit when the one or more mobile stations are operating in MU-MIMO operation, the first communication resource unit comprising Orthogonal Frequency Division Multiplexing (OFDM) subcarriers and slot data; and

in response to switching a first mobile station of the pair of mobile stations from single-user MIMO operation to multi-user MIMO operation, predicting CQI degradation for each of the first and second mobile stations based on precoding scheme data and data indicative of channel quality corresponding to each of the mobile stations sharing the first communication resource unit using the knowledge base; and

selecting a modulation scheme for downlink communication for each of the mobile stations sharing the first communication resource unit as a function of the predicted CQI degradation and the CQI of the single-user MIMO.

12. A method as defined in claim 11, wherein, for each mobile station sharing the first communication resource unit in MU-MIMO operation, the mapping uses a knowledge base to predict an effect on CQI of: (a) data indicative of channel quality for that mobile station and (b) interference effects of the selected precoding scheme for each other mobile station sharing the first communications resource element.

13. The method of claim 11, wherein selecting a modulation scheme comprises selecting a modulation scheme according to another function of CQI for single-user MIMO for the respective mobile station.

14. The method of claim 11, further comprising:

comparing HARQ statistics for communications between the base station and the first mobile station under single-user MIMO operation with HARQ statistics for communications between the base station and the first mobile station under multi-user MIMO operation sharing communication resource units with at least one second mobile station, the at least one second mobile station using a particular precoding scheme to generate comparison data; and

estimating an effect of a particular precoder scheme of the at least one second mobile station on the CQI of the first mobile station using comparison data, wherein the data in the knowledge base regarding the correlation of the CQI to the precoding scheme of the at least one second mobile station with which the first mobile station shares the communication resource unit is generated at least in part from the comparison data.

15. A method of estimating downlink channel quality between a base station and a mobile station in a wireless communication network when in multi-user multiple-input multiple-output, MU-MIMO, operation, the network using orthogonal frequency division multiple access, OFDMA, modulation in the downlink communication channel and supporting single-user MIMO su-MIMO and MU-MIMO communication protocols, the method comprising:

maintaining a knowledge base of the data indicative of the channel quality of the mobile stations and the correlation of at least the precoding scheme of the second mobile station and the difference of the downlink communication channel quality index, CQI, between the base station and each mobile station for SU-MIMO operation versus MU-MIMO operation;

determining a precoding scheme selected by each mobile station for multi-user MIMO operation, CQI for single-user MIMO for each mobile station, and data indicative of channel quality for each mobile station;

determining one or more mobile stations to share a first communication resource unit when the one or more mobile stations are operating in MU-MIMO operation, the first communication resource unit comprising Orthogonal Frequency Division Multiplexing (OFDM) subcarriers and slot data; and

in response to switching a first mobile station of the pair of mobile stations from SU-MIMO operation to MU-MIMO operation, predicting CQI degradation for each of the first and second mobile stations based on precoding scheme data and data indicative of channel quality corresponding to each of the mobile stations sharing the first communication resource unit using the knowledge base.

16. The method of claim 15, wherein the mapping uses a knowledge base to predict an impact on CQI of the following due to switching from single-user MIMO to multi-user MIMO for each mobile station sharing the first communication resource unit while in MU-MIMO operation: (a) data indicative of channel quality for that mobile station and (b) interference effects of the selected precoding scheme for each other mobile station sharing the first communications resource element.

17. The method of claim 15, further comprising:

selecting a modulation scheme for downlink communications in the first communication resource for each of the mobile stations sharing the first communication resource unit when in MU-MIMO operation as a function of the predicted CQI degradation for that mobile station and the CQI for SU-MIMO.

18. The method of claim 15, further comprising:

comparing HARQ statistics for communications between the base station and the first mobile station when in single-user MIMO operation with HARQ statistics for communications between the base station and the first mobile station when in multi-user MIMO operation sharing communication resource units with at least one second mobile station, the at least one second mobile station using a particular precoding scheme to generate comparison data; and

estimating an effect of a particular precoder scheme of the at least one second mobile station on the CQI of the first mobile station using comparison data, wherein the data in the knowledge base regarding the correlation of the CQI to the precoding scheme of the at least one second mobile station with which the first mobile station shares the communication resource unit is generated at least in part from the comparison data.