Distributed antenna wireless communication system and method

Abstract

A multi-user multi-antenna system (MU-MAS) is provided that estimates uplink channel state information (CSI) between distributed transceiver stations and client devices and CSI between distributed transceiver stations, and utilizes channel reciprocity to derive downlink CSI. [Solution] The MU-MAS comprises one or more centralized processing units, a plurality of distributed input distributed output (DIDO) base transceiver stations (BTSs) communicatively coupled to the one or more centralized processing units via a network, and a plurality of client devices, and estimates uplink CSI between the DIDO BTSs and the client devices and CSI between the DIDO BTSs, and derives downlink CSI using channel reciprocity.

Description

[Technical Field]

[0001] [Related Applications] This application is a continuation-in-part of the following currently pending U.S. patent applications:

[0002] U.S. patent application Ser. No. 12 / 802,988, filed June 16, 2010, entitled "Interference Management, Handoff, Power Control, and Link Adaptation in a Distributed Input Distributed Output (DIDO) Communication System."

[0003] U.S. patent application Ser. No. 12 / 802,976, filed June 16, 2010, entitled "System and Method for Adjusting DIDO Interference Cancellation Based on Signal Strength Measurements."

[0004] U.S. patent application Ser. No. 12 / 802,974, filed June 16, 2010, entitled "System and Method for Managing Inter-Cluster Handoff of Clients Across Multiple DIDO Clusters."

[0005] U.S. patent application Ser. No. 12 / 802,989, filed June 16, 2010, entitled "System and Method for Managing Client Handoff Between Different Distributed Input Distributed Output (DIDO) Networks Based on the Client's Detected Speed."

[0006] U.S. patent application Ser. No. 12 / 802,958, filed June 16, 2010, entitled "System and Method for Power Control and Antenna Grouping in Distributed Input Distributed Output (DIDO) Networks."

[0007] U.S. patent application Ser. No. 12 / 802,975, filed June 16, 2010, entitled "System and Method for Link Adaptation in DIDO Multi-Carrier Systems."

[0008] U.S. patent application Ser. No. 12 / 802,938, filed June 16, 2010, entitled "System and Method for DIDO Precoding Interpolation in Multi-Carrier Systems."

[0009] U.S. patent application Ser. No. 12 / 630,627, filed Dec. 3, 2009, entitled "System and Method for Distributed Antenna Wireless Communications."

[0010] U.S. patent application Ser. No. 12 / 143,503, filed June 20, 2008, entitled "System and Method for Distributed Input Distributed Output Wireless Communications."

[0011] U.S. patent application Ser. No. 11 / 894,394, filed Aug. 20, 2007, entitled "System and Method for Distributed Input Distributed Output Wireless Communications."

[0012] U.S. patent application Ser. No. 11 / 894,362, filed Aug. 20, 2007, entitled "System and Method for Distributed Input Distributed Wireless Communications."

[0013] U.S. patent application Ser. No. 11 / 894,540, filed Aug. 20, 2007, entitled "System and Method for Distributed Input Distributed Output Wireless Communications."

[0014] U.S. patent application Ser. No. 11 / 256,478, filed Oct. 21, 2005, entitled "System and Method for Spatially Multiplexed Tropospheric Scatter Communications."

[0015] U.S. patent application Ser. No. 10 / 817,731, filed April 2, 2004, entitled "System and Method for Enhancing Near-Normal Incidence Ionospheric Wave ("NVIS") Communications Using Space-Time Coding." [Background technology]

[0016] Prior art multi-user wireless systems may include only a single base station or several base stations.

[0017] A single WiFi base station (e.g., using 2.4 GHz 802.11b, g, or n protocols) attached to a broadband wired Internet connection in an area with no other WiFi access points (e.g., a WiFi access point attached to a rural residential DSL) is an example of a relatively simple multi-user wireless system, with a single base station shared by one or more users within its transmission range. If the user is in the same room as the wireless access point, the user typically receives a high-speed link with few disruptions (e.g., there may be packet loss from 2.4 GHz interference, such as from a microwave oven, but not from spectrum shared with other WiFi devices). If the user is a moderate distance away, or if several obstacles are in the path between the user and the WiFi access point, the user will likely receive a medium-speed link. If the user is near the edge of the WiFi access point's range, the user will likely receive a low-speed link and may be susceptible to periodic dropouts due to channel changes as a result of the signal SNR dropping below a usable level. Finally, if the user is beyond the range of the WiFi base station, the user will have no link at all.

[0018] When multiple users simultaneously access a WiFi base station, the available data throughput is shared among them. Different users typically have different throughput requirements for the WiFi base station at a given time, but when the total throughput requirement adds up to the available throughput from the WiFi base station to the users, some or all users may receive less data throughput than they desire. In extreme situations where a WiFi access point is shared among a large number of users, the throughput to each user may be slow; even worse, the data throughput to each user may all arrive in short bursts separated by long periods of no data throughput while other users are served. As with media streaming, this "choppy" data delivery can impair certain applications.

[0019] Adding more WiFi base stations to a large number of users only has a limited effect up to a certain point. In the 2.4 GHz ISM band in the United States, there are three non-interfering channels available for WiFi use. If three WiFi base stations in the same coverage area are each configured to use a different non-interfering channel, the total yield of the coverage area among multiple users increases by up to three times. However, beyond that, adding more WiFi base stations in the same coverage area does not increase the total yield because WiFi base stations begin to share the same available spectrum by utilizing time division multiple access (TDMA) through "rotating" spectrum use. This situation is often seen in densely populated coverage areas, such as in apartment buildings. For example, a user in a large apartment building with a WiFi adapter may well experience very poor yield due to dozens of other interfering WiFi networks (e.g., other apartment buildings) serving other users in the same coverage area, even if the user's access point is in the same room as the client device accessing the base station. Although the link quality is likely to be good in that situation, the user will experience interference from neighboring WiFi adapters operating in the same frequency band, reducing the effective throughput to the user.

