CA2465775C - Multiple input multiple output (mimo) wireless communications system - Google Patents

Abstract

A MIMO wireless communications system includes first and second transceivers and respective first and second antennas, wherein the first transceiver and the first antenna are operable to transmit and receive signals propagating along at least one signal path using at least one coding and modulation scheme, and the second transceiver and the second antenna are also operable to transmit and receive signals propagating along the at least one signal path using the at least one coding and modulation scheme. Each first and second antenna includes dual polarized antennas for operating at different polarizations to transmit first and second signals, one signal being a delayed version of another one of the first and second signals. Further, the system may include another one of the first and second antennas having first and second spatially separated antenna elements operating at different polarizations configured to provide signal delay diversity.

Description

TITLE OF THE INVENTION
Multiple Input Multiple Output (MIMO) Wireless Communications System FIELD OF THE INVENTION
The present application relates generally to wireless communication systems and methods, and more specifically to a low-cost high performance wireless communication system capable of providing non-line-of-sight communications.
Because wireless communications systems are increasingly being installed by data communications installers rather than highly trained wireless communications installers, the present application further relates to wireless communications systems that allow easier installation.

BRIEF SUMMARY OF THE INVENTION
Installation Ease of installation of the presently disclosed wireless communications system is achieved by (1) a diagnosis lamp such as a light emitting diode (LED), (2) a special acquisition mode, (3) audio feedback, and (3) dynamic frequency selection (DFS). These features come together to make the radio and electrical aspects of system installation easier, thereby allowing the installer to concentrate on the physical aspects of installation.

Link Optimization To make non-line-of-sight (NLoS) communications possible, the following three system capabilities are provided:
(1) The ability to operate in a dispersive channel.
This is provided by the use of orthogonal frequency division modulation (OFDM) with a particular set of parameters to accommodate the amount of dispersion and the channel dynamics.

(2) The ability to overcome losses due to obstructions such as trees and buildings. The loss is multiplicative due to the free space loss and is usually called excess path loss. This requires a high system gain, which is achieved by high power output, a low noise receiver, low system loss, and high gain antennas.

(3) The ability to mitigate fading, which accompanies the excess path loss. This is achieved through the use of Space Time Coding (STC).
Once the" system has the ability to perform basic communications in the NLoS environment, some other features are required to assure that, at any instant, communication takes place at the fastest possible rate. These features include the following:
(a) Adaptive modulation, which adjusts the modulation mode used on the link based upon the signal relative to the noise and distortion (SINAD) that would be available when operating in the next mode (up or down).
(b) Dynamic frequency selection (DFS), which measures the interference on each channel and selects the best channel available, taking account of the receiver having automatic retransmission request (ARQ).
Other features, functions, and aspects of the invention will be evident from the Detailed Description of the Invention that fo_lows.

BRIEF DESCRIPTION-OF THE SEVERAL VIEWS OF THE DRAWINGS
The invention will be more fully understood with reference to the following Detailed Description of the Invention in conjunction with the drawings of which:
Fig. 1 is a- diagram of an illustrative radio path configuration employed by the wireless communications system according to the present invention;
Fig. 2 is a diagram of typical non-line-of-sight fading signals, before and after combining;
Fig. 3 is a diagram illustrating line-of-sight signal propagation over water;
Fig. 4 is a vector diagram of the sum of two signals received at point B of Fig. 3;
Fig. 5 is a diagram of the resultant variation in amplitude of the signal at point B of Fig. 3 versus height;
Fig. 6 is a block diagram of an adaptive modulation system according to the present invention;
Fig. 7 is a diagram illustrating vector error;
Fig. 8 is a diagram illustrating a measurement of variability of a channel;
Fig. 9 is a diagram illustrating a signal level-to-audio transfer characteristic;
Fig. 10 is a block diagram of a wireless system according to the present invention;

Fig. 11 is a block diagram of a representative signal processor employed in the wireless system of Fig. 10; and Fig. 12 is a block diagram of a representative transceiver employed in the wireless system of Fig. 10.

