CN1868147B - System and method for channel-adaptive antenna selection - Google Patents

Abstract

Systems and methods that provide channel-adaptive antenna selection in multi-antenna-element communication systems are provided. In one embodiment, a method that selects a subset of receive antennas of a receiver to receive a transmitted RF signal may include, for example, one or more of the following: establishing possible subsets of the receive antennas; determining sets of channel parameter statistics corresponding to the possible subsets of the receive antennas; computing output bit error rates of the receiver, each output bit error rate being computed based on at least one set of channel parameter statistics; selecting a particular possible subset of the receive antennas based upon a criterion predicated on the computed output bit error rates; and connecting one or more RF chains of the receiver to the receive antennas of the selected particular possible subset.

Description

Method and system for channel adaptive antenna selection

Technical Field

The present invention relates to communication systems, and more particularly, to a method and system for antenna selection in a multiple antenna element communication system.

Background

Most current wireless communication systems consist of nodes equipped with a single transmit and receive antenna. It is predicted that the performance, including capacity, of many communication systems will be substantially improved by using multiple transmit and/or multiple receive antennas. This antenna configuration forms the basis of "smart" antenna technology. Smart antenna techniques combined with space-time signal processing can be used to reduce the deleterious effects of multipath fading of the incoming signal and to suppress interfering signals. In this way, the performance and capacity of existing or deploying digital wireless systems (e.g., CDMA-based systems, TDMA-based systems, WLAN systems, and OFDM-based systems such as IEEE 802.11a/g) will be improved.

At least some of the performance penalties of wireless systems of the type described above are partially improved by the use of multiple-element antenna systems designed to introduce diversity gain and suppress interference during signal reception processing. Winters et al, in 2.1994, IEEE letters 42, 2/3/4, pages 1750, The "The image of Antenna Diversity On The Capacity of Wireless Communication Systems". Such diversity gain improves system performance by mitigating multipath for more equalized coverage, increasing received signal-to-noise ratio for greater distance or reducing required transmit power, providing greater robustness against interference, or allowing more frequency reuse for greater capacity.

In a communication system incorporating a multi-antenna receiver, a set of M receive antennas can cancel M-1 interferers. Thus, N signals can be transmitted simultaneously over the same bandwidth with N transmit antennas, which are then split into N separate signals by a set of N antennas of the receiver. Such systems are commonly referred to as Multiple Input Multiple Output (MIMO) systems and have been widely studied. See, for example, J.H Winters "IEEE communications journal, 11.11 COM-35, 1987," Optimum combining for index radio systems with multiple users "; chuan et al, 11.1998, "Global communication Systems society' 98, Australia, Sydney, IEEE 1998," Capacity of Multi-Antenna Array Systems In Indor WirelessEnvironment ", 1894-1899; shiu et al, IEEE letters, 2000, volume 3, 48, volume 3, page 502 and 513, "diagnosis and Its Effect on the Capacity of Multi-Element Antenna Systems".

Some multiple antenna arrays (e.g., MIMO) add capacity to the system, which can be achieved using the above-mentioned configuration. In a MIMO system, the received signal is decomposed into M "spatially multiplexed" independent channels under the assumption of ideal estimation of the receiver-applicable channel. This allows for an increase in capacity, equivalent to M times that of a single antenna system. The capacity provided by MIMO scales linearly with the number of antenna elements, given a fixed total transmit power. Specifically, it was confirmed that: the data rate with N transmit and N receive antennas may be N times that of a single antenna system without an increase in the total bandwidth and total transmit power. See, for example, G.J.Foschini et al, "On Limits of Wireless Communications in a Wireless environmental protection in a family of environments Using Multipleantennas", Vol.6, No. 3, page 311, 335, of Kluwer Academic Publishers, 3.1998. In experimental MIMO systems based on N-fold spatial multiplexing, typically more than N antennas are used at a given transmitter or receiver. Since each additional antenna adds diversity gain and antenna gain, interference suppression is applied to all N spatially multiplexed signals. See "Simplified processing for high speed pixel efficiency communicating multi-elementary systems" by IEEE Journal on Selected Areas in Communications 1999, 11.11.11.1841 and 1852, Foschini et al.

While increasing the number of transmit and/or receive antennas enhances the multi-aspect performance of a MIMO system, providing separate RF chains for each transmit and receive antenna increases cost. Typically each RF link includes a low noise amplifier, a filter, a down-converter and an analog-to-digital converter (a/D), with the latter three accounting for a major portion of the cost of the RF link. In some existing single antenna wireless receivers, the cost of the single RF link required exceeds 30% of the total receiver cost. It is clear that as transmit and receive antennas increase, the overall system cost and power consumption will increase significantly.

Several approaches to address the above deficiencies have been found, such as U.S. patent publication No. 20020102950 entitled "Method and apparatus for selection and use of optical antipans in wireless systems"; molisch et al, IEEE ICC institute at 6 months 2001, "Capacity of MIMOSystems with anti-senta selection" at page 570-574 of Finland, Helsinki, Vol.2; and "Onoptimum MIMO with 324 of No. 8, No. 322 of No. 8 of No. 6 of No. 8 of 2002 of" IEEEcommunications Letters "of R.S. Blum et alantenna selection ", in which a subset of transmit/receive antennas is selected from a multitude of antennas. Since there is N times spatial multiplexing, at least N RF chains are used, typically N antennas are selected from a total of M antennas at the receiver and/or a total of N antennas from the transmitter_TSelecting N antennas from the antennas, where M > N and N_T＞N。

The performance of a system with antenna selection depends on the criteria used in the selection process. The different criteria used may result in different antenna subsets being selected even under the same channel conditions, resulting in different performance. Some of the above-mentioned documents claim to select a subset of antennas using criteria that maximizes capacity. However, capacity is an ideal amount that may not be achievable because she requires perfect coding and/or equalization and/or continuous modulation. In practice, limited (or even no) coding and quantization modulation are used, and the equalizer is not ideal.

Disclosure of Invention

Some aspects of the present invention relate to systems and methods for providing channel-adaptive antenna selection in a multiple antenna element communication system.

In one embodiment according to some aspects of the present invention, a system for selecting N antennas from a transmitter having M antenna elements or a receiver having M antenna elements, where N is less than M, comprises: for example, M antenna elements in M antenna element transmitters or M antenna element receivers, N RF chains, and a switch connected to the N RF chains. The M antenna element receivers calculate an output error rate for each possible subset of N antenna elements of the M antenna elements. Each output bit error rate is calculated based on at least one set of channel parameter statistics. The M antenna element receivers select a particular subset of N antenna elements based on a criterion based on the calculated output error rate. In response to selecting a particular subset of N antenna elements based on the criteria, the switch connects the N RF chains to the N antenna elements of the particular subset of N antenna elements.

In another embodiment according to some aspects of the invention, the method of selecting a subset of receive antennas of the receiver to receive the transmitted RF signal comprises, for example, one or more of: establishing possible subsets of receive antennas; determining channel parameter statistics corresponding to the possible subset of receive antennas; calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics; selecting a particular possible subset of receive antennas based on a criterion based on the calculated output error rate; one or more RF chains of the receiver are connected to the receive antennas of the selected particular possible subset.

In yet another embodiment according to some aspects of the present invention, a method of selecting a subset of transmit antennas of a transmitter to transmit an RF input signal as a plurality of output RF signals for subsequent reception by a receiver includes, for example, one or more of: establishing possible subsets of transmit antennas; determining a set of channel parameter statistics corresponding to the possible subset of transmit antennas; selecting a transmission mode corresponding to each group of channel parameter statistics; calculating output bit error rates of the receiver, each output bit error rate calculated based on at least one set of channel parameter statistics and at least one selected transmission mode; selecting a particular possible subset of transmit antennas based on a criterion based at least on the calculated output error rate; one or more RF chains of the transmitter are connected to the transmit antennas of the selected particular possible subset.

In another embodiment according to some aspects of the present invention, there is provided a method of selecting an antenna in a communication system including a transmitter and a receiver, the transmitter including a transmit antenna that transmits a set of spatially multiplexed RF output signals over a channel using two or more RF transmit chains. The receiver includes a receive antenna for receiving the set of spatially multiplexed RF output signals and correspondingly generating a set of spatially multiplexed receive RF signals processed by two or more RF chains. The method comprises one or more of: establishing possible subsets of the transmit antennas and possible subsets of the receive antennas; determining channel parameter statistics corresponding to various combinations of one of the possible subsets of transmit antennas and one of the possible subsets of receive antennas; selecting transmission modes respectively corresponding to various channel parameter statistics; calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics and the selected corresponding transmission mode; selecting a particular possible subset of transmit antennas and a particular possible subset of receive antennas based on a criterion based at least on the calculated output error rate; connecting the two or more RF transmit chains to a possible subset of the selected particular transmit antennas; two or more RF receive chains are connected to a possible subset of the selected particular receive antennas.