[0020] Current multi-user wireless systems, including unlicensed and licensed spectrum such as Wi-Fi, have several limitations. These include coverage area, downlink (DL) data rate, and uplink (UL) data rate. An important goal of next-generation wireless systems (such as WiMax and LTE) is to improve coverage area and DL and UL data rate through multiple-input multiple-output (MIMO) technology. MIMO uses multiple antennas at the transmitting and receiving ends of a wireless link to improve link quality (resulting in increased coverage) or data rate (by creating multiple non-interfering spatial channels for every user). However, when sufficient data rate is available for every user (Note: the terms "user" and "client" are used interchangeably herein), it would be desirable to utilize channel spatial diversity to create non-interfering channels for multiple users (rather than a single user) according to multi-user MIMO (MU-MIMO) techniques. See, for example, the following references:

[0021] G. Caire and S. Shamai, "On the achievable yield of multi-antenna Gaussian broadcast channels," IEEE Transactions on Information Theory, Vol. 49, pp. 1691-1706, July 2003.

[0022] P. Viswanath and D. Tse, "Sum Capacity of Vector Gaussian Broadcast Channels and Uplink-Downlink Duality," IEEE Transactions on Information Theory, Vol. 49, pp. 1912-1921, August 2003.

[0023] S. Vishwanath, N. Jindal, and A. Goldsmith, "Duality, Achievable Rate, and Sum-Rate Capacity of Gaussian MIMO Broadcast Channels," IEEE Transactions on Information Theory, vol. 49, pp. 2658-2668, October 2003.

[0024] W. Yu and J. Cioffi, "Sum Capacity of the Gaussian Vector Broadcast Channel," IEEE Transactions on Information Theory, Vol. 50, pp. 1875-1892, September 2004.

[0025] M. Costa, "Writing on Dirty Paper," IEEE Transactions on Information Theory, Vol. 29, pp. 439-441, May 1983.

[0026] M. Bengtsson, "Practical Methods for Multiuser Spatial Multiplexing," Proceedings of the Sensor Array Conference and Multichannel Signal Processing Workshop, pp. 130-134, August 2002.

[0027] K.-K. Wong, R.D. Murch, and K.B. Letaief, "Performance Enhancement of Multiuser MIMO Wireless Communication Systems," IEEE Transactions on Communications, vol. 50, pp. 1960-1970, December 2002.

[0028] M. Sharif and B. Hassibi, "On the Performance of MIMO Broadcast Channels with Partial Side Information," IEEE Transactions on Information Theory, Vol. 51, pp. 506-522, February 2005. Summary of the Invention [Problem to be solved by the invention]

[0029] For example, in a MIMO 4x4 system (i.e., four transmit antennas and four receive antennas), with a 10 MHz bandwidth, 16-QAM modulation rate 3 / 4 (resulting in a spectral efficiency of 3 bps / Hz) and forward error correction (FEC) coding, the ideal peak data rate achievable at the physical layer for every user is 4 x 30 Mbps = 120 Mbps, which is far higher than the peak data rate required to deliver high-definition video content (which is expected to require only 10 Mbps). In a MU-MIMO system with four transmit antennas, four users, and a single antenna per user in an ideal scenario (i.e., independent and identically distributed, iid, channels), the downlink data rate can be shared across the four users, and channel spatial diversity can be utilized to create four parallel 30 Mbps data links for the users. Different MU-MIMO schemes have been proposed as part of the LTE standard, as described, for example, in 3GPP, "Multiple Input Multiple Output in UTRA," 3GPP TR 25.876 V7.0.0, March 2007; 3GPP, "Base Physical Channel and Modulation," TS 36.211, V8.7.0, May 2009; and 3GPP, "Multiplexing and Channel Coding," TS 36.212, V8.7.0, May 2009. However, these schemes can only achieve a maximum improvement of 2x data rate in DL with four transmit antennas. ArrayComm (e.g., ArrayComm, "Field Validation Results," http: / / www.arraycomm.com / serve.php?page=proofPractical examples of MU-MIMO techniques in standard and proprietary cellular systems by companies such as (see e.g., IEEE 802.11a, 802.11b, 802.11c, 802.11d, 802.11d, 802.11e, 802.11f, 802.11f, 802.11g ...

[0030] Other practical problems with cellular system deployment include the excessive expense and limited availability of locations for cellular antenna locations (e.g., due to local government restrictions on antenna placement, real estate costs, physical obstructions, etc.), and the cost and / or availability of network connectivity to transmitters (referred to herein as "backhaul"). Additionally, cellular systems often have difficulty reaching clients deep inside buildings due to losses through walls, ceilings, floors, furniture, and other obstructions.

[0031] Indeed, the entire concept of cell structure for wide area wireless networks presupposes fairly robust placement of cellular towers, frequency alternation between adjacent cells, and frequent sectorization to avoid interference between transmitters (base stations or users) using the same frequency. As a result, a given sector of a given cell results in the sharing of blocks of UL and DL spectrum among all of the users within the cell sector, which in turn is shared among these users primarily in the time domain. For example, cellular systems based on time division multiple access (TDMA) and code division multiple access (CDMA) both share spectrum among users in the time domain. Overlapping such cellular systems with sectorization can potentially provide a two- to three-fold spatial domain advantage. Then, overlaying such cellular systems with MU-MIMO systems as described above can potentially provide another two- to three-fold space-time domain advantage. However, given that cellular system cells and sectors are typically in fixed locations, often dictated by where towers can be located, even this limited benefit can be difficult to utilize if the user density (or data rate requirements) at a given time is poorly matched to the tower / sector placement. Mobile smartphone users today often suffer this effect: they may be having a phone conversation or downloading a web page without any trouble, and then, after driving (or walking) to a new location, suddenly find their voice quality degraded, or the web page speed slows down, or they even lose their connection completely. However, from one day to the next, a user may experience the exact opposite at each location. Assuming the environmental conditions are the same, what users likely experience is a very variable user density (or data rate requirements). However, the total available spectrum (and therefore the total data rate using conventional techniques) to be shared among users at a given location is largely fixed.

[0032] Furthermore, prior art cellular systems rely on using different frequencies, typically three different frequencies, in different adjacent cells, which for a given amount of spectrum reduces the available data rate by a factor of three.