DETAILED DESCRIPTION OF THE INVENTION
In accordance with the present invention, two software-defined radios are connected as standard to a horizontally (A and X) and vertically (B and Y) polarized antenna occupying the same space (see Fig. 1). This arrangement is capable of operating in a number of interesting ways, depending upon the environment into which the link is being deployed. These ways are easy to understand in a Point-to-Point system.

Non-line-of-sight In non-line-of-sight (NLoS) environments, the radios optimally operate as one radio system, in which the two transmitters transmit signals that can be combined at the other end of the link. These signals propagate through the NLoS environment, twisting on reflection from various objects. When received at the other end of the link, the signals from the two antennas are combined using an optimum combiner or Maximum Ratio Combiner (MRC), which combines two signals (a and b) by rotating the phase of each signal to be in alignment and adjusting the amplitude of each signal to maximize the combined signal to noise ratio.
The resulting signal-to-noise ratio is c = (a2+b2) .

4-There are four paths in this arrangement, i.e., A to X, A to Y, B to X, and B to Y, in which A and X are vertically polarized antennas and B and Y are horizontally polarized antennas, as illustrated in Fig. 1. Each of these paths fade independently. In NLoS communications, the fading follows a Rician distribution. When the excess path loss is greater than about 30 dB, the fading is almost Rayleigh, which means that there is a 40 dB loss for 0.01% of the time.
The fading of each of the four paths are normally well decorrelated - the mean signal loss from A to Y and B to X
is typically 6 dB lower than the mean signal level from A to X and B to Y. The overall diversity order of this arrangement operates as though there were three full ways of diversity. This has a benefit in the NLoS environment of 25 dB.
A known transmit signal for use in this type of system is described in the document by Siavash M. Alamouti entitled A SIMPLE TRANSMIT DIVERSITY TECHNIQUE FOR WIRELESS

COMMUNICATIONS, IEEE Journal on Select Areas in Communications, Vol. 16, No. 8, October 1998, (the "Alamouti system"). In the Alamouti system, two symbols are transmitted in a special combination from the two antennas. The receiver can optimally combine each of the two transmitted signals for each of the symbols. The consecutive symbols can be combined to complete the four-way diversity. The Alamouti system is fairly complex, however, and needs a large amount of processing resources.

-5-Nearly all of the gain of the Alamouti system can be obtained by transmitting from channel B a delayed version of the signal from channel A. In the system where the transmission is mainly horizontal or vertical, the optimum combiner collects all of the main signals. The generally smaller cross polar signals add to the main signals at important moments when the signals from the co-polar have faded without much overall loss. The delay assures that the two signals are separate and do not suffer fading simultaneously on all frequencies. The FEC coding applied in the frequency domain assures that if a null occurs in the frequency domain, then that null does not cause much damage to the signal. The implementation of delay diversity is achieved as follows. At the transmitter a delay can be applied, while at the receiver a greater dispersion of the channel is allowed.
Fig. 2 shows, on the left side of the diagram, a single signal from a single channel system fading in a typical NLoS
channel - this signal has a 0.01% chance of fading 40 dB.
The time scale may be a few seconds in a windy tree environment, while being many weeks in a high rise concrete city environment. On the right side of the diagram, the four signals from the four channels (A to X, etc.) described in Fig. 1 are fading in the NLoS way, but are then combined using the optimum combiner to form the top line, which has a 0.01% chance of fading 15 dB.

Line-of-sight Diversity Although the arrangement depicted in Fig. 1 has diversity, in line-of-sight use there are only two paths,

-6-and there is a high degree of correlation between them.
When longer distances are required, there are the problems of ground/water reflection and atmospheric scintillation to overcome. Each of these can be overcome by the use of separate antennas at one end of the kink. For this purpose, a product can be supplied as integrated with the antenna at one end of the link and with a separately supplied single polarization antennas at the other end of the link. The two single polarized antennas are deployed at different heights, assuring decorrelation of the two paths.
In the case of ground/water reflection, the height can be chosen to assure that while one antenna is in a null, the antenna is simultaneously deployed near the peak. Simple geometry can be used to determine this antenna separation.
In this application, using two transmitters and two receivers operating using the horizontal and vertical polarization optimizes this communication system by allowing one end to be a small integrated antenna unobtrusively mounted on a building, while the other end can be a vertically separated pair of antennas mounted on a mast or other suitable structure(s).
Fig. 3 shows the geometry of a typical line-of-sight (LoS) path over water. The water will often create a reflection at C that is substantially equal in amplitude to the LoS path from point A to point B via path D. The signal received at point B is the vector sum of the signal travelling the two paths D and C (see Fig. 4) Note that LoS paths are not straight, but are bending due to the change in refractive index with height.