In yet another embodiment according to some aspects of the present invention, a method of selecting an antenna in a communication system including a transmitter and a receiver is provided. The transmitter includes a transmit antenna that transmits a set of RF output signals over a channel using one or more RF transmit chains. The receiver includes a receive antenna for receiving the set of RF output signals and correspondingly generating a set of receive RF signals processed through one or more RF receive chains. The method comprises one or more of: establishing possible subsets of the transmit antennas and possible subsets of the receive antennas; determining channel parameter statistics corresponding to various combinations of the possible subset of transmit antennas and the possible subset of receive antennas; selecting transmission modes respectively corresponding to various channel parameter statistics; calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics and the selected corresponding transmission mode; selecting a particular possible subset of transmit antennas and a particular possible subset of receive antennas based on a criterion based at least on the calculated output error rate; connecting the one or more RF transmit chains to the selected possible subset of specific transmit antennas; the one or more RF receive chains are connected to the selected possible subset of the particular receive antennas.

The various advantages, features, aspects and details of the embodiments of the present invention can be more fully understood from the following description and accompanying drawings. In which the same reference numerals are used throughout the drawings to refer to the same or like parts.

Drawings

Fig. 1A and 1B are schematic diagrams of a conventional MIMO system;

fig. 2A and 2B are schematic diagrams of a MIMO system according to some aspects of the present invention;

fig. 3 is a schematic diagram of an SM-MIMO-OFDM system in accordance with some aspects of the present invention;

fig. 4A and 4B are flow diagrams of antenna selection according to some aspects of the present invention;

FIG. 5 is a schematic illustration of performance changes caused according to aspects of the present invention;

FIG. 6 is a schematic diagram of an SC-MIMO-OFDM system in accordance with some aspects of the present invention;

fig. 7 is a schematic diagram of a receiver in a DS-SS SIMO system in accordance with aspects of the invention.

Detailed Description

Summary of the invention

Some embodiments of some aspects of the present invention relate to communication systems using transmitters and/or receivers with multiple antenna elements.

Some embodiments of some aspects of the present invention relate to methods and systems for providing channel-adaptive antenna selection in a multiple antenna element communication system.

Some embodiments implementing some aspects of the present invention facilitate selecting a subset of antenna elements in one or more multi-antenna wireless communication devices based at least in part on a calculated output error rate.

Some embodiments implementing some aspects of the present invention facilitate selecting a subset of antenna elements in one or more multi-antenna wireless communication devices to minimize or optimize Bit Error Rate (BER).

Some embodiments implementing some aspects of the present invention facilitate channel-adaptive selection of a subset of antenna elements in one or more multi-antenna wireless communication devices according to a criterion based at least in part on a calculated output error rate.

Some embodiments implementing some aspects of the present invention can use more antenna elements than the number of Radio Frequency (RF) links. Some embodiments of some aspects of the invention improve system performance in a cost-effective manner.

Some embodiments of some aspects of the present invention may be used to select a subset of antenna elements in a multi-antenna transmitter to transmit signals and/or to select a subset of antenna elements in a multi-antenna receiver to receive signals.

In some embodiments of some aspects of the present invention, the subset of antenna elements is selected based at least in part on a criterion based at least in part on the calculated output error rate.

In some embodiments of some aspects of the present invention, the subset of antenna elements is selected based at least in part on a minimum bit error rate.

Some embodiments of some aspects of the present invention are applicable to communication systems using, for example, code division multiple access signals, spread spectrum signals, single carrier signals, multi-carrier signals, orthogonal frequency division multiplexed signals and ultra wideband signals, space time diversity signals, and spatially multiplexed signals.

In some embodiments of some aspects of the present invention, the subset of antenna elements is selected based at least in part on a minimum bit error rate based on parameters (e.g., statistical parameters) associated with the one or more communication channels. In some embodiments of some aspects of the present invention, the subset of antenna elements is selected based at least in part on a minimum bit error rate calculated based on parameter statistics of one or more applicable communication channels.

Some embodiments of some aspects of the present invention may be used to select antennas in a multiple-input multiple-output (MIMO) communication system. For example, MIMO communication systems provide a transmitter that broadcasts multiple (N) spatially multiplexed signals over N transmit antenna elements from a set of N_TSelected from an antenna element, where n_TIs greater than N. For example, a MIMO communication system provides a receiver that has M antenna elements in total, from which N receive antenna elements are selected, where M > N, the N receive antenna elements generating a number of output signals equal to the number of spatially multiplexed signals. The output signal is then provided to a corresponding RF chain for processing in baseband. As such, some embodiments of some aspects of the present invention may facilitate minimizing BER (e.g., minimizing channel-adaptive BER) and/or reducing the processing cost of RF signals in a multi-antenna system.

Some embodiments according to some aspects of the present invention provide a transmitter comprising one or more RF chains and a certain number of transmit antennas, wherein the number of RF chains is equal to or less than the number of transmit antennas. Some embodiments according to some aspects of the present invention provide a system and/or method, for example, selecting a particular subset of the plurality of transmit antennas, the selected subset of antennas transmitting an RF input signal, the transmitted RF signal being subsequently received by a receiver as a plurality of RF output signals. A number of possible subsets of a particular number of transmit antennas (e.g., equal to or less than the number of possible subsets including a particular number of transmit antennas or other type of combining characteristic) is established. Sets of channel parameter statistics or other parameters corresponding to possible subsets of the particular plurality of transmit antennas are determined. A transmission mode or other type of mode corresponding to the corresponding channel parameter statistics is selected. The transmission mode may include, for example, a modulation level and/or a coding rate. For example, a plurality of BERs (e.g., output BERs) of the receiver are calculated. Wherein each output BER is calculated based at least in part on a set of channel parameter statistics and/or a corresponding selected transmission mode. For example, a possible subset of a particular plurality of transmit antennas may be selected based on a criterion based at least in part on the output BER and/or the selected transmission mode. One or more RF chains are then connected to the transmit antenna or transmit antennas of a selected possible subset of the particular plurality of transmit antennas.

Some embodiments according to some aspects of the present invention provide channel parameter statistics including first order statistics, second order statistics, and higher order statistics. The channel parameter statistics include one or more of: such as the output signal-to-noise ratio of the receiver, the output signal-to-interference-and-noise ratio, the likelihood (e.g., log-likelihood), the Euclidean distance of the signal constellation (Euclidean distance). For example, the channel parameter statistics may be calculated in the frequency domain or the time domain.

Some embodiments of the antenna selection method according to some aspects of the present invention may be used in different types of multi-antenna communication systems. In particular embodiments, some embodiments of the antenna selection method according to some aspects of the present invention may be applied in a multi-antenna receiver of a Single Channel (SC) system (e.g., a system without spatial multiplexing), in a multi-antenna transmitter of a single channel system or in a transmitter and/or receiver of a MIMO system using Spatial Multiplexing (SM) or a single channel.

For example, according to some embodiments of some aspects of the present invention, N receive antenna elements (where M > N) are selected from a set of M available antenna elements, e.g., a subset of the selected antenna elements minimizes BER taking into account various channel parameter statistics. The realization process is as follows: a possible subset of the plurality of transmit antennas and a possible subset of the M receive antennas are first established. The method also includes determining sets of channel parameter statistics corresponding to various combinations of one of the possible subsets of transmit antennas and one of the possible subsets of receive antennas. Selecting a plurality of transmission modes respectively corresponding to the plurality of sets of channel parameter statistics. In addition, a plurality of output error rates of the receiver are calculated based at least in part on one of the sets of channel parameter statistics and a corresponding one of a plurality of transmission modes. Next, a possible subset of the plurality of transmit antennas and a possible subset of the plurality of receive antennas are selected based on a criterion, wherein the criterion is determined based at least in part on the plurality of output error rates and a plurality of transmit modes. The method further comprises the following steps: one or more RF transmit chains are connected to the one possible subset of the plurality of transmit antennas and one or more RF receive chains are connected to the one possible subset of the plurality of receive antennas.

Some embodiments according to some aspects of the invention provide a criterion, for example, the criterion is based on one or more of: an output error rate, a transmission mode, a minimum value of the output error rate, a maximum value of the data rate at a first stage of the receiver, and a minimum value of the output error rate at a second stage of the receiver.

In the case where a single-channel or spatial multiplexing MIMO system uses a plurality of RF chains at a transmitting end and/or a receiving end, the determined baseband weighting and combining scheme may be applied to a transmitter (e.g., precoding) and/or a receiver in combination with an antenna selection method. For example, both baseband weighting and antenna selection are designed to minimize BER. In another embodiment, the baseband weights may be designed to maximize the output signal-to-noise ratio (SNR), signal-to-interference-and-noise ratio (SINR), or capacity, for example, while minimizing BER through proper antenna selection.

Some embodiments according to some aspects of the present invention relate to channel-adaptive methods and/or systems for selecting antennas based on minimizing BER in multi-antenna systems, including multi-antenna systems such as N-fold spatial multiplexing. In order to facilitate understanding of some embodiments according to some aspects of the invention, an overview of a typical architecture for implementing antenna selection in a multi-antenna system is provided below. Additional details regarding systems and methods for channel-adaptive antenna selection based on minimizing BER are also provided later herein.

Second, framework for antenna selection

Some embodiments according to some aspects of the present invention may be used in a wireless communication system in which the number of RF chains used by a transmitter and/or receiver is less than the number of transmit/receive antennas used. In some embodiments according to some aspects of the present invention, N receive antenna elements are selected from a total of M receive antenna elements, where M > N. This generates N RF output signals which then pass through N RF chains. In a typical embodiment, each RF chain includes, for example, a filter, a down-converter, and an a/D converter. The output signals produced by the a/D converters of each RF chain are then digitally processed to produce N spatially multiplexed output signals. By performing the necessary selection of a subset of antennas at RF, the implementation cost of an N-fold spatially multiplexed system with more than N receive antennas, but only N RF chains, is similar to the cost of a system with N receive antennas. Thus, receiver performance can be improved by using additional antennas at a relatively low cost.