[0033] So, in summary, a prior art cellular system can lose perhaps three times its spectrum utilization due to cellization, and can improve spectrum utilization by perhaps three times through sectorization and perhaps another three times through MU-MIMO techniques, resulting in a net gain of 3 * This results in 3 / 3 = 3 times the potential spectrum utilization. That bandwidth is then divided among users, typically in the time domain, based on which sector of which cell the user falls into at a given time. Yet another inefficiency exists because a given user's data rate requirements are typically independent of the user's location, but the available data rate varies based on the link quality between the user and the base station. For example, users farther from a cellular base station typically have a lower available data rate than users closer to the base station. Because data rates are typically shared among all of the users within a given cellular sector, this results in all users being subject to a high data rate request from a distant user with poor link quality (e.g., at the edge of the cell), because such a user would still require the same amount of data rate and would consume a greater amount of the shared spectrum to obtain the same amount of data rate.

[0034] Other proposed spectrum sharing systems, such as those used by WiFi (e.g., 802.11b, g, and n) and those proposed by the White Space Coalition, have very inefficient spectrum sharing because simultaneous transmissions by base stations within a user's range result in interference, and therefore the systems utilize anti-collision and sharing protocols. These spectrum sharing protocols are in the time domain, and therefore, when there are many interfering base stations and users, collectively, base stations are limited to time-domain sharing of spectrum between each other, regardless of how efficient each base station's own spectrum usage. Other prior art spectrum sharing systems similarly rely on similar methods to mitigate interference between base stations (in the case of cellular base stations or small base stations (such as WiFi access points (APs)) with tower-mounted antennas). These methods include limiting the transmit power from base stations to limit the range of interference, beam pulse forming (through synthesis or physical means) to narrow the interference area, time-domain multiplexing of spectrum, and / or MU-MIMO methods with multiple clustered antennas on user devices, base stations, or both. In the case of advanced cellular networks that already exist or are being planned today, many of these technologies are likely to be in use at once.

[0035] However, all of these technologies do little to increase the aggregate data rate among shared users for a given area of ​​coverage, as evidenced by the fact that even advanced cellular systems can only increase spectrum utilization by about three times compared to a single user using the spectrum. In particular, as a given coverage area expands from a user perspective, it becomes increasingly difficult to increase the available data rate within a given amount of spectrum to keep pace with user growth. For example, to increase the aggregate data rate within a given area for cellular systems, cells are typically subdivided into smaller cells (often called nanocells or femtocells). Such small cells can be prohibitively expensive given the "dead zones" that result in minimal coverage and, further, limitations on where towers can be placed to avoid interference between nearby cells using the same frequencies, and the requirement that towers be deployed in a fair, structured pattern. Essentially, coverage areas must be meticulously planned, available locations for towers or base stations identified, and then, given these constraints, cellular system designers must make do with the best they can. Needless to say, as user data rate requirements increase over time, cellular system designers must again remap coverage areas, seek to locate towers or base stations, and once again address situational constraints. Very often, there are simply no good solutions, resulting in dead zones or inadequate total data rate capacity within the coverage area. In other words, the cellular system's strong physical location requirements to avoid interference between towers or base stations utilizing the same frequency create significant challenges and constraints in cellular system design, and these requirements often fail to meet user data rate and coverage requirements.

[0036] So-called prior art "cooperative" and "cognitive" radio systems attempt to increase spectrum utilization in a given area by using intelligent algorithms within the radios to potentially "listen" for other spectrum usage so that they can minimize interference with each other and / or wait for a channel to become free. Systems have been proposed to increase spectrum utilization in such spectrum, particularly for use in unlicensed spectrum.

[0037] Mobile Ad Hoc Network (MANET) http: / / en.wikipedia.org / wiki / Mobile_ad_hoc_network (see [1]) is an example of a cooperative self-configuring network aimed at providing peer-to-peer communication, and can be used to establish radio-to-radio communication without cellular infrastructure, and with sufficiently low power communication to potentially mitigate interference between simultaneous transmissions that are out of range of each other. Although numerous routing protocols have been proposed and implemented for MANET systems (for a list of dozens of routing protocols across a wide range of classes, see [1]). http: / / en.wikipedia.org / wiki / List_of_ad-hoc_routing_protocols , a common theme among routing protocols is that they are all techniques for routing (e.g., repeating) transmissions to minimize transmitter interference within the available spectrum toward the goal of a given efficiency or reliability paradigm.

[0038] All prior art multi-user wireless systems attempt to improve spectrum utilization within a given coverage area by utilizing techniques that enable simultaneous spectrum utilization between a base station and multiple users. In particular, in all of these cases, the techniques utilized for simultaneous spectrum utilization between a base station and multiple users result in simultaneous spectrum utilization by multiple users by mitigating interference between waveforms for the multiple users. For example, in the case of three base stations, each using a different frequency to transmit to one of the three users, interference is mitigated because the three transmissions are on three different frequencies. In the case of sectorization from a base station to three different users, each 180° apart relative to the base station, interference is mitigated because beam pulse forming prevents the three transmissions from overlapping at the user.

[0039] When such techniques are augmented with MU-MIMO, and each base station has, for example, four antennas, this can potentially increase downlink yield by a factor of four by creating four non-interfering spatial channels for users in a given coverage area, but it is still the case that some technique should be employed to mitigate interference between multiple simultaneous transmissions to multiple users in different coverage areas.

[0040] As discussed above, such conventional techniques (e.g., cellularization, sectorization) typically suffer from problems not only due to increased costs and / or deployment flexibility of multi-user wireless systems, but also due to physical or practical limitations on total throughput within a given coverage area. For example, a cellular system may not have enough available locations to deploy more base stations to achieve small cell deployment. MU-MIMO systems, given the spacing of clustered antennas at each base station location, experience asymptotically diminishing throughput returns as more antennas are added to a base station due to limited spatial diversity.