-7-Fig. 5 shows the way in which the vectors from the main path D and the reflected path C add together. Specifically, Fig. 5 shows the variation in amplitude of the signal at point B with the height of B. If the angle 0 were constant with time, then a simple solution would be to place the antenna at the height where 0 = 0. Unfortunately, two effects take place that cause the effective path lengths ADB
and ADC to vary relative to one another. One is variation in the height of the water, and the other is the variation in the refractive index of the air. A solution is to place two antennas at B at different heights and arrange that the diversity receivers collect the optimum sum of the signals from the two antennas. The optimum spacing of these antennas for this purpose is where there is a peak on one antenna for the condition where there is a trough on the other antenna. This is shown in Fig. 3.
The International Telecommunications Union (ITU) published a recommendation ITU-R P.530, which treats this subject in more detail. In addition, that document gives details on the reduction in fading achieved by spacing the antennas where the fading is due to the scintillation of the atmosphere.

Adaptive Modulation Adaptive modulation has been previously deployed, e.g., by Ensemble Communications Inc., San Diego, California, U.S.A. In the presently disclosed embodiment, adaptive modulation is enhanced with respect to the speed of operation, automatic allowances for other noise and

-8-distortion, and automatic hysteresis based upon the recent history of the channel variability.
Fig. 6 shows the overall block diagram of the adaptive modulation system. The system operation is described in the following subsections:

Automatic allowance for distortion The demodulator is continuously measuring the ratio of the signal to the noise plus distortion. This measurement is often called "vector error". If one is to operate at the maximum possible power output, some distortions are forced upon the system by the FCC regulations limiting the power output to 1 Watt. The distortion that is applied is at a known level for each of the modulation modes. This distortion is subtracted from the measured vector error to establish the environmental noise plus front end thermal noise. This then enables the adaptive modulation controller to know whether the link will work in a higher mode despite the fact that the direct measurement would suggest that it would not. This enables the maximum power to be transmitted in all modes, and the adaptive modulation to change up in speed optimally.
Fig. 7 shows diagrammatically the various measures of the vector errors. It can be seen that a change up to the next mode is possible before the vector error is available.
The signal to noise thresholds can be pre-computed using the following equation and knowing the required vector error r, 10"h10 = 10 10 + lOs/'1D.

-9-Automatic allowance for path variability The adaptive modulation controller allows a continuous measure of the dynamics of the channel, in order to switch modulation early enough to assure that the switch will take place before the signal has decayed too far. This effectively takes the form of variable hysteresis dependent upon the channel dynamics. Point-to-point NLoS channels fade at a rate that depends upon the movement of objects surrounding the path. These objects will move at different rates, depending upon (1) the stiffness of the objects, (2) where they are in the path, and (3) the variability of the wind. The technique is to measure the variation over the interval that it takes to change mode, and use three times the standard deviation of that value to set the back-off..
The averaging time should be about one minute, as the wind can vary in velocity over a period that is of the order of one minute.
Fig. 8 shows the measurement of the variability of the channel. Over a period of about one minute prior to the adaptive modulation decision, the standard deviation of di should be computed where ti is the time taken to perform a change of modulation. Note that the modulation is changed by a process whereby the receiver requests a change of modulation, and at some known time later the transmitter changes modulation. In the previous section, it was known what modulation would be required. Here, because of the time taken to change modulation, a back-off should be implemented, which is three times the standard deviation of di.