Can be used in a system having N RF chains and N_TUsing similar techniques on the transmitter of a transmitting antenna, where n_TIs greater than N. In one embodiment, the N RF chains are followed by a switch that connects each RF chain to the slave N_TOn a selected subset of the N transmit antennas. As in the receiver, by performing the selection of the necessary subset of antennas at RF, the implementation cost of a system with more than N transmit antennas, but only N times the spatial multiplexing of N RF chains, is similar to the cost of a system with N transmit antennas and N RF chains. Thus, transmitter performance can be improved by using additional antennas at relatively low cost.

A. Spatial multiplexing

According to some embodiments of some aspects of the present invention, Spatial Multiplexing (SM) provides a signal transmission mode based on the use of multiple antennas at both the transmitter and receiver, which can increase the bit rate of the wireless link without a corresponding increase in power or bandwidth consumption. In the case where both the receiver and the transmitter use N antennas, the input stream of information symbols provided to the transmitter is divided into N independent subcode streams. Spatial multiplexing attempts to have each sub-stream occupy the same "channel" (e.g., time slot, frequency, or code/key sequence) in the applicable multiple access protocol. In the transmitter, each of the subcode streams is applied to the N transmit antennas and propagated to the receiver via an intervening multipath communication channel. The composite multipath signal is then received by a receiving array of N receive antennas of the receiver arrangement. At the receiver, the "spatial signature" defined by N phases and N amplitudes relative to a given subcode stream from the receive antenna array is estimated. Next, a signal processing technique is applied to separate the received signals, which enables the original subcode stream to be restored and synthesized into the original input symbol stream. Winters further elaborates the principles of spatial multiplexing communications and typical system implementation in the IEEE communications journal, 11 COM-35, 1987, article "optimal combining for index radio systems with multiple users". The present invention makes full reference to this paper.

B. Conventional MIMO system

Some aspects of the present invention can be fully elucidated by considering a conventional MIMO system as shown in fig. 1. As shown in fig. 1, the MIMO system 100 in fig. 1 includes a transmitter 110 shown in fig. 1A and a receiver 130 shown in fig. 1B. The transmitter 110 and receiver 130 include a T number of set of RF transmit chains and a R number of set of RF receive chains, respectively, configured to transmit and receive a N number of spatially multiplexed signals. In the system 100, assume one of the following: (i) t is greater than N, R is equal to N; (ii) t is equal to N, R is greater than N; or (iii) both T and R are greater than N.

Referring to fig. 1A, an input signal S to be transmitted, typically comprising a stream of digital symbols, is separated into N independent sub-streams S by a demultiplexer 102_{1，2……，N}. Then, the subcode stream S_{1，2……，N}Is sent outTo a Digital Signal Processor (DSP)105, the DSP 105 generates T output signals T_{1，2……，T}. The T output signals T_{1，2……，T}Usually by dividing N subcode streams S_{1，2……，N}Generated by weighting, i.e. by weighting the N subcode streams S with T different weighting factors_1，2…，NEach sub-stream weight in (a) is weighted (e.g., multiplied by a complex number) to form NT streams. Then, the N.T code streams are combined to form T output signals T_{1，2……，T}. The T output signals T are then converted by T digital-to-analog (D/A) converters 108_{1，2……，T}Conversion into T analog signals A_{1，2……，T}. By mixing the signal generated by the local oscillator 114 with the T analog signals a in the mixer 112_{1，2……，T}Mixing the T analog signals A_1，2…，TUp-converted to the appropriate transmit carrier RF frequency. Then, the T RF signals (e.g., RF)_{1，2……，T}) Amplified by a respective amplifier 116 and transmitted by a respective antenna 118.

Referring now to fig. 1B, the RF signal transmitted by the transmitter 100 is received by R receiving antennas 131 mounted on the receiver 130. Each of the R signals received by the receiving antenna 131 is amplified by a corresponding low noise amplifier 133 and filtered by a filter 135. The resulting filtered signal is then downconverted from RF to baseband using mixers 137, each of which is provided with a local oscillator signal by a local oscillator 138. Although the receivers in fig. 1B are all configured as zero difference receivers, heterodyne receivers featuring Intermediate Frequency (IF) frequencies may also be used. The R baseband signals generated by the mixer 137 are then converted to digital signals using a corresponding set of R analog-to-digital (a/D) converters 140. Then, the digital signal processor 142 converts the R digital signals D_{1，2……，R}Weighted and combined to form N spatially multiplexed output signals S'_{1，2……，N}The N spatially multiplexed output signals S'_{1，2……，N}Comprising a transmission signal S_{1，2……，N}And (4) estimating. Multiplexer 155 then outputs the N output signals S'_{1，2……，N}Multiplexing to produce an estimate of the original input signal SAnd calculating 160 (S').

C. Antenna selection for RF in spatial multiplexing communication systems

Referring now to fig. 2, fig. 2 illustrates a MIMO communication system 200 having a transmitter 210 and a receiver 250, the transmitter 210 and the receiver 250 configured to achieve N-fold spatial multiplexing using only N transmit/RF receive chains, even though more than N transmit/receive antennas are equipped on the transmitter 210 and the receiver 250, respectively. In particular, the transmitter 210 includes a set of MT transmit antennas 240 and the receiver includes a set of MR receive antennas 260, in some embodiments of some aspects of the present invention, the MT and/or MR are greater than or equal to N. For example, (i) MT is greater than N, MR is equal to N; (ii) MT equals N, MR is greater than N; or (iii) MT and MR are both greater than N.

As shown in fig. 2A, an input signal S to be transmitted is separated by a demultiplexer 202 into N independent subcode streams SS_{1，2……，N}. Then, the corresponding D/A converter group 206 converts the N independent sub-stream SS_{1， 2……，N}Converting into N analog subcode streams AS_{1，2……，N}. The local oscillator 214 then provides a signal to the mixer 212, and the mixer 212 converts the N analog subcode streams AS_{1，2……，N}Up-converted to the appropriate transmit carrier RF frequency. Next, the switch 218 will generate N RF signals (e.g., RF signals)_{1，2……，N}) Each RF signal of which is connected to a selected subset of the N transmit antenna elements. The switch 218 switches the N RF signals (e.g., RF signals)_{1，2……，N}) Is connected to N transmit antennas selected from the MT available transmit antennas 240 to obtain a set of N output signals. Then, the N corresponding amplifiers 234 amplify the N output signals, and the amplified output signals are transmitted by the selected N transmitting antennas 240. In another embodiment, amplifier 234 is located before switch 218. In this configuration, only N amplifiers are required instead of MT; if one amplifier is provided for each of the MT transmit antennas, MT amplifiers are required. The N antennas are chosen to minimize the BER of the output signal of the receiver.

Referring to fig. 2B, the transmitter 210 transmitsAre received by MR receive antennas 260 disposed on the receiver 250. Each of the MR receive signals is amplified by a corresponding Low Noise Amplifier (LNA)264, and then a switch 276 connects a subset of the N receive signals to the N RF chains to form N RF signals, which are passed through a corresponding N filters 280. In another embodiment, the low noise amplifier 264 may be located after the switch 276, so that the total number of LNAs used is N instead of MR; if one LNA is provided for each of the MR receive antennas, MR LNAs are required. The generated N filtered signals are then downconverted to baseband using N mixers 282, where a local oscillator 284 provides a carrier signal to each mixer. In the embodiment of fig. 2B, although receiver 250 is implemented as a homodyne receiver, receiver 250 can also be implemented as a heterodyne receiver, which features an Intermediate Frequency (IF) frequency. (indeed, any of the embodiments according to some aspects of the invention may incorporate a homodyne or heterodyne configuration). A corresponding set of N a/D converters 286 converts the N baseband signals generated by the mixer 282 into digital signals. Digital signal processor 288 further processes the N digital signals to form N spatially multiplexed output signals SS'_{1，2……， N}The N spatially multiplexed output signals SS'_{1，2……，N}Is N independent subcode streams SS_{1，2……，N}And (4) estimating. Then, the multiplexer 292 pairs the N output signals SS'_{1，2……，N}The multiplexing is performed to produce an output signal S', which is an estimate of the input signal S.

In some embodiments of some aspects of the present invention, a baseband weighting and combining (e.g., "precoding") scheme is added at the transmitting end, for use in conjunction with the antenna selection method described below. In this case, the DSP module is disposed between the demultiplexer 202 and the D/A converter 206, so that the N independent subcode streams SS_{1，2……，N}The sums weighted by the complex factors are combined to form a set of N output signals. Next, a corresponding set of D/A converters 206 converts the N output signals into an analog signal AS_{1，2……，N}。

In some embodiments of aspects of the invention, a typical antenna selection method may be used in conjunction with increasing space-time coding at the transmitting end. In this case, the demultiplexer 202 is replaced by a DSP module which processes the input signal S in the spatial and temporal domain to form a set of N output signals. Next, a corresponding set of D/A converters 206 converts the N output signals into an analog signal AS_{1，2……，N}. Two spatio-temporal techniques that are commonly used are: 1) introducing a time delay (or equivalently a phase offset) in one or more of the N output signals; 2) the use of transmit diversity techniques is described in IEEE Journal on Selected area Communications, 1998, 8.10.16, pages 1451 and 1458, a paper written by S.M. Alamouti "A Simple transmit diversity technique for wireless Communications". This article is referred to in the present document.