[0041] Furthermore, in multi-user wireless systems where user locations and densities are unpredictable, limited spatial diversity results in unpredictable throughput (with rapid changes in frequency), which is inconvenient for users and makes some applications (e.g., delivery of services requiring predictable throughput) impractical or of poor quality. Thus, prior art multi-user wireless systems leave much to be desired in terms of their ability to provide predictable and / or high-quality service to users.

[0042] Despite the tremendous enhancements and complexities that have been developed over time for prior art multi-user wireless systems, there is a common theme in that transmissions are distributed to different base stations (or ad-hoc transceivers) and are configured and / or controlled to avoid RF waveform transmissions from different base stations and / or different ad-hoc transceivers interfering with each other at a particular user's receiver.

[0043] Or, stated another way, if a user happens to receive transmissions from more than one base station or ad hoc transceiver at the same time, the interference from the multiple simultaneous transmissions will cause a reduction in the SNR and / or bandwidth of the signal to the user, which, if severe enough, will result in the loss of all or part of the potential data (or analog information) that would otherwise be received by the user.

[0044] Therefore, multi-user wireless systems must utilize one or more spectrum sharing techniques or alternatives to avoid or mitigate such interference to users from multiple base stations or ad-hoc transceivers transmitting simultaneously on the same frequency. There are numerous prior art techniques for avoiding such interference, including controlling the physical location of base stations (e.g., celling), limiting the power output of base stations and / or ad-hoc transceivers (e.g., limiting transmission range), beam pulse forming / sectorization, and time-domain multiplexing. That is, all of these spectrum sharing systems attempt to address a limitation of multi-user wireless systems in which data yield for affected users is reduced or destroyed by the resulting interference when multiple base stations and / or ad-hoc transceivers transmitting simultaneously on the same frequency are received by the same user. Situations can arise where the total yield of the multi-user wireless system is severely reduced or even destroyed when most or all of the users in the multi-user wireless system experience interference from multiple base stations and / or ad-hoc transceivers (e.g., in the event of malfunctioning components of the multi-user wireless system).

[0045] Prior art multi-user wireless systems add complexity and introduce limitations to wireless networks, often resulting in situations where a given user's experience (e.g., available bandwidth, latency, predictability, reliability) is impacted by spectrum usage by other users in the area. Given the increasing demand for total bandwidth in the wireless spectrum shared by multiple users and the increasing growth of applications that may depend on multi-user wireless network reliability, predictability, and low latency for a given user, it is clear that prior art multi-user wireless technologies suffer from many limitations. Indeed, due to the limited availability of spectrum suitable for certain types of wireless communications (e.g., at wavelengths that are efficient at passing through building walls), it may be the case that prior art wireless technologies are inadequate to meet the growing demand for reliable, predictable, and low-latency bandwidth. [Means for solving the problem]

[0046] Prior art related to the present invention describes beam-pulse forming systems and methods for null steering in multi-user scenarios. Beam-pulse forming is originally conceived to maximize the received signal-to-noise ratio (SNR) by dynamically adjusting the phase and / or amplitude of signals fed to the antennas of an array (i.e., beam-pulse forming weights) to thereby focus energy toward the user. In multi-user scenarios, beam-pulse forming can be used to suppress interference sources and maximize the signal-to-interference-and-noise ratio (SINR). For example, when beam-pulse forming is used at the receiver of a wireless link, weights are calculated to create a null in the direction of the interference source. When beam-pulse forming is used at the transmitter in a multi-user downlink scenario, weights are calculated to pre-cancel inter-user interference and maximize the SINR for all users. An alternative technique for BD pre-coding multi-user systems is to calculate pre-coding weights to maximize throughput in the downlink broadcast channel. The co-pending patent applications, which are incorporated herein by reference, describe the above-mentioned technology (see co-pending patent applications for specific citations).

[0047] The present invention can be better understood from the following detailed description taken in conjunction with the drawings. [Prior art documents] [Patent documents]

[0048] [Patent Document 1] U.S. Patent Application Serial No. 12 / 802,988 [Patent Document 2] U.S. Patent Application Serial No. 12 / 802,976 [Patent Document 3] U.S. Patent Application Serial No. 12 / 802,974 [Patent Document 4] U.S. Patent Application Serial No. 12 / 802,989 [Patent Document 5] U.S. Patent Application Serial No. 12 / 802,958 [Patent Document 6] U.S. Patent Application Serial No. 12 / 802,975 [Patent Document 7] U.S. Patent Application Serial No. 12 / 802,938 [Patent Document 8] U.S. Patent Application Serial No. 12 / 630,627 [Patent Document 9] U.S. Patent No. 7,599,420 [Patent Document 10] U.S. Patent No. 7,633,994 [Patent Document 11] U.S. Patent No. 7,636,381 [Patent Document 12] U.S. Patent Application Serial No. 12 / 143,503 [Patent Document 13] U.S. Patent Application Serial No. 11 / 256,478 [Patent Document 14] U.S. Patent No. 7,418,053 [Patent Document 15] U.S. Patent Application Serial No. 10 / 817,731 [Patent Document 16] U.S. Patent No. 4,003,016 [Patent Document 17] U.S. Patent No. 4,771,289 [Patent Document 18] U.S. Patent No. 5,600,326 [Non-patent literature]