-10-For example, modulation modes may be described by a modulation, e.g., 16QAM, and a FEC code rate, e.g., 2/3.
This will result in a dB value for the SINAD required. When changing from 16QAM 1/2 to QPSK 2/3, this dB value may change from 12dB to 8dB. The time taken from request to action may be 40ms. If three times the standard deviation of differences in signal level in the 40ms periods were 2dB, then it would be necessary to change mode down at 12+2=14dB
SINAD. Changing up should take place at 12+2*2=16dB because the signal level could reverse as soon as the decision is made, causing two time periods to be necessary - one to change up and the other to change back down again.

Automatic allowance for Interference In this wireless system, which is operating in a shared band, there is likely to be interference from other communications users and radars. Since the statistics of these interferers are unknown, an automatic correction to the adaptive modulation must be applied which assures that the number of errors does not exceed the capability of any correction system of the communications traffic. In this case, two key correction systems are Transport Control Protocol (TCP) and Automatic Retransmission request (ARQ).
Within this wireless system, there is a means to detect errors as they occur in code words. If the error rate increases to a state where the number of errors can be expected to be too high, then the modulation mode is reduced. For example, an illustrative reduction in modulation may be from 16QAM 1/2 to QPSK 2/3. Typically, this will occur when adjacent errors occur, which

-11-demonstrate that the error rate is greater than 1:1000 code words. If the number of errors in the new mode is less than 1:10000, then the interference is assumed to have passed and the mode is allowed to adjust itself in the method described in the above sections.

Power Control Power reduction is applied as a function of the mode being transmitted. Also, once the received signal level is stronger than required for reliable reception, the power is reduced to lower the interference in the band for other users of the band.
Equipment with adaptive modulation. may have, for example, between 3 and 10 modes. Adaptive power control is simplified considerably if it operates at fixed power by mode until the mode reaches the highest (fastest) data rate.
(When in this fastest data rate, the power control should be allowed to vary to keep the received signal level constant.
Also, the power margin that should be applied to the received signal level should be sufficient to overcome the dynamics of the channel.) Dynamic Frequency Selection (DFS) The object for dynamic frequency selection is for the radio to work on the best channel available. In one embodiment, there are 19 frequency channels employed for transmission. Generally, the radio either instantaneously works or does not work, depending upon the signal to noise received. Automatic retransmission request (ARQ) causes the overall transmission to be about 1::106 packet error rate

-12-until the code word error rate increases above 1:1000. The noise and interference environment can vary from a high mean noise level to one where the mean noise is low, but there are very high spikes of noise that could be generated by a radar, for example. What is needed is to find the 99.9 percentile for each channel, and then choose that channel with the lowest 99.9 percentile.
The DFS measurement samples each channel regularly including the operating frequency. It does this at a moment between transmission and reception when the transmitter and receiver are not otherwise active. The measurements are passed for subsequent processing, as described below.
Accordingly, the DFS collects the measurements into 1 dB bins to form a histogram. This histogram is used to find the 99.9 percentile level, and this is reported as the level for that channel. DFS selects for use the channel that has the lowest noise/interference level using the above measurement technique for each frequency.
It is generally not desirable to have each link changing frequency often to keep up with the best perceived channel because if each radio keeps moving in an area, the environment will take a long time to settle down. The DFS
system therefore has averaging of about 20 minutes and hysteresis of about 2 dB built-in to assure that the frequency does not change until there is a good. reason to do so.

Install DFS

Automatic installation of a link, without the user needing to know the frequency to use, requires an algorithm

-13-to determine the best channel to use for the installation process. This is particularly important for unlicensed band operation.
One unit is designated as the master unit and the other unit is designated as the slave unit. The master unit of the pair searches through the frequency channels to find the quietest n channels, in which n is typically three from a possible m independent channels. The master unit then transmits sequentially on each of the three channels, waiting for its peer slave device to communicate. More than one channel is required because the slave may have a lot of noise on the quietest frequency at the master location.
The slave searches through all of the channels, listening for the master transmission showing the correct security code. The transmission is a special OFDM
transmission described below, which enables the slave to understand the signal even when the frequency or timing is in error. The slave is able to quickly lock-on to the correct frequency and timing.
This special OFDM signal contains alternately pilots and data carriers. For example, typical OFDM waveforms contain pilots. If the number of possible tones in OFDM
were 1024, then one may choose one in 16 not to be data, but instead to be a carrier frequency (i.e., a pilot) . These are similar to the carrier in Amplitude Modulation (AM) or a suppressed carrier in SSB. They provide a phase reference for the data carriers to enable accurate demodulation. The signal has a cyclic prefix of one complete OFDM symbol. The phase of the pilots are arranged to be the result of Fourier transforming a chirp signal. In this way, the pilots do not