For example, space-time coding techniques may be applicable to SC MIMO systems and/or systems designed to achieve diversity gains. The precoding technique is applicable to SC-based or spatial multiplexing-based MIMO systems or systems designed to achieve both data rate and diversity gain.

Thirdly, based on the minimum bit error rate, the method is carried out in the RF frequency band

Method and/or system for channel adaptive antenna selection

A. Overview

Some embodiments according to some aspects of the present invention relate to a method for channel-adaptively selecting antennas based on minimizing bit error rates in a multi-antenna communication system and/or a channel-adaptive antenna selection system. For example, according to some embodiments of some aspects of the present invention, in a multi-antenna communication system, a subset of antenna elements are selected to transmit and/or receive signals to minimize error rates. For example, some embodiments according to some aspects of the present invention, in a multi-antenna communication system, a subset of antenna elements are selected to transmit and/or receive signals to minimize the error rate over the entire time period in the presence of variations in one or more applicable communication channels. When multiple antennas are used for transmission, some embodiments according to some aspects of the present invention may be used to select the antennas of the transmitter. Some embodiments according to some aspects of the present invention may be used to select an antenna of a receiver when multiple antennas are used for reception.

For example, some embodiments according to some aspects of the invention are applicable to: (i) receivers using multiple antennas in what is known as a single channel system (e.g., systems that do not employ spatial multiplexing); (ii) a transmitter using a plurality of antennas in a single channel system; and (iii) systems in which the transmitter and/or receiver uses fewer RF chains than transmit and/or receive antennas in a spatially multiplexed or single channel MIMO system.

Some embodiments of some aspects of the invention will be described below in the following exemplary ranges with reference to fig. 3 to 7: 1) MIMO systems with spatial multiplexing, where the number of RF chains used by the transmitter and receiver is smaller than the number of transmitter/receiver antennas; 2) single channel MIMO systems without spatial multiplexing, where the number of RF chains used by the transmitter and receiver is less than the number of transmitter/receiver antennas; and 3) single channel SIMO systems without spatial multiplexing, including receivers using multiple antenna elements. Some embodiments according to some aspects of the present invention may also be used in Single Channel (SC) multiple-in single-out (MISO) systems without spatial multiplexing, where the transmitter uses multiple antenna elements.

The following embodiments are set forth, by way of example, for a system using orthogonal frequency division multiplexing modulation (OFDM) (e.g., compliant with the 802.1-1a WLAN standard) or a system based on direct sequence spread spectrum (DS-SS) (e.g., compliant with the WCDMA standard). According to some embodiments of some aspects of the present invention, the processing capacity of a DS-SS receiver can be extended to the spatial domain by incorporating a space-time Rake receiver that can efficiently combine multipath "taps" corresponding to both the time and spatial domains. This extension indicates that the techniques described herein can be generalized to virtually any system that uses time and/or frequency domain processing in a frequency selective fading environment.

B. Antenna selection in SM-MIMO-OFDM systems

Fig. 3 illustrates a transmitter and receiver structure of an SM-mimo ofdm system 300 using antenna selection according to some embodiments of aspects of the invention. As shown, two independent substreams 304 (e.g., spatially multiplexed signals) are OFDM modulated onto Nt frequency subcarriers and are prepared for transmission over two RF chains 308. In this regard, the switching module 312 selects two antenna elements from the 4 transmit antenna elements 316 and connects the two antenna elements to the two RF chains 308. Since only two of the 4 antenna elements 316 of the transmitter 302 are selected, the number of RF transmit chains is advantageously reduced to the number of spatially multiplexed signals.

In the embodiment shown in fig. 3, at any given moment, the switching module 312 contains information identifying the antenna element pair 316 to be used for transmission. The switch module 312 may itself calculate this information according to an algorithm based on a minimum BER criterion (e.g., in the case where the channel 318 is interchangeable). In another embodiment, the module 312 may receive information from the receiver 330 through a feedback path (not shown). The latter approach may be used in cases where the channel 318 is not interchangeable (reciprocal), for example, in interference-limited environments.

In the receiver 330, the switching module 334 selects two antenna elements from the 4 antenna elements 338 for receiving the incident signal transmitted by the transmitter 302. The switch module 334 connects the selected 2 antennas 338 to two RF chains 342, the two RR chains 342 being used to convert the two signals into the digital domain for baseband processing. Next, a weighting matrix 346 is applied to the received signal on each tone (at each tone) to separate and recover each transmitted spatially multiplexed signal.

In the exemplary embodiment, switch module 334 is configured to self-calculate which pair of antenna elements 338 should be selected for reception by executing an algorithm based on a minimum BER criterion. Where the channels are not interchangeable, module 334 may be further configured to calculate which pair of antenna elements 316 should be used by transmitter 302 and provide this information to transmitter 302. Two possible antenna selection algorithms implemented by the switch modules 312, 334 are described below in conjunction with fig. 4A and 4B.

Turning to fig. 4A, fig. 4A is a flow diagram of an antenna selection algorithm 400 in which the coding/modulation scheme (e.g., data rate or throughput) is fixed or adapted to long-term principles (e.g., to accommodate wide variations in SNR). The task of the antenna selection algorithm is to determine for each data packet which pair of antenna elements 316 should be used by transmitter 302 and which pair of antenna elements 338 should be used by transmitter 330 in a given mode. For example, in this selection process, it is assumed that channel 318 is quasi-stationary (e.g., channel 318 is constant during data packet transmission and varies independently between two adjacent data packets). Although the channel 318 exhibits some frequency selectivity, the antenna selection is common to the entire frequency bandwidth.

Referring to fig. 4A, when the transmitter 302 has just been activated (step 401) and the state of the channel 318 is not yet known, a subset of a default set of two of the antenna elements 316 is used to transmit wireless signals. Similarly, receiver 330 uses a default set of a subset of two antenna elements in antenna element 338 to obtain synchronization. Next, Channel State Information (CSI) is obtained (step 402). In some embodiments according to some aspects of the present invention, the operation of obtaining CSI is performed by receiver 330. A training sequence consisting of known symbols is sent from the transmitter 302 to the receiver 330. In the receiver 330, the channel 318 is estimated based on the received signal and the known symbol sequence. This estimation operation is performed each time the channel 318 changes, as is done for each data packet. To successfully perform this selection method, a complete channel matrix estimation should be performed over the entire frequency bandwidth (e.g., channel path gain estimation at all tones (errors all tones) from all antenna elements 316 of transmitter 302 to all antenna elements 338 of receiver 330). J.J.Van de Beek et al set forth training sequence-based channel estimation techniques applicable to MIMO Systems in the IEEE 45 th "Vehicular technology conference" 1995, 25-28.7.2, 815. multidot. On channel estimation in OFDM Systems "and in the IEEE Globecom2001, first volume, 509. multidot. 513, the" Synchronization for MIMO OFDM Systems "by A.N.Mody and G.L.Stuber. These papers are referred to throughout this specification.

Referring again to FIG. 4, the mode information is obtained by executing a link adaptation algorithm (step 404). In the embodiment shown in fig. 4A, the mode change occurs slowly. A link adaptation algorithm can thus be used to determine which of the possible candidate modes is the most suitable to use from the long term average SNR point of view. Using the link adaptation algorithm ensures that the most efficient mode is used, given the mode selection criteria (e.g., maximum data rate and minimum transmit power), from the perspective of long-term fluctuating channel/SNR conditions. A typical link adaptation algorithm that can be used in a frequency selective MIMO system is described in the IEEE journal of communications 2002, volume 40, No. 6, page 108-. In general, mode selection is independent of the method of selecting the transmitter/receiver antenna elements. The selection of the mode may be based exclusively on long-term SNR statistics. Thus, the rate of mode change is much slower than the rate of antenna selection. In other words, the selection algorithm may select a new subset of antennas for each data packet implementation, with the pattern change as a function of the long-term SNR change.

Steps 406, 408 and 410 are repeated in a round-robin fashion until all possible combinations of subsets of transmit/receive antenna elements are estimated (step 411). For example, considering a MIMO-OFDM system of the type shown in fig. 3 (e.g., configured with 4 transmit antenna elements 316 and 4 receive antenna elements 338), the complete channel matrix may be represented in the frequency domain by a 4 x 4 matrix, denoted by H, in the k (tone k) modulation_kAnd (4) showing. After selecting a subset of two antennas at each end, the sub-signalsThe size of the track matrix being reduced to a 2 x 2 matrix, usingAnd (4) showing. A combination of 2 elements selected from a total of 4 antenna elements $\left(\begin{array}{c}4\\ 2\end{array}\right)=6$ And (4) selecting the method possibly. Because antenna selection is applied to both the transmitter 302 and the receiver 330

The total number of possible combinations of (a) is 36. In general, in the case where the size of an M × M MIMO system is reduced to an n × n MIMO system (M > n), a possible combination of n antenna elements selected from M possible antenna elements is $\left(\begin{array}{c}M\\ n\end{array}\right)=\frac{M!}{n!(M-n)!}$ And (4) seed preparation. When antenna selection is performed at both the transmitter and the receiver,

is that the total number of possible combinations ofThis corresponds to the number of times the loop including steps 406, 408 and 410 is repeated. This iterative process may be performed in a serial manner (e.g., reusing common processing resources) or in a parallel manner (e.g., expending additional processing resources). In a typical embodiment, all possible antenna combinations can be processed simultaneously, where each possible antenna combination uses independent processing resources.