[0049] [Non-Patent Document 1] G. Caire and S. Shamai, "On the Achievable Yield of Multi-Antenna Gaussian Broadcast Channels," IEEE Transactions on Information Theory, Vol. 49, pp. 1691-1706, July 2003. [Non-patent document 2] P. Viswanath and D. Tse, "Sum Capacity of Vector Gaussian Broadcast Channels and Uplink-Downlink Duality," IEEE Transactions on Information Theory, Vol. 49, pp. 1912-1921, August 2003. [Non-patent document 3] S. Vishwanath, N. Jindal, and A. Goldsmith, "Duality, Achievable Rate, and Sum-Rate Capacity of Gaussian MIMO Broadcast Channels," IEEE Transactions on Information Theory, Vol. 49, pp. 2658-2668, October 2003. [Non-patent document 4] W. Yu and J. Cioffi, "Sum Capacity of the Gaussian Vector Broadcast Channel," IEEE Transactions on Information Theory, Vol. 50, pp. 1875-1892, September 2004. [Non-Patent Document 5] M. Costa, "Writing on Dirty Paper," IEEE Transactions on Information Theory, Vol. 29, pp. 439-441, May 1983. [Non-patent document 6] M. Bengtsson, "Practical Methods for Multiuser Spatial Multiplexing," Proceedings of the Sensor Array Conference and the Multichannel Signal Processing Workshop, pp. 130-134, August 2002. [Non-Patent Document 7] K.-K. Wong, R.D. Murch, and K.B. Letaief, "Performance Enhancement of Multiuser MIMO Wireless Communication Systems," IEEE Transactions on Communications, Vol. 50, pp. 1960-1970, December 2002. [Non-patent document 8] M. Sharif and B. Hassibi, "On the Performance of MIMO Broadcast Channels with Partial Side Information," IEEE Transactions on Information Theory, Vol. 51, pp. 506-522, February 2005. [Non-Patent Document 9] 3GPP, "Multiple Input Multiple Output in UTRA," 3GPP TR 25.876 V7.0.0, March 2007 [Non-Patent Document 10] 3GPP, "Base Physical Channels and Modulation," TS 36.211, V8.7.0, May 2009 [Non-Patent Document 11] 3GPP, "Multiplexing and Channel Coding," TS 36.212, V8.7.0, May 2009 [Non-Patent Document 12] FCC, "Evaluating Compliance with FCC Guidelines for Human Exposure to Radio Frequency Electromagnetic Fields," OET Bulletin 65, January 1997, August 1997 [Non-Patent Document 13] 3GPP, "Spatial Channel Model AHG (Ad Hoc Combination from 3GPP and 3GPP2)", SCM, Text V6.0, April 22, 2003 [Non-Patent Document 14] 3GPP TR 25.912: "Achievability Study for Evolved UTRA and UTRAN", V9.0.0 (2009-10) [Non-Patent Document 15] 3GPP TR 25.913: "Evolved UTRA (E-UTRA) and Evolved UTRAN (E-UTRAN) Requirements", V8.0.0 (2009-01) [Non-Patent Document 16] "Microwave Mobile Communications," by W.C. Jakes, IEEE Press, 1974 [Non-Patent Document 17] K.K. Wong et al., "Joint Channel Orthogonalization for Multiuser MIMO Antenna Systems," IEEE Transactions on Wireless Communications, Vol. 2, pp. 773-786, July 2003. [Non-Patent Document 18] P. Viswanath et al., "Beam Pulse Forming with Primary User Awareness by Secondary Users Using Dumped Antennas," IEEE Transactions on Information Theory, Vol. 48, pp. 1277-1294, June 2002. [Non-Patent Document 19] A. A. M. Saleh et al., "Statistical Model of Indoor Multipath Propagation," Transactions of the IEEE Society for Selected Areas in Communications, Vol. 195, SAC-5, No. 2, pp. 128-137, February 1987. [Non-Patent Document 20] A. Paulraj et al., "Introduction to Space-Time Radio Communications," Cambridge University Press, 40 West 20th Street, New York, NY, USA, 2003 [Non-Patent Document 21] J. Choi et al., "Interpolation-Based Transmit Beampulse Forming for Feedback-Limited MIMO-OFDM," IEEE Transactions on Signal Processing, Vol. 53, No. 11, pp. 4125-4135, November 2005. [Non-Patent Document 22] I. Wong et al., "Long-Range Channel Prediction for Adaptive OFDM Systems," Proceedings of the IEEE 2006 Asilomar Conference on Signals, Systems, and Computers, Vol. 1, pp. 723-736, Pacific Grove, California, USA, November 7-10, 2004. [Non-Patent Document 23] J.G. Proakis, "Communications Systems Engineering," Prentice Hall, 1994 [Non-Patent Document 24] "Beam Pulse Forming: A Versatile Approach to Spatial Filtering Applications," IEEE ASSP Magazine, April 1988. [Non-Patent Document 25] R.G. Vaughan, "On Optimal Coupling in Mobile," IEEE Transactions on Vehicular Technology, Vol. 37, No. 4, pp. 181-188, November 1988. [Non-Patent Document 26] F. Qian, "Partially Adaptive Beam Pulse Forming for Correlated Interference Rejection," IEEE Transactions on Signal Processing, Vol. 43, No. 2, pp. 506-515, February 19, 1995. [Non-Patent Document 27] H. Krim et al., "Twenty Years of Array Signal Processing Research," IEEE Signal Processing Journal, pp. 67-94, July 1996. [Non-patent document 28] H. Boche et al., "Analysis of Different Precoding / Decoding Strategies for Multiuser Beampulse Forming," IEEE Vehicle Technology Conference, Vol. 1, April 2003. [Non-Patent Document 29] M. Schubert et al., "'Joint' Dirty Paper Precoding and Downlink Beam Pulse Forming," Vol. 2, pp. 536-540, December 2002. [Non-Patent Document 30] H. Boche et al., "General Duality Theory of Uplink and Downlink Beam Pulse Forming," Vol. 1, pp. 87-91, December 2002. [Non-Patent Document 31] K.K. Wong, R.D. Murch, and K.B. Letaief, "Joint Channel Orthogonalization for Multiuser MIMO Antenna Systems," IEEE Transactions on Wireless Communications, Vol. 2, pp. 773-786, July 2003. [Non-Patent Document 32] Q. H. Pencer, A. L. Windlehurst, and M. Haardt, "Zero-Forcing Methods for Downlink Spatial Multiplexing in Multiuser MIMO Channels," IEEE Transactions on Signal Processing, Vol. 52, pp. 461-471, February 2004. [Non-Patent Document 33] S. Robinson, "Towards an Optimal Algorithm for Matrix Multiplication," SIAM News, Vol. 38, No. 9, November 2005. [Non-Patent Document 34] D. Coppersmith and S. Winograd, "Matrix Multiplication via Arithmetic Sequences," J. Symb. Comp., vol. 9, pp. 251-280, 1990. [Non-Patent Document 35] H. Cohn, R. Kleinberg, B. Szegedy, and C. Umans, "A Group-Theoretic Algorithm for Matrix Multiplication," pp. 379-388, November 2005. [Non-Patent Document 36] WH Press, sa Teukolsky, WT Etterling, and B.P. Flannery, Numerical Computing Recipes in C: Scientific Computing Techniques, Cambridge University Press, 1992. [Non-Patent Document 37] Per-Erik Eriksson and Bjorn Odenhammar, "VDSL2: The Next Key Broadband Technology," Ericsson Review, No. 1, 2006. [Non-Patent Document 38] ArrayComm, "Field Proof Results," http: / / www.arraycomm.com / serve.php?page=proof [Non-Patent Document 39] Mobile ad hoc network (MANET), http: / / en.wikipedia.org / wiki / Mobile_ad_hoc_network [Non-Patent Document 40] http: / / en.wikipedia.org / wiki / List_of_ad-hoc_routing_protocols [Brief explanation of the drawings]