-14-substantially add to the peak excursion of the signal. The data carrier phases are BPSK modulated with the result of hashing with the MAC address of the transmitter.
Timing of the signal is recovered by performing a sliding correlation of samples that. are exactly one OFDM
symbol length. This correlation achieves a peak value when the timing is correct. The frequency offset is proportional to the phase of this correlation times the tone spacing.
Determination of the validity of the peer unit is accomplished by establishing that the data is sufficiently correct. This is done by counting the number of correctly demodulated data carriers. When using an OFDM signal with about 800 used carriers, only 60% are necessary to be correct to be able to determine that the peer unit is the correct one with probability 1:1000. This can then be achieved at very low signal to noise level, -5 dB.
Once the slave starts communicating with the master, it quickly establishes the noise level on each of the channels.
The signal levels are communicated to the master, and after a short period a decision is made about the quietest channel for the communication.
Another mode is available where the receiver demands that communication occur simply on the quietest channel for it.
In summary, the important features of automatic installation of a link include (1) DFS quiet channel search, (2) special OFDM symbols, and (3) hashing. the MAC address to produce a pseudo-random signal to detect.

-15-Install Tones Installation of any wireless system should be achievable without any wireless-specific test equipment. To that end, the equipment must provide an observable and suitably sensitive indication that the link has been optimized by the installer. The indicator should not require any hands to operate, and should provide an indication in a noisy environment such as near a road while the installer is up a mast. Typically, an installer is required to direct the antenna within 20 of the direction of the peer equipment in both the vertical and horizontal domains. In general, this is not easy, and an important extra functionality is to detect the peer equipment at least dB below the point at which the equipment could be 15 operational.
In the preferred embodiment, an audio tone is employed.
This tone is slowly stuttered when the correct address is not seen. Slow stuttering is essential since people are unable to hear the differences in frequency when fast 20 stuttering is used. Slow stuttering in this context is defined as 2 seconds "on" and 0.5 seconds "off".
Once the correct address is seen, the audio tone is continuous, giving greater sensitivity to the frequency and therefore the signal level. Sensitivity to this level is very important, particularly in the region where we expect the most important use. Therefore, we have changed the characteristic of the tone generator to have more sensitivity in this region, as shown in Fig. 9.
The housing of the wireless unit is substantially sealed, but there is an opening in the housing through which

-16-sound can emanate. Although the tone can be arranged to be fairly loud, there are noisy environments where this cannot be easily heard. In the preferred embodiment, a simple stethoscope is provided with an adaptor to connect to the equipment opening. In this way, the tone can be heard above typical street and motorway noise.

Ethernet LED

The Ethernet LED is mounted in an indoor unit (IDU; see Fig. 10), which has no active electronics. The connection to the outdoor unit (ODU; see Fig. 10) is via a CAT 5 cable, which has four pairs of cores - two of these are used for Ethernet, one for power, and the final one for this indicator - and a reset button that also has multiple functions.
The communication of functions is achieved using two AC
frequencies, in one illustrative implementation, 50 kHz and 1.5 MHz. Presence or otherwise of one frequency (e.g., 50 kHz) is used to drive the lamp, and the other frequency (e.g., 1.5 MHz) is used to detect a short circuit for the reset by detecting a load on that frequency. In this way, the function can be achieved without any active electronics.
The indication on this LED can be used to diagnose several stages of potential problems with the wiring to the wireless unit, which may be mounted 300 feet up on a mast. These stages are identified as an early stage when the LED is off, a regular flashing stage for 10 seconds when the software has successfully loaded, and an intermittent flashing stage when there is data on the Ethernet.