Comprising steps 406, 408 and 410Each repetition of the cycle affects the processing of one antenna subsystem. First, a 2 × 2 matrix of (k ═ 1, … …, Nt) on all tones (tones) is obtained for the desired subsystem

(step 406). Next, a processed signal to interference plus noise ratio (SINR) is calculated at each tone k and for each transmitted spatially multiplexed signal (step 408). The SINR is often solved using a closed-form solution based on the signal processing techniques used by the transmitter 302 and/or receiver 330, such as Maximum Ratio Combining (MRC), Minimum Mean Square Error (MMSE), eigen-beamforming, and Maximum Likelihood (ML). For example, if the transmitter 302 does not perform spatial processing and the receiver 330 uses MMSE, the SINR may be determined as follows:

computing <math><mrow> <msub> <mi>B</mi> <mi>k</mi> </msub> <mo>=</mo> <msubsup> <mover> <mi>H</mi> <mo>~</mo> </mover> <mi>k</mi> <mi>H</mi> </msubsup> <msub> <mover> <mi>H</mi> <mo>~</mo> </mover> <mi>k</mi> </msub> <mo>+</mo> <mfrac> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> </mfrac> <msub> <mi>I</mi> <mn>2</mn> </msub> </mrow></math> Wherein, $I_2=\left[\begin{array}{c}10\\ 01\end{array}\right],$ σ²and σ_s ²Representing noise and signal power, respectively, k being 1, … …, Nt (step 408-1).

Computing $C_k=1/\mathrm{diag}\left(B_k^{-1}\right),$ For each k equal to 1, … …, Nt, C_KIs a vector of Nx 1 (step)408-2)。

Computing <math><mrow> <msub> <mi>SINR</mi> <mi>k</mi> </msub> <mo>=</mo> <mfrac> <msubsup> <mi>&sigma;</mi> <mi>s</mi> <mn>2</mn> </msubsup> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mfrac> <msub> <mi>C</mi> <mi>k</mi> </msub> <mo>-</mo> <mn>1</mn> <mo>,</mo> </mrow></math> SINR for each k equal to 1, … …, Nt_kIs a vector of Nx 1

(step 408-3).

In step 410, the SINR information is converted to BER information (as in step 404) taking into account the current mode. Since the BER may be a complex function of the channel 318 and the coding/modulation and antenna combining techniques used, an approximate expression of BER is employed. The approximate expression may also be a function of the channel 318 and the applicable coding/modulation and antenna combination technique. The BER on a data packet (e.g., the output of a Viterbi decoder if coding is used) of a transmitted sub-stream i may be represented as a set of SINRs_kK — 1, … …, a non-linear unknown function f of Nt, such as:

<math><mrow> <mover> <msub> <mi>BER</mi> <mi>i</mi> </msub> <mo>&OverBar;</mo> </mover> <mo>=</mo> <mi>f</mi> <mrow> <mo>(</mo> <mo>{</mo> <msubsup> <mi>SINR</mi> <mi>k</mi> <mi>i</mi> </msubsup> <mo>}</mo> <mo>)</mo> </mrow> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <mi>N</mi> <mo>;</mo> <mi>k</mi> <mo>=</mo> <mi>i</mi> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msub> <mi>N</mi> <mi>t</mi> </msub> </mrow></math>

the function f is then approximated by some known function. Specifically, the output BER is approximately represented by the average bit error rate of the channel, such as:

<math><mrow> <mover> <msub> <mi>BER</mi> <mi>t</mi> </msub> <mo>&OverBar;</mo> </mover> <mo>&ap;</mo> <mn>1</mn> <mo>/</mo> <msub> <mi>N</mi> <mi>t</mi> </msub> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>N</mi> <mi>t</mi> </msub> </munderover> <msubsup> <mi>BER</mi> <mi>k</mi> <mi>i</mi> </msubsup> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow></math>

wherein BER_k ⁱIs the bit error rate of the spatial multiplexing sub-code stream i at a given SINR on the modulation k. In another embodiment, BER_kIs the bit error rate at a given signal-to-noise ratio (SNR) on the k-tune. When BER_kIs the error rate at a given SINR over the channel time samples k, the average can also be taken in the time domain. BER_kMay be a bit error rate corresponding to a given signal component, such as a channel modulation or control delay.

Alternatively, the BER can be approximated by some simple closed-form function_k. Simulations have shown that the behavior of the average bit error rate BER in 802.1-1a mode 1 (e.g. BPSK, R1/2) with respect to SINR or SNR (in some embodiments, the BER normalization factor 1/Nt can be ignored because this factor does not affect antenna selection) can be simulated, for example: by passing

<math><mrow> <mover> <mi>BER</mi> <mo>&OverBar;</mo> </mover> <mo>&ap;</mo> <mo>-</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>N</mi> <mi>t</mi> </msub> </munderover> <mi>tanh</mi> <mrow> <mo>(</mo> <msubsup> <mi>SINR</mi> <mi>k</mi> <mi>i</mi> </msubsup> <mo>)</mo> </mrow> <mo>;</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <mi>N</mi> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow></math>

Available-tanh (SINR)_k) Approximately representing the BER of the signal component k.

The function tanh is not always able to adequately approximate the BER, especially for different modulation techniques. When using a particular technique, the BER can be better approximated by the following function:

1) BER of uncoded BPSK modulation in AWGN channels is (see, e.g., J.G.proakis, Digital Communications, 3)^rdED.McGraw-Hill Series，1995)：

<math><mrow> <msub> <mi>BER</mi> <mi>BPSK</mi> </msub> <mo>=</mo> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mrow> <mn>2</mn> <mi>E</mi> </mrow> <mi>b</mi> </msub> <msub> <mi>N</mi> <mi>o</mi> </msub> </mfrac> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <msub> <mrow> <mn>2</mn> <mi>&gamma;</mi> </mrow> <mi>b</mi> </msub> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <msqrt> <msub> <mi>&gamma;</mi> <mi>b</mi> </msub> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <msqrt> <msub> <mi>&gamma;</mi> <mi>s</mi> </msub> </msqrt> <mo>)</mo> </mrow> </mrow></math>

The form of the function erfc can be reasonably approximated by the following function (compared to y ═ tanh (x)):

$y=-[(1-e^{-2\sqrt{x}})+(1-e^{-1.8x})]$

2) BER of uncoded QPSK modulation in AWGN channel is (see, e.g., J.G.proakis, Digital Communications, 3)^rdED.McGraw-Hill Series，1995)

<math><mrow> <msub> <mi>BER</mi> <mi>QPSK</mi> </msub> <mo>=</mo> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mrow> <mn>2</mn> <mi>E</mi> </mrow> <mi>b</mi> </msub> <msub> <mi>N</mi> <mi>o</mi> </msub> </mfrac> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <mn>2</mn> <msub> <mi>&gamma;</mi> <mi>b</mi> </msub> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <msqrt> <msub> <mi>&gamma;</mi> <mi>b</mi> </msub> </msqrt> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mi>&gamma;</mi> <mi>s</mi> </msub> <mn>2</mn> </mfrac> </msqrt> <mo>)</mo> </mrow> <mo>.</mo> </mrow></math>

The following function can be used to better approximate y-tanh (x)

In the form of:

$y=-[(1-e^{-1.3\sqrt{x}})+(1-e^{-x})]$

3) the uncoded 16QAM modulated BER in an AWGN channel can be derived from a given symbol error rate, for example from j.g. proakis, Digital Communications，3^rdmcgraw-Hill Series, 1995, derives the symbol error rate given:

<math><mrow> <msub> <mi>BER</mi> <mrow> <mn>16</mn> <mi>QAM</mi> </mrow> </msub> <mo>=</mo> <mn>1</mn> <mo>-</mo> <msqrt> <mn>1</mn> <mo>-</mo> <mfrac> <mn>3</mn> <mn>2</mn> </mfrac> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mrow> <mn>3</mn> <mi>E</mi> </mrow> <mi>s</mi> </msub> <mrow> <mn>15</mn> <msub> <mi>N</mi> <mi>o</mi> </msub> </mrow> </mfrac> </msqrt> <mo>)</mo> </mrow> </msqrt> <mo>=</mo> <mn>1</mn> <mo>-</mo> <msqrt> <mn>1</mn> <mo>-</mo> <mfrac> <mn>3</mn> <mn>4</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <mfrac> <msub> <mi>&gamma;</mi> <mi>s</mi> </msub> <mn>10</mn> </mfrac> <mo>)</mo> </mrow> </msqrt> </mrow></math>

one suitable fitting function is y ═ - (1-e)^-02x)。

4) The uncoded 64QAM modulated BER in an AWGN channel can be derived from a given symbol error rate, for example from j.g. proakis, Digital Communications, 3^rdmcgraw-Hill Series, 1995, derives the symbol error rate given:

<math><mrow> <msub> <mi>BER</mi> <mrow> <mn>64</mn> <mi>QAM</mi> </mrow> </msub> <mo>=</mo> <mn>1</mn> <mo>-</mo> <msup> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mfrac> <mn>7</mn> <mn>4</mn> </mfrac> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mi>&gamma;</mi> <mi>s</mi> </msub> <mn>21</mn> </mfrac> </msqrt> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mrow> <mn>1</mn> <mo>/</mo> <mn>3</mn> </mrow> </msup> <mo>=</mo> <mn>1</mn> <mo>-</mo> <msup> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mfrac> <mn>7</mn> <mn>8</mn> </mfrac> <mi>erfc</mi> <mrow> <mo>(</mo> <msqrt> <mfrac> <msub> <mi>&gamma;</mi> <mi>s</mi> </msub> <mn>42</mn> </mfrac> </msqrt> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mrow> <mn>1</mn> <mo>/</mo> <mn>3</mn> </mrow> </msup> </mrow></math>

a suitable fitting function is $y=-(1-e^{-0.35\sqrt{x}}).$

It will be appreciated that any fitting function that reasonably models the relationship of BER to SINR may be used in equation (2). The number of suitable fitting functions is not limited to the examples listed above.