[0050] [Figure 1] FIG. 1 shows a main DIDO cluster surrounded by neighboring DIDO clusters in one embodiment of the present invention. [Figure 2] FIG. 1 illustrates a frequency division multiple access (FDMA) method used in one embodiment of the present invention. [Figure 3] FIG. 1 illustrates a time division multiple access (TDMA) method used in one embodiment of the present invention. [Figure 4] FIG. 1 illustrates different types of interference zones addressed in one embodiment of the present invention. [Figure 5] FIG. 1 illustrates a framework used in one embodiment of the present invention. [Figure 6] 10 is a graph showing SER as a function of SNR assuming SIR=10 dB for a target client in an interference zone. [Figure 7] 1 is a graph showing the SER derived from two IDCI encoding methods. [Figure 8] FIG. 10 illustrates an exemplary scenario in which a target client moves from a primary DIDO cluster to an interfering cluster. [Figure 9] FIG. 1 shows the signal-to-interference-and-noise ratio (SINR) as a function of distance (D). [Figure 10] FIG. 10 illustrates the bit error rate (SER) performance of three scenarios of 4QAM modulation in a flat-fading narrowband channel. [Figure 11] FIG. 2 illustrates a method for IDCI encoding according to one embodiment of the present invention. [Figure 12] FIG. 10 illustrates SINR variation as a function of a client's distance from the center of the main DIDO cluster in one embodiment. [Figure 13] FIG. 1 illustrates an embodiment in which SER is derived for 4-QAM modulation. [Figure 14] FIG. 1 illustrates one embodiment of the present invention in which a finite state machine executes the handoff algorithm. [Figure 15] FIG. 1 illustrates one embodiment of a handoff strategy in the presence of shadowing. [Figure 16] FIG. 15 is a diagram showing the hysteresis loop mechanism when switching between any two states in FIG. 14. [Figure 17] FIG. 1 illustrates an embodiment of a DIDO system with power control. [Figure 18] FIG. 10 shows SNR versus SER assuming four DIDO transmit antennas and four clients in different scenarios. [Figure 19] FIG. 10 illustrates MPE power density as a function of distance from an RF radiation source for different values ​​of transmit power according to one embodiment of the present invention. [Figure 20a] FIG. 1 illustrates the distribution of low-power DIDO distributed antennas. [Figure 20b] FIG. 1 illustrates the distribution of high-power DIDO distributed antennas. [Figure 21a] FIG. 20b shows the power distribution corresponding to the configuration of FIG. 20a. [Figure 21b] FIG. 20b shows the power distribution corresponding to the configuration of FIG. [Figure 22a]FIG. 20b shows velocity distributions for the scenario shown in FIG. 20a. [Figure 22b] FIG. 20b shows velocity distributions for the scenario shown in FIG. [Figure 23] FIG. 1 illustrates an embodiment of a DIDO system with power control. [Figure 24] 1 illustrates an embodiment of a method for iterating over all antenna groups according to a round-robin scheduling policy for transmitting data. [Figure 25] FIG. 1 shows a comparison of uncoded SER performance of power control with antenna grouping versus conventional eigenmode selection in U.S. Pat. No. 7,636,381. [Figure 26a] FIG. 1 illustrates a scenario in which BD coding dynamically adjusts coding weights to accommodate different power levels across the wireless link between the DIDO antenna and the client. [Figure 26b] FIG. 1 illustrates a scenario in which BD coding dynamically adjusts coding weights to accommodate different power levels across the wireless link between the DIDO antenna and the client. [Figure 26c] FIG. 1 illustrates a scenario in which BD coding dynamically adjusts coding weights to accommodate different power levels across the wireless link between the DIDO antenna and the client. [Figure 27] 1 shows the amplitude of the low frequency selective channel (assuming β=1) across the delay domain or instantaneous PDP (upper plot) and frequency domain (lower plot) for a DIDO 2x2 system. [Figure 28] FIG. 10 illustrates one embodiment of the channel matrix frequency response for DIDO 2x2 with one antenna per client. [Figure 29] FIG. 10 illustrates one embodiment of a channel matrix frequency response for DIDO 2x2 with one antenna per client for a channel characterized by high frequency selectivity (eg, with β=0.1). [Figure 30]FIG. 1 illustrates exemplary SER for different QAM schemes (i.e., 4-QAM, 16-QAM, 64-QAM). [Figure 31] FIG. 1 illustrates an embodiment of a method for performing a link adaptation (LA) method. [Figure 32] FIG. 1 illustrates the SER performance of one embodiment of a link adaptation (LA) method. [Figure 33] FIG. 28 shows the matrix entries in equation (28) as a function of OFDM tone index for a DIDO 2x2 system with NFFT=64 and L0=8. [Figure 34] 10 shows SER vs. SNR for L0=8, M=Nt=2 transmit antennas, and varying number of P. [Figure 35] FIG. 10 illustrates the SER performance of an embodiment of the interpolation method for different DIDO orders and L0=16. [Figure 36] FIG. 1 illustrates one embodiment of a system using super clusters, DIDO clusters, and user clusters. [Figure 37] FIG. 1 illustrates a system having user clusters according to one embodiment of the present invention. [Figure 38a] FIG. 2 illustrates link quality metric thresholds used in one embodiment of the present invention. [Figure 38b] FIG. 2 illustrates link quality metric thresholds used in one embodiment of the present invention. [Figure 39] FIG. 10 illustrates an example of a link quality matrix for establishing user clusters. [Figure 40] FIG. 10 illustrates an example of a link quality matrix for establishing user clusters. [Figure 41] FIG. 10 illustrates an example of a link quality matrix for establishing user clusters. [Figure 42] FIG. 1 illustrates an embodiment in which a client traverses different DIDO clusters. DETAILED DESCRIPTION OF THE INVENTION