-17-The multiple functions for reset include (1) resetting for 30 seconds during power up, causing the unit to go to factory defaults, or (2) resetting for 30 seconds in operation when the wireless unit changes, its IP address to 10.10.10.10. This is useful for the occasions when the IP
address has been forgotten. Maintenance can be restored without interrupting service.

Complete Transceiver Fig. 12 depicts an illustrative embodiment of the communications system. In the illustrated embodiment, the communications system is Time Division Duplex (TDD), in which the direction of communication changes alternately from one direction to the other. Some components can be shared since at each end of the wireless link transmission and reception is not taking place simultaneously. For example, the transmit modulator and the receiver signal processing may comprise the same hardware re-programmed.
Fig. 11 depicts a diagram of a signal processor employed in the transceivers of Fig. 12. In addition to the functions described above, the signal processor performs the modulation and demodulation of signals, as well as the control of the synthesizer and RF gain.
An illustrative method of performing wireless communications over a wireless link is described below with reference to Fig. 12. First, a plurality of wireless signals (sigAl, sigA2) is transmitted over the wireless link by a first wireless transceiver (Transceiver A) for, receipt by a second wireless transceiver (Transceiver B) Each wireless transceiver includes a plurality of simultaneous

-18-transmitters or receivers, and the wireless link comprises a plurality of frequency channels. The step of transmitting includes modulating at least one wireless signal (sigAl) by an orthogonal frequency division modulation technique, transmitting the modulated wireless signal over a first channel via first and second antennas (i.e., Antenna(s) 1 and Antenna(s) 2) by a first transmitter of Transmitter A, and transmitting at least one other modulated wireless signal (sigA2) via the first and second antennas (i.e., Antenna(s) 2 and Antenna(s) 1) by a second transmitter of Transmitter A by a space-time coding technique. It is noted that at any point in time, a receiver such as Receiver B is listening to one frequency channel or another frequency channel. If the receiver is being transmitted to by the peer transmitter, then communication takes place according to the wireless system diagram of Fig. 12. At other times, the receiver is listening to one of the other frequencies to determine the interference level for later comparison.
The strength of the transmitted wireless signals is then determined by, e.g., Receiver B. In the event the wireless signal strength is less than a predetermined acceptable signal strength, the modulation of the wireless signals is dynamically modified by an adaptive modulation technique to increase the probability of reliable detection of the transmitted wireless signals. It is noted that there may be an increase in the power of the transmitter, but primarily one is reducing the required SINAD necessary.
Next, respective interference levels associated with the plurality of frequency channels of the wireless link are determined. In the event the interference level associated

-19-with the first channel is greater than the interference level associated with a second channel of the wireless link, at least one modulated wireless signal is transmitted over the second frequency channel by the first transceiver by the space-time coding technique.
It should be appreciated that in non-line-of-sight communications, the two antennas Antenna 1, Antenna 2 optimally comprise one dual polarized antenna, horizontal and vertical, to reduce space requirements. In line-of-sight communications, one end will be dual polarized, while the other end will be two antennas vertically separated, one polarized horizontally and the other polarized vertically.
It will further be appreciated by those of ordinary skill in the art that modifications to and variations of the above-described method and system may be made without departing from the inventive concepts disclosed herein.

-20-

Claims (24)

We claim:

1. A multiple input multiple output (MIMO) wireless communications system, comprising: a first transceiver; at least one first antenna operatively connected to the first transceiver, wherein the first transceiver and the first antenna are operable to transmit and receive signals propagating along at least one signal path using at least one coding and modulation scheme; a second transceiver; and at least one second antenna operatively connected to the second transceiver, wherein the second transceiver and the second antenna are operable to transmit and receive signals propagating along the at least one signal path using the at least one coding and modulation scheme, wherein each of the first and second antennas comprises a respective dual polarized antenna including first and second co-located antenna elements operating at different polarizations, and wherein the first and second co-located antenna elements are operable to transmit first and second signals, respectively, a selected one of the first and second signals being a delayed version of another one of the first and second signals.