As indicated above, steps 406 to 410 are repeated until all possible combinations of antenna subsets have been considered (step 411). As a result of the repetition process, all of the(orA set of N estimates of BER values for each possible antenna combination (one estimate for each spatially multiplexed signal). Next, a subset of antennas is selected that minimizes the mean (mean) of the set of BERs, the maximum (max) of the set of BERs, or the minimum (min) of the set of BERs (step 412).

Fig. 4B is a flow chart of an antenna selection algorithm 500 in which the coding/modulation scheme is changed the same number of times as is implemented per data packet in response to a corresponding change in the channel 318. In this embodiment, the frequency of the coding/modulation mode adjustment is the same as the frequency of the antenna selection implementation.

Referring to fig. 4B, steps 501 and 502 are similar to steps 401 and 402, respectively. As shown, steps 504 through 510 comprise a loop that is repeated until all possible antenna subset combinations have been evaluated. The number of times the cycle is repeated is equal to(selection at one end of the link) or

(selection is made at both ends of the link). In this regard, steps 504 and 506 are similar to steps 406 and 408, respectively. Based on knowledge of the instantaneous SINR at all tones, the link adaptation module determines the most efficient mode for each spatially multiplexed signal given the mode selection criteria (e.g., maximum data rate and minimum transmit power) (step 508). This step is similar to step 404 except that the mode determination is based on instantaneous SNR (or SINR) statistics rather than long-term SNR (or SINR) statistics. Thus, different combinations of antenna subsets may result in different mode determinations. Finally, step 510 calculates or determines the corresponding BER in the manner described in step 410, given the instantaneous SINR and mode information.

Again, steps 504 through 510 are performed until all possible antenna combinations have been considered (step 511). Once the repeated execution is completed, all the data are obtained(or

A set of N estimates of BER values for a possible antenna combination (e.g., one estimate for each spatially multiplexed signal). The difference between the selection algorithm 500 and the selection algorithm 400 is that(orOne possible antenna combination need not use the same coding/modulation mode. Therefore, which subset of antennas is selected depends not only on the minimization of BER but also on the mode (e.g., data rate or throughput). Following step 512 in the selection algorithm 500, in selecting the subset of antenna elements to make the final determination, there are several exemplary options:

option 1

1) All combinations of antenna subsets using the same pattern are grouped into a common pool (common pool).

2) The pool (pool) corresponding to the highest mode (which achieves the maximum data rate) is selected.

3) In the pool, the combination of antenna subsets that minimizes BER is selected in a substantially similar manner as in step 412.

Option 2

Regardless of the mode used for each combination, the combination of antenna subsets that minimizes BER is selected in a substantially similar manner as step 412.

Option 3

A mixed version of option 1 and option 2 is implemented, for example:

1) all combinations of antenna subsets using the same pattern are grouped into a common storage pool.

2) X storage pools are selected that correspond to the X highest modes (which achieve the maximum data rate), where X is an integer, equal to 1 or 2 or 3, etc.

3) In these storage pools, the combination of antenna subsets that minimizes BER is selected in a substantially similar manner as in step 412.

Fig. 5 is a diagram illustrating Packet Error Rate (PER) as a function of SNR after a typical antenna selection technique is employed in an SM-MIMO-OFDM system in a noise-limited environment. For example, the results of fig. 5 may be applied to a system using 4 transmit and receive antennas in the exemplary manner shown in fig. 3. The result reflects only the case of a data packet of size 1000 bytes and a fixed coding/modulation pattern in some examples. The results also reflect that both the applicable transmitter and receiver incorporate two typical RF chains. In addition, the results of fig. 5 use BPSK modulation, the code rate is 1/2 (e.g., mode 1 of 802.11 a), the channel model is characterized as "channel a" (e.g., 50ns rms delay spread, 0.5 antenna correlation), and the fitting function is tanh.

The legend for the curves in fig. 5 is as follows:

2 × 22SM-MIMO MMSE: this system corresponds to a 2 Spatial Multiplexing (SM) -MIMO-OFDM system using 2 transmit antennas and 2 receive antennas with 2SM signals. Since the number of antennas is equal to the number of SM signals, antenna selection is not applied. The baseband combining scheme is used in the receiver to separate the 2 sub-streams, such as MMSE.

4 × 42SM-MIMO sel mcap MMSE: this system corresponds to a space-multiplexed (SM) -MIMO-OFDM system using 4 transmit antenna elements and 4 receive antenna elements with 2SM signals. Conventional selection methods are applied at both the transmitter and the receiver to select a subset of 2 antenna elements among the 4 antenna elements according to a maximum capacity criterion. After selecting a subset of antennas at the receiver, MMSE is applied to the baseband to separate 2 sub-streams.

2 × 42SM-MIMO sel number MMSE (bound): this system corresponds to a 2 Spatial Multiplexing (SM) -MIMO-OFDM system using 2 transmit antenna elements and 4 receive antenna elements with 2SM signals. The selection method is applied only at the receiving end to select a subset of 2 antenna elements among the 4 antenna elements according to a minimum BER criterion. In this case, the BER is not approximated using a quasi-holofunction. Instead, it is assumed that the BER is already fully known. This is not easily achieved but provides a range of performance that can be achieved by using some embodiments of aspects of the present invention.

4 × 42SM-MIMO sel number MMSE (bound): this system corresponds to a space-multiplexed (SM) -MIMO-OFDM system using 4 transmit antenna elements and 4 receive antenna elements with 2SM signals. A selection method is applied at both the transmitting end and the receiving end to select a subset of 2 antenna elements out of 4 antenna elements according to a minimum BER criterion. In this case, the BER is not approximated using a fitting function. Instead, it is assumed that the BER is already fully known. This is not easily achieved but provides a range of performance that can be achieved by using some embodiments of aspects of the present invention.

4 × 42SM-MIMO sel number MMSE (initialization tanh): this system corresponds to a space-multiplexed (SM) -MIMO-OFDM system using 4 transmit antenna elements and 4 receive antenna elements with 2SM signals. The selection method according to some embodiments of the invention is applied at both the transmitting end and the receiving end to select a subset of 2 antenna elements out of 4 antenna elements according to a minimum BER criterion. Tanh is used as a fitting function to approximate the BER.

The results of fig. 5 show that all systems using some type of antenna selection provide gain relative to systems not using antenna selection, and that antenna selection based on the minimum BER criteria provides more gain than antenna selection based on the maximum capacity criteria. Specifically, a system applying antenna selection according to the present invention at both the transmitter and receiver achieves a gain of 7.6dB at a PER level of 10e-2 relative to a system not using antenna selection; a system applying antenna selection according to the present invention at both the transmitter and the receiver achieves a gain of 4.2dB at a PER level of 10e-2 relative to a system using antenna selection based on a maximum capacity criterion. It can be observed that the gain obtained by applying the selection method according to some embodiments of some aspects of the invention only at the receiver is smaller than the gain obtained by applying the selection method at both the transmitter and the receiver, but larger than the gain obtained without using the selection method. Finally, the performance of the system according to some embodiments of aspects of the invention closely approaches the theoretical performance range shown in fig. 5.

C. Antenna selection for SC-MIMO-OFDM systems

Shown in fig. 6 is an SC-MIMO-OFDM system 600 that uses precoding techniques in addition to antenna selection methods according to some embodiments of the present invention. In the embodiment of fig. 6, precoding involves various baseband weighting and combining schemes performed at the transmitter 602. Referring to fig. 6, the single-stream symbols 604 are weighted and combined by a set of complex factors 608 to generate a set of N output signals, where N is related to the number of RF chains 612 used in the transmitter 602. The N output signals are then passed through N RF chains 612 to generate N RF signals. The N RF signals are then coupled by switch 620 to a corresponding set of N transmit antenna elements 616 and transmitted over channel 624.

In receiver 622, a switch 630 selects N receive antenna elements from M receive antenna elements 626 to receive incoming signals over channel 624. The N RF receive signals are then processed by N RF chains 634 and converted to the digital domain for baseband processing to recover the original transmit signal.

