MIMO demodulator and method for the same

This is a continuation of International Application PCT/JP2006/314424, published in Japanese, with an international filing date of Jul. 20, 2006.

The present invention relates to an MIMO demodulator and MIMO demodulation method for performing channel demultiplexing of received data in an MIMO (Multiple Input Multiple Output) system in which different data streams are transmitted from a plurality of transmitting antennas using the same bandwidth and the transmitted data streams are received by a plurality of antennas.

In MIMO multiplexing transmission, by transmitting different channel-coded data streams from a plurality of transmitting antennas using the same bandwidth, the data rate (i.e., the throughput) can be increased in proportion to the number of transmitting antennas within the given bandwidth. In S-PARC (Selective Per-Antenna Rate Control) employed in such an MIMO system, in addition to controlling the data rate adaptively in accordance with the change in radio wave propagation conditions, control is performed to maintain reception quality by stopping signal transmission from any transmitting antenna for which the received SINR (Signal-to-Interference and Noise power Ratio) has dropped below a predetermined value and by concentrating the transmitter power to the remaining transmitting antennas.

In this case, when the number of antennas used for transmitting the data decrease, the number of channel estimates needed for computation at the receiving end decrease, and as a result, the amount of computation decreases. In other words, the computing resources allocated to the path no longer in use are rendered idle. On the other hand, the channel estimates are periodically calculated from the received data and used for the computation for channel demultiplexing. For example, as described in JP 2001-522166A, if the calculated channel estimates are linearly interpolated to reduce the range of received data to which the same channel estimate is applied, it is possible to quickly cope with fast fading, but this requires a large amount of computing resources.

According to the present invention, there is provided an MIMO (Multi Input Multi Output) demodulator for performing channel demultiplexing of data transmitted from a plurality of transmitting antennas and received by a plurality of receiving antennas, comprising: a channel estimator calculating a channel estimate based on the received data; a channel estimate adjusting circuit changing, for the channel estimate calculated by the channel estimator, a range of data to which the same channel estimate is to be applied, based on the number of transmitting antennas used for transmitting the data; and an MIMO demultiplexer performing MIMO demultiplexing on the received data by using the channel estimate whose applicable data range has been adjusted by the channel estimate adjusting circuit.

For example, when the number of transmitting antennas is reduced, the channel estimate adjusting circuit reduces the range of data to which the same channel estimate is to be applied.

When the number of transmitting antennas is reduced, the range of data to which the same channel estimate is to be applied is reduced by quickly responding to the change in the propagation environment. Then, by making effective use of the computing resources rendered idle when the propagation environment deteriorates, it is possible to accurately cope with fast fading.

FIG. 1 is a diagram showing one example of an MIMO system;

FIG. 2 is a diagram explaining one example of frame format and channel estimate;

FIG. 3 is a diagram explaining the case where transmissions from some of transmitting antennas are stopped;

FIG. 4 is a diagram explaining channel estimates for the case where transmissions from some of transmitting antennas are stopped;

FIG. 5 is a diagram showing the configuration of a receiver according to one embodiment of the present invention.

The following description is given by taking as an example a case where transmissions from three antennas are stopped in a system that transmits signals from four antennas and receives the signals by four antennas, but generally, the following description is applicable to a case where a transmission or transmissions from selected antenna or antennas are stopped in a system that transmits signals from a plurality of antennas and receives the signals by a plurality of antennas.

FIG. 1 shows one example of an MIMO system to which the present invention is applied. A transmitter 10 transmits different channel-coded data streams from four transmitting antennas 12 using the same frequency band. At the receiving end, the data streams are received by four receiving antennas 14 ; in a receiver 16 , channel estimates that can be expressed in the form of a 4×4 matrix, whose elements consist of 16 complex numbers h ij (i, j=1 to 4) as shown in FIG. 1 , are calculated by using pilot signals contained in the respective received data streams, and the received data channels are demultiplexed by using the channel estimates, to reconstruct the original data.

FIG. 2 shows one example of the frame format of the data transmitted from the transmitter 10 to the receiver 16 . One frame consists of a data channel (CH) 20 and a pilot CH 22 preceding it. In the pilot CH section, the pilot signals comprising mutually orthogonal codes (known patterns) from the respective transmitting antennas are received in multiplexed form by the respective receiving antennas; therefore, by taking correlation with each known pattern for each received signal, the complex number indicating the amplitude and phase of the pilot signal from each transmitting antenna is determined and, by dividing it by a predetermined value, the channel estimate h ij is calculated. Then, an average value ξ 0 (n) taken between the channel estimate h(n) determined for the pilot CH in the n-th frame and the channel estimate h(n+1) determined for the pilot CH in the (n+1)th frame is used as the channel estimate ξ(n) that is used for channel demultiplexing of the data CH in the n-th frame (n). Such channel estimates ξ 0 (n) also comprise 16 complex numbers h ij (i, j=1 to 4) and can be expressed in the form of a 4×4 matrix having these complex numbers as the elements.

In the receiver 16 , computing resources are made available for each frame so that the computation for the channel demultiplexing using the channel estimates ξ 0 (n) completes within one frame period.

Here, if transmissions from three of the four transmitting antennas are stopped due to a degradation in SNIR, as shown in FIG. 3 , the channel estimates needed for the computation are only h 11 , h 21 , h 31 , and h 41 , and thus the amount of computation decreases, rendering some of the computing resources idle.

In view of this, in the present invention, as shown in FIG. 4 , the period of the data CH 20 is divided into four sections, and the computation is performed using four different channel estimates ξ 0 (n), ξ 1 (n), ξ 2 (n), and ξ 3 (n) calculated for the respective sections by linear interpolation from h(n) and h(n+1). In other words, the data of N symbol intervals within one frame is divided into four parts, and the four different channel estimates are sequentially applied to the respective data of N/4 symbol intervals. This achieves highly accurate computation that can cope with fast fading by making effective use of the computing resources made available as a result of the number of transmitting antennas reduced due to a deterioration of the propagation environment.

FIG. 5 shows one configuration example of the receiver 16 that implements the above process. A channel estimator 32 calculates the channel estimate ξ 0 (n) from the amplitude and phase of the pilot signal contained in the received data in the pilot CH from among the data received by a receiving section 30 , and a linear interpolator 34 calculates the channel estimates ξ 0 (n), ξ 1 (n), ξ 2 (n), and ξ 3 (n) by linear interpolation. A channel estimate selector 36 selects the channel estimates ξ 0 (n), ξ 1 (n), ξ 2 (n), and ξ 3 (n) obtained by linear interpolation from adjacent channel estimates when the number of transmitting antennas is reduced to one, or otherwise selects the channel estimate ξ 0 (n) obtained by taking an average between the adjacent channel estimates, and supplies the selected channel estimates to an MIMO demultiplexer 38 . Using the channel estimates selected by the channel estimate selector 36 , the MIMO demultiplexer 38 performs the channel demultiplexing by applying a demultiplexing algorithm, such as an MMSE (Minimum Mean Square Error) algorithm or an MLD (Maximum Likelihood Detection) algorithm, to the received data in the data CH period.

As the number of transmitting antennas that is used by the channel estimate selector 36 , use may be made of the number of transmitting antennas that is transmitted from the transmitting end over the control channel, or alternatively, the number of pilot signals detected by the channel estimator 32 as having correlation greater than a threshold value with a known pattern may be used as the number of transmitting antennas.