[0051] One solution that overcomes many of the limitations of the previous prior art is the implementation of Distributed Input Distributed Output (DIDO) technology. DIDO technology is described in the following patents and patent applications, all of which are assigned to the assignee of this patent and incorporated by reference. These patents and applications are sometimes collectively referred to herein as the "related patents and applications."

[0052] U.S. patent application Ser. No. 12 / 802,988, filed June 16, 2010, entitled "Interference Management, Handoff, Power Control, and Link Adaptation in a Distributed Input Distributed Output (DIDO) Communication System."

[0053] U.S. patent application Ser. No. 12 / 802,976, filed June 16, 2010, entitled "System and Method for Adjusting DIDO Interference Cancellation Based on Signal Strength Measurements."

[0054] U.S. patent application Ser. No. 12 / 802,974, filed June 16, 2010, entitled "System and Method for Managing Inter-Cluster Handoff of Clients Across Multiple DIDO Clusters."

[0055] U.S. patent application Ser. No. 12 / 802,989, filed June 16, 2010, entitled "System and Method for Managing Client Handoff Between Different Distributed Input Distributed Output (DIDO) Networks Based on the Client's Detected Speed."

[0056] U.S. patent application Ser. No. 12 / 802,958, filed June 16, 2010, entitled "System and Method for Power Control and Antenna Grouping in Distributed Input Distributed Output (DIDO) Networks."

[0057] U.S. patent application Ser. No. 12 / 802,975, filed June 16, 2010, entitled "System and Method for Link Adaptation in DIDO Multi-Carrier Systems."

[0058] U.S. patent application Ser. No. 12 / 802,938, filed June 16, 2010, entitled "System and Method for DIDO Precoding Interpolation in Multi-Carrier Systems."

[0059] U.S. patent application Ser. No. 12 / 630,627, filed Dec. 3, 2009, entitled "System and Method for Distributed Antenna Wireless Communications."

[0060] U.S. Patent No. 7,599,420, entitled "System and Method for Distributed Input / Distributed Output Wireless Communication," filed August 20, 2007, and granted October 6, 2009.

[0061] U.S. Patent No. 7,633,994, entitled "System and Method for Distributed Input / Distributed Output Wireless Communication," filed August 20, 2007, and granted December 15, 2009.

[0062] U.S. Patent No. 7,636,381, entitled "System and Method for Distributed Input / Distributed Output Wireless Communication," filed August 20, 2007, and granted December 22, 2009.

[0063] U.S. patent application Ser. No. 12 / 143,503, filed June 20, 2008, entitled "System and Method for Distributed Input / Distributed Output Wireless Communications."

[0064] U.S. patent application Ser. No. 11 / 256,478, filed Oct. 21, 2005, entitled "System and Method for Spatially Multiplexed Tropospheric Scatter Communications."

[0065] U.S. Patent No. 7,418,053, entitled "System and Method for Distributed Input / Distributed Output Wireless Communications," filed July 30, 2004, and granted August 26, 2008.

[0066] U.S. patent application Ser. No. 10 / 817,731, filed April 2, 2004, entitled "System and Method for Enhancing Near-Normal Incidence Ionospheric Wave ("NVIS") Communications Using Space-Time Coding."

[0067] In order to reduce the size and complexity of this patent application, the disclosures of some of the related patents and applications are not explicitly described below, and reference is made to the related patents and applications for a complete detailed description of the disclosure of the present invention.

[0068] Note that Section I below (disclosure from related application Ser. No. 12 / 802,988) utilizes a unique set of endnotes that refer to prior art references and prior applications assigned to the assignee of the present application. The endnote citations are explained at the end of Section I (immediately preceding the heading of Section II). A citation used in Section II may have a numerical identification for that source that overlaps with the numerical identification used in Section I, even though these numerical identifications identify different references (listed at the end of Section II). Thus, a reference identified by a given numerical identification can be identified within the section in which the numerical identification is used.

[0069] I. DISCLOSURE FROM RELATED APPLICATION SERIAL NO. 12 / 802,988 1. How to remove inter-cluster interference A wireless radio frequency (RF) communication system and method using multiple distributed transmit antennas to create locations in space with zero RF energy is described below. When M transmit antennas are used, up to (M-1) points of zero RF energy can be created at a given location. In one embodiment of the present invention, the points of zero RF energy are wireless devices, and the transmit antennas are aware of channel state information (CSI) between the transmitter and receiver. In one embodiment, the CSI is calculated at the receiver and fed back to the transmitter. In another embodiment, the CSI is calculated at the transmitter through training from the receiver, assuming channel reciprocity is utilized. The transmitter can use the CSI to determine simultaneously transmitted interfering signals. In one embodiment, block diagonalization (BD) precoding is used at the transmit antennas to create the points of zero RF energy.

[0070] The systems and methods described herein are different from the conventional receive / transmit beamforming methods described above. In fact, in receive beamforming, weights are calculated (through null steering) to suppress interference on the receive side, while some embodiments of the invention described herein apply weights on the transmit side to create an interference pattern that results in one or more locations in space having "zero RF energy". Each, unlike conventional transmit beam pulse forming or BD precoding designed to maximize signal quality (or SINR) or downlink throughput for every user, the systems and methods described herein minimize signal quality under certain circumstances and / or from certain transmitters, and thus, points of zero RF energy are created at client devices (sometimes referred to herein as "users"). Further, in the context of a distributed input distributed output (DIDO) system (described in the inventor's related patents and applications), an increase in degrees of freedom (i.e., an increase in channel spatial diversity) can be obtained that can be utilized to create multiple points of zero RF energy and / or maximum SINR for different users by transmit antennas distributed in space. For example, with M transmit antennas, up to (M - 1) points of RF energy can be created, whereas a practical beam pulse forming or BD multi-user system is typically designed with closely spaced antennas that limit the number of simultaneous users that can be handled on a wireless link such that any number of transmit antennas M can be obtained.