For example, according to some embodiments of some aspects of the present invention, the baseband weighting 608 and the antenna selection method are designed together to collectively minimize the BER. For example, in some embodiments according to some aspects of the present invention, the baseband weights 608 are selected to maximize output SNR (or SINR) or capacity while antenna selection is implemented to minimize BER. Subchannel matrix corresponding to maximum singular value (singular value)

May be used to select the best subset of transmit antennas 616, the subset of receive antennas 626, and the appropriate transmit baseband weights 608 and receive baseband weights 640.Andersen, at IEEE Antennas and Propagation Magazine, vol.4/2000, 42, pages 13-16, describes the determination of baseband weighting values in the context of MIMO systems that do not employ antenna selection, and is incorporated herein by reference in its entirety.

The embodiment shown in fig. 6 may be modified by replacing the baseband weighting 608 in the transmitter 602 with a space-time coding module. In this case, according to some embodiments of the present invention, an antenna selection method is used at both the transmitter and the receiver to select the subset of antennas. In addition, the spatio-temporal coding module processes the input stream of symbols as described in "A simple transmit diversity technology for Wireless Communications" in IEEE Journal on Selected Areas in Communications "1998, volume 8, part 10, part 16, page 1451, 1458.

D. Antenna selection for DS-SS-SIMO systems

FIG. 7 is a diagram of a DS-SS SIMO system with 2 receive antenna elements 704 (n)_R2) of the receiver 700. Receiver 700 combines RAKE receiver functionality with typical antenna selection processing. As shown, the receiver 700 is provided with only a single RF link 708, and at any one time, a switch 712 connects the RF link 708 to one of the two receiving antenna units 704. The selection of which of the two antenna elements 704 to connect to the RF link 708 is determined based on a minimum BER criterion. In some embodiments according to some aspects of the present invention, BER values for the received signals corresponding to each receive antenna element 704 are calculated, and the antenna element 704 that achieves the smallest BER value is selected. Since BER is generally a complex function of the applicable channel and coding/modulation and antenna combining technique used, BER is approximately represented given the channel and antenna combining technique, making the BER variable as a function of the coding/modulation method used.

Once the best antenna in antenna unit 704 is selected, the RAKE receiver operates in the same manner in a single-input single-output (SISO) system (e.g., one antenna at each end of the link). The RAKE receiver uses J correlators 720 (J-2 in fig. 7), eachThe correlators all correspond to one of the first J independent multipath components. Each such component is associated with a respective time delay τ_JAnd J is 1, … …, J. The output (e.g., fingers) of each correlator 720 is then weighted 730 and combined 740 to form a single output 750 of the received signal, which output 750 comprises an estimate of the transmitted signal.

In one embodiment, at the input of the RAKE receiver, the received signal corresponding to the ith antenna element 704 may be represented as:

<math><mrow> <msub> <mi>r</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>l</mi> <mo>=</mo> <mn>1</mn> </mrow> <msub> <mi>L</mi> <mi>i</mi> </msub> </munderover> <msub> <mi>h</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </msub> <msqrt> <mn>2</mn> <mi>P</mi> </msqrt> <mi>d</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <msub> <mi>&tau;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </msub> <mo>)</mo> </mrow> <mi>p</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <msub> <mi>&tau;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </msub> <mo>)</mo> </mrow> <mi>cos</mi> <mrow> <mo>(</mo> <msub> <mi>w</mi> <mn>0</mn> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <msub> <mi>&tau;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&theta;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>l</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>+</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow></math>

wherein L is_iIs the number of taps (tap) in the channel received by the ith antenna element 704, h_i，lIs antenna i and tap

P is the signal transmission power, d is the data sequence comprising symbols of period T, P is the spreading sequence comprising chips of period Tc T/G, where G is the spreading factor. In addition, tau_i，lCorresponding to the tap

And path delay of antenna i, w_oCorresponding to the carrier frequency, w_o＝2πf₀，θ_i，lCorresponding to the tap

And the phase shift of antenna i. Noise n measured at the i-th antenna element 704_iIs modeled as a two-terminal spectral density of N₀AWGN processing of/2. For simplicity and clarity of expression, equations

(3) Assume a single user context. However, the present invention is not limited to this assumption, and the present invention can be applied to a multi-user environment.

At the output of the j-th finger correlator 720, the received signal is represented as follows:

<math><mrow> <msub> <mi>r</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>2</mn> <mi>T</mi> </mfrac> </msqrt> <msubsup> <mo>&Integral;</mo> <msub> <mi>&tau;</mi> <mi>j</mi> </msub> <msub> <mi>&tau;</mi> <mi>j</mi> </msub> </msubsup> <msub> <mi>r</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mi>p</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <msub> <mi>&tau;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mi>cos</mi> <mrow> <mo>(</mo> <msub> <mi>w</mi> <mn>0</mn> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <msub> <mi>&tau;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mo>-</mo> <msub> <mi>&theta;</mi> <mi>j</mi> </msub> <mo>)</mo> </mrow> <mi>dt</mi> <mo>=</mo> <msqrt> <mi>PT</mi> </msqrt> <msub> <mi>h</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <msub> <mi>d</mi> <mn>0</mn> </msub> <mo>+</mo> <msub> <mi>n</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>.</mo> <mo>)</mo> </mrow> </mrow></math>

wherein d is₀Is the desired symbol to be modulated, n_i，lIs a mean value of 0 and a spectral density of N at both ends₀AWGN noise component of/2. Also for the sake of simplicity and clarity of expression, assume that there is no in-path interference (IPI) in equation (4). However, the present invention can also be used in environments where IPIs are present.

After diversity combining, the final output of the RAKE receiver corresponding to the ith antenna element 704 is:

<math><mrow> <msub> <mi>r</mi> <mi>i</mi> </msub> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>J</mi> </munderover> <msub> <mi>w</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <msub> <mi>r</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>5</mn> <mo>.</mo> <mo>)</mo> </mrow> </mrow></math>

where J is the number of fingers of the RAKE receiver, the most appropriate combining weights are typically chosen to match the channel, such as:

$w_{i,j}=h_{i,j}^{*}---(6.)$

in this environment, RAKE performs the maximum ratio combining, and the SNR at the RAKE output corresponding to the i-th antenna element 704 is:

<math><mrow> <msub> <mi>&gamma;</mi> <mi>i</mi> </msub> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>J</mi> </munderover> <msub> <mi>&gamma;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>7</mn> <mo>.</mo> <mo>)</mo> </mrow> </mrow></math>

wherein, γ_i，jIs the combined SNR on the jth path associated with the ith antenna element 704. Based on equation (4), it can be expressed by the following equation:

<math><mrow> <msub> <mi>&gamma;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>=</mo> <mfrac> <mrow> <msup> <mrow> <mo>|</mo> <msub> <mi>h</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mo>|</mo> </mrow> <mn>2</mn> </msup> <mi>P</mi> </mrow> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>8</mn> <mo>.</mo> <mo>)</mo> </mrow> </mrow></math>

wherein, <math><mrow> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> <mo>=</mo> <mfrac> <msub> <mi>N</mi> <mn>0</mn> </msub> <mn>2</mn> </mfrac> <mo>&CenterDot;</mo> <mfrac> <mn>2</mn> <mi>T</mi> </mfrac> </mrow></math> is the noise power.

Can be selected from pairs of gamma_iThe BER of the output of the RAKE receiver corresponding to the i-th antenna unit 704 is obtained from the knowledge of the likelihood density function (PDF). For example, if no coding is used for the data sequence, BPSK modulation is applied, e.g. J.G.Proakis, McGraw-Hill series 1995, thThe method described in "Digital communications" version 3, by way of expression

Is combined to gamma_iThe BER can be found in the likelihood density function (PDF) of (1), such as:

<math><mrow> <msub> <mi>BER</mi> <mi>i</mi> </msub> <mo>=</mo> <msubsup> <mo>&Integral;</mo> <mn>0</mn> <mo>&infin;</mo> </msubsup> <mi>Q</mi> <mrow> <mo>(</mo> <msqrt> <msub> <mrow> <mn>2</mn> <mi>&gamma;</mi> </mrow> <mi>i</mi> </msub> </msqrt> <mo>)</mo> </mrow> <msub> <mi>p</mi> <mi>r</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>&gamma;</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mi>d</mi> <msub> <mi>&gamma;</mi> <mi>i</mi> </msub> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>9</mn> <mo>.</mo> <mo>)</mo> </mrow> </mrow></math>

once the BERs of all the receiving antennas have been estimated, the antenna element that achieves the minimum BER is selected 704:

$\underset{i=1,...n_R}{\min }\left\{{\mathrm{BER}}_i\right\}---\left(10\right)$

wherein n is_RRepresenting the total number of receive antenna elements.

It is clear that the analog function used in equation (9) for estimating BER will need to be changed when coding (e.g., turbo coding, convolutional coding) and other modulation stages are added to the system. In some embodiments according to aspects of the present invention, the exemplary antenna selection algorithm can use any fitting function that accurately models the BER behavior in a given system. Typically, the fitting function depends on one or more of the following parameters, for example: channel, coding and/or modulation used, signal processing at the transmitting and/or receiving end, receiver SNR, and other parameters.

The exemplary embodiment shown in fig. 7 may be extended to a two-dimensional RAKE receiver in which the processing is performed in both the spatial and temporal domains. Within this range, a subset of N (N > 1) antennas (M > N) from the total of M antennas may be selected in conjunction with a typical antenna selection algorithm, the selected subset minimizing the BER at the two-dimensional RAKE output.