2. The system of claim 1 wherein the first and second co-located antenna elements operate at substantially orthogonal polarizations.

3. The system of claim 2 wherein the first and second co-located antenna elements operate at linear orthogonal polarizations.

4. The system of claim 2 wherein the first and second co-located antenna elements operate at circular orthogonal polarizations.

5. The system of claim 1 wherein a selected one of the first and second co-located antenna elements operates at a horizontal polarization, and another one of the first and second co-located antenna elements operates at a vertical polarization.

6. The system of claim 1 wherein the at least one signal path comprises a plurality of signal paths, wherein the first transceiver and the first antenna are operable to transmit and receive at least one co-polar signal and at least one cross polar signal propagating along respective ones of the plurality of signal paths, and wherein the second transceiver and the second antenna are operable to transmit and receive at least one co-polar signal and at least one cross polar signal propagating along the respective ones of the plurality of signal paths.

7. The system of claim 6 wherein each of the first and second transceivers includes a maximum ratio combiner operable to optimally combine the cross polar signals with the co-polar signals.

8. The system of claim 1 wherein the at least one coding and modulation scheme includes a space-time coding technique.

9. The system of claim 1 wherein the at least one coding and modulation scheme includes an orthogonal frequency division modulation (OFDM) technique.

10. A multiple input multiple output (MIMO) wireless communications system, comprising: a first transceiver; at least one first antenna operatively connected to the first transceiver, wherein the first transceiver and the at least one first antenna are operable to transmit and receive signals propagating along at least one signal path using at least one coding and modulation scheme; a second transceiver; and at least one second antenna operatively connected to the second transceiver, wherein the second transceiver and the at least one second antenna are operable to transmit and receive signals propagating along the at least one signal path using the at least one coding and modulation scheme, wherein a selected one of the first and second antennas comprises a respective dual polarized antenna including first and second co-located antenna elements operating at different polarizations, and another one of the first and second antennas comprises first and second spatially separated antenna elements operating at different polarizations, and wherein each of the first and second antennas is configured to provide signal delay diversity.

11. The system of claim 10 wherein the first and second spatially separated antenna elements have a predetermined spatial separation.

12. The system of claim 11 wherein the predetermined spatial separation of the first and second spatially separated antenna elements corresponds to n plus one half of a distance between successive troughs of a received signal, and wherein n is a non-negative integer value.

13. The system of claim 10 wherein the first and second spatially separated antenna elements operate at substantially orthogonal polarizations.

14. The system of claim 13 wherein the first and second spatially separated antenna elements operate at linear orthogonal polarizations.

15. The system of claim 13 wherein the first and second spatially separated antenna elements operate at circular orthogonal polarizations.

16. The system of claim 10 wherein a selected one of the first and second spatially separated antenna elements operates at a horizontal polarization, and another one of the first and second co-located antenna elements operates at a vertical polarization.

17. The system of claim 10 wherein the first and second co-located antenna elements operate at substantially orthogonal polarizations.

18. The system of claim 17 wherein the first and second co-located antenna elements operate at linear orthogonal polarizations.

19. The system of claim 17 wherein the first and second co-located antenna elements operate at circular orthogonal polarizations.

20. The system of claim 10 wherein a selected one of the first and second co-located antenna elements operates at a horizontal polarization, and another one of the first and second co-located antenna elements operates at a vertical polarization.

21. The system of claim 10 wherein the at least one signal path comprises a plurality of signal paths, wherein the first transceiver and the first antenna are operable to transmit and receive at least one co-polar signal and at least one cross polar signal propagating along respective ones of the plurality of signal paths, and wherein the second transceiver and the second antenna are operable to transmit and receive at least one co-polar signal and at least one cross polar signal propagating along the respective ones of the plurality of signal paths.

22. The system of claim 21 wherein each of the first and second transceivers includes a maximum ratio combiner operable to optimally combine the cross polar signals with the co-polar signals.

23. The system of claim 10 wherein the at least one coding and modulation scheme includes a space-time coding technique.

24. The system of claim 10 wherein the at least one coding and modulation scheme includes an orthogonal frequency division modulation (OFDM) technique.