[0071] Consider a system with M transmit antennas and K users where K < M. Assume that the transmitter recognizes the CSI ( TIFF0007621401000001.tif515) between the M transmit antennas and the K users. For simplicity, assume that each user is equipped with a single antenna, although the same method can be extended to multiple receive antennas per user. The precoding weights ([[]] TIFF0007621401000002.tif515) is calculated to satisfy the following conditions: TIFF0007621401000003.tif420 where, TIFF0007621401000004.tif46 is a vector with all zero entries, and H is TIFF0007621401000005.tif2115 is a channel vector ( TIFF0007621401000006.tif416). In one embodiment, we compute the singular value decomposition (SVD) of the channel matrix H and define the precoding weights w as the right singular vectors corresponding to the null subspace of H (identified by the single value 0). The transmit antenna is As shown by TIFF0007621401000007.tif447, the weight vector defined above is used to transmit RF energy while creating K points of zero RF energy at K user locations, where: TIFF0007621401000008.tif515 is additive white Gaussian noise (AWGN) at the kth user. In one embodiment, we compute the singular value decomposition (SVD) of the channel matrix H, and define the precoding weights w as the right singular vectors corresponding to the null subspace of H (identified by 0 singular values).

[0072] In another embodiment, the wireless system is a DIDO system, and points of zero RF energy are created between different DIDO coverage areas to proactively cancel interference to clients. In U.S. patent application Ser. No. 12 / 630,627, a DIDO system is described that includes: DIDO Client DIDO Distributed Antenna DIDO Base Transceiver Station (BTS) DIDO Base Station Network (BSN): Every BTS is connected through the BSN to multiple distributed antennas that serve a given coverage area called a DIDO cluster. This patent application describes a system and method for eliminating interference between adjacent DIDO clusters. As shown in Figure 1, it is assumed that a primary DIDO cluster serves clients (i.e., user devices served by the multi-user DIDO system) (or target clients) that are affected by interference from neighboring clusters.

[0073] In one embodiment, neighboring clusters operate on different frequencies according to a frequency division multiple access (FDMA) method similar to conventional cellular systems. For example, with a frequency reuse factor of three, the same carrier frequency is repeated for every three DIDO clusters, as shown in FIG. 2. In FIG. 2, the different carrier frequencies are identified as F1, F2, and F3. While this embodiment can be used in some instances, this solution results in reduced spectral efficiency because the available spectrum is divided into multiple subbands and only a subset of the DIDO clusters operate in the same subband. Furthermore, complex cell designs are required to associate different DIDO clusters with different frequencies, thus preventing interference. As with prior art cellular systems, such cell designs require predetermined antenna placement and transmit power limitations to avoid interference between clusters using the same frequency.

[0074] In another embodiment, neighboring clusters operate in the same frequency band but in different time slots according to a time division multiple access (TDMA) method. For example, as shown in FIG. 3, DIDO transmissions are only permitted for a given cluster in time slots T1, T2, and T3, as shown. Time slots are equally assigned to different clusters such that the different clusters are scheduled according to a round-robin policy. If different clusters are characterized by different data rate requirements (i.e., a cluster in a rural area with a low number of clients per coverage area versus a cluster in a congested urban environment), different priorities are assigned to the different clusters such that the greater the data rate requirement, the more time slots are assigned. While TDMA as described above can be used in one embodiment of the present invention, the TDMA approach may require time synchronization across different clusters and may result in reduced spectral efficiency because interfering clusters cannot use the same frequency at the same time.

[0075] In one embodiment, all neighboring clusters transmit simultaneously in the same frequency band and use spatial processing across the clusters to avoid interference. In this embodiment, the multi-cluster DIDO system (i) uses conventional DIDO precoding in the primary cluster to transmit simultaneous non-interfering data streams in the same frequency band to multiple clients (e.g., as described in related patents and applications, including U.S. Pat. Nos. 7,599,420, 7,633,994, 7,636,381, and U.S. patent application Ser. No. 12 / 143,503), and (ii) uses DIDO precoding with interference cancellation in neighboring clusters to avoid interference to clients located in the interference zone 8010 in FIG. 4 by creating a point of zero radio frequency (RF) energy at the location of the target client. When a target client is located in the interference zone 410, it receives zero RF energy from the primary cluster 411, resulting in the sum of the RF containing data streams from the interfering clusters 412-413 being simply the RF containing data stream from the primary cluster. Therefore, nearby clusters can utilize the same frequency at the same time without target clients in the interference zone experiencing interference.

[0076] In practical systems, the performance of DIDO precoding may be affected by different factors such as channel estimation errors or Doppler effects (which cause stale channel state information in DIDO distributed antennas), intermodulation distortion (IMD) in multi-carrier DIDO systems, and time or frequency offsets. As a result of these effects, it may be impractical to achieve a point of zero RF energy. However, as long as the RF energy at the target client from the interference cluster is negligible compared to the RF energy from the main cluster, the relative performance at the target client will not be affected by the interference. For example, -6Assume that forward error correction (FEC) coding is used to produce a target bit error rate (BER) of 0.01 and that a client requires a signal-to-noise ratio (SNR) of 20 dB to demodulate a 4-QAM satellite constellation. If the RF energy at the target client received from the interfering cluster is 20 dB below the RF energy received from the main cluster, the interference is negligible and the client can successfully demodulate data within the predetermined BER target. Thus, as used herein, the term "zero RF energy" does not necessarily mean that the RF energy from the interfering RF signal is zero. Rather, it means that the RF energy is sufficiently