While the invention has been described with reference to certain embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.

Claims (10)

1. A method of selecting a subset of receive antennas of a receiver to receive a transmitted RF signal, comprising:

establishing a possible subset of the receive antennas;

determining a set of channel parameter statistics corresponding to the possible subset of receive antennas;

calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics;

selecting a particular possible subset of receive antennas based on criteria based on the calculated output error rate, the transmission mode, a minimum value of the output error rate, and a maximum value of the data rate;

connecting one or more RF chains of the receiver to the receive antennas of the selected particular possible subset; wherein,

at least one of said output bit error rates is approximated by a closed-form expression, said closed-form expression including at least one of:

a)y＝-tanh(x)，

b)

c)

d)y＝-(1-e^-0.2x) Or is or

e)

And using, in the receiver, receive baseband weighting to maximize output signal-to-noise ratio, signal-to-interference-and-noise ratio, or capacity, while selecting receive antennas to minimize output error rates; wherein,

said selecting a particular possible subset of receive antennas comprises the steps of:

grouping all combinations of antenna subsets that use the same data rate into a common storage pool;

selecting n storage pools corresponding to n highest data rates, wherein n is 1, 2 or 3;

and selecting the antenna subset which can minimize the average value of the error rates, the maximum value of the error rates or the minimum value of the error rates in the storage pool.

2. The method of claim 1, further comprising:

selecting transmission modes, wherein the selection of each transmission mode is based on at least one set of channel parameter statistics.

3. The method of claim 1, wherein each set of channel parameter statistics comprises at least one of an output signal-to-noise ratio of the receiver, an output signal-to-interference-and-noise ratio, a log-likelihood ratio, and a euclidean distance of a signal constellation.

4. The method of claim 1, wherein each set of channel parameter statistics comprises at least one of first order statistics, second order statistics, and higher order statistics.

5. A method of selecting a subset of transmit antennas of a transmitter to transmit an RF input signal as a plurality of output RF signals for subsequent reception by a receiver, the method comprising:

establishing possible subsets of transmit antennas;

determining a set of channel parameter statistics corresponding to the possible subset of transmit antennas;

selecting a transmission mode corresponding to each group of channel parameter statistics;

calculating output bit error rates of the receiver, each output bit error rate calculated based on at least one set of channel parameter statistics and at least one selected transmission mode;

selecting a particular possible subset of transmit antennas based on criteria based on at least the calculated output error rate, the transmit mode, a minimum value of the output error rate, and a maximum value of the data rate;

connecting one or more RF chains of the transmitter to the transmit antennas of the selected particular possible subset;

wherein,

at least one of said output bit error rates is approximated by a closed-form expression, said closed-form expression including at least one of:

a)y＝-tanh(x)，

b)

c)

d)y＝-(1-e^-0.2x) Or is or

e)

And using transmit baseband weighting in the transmitter to maximize output signal-to-noise ratio, signal-to-interference-and-noise ratio, or capacity, while selecting transmit antennas to minimize output error rates;

said selecting a particular possible subset of transmit antennas comprises the steps of:

grouping all combinations of antenna subsets that use the same data rate into a common storage pool;

selecting n storage pools corresponding to n highest data rates, wherein n is 1, 2 or 3;

and selecting the antenna subset which can minimize the average value of the error rates, the maximum value of the error rates or the minimum value of the error rates in the storage pool.

6. The method of claim 5, wherein each set of channel parameter statistics comprises at least one of an output signal-to-noise ratio of the receiver, an output signal-to-interference-and-noise ratio, a log-likelihood ratio, and a Euclidean distance of a signal constellation.

7. In a communication system including a transmitter and a receiver, the transmitter including a transmit antenna for transmitting a set of spatially multiplexed RF output signals over a channel using two or more RF transmit chains, the receiver including a receive antenna for receiving the set of spatially multiplexed RF output signals and correspondingly generating a set of spatially multiplexed received RF signals processed by the two or more RF chains, a method of selecting an antenna, the method comprising:

establishing possible subsets of the transmit antennas and possible subsets of the receive antennas;

determining channel parameter statistics corresponding to various combinations of the possible subsets of the transmit antennas and the possible subsets of the receive antennas;

selecting transmission modes respectively corresponding to various channel parameter statistics;

calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics and the selected corresponding transmission mode;

selecting a particular possible subset of transmit antennas and a particular possible subset of receive antennas based on a criterion based on at least the calculated output error rate, the transmit mode, a minimum value of the output error rate, and a maximum value of the data rate;

connecting the two or more RF transmit chains to a possible subset of the selected particular transmit antennas;

connecting the two or more RF receive chains to a possible subset of the selected particular receive antennas;

wherein,

at least one of said output bit error rates is approximated by a closed-form expression, said closed-form expression including at least one of:

a)y＝-tanh(x)，

b)

c)

d)y＝-(1-e^-0.2x) Or is or

e)

And using transmit baseband weighting and receive baseband weighting in the transmitter and receiver, respectively, to maximize output signal-to-noise ratio, signal-to-interference-and-noise ratio, or capacity, while selecting transmit and receive antennas to minimize output error rates;

said selecting a particular possible subset of said transmit antennas and a particular possible subset of said receive antennas comprises the steps of:

grouping all combinations of antenna subsets that use the same data rate into a common storage pool;

selecting n storage pools corresponding to n highest data rates, wherein n is 1, 2 or 3;

and selecting the antenna subset which can minimize the average value of the error rates, the maximum value of the error rates or the minimum value of the error rates in the storage pool.

8. The method of claim 7, wherein each set of channel parameter statistics comprises at least one of an output signal-to-noise ratio of a receiver, an output signal-to-interference-and-noise ratio, a log-likelihood ratio, and a euclidean distance of a signal constellation.

9. In a communication system including a transmitter and a receiver, the transmitter including a transmit antenna for transmitting a set of RF output signals over a channel using one or more RF transmit chains, the receiver including a receive antenna for receiving the set of RF output signals and correspondingly generating a set of received RF signals for processing by one or more RF receive chains, a method of selecting an antenna, the method comprising:

establishing possible subsets of the transmit antennas and possible subsets of the receive antennas;

determining channel parameter statistics corresponding to various combinations of the possible subsets of the transmit antennas and the possible subsets of the receive antennas;

selecting transmission modes respectively corresponding to various channel parameter statistics;

calculating output error rates of the receiver, each output error rate calculated based on at least one set of channel parameter statistics and the selected corresponding transmission mode;

selecting a particular possible subset of transmit antennas and a particular possible subset of receive antennas based on a criterion based on at least the calculated output error rate, the transmit mode, a minimum value of the output error rate, and a maximum value of the data rate;

connecting the one or more RF transmit chains to the selected possible subset of specific transmit antennas;

connecting the one or more RF receive chains to the selected possible subset of the particular receive antennas;

wherein,

at least one of said output bit error rates is approximated by a closed-form expression, said closed-form expression including at least one of:

a)y＝-tanh(x)，

b)

c)

d)y＝-(1-e^-0.2x) Or is or

e)

And using transmit baseband weighting and receive baseband weighting in the transmitter and receiver, respectively, to maximize output signal-to-noise ratio, signal-to-interference-and-noise ratio, or capacity, while selecting transmit and receive antennas to minimize output error rates;

said selecting a particular possible subset of said transmit antennas and a particular possible subset of said receive antennas comprises the steps of:

grouping all combinations of antenna subsets that use the same data rate into a common storage pool;

selecting n storage pools corresponding to n highest data rates, wherein n is 1, 2 or 3;

and selecting the antenna subset which can minimize the average value of the error rates, the maximum value of the error rates or the minimum value of the error rates in the storage pool.

10. A system for selecting N antenna elements from a transmitter of M antenna elements or a receiver of M antenna elements, where N is less than M, comprising:

m antenna elements in a transmitter of M antenna elements or a receiver of M antenna elements;

n RF links; and

a switch connected to the N RF chains,

wherein the receiver of the M antenna elements calculates an output error rate for each possible subset of N antenna elements of the M antenna elements, each output error rate calculated based at least on a set of channel parameter statistics,

the receiver of the M antenna elements selects a particular subset of N antenna elements based on criteria based on the calculated output error rate, transmission mode, minimum of output error rates, maximum of data rates; and

in response to selecting a particular subset of N antenna elements based on the criteria, the switch connects the N RF chains to the N antenna elements of the particular subset of N antenna elements;

wherein,

at least one of said output bit error rates is approximated by a closed-form expression, said closed-form expression including at least one of:

a)y＝-tanh(x)，

b)

c)

d)y＝-(1-e^-0.2x) Or is or

e)

And using transmit baseband weighting and receive baseband weighting in the transmitter and receiver, respectively, to maximize output signal-to-noise ratio, signal-to-interference-and-noise ratio, or capacity, while selecting transmit and receive antennas to minimize output error rates;

said selecting a particular possible subset of said transmit antennas or a particular possible subset of said receive antennas comprises the steps of:

grouping all combinations of antenna subsets that use the same data rate into a common storage pool;

selecting n storage pools corresponding to n highest data rates, wherein n is 1, 2 or 3;

and selecting the antenna subset which can minimize the average value of the error rates, the maximum value of the error rates or the minimum value of the error rates in the storage pool.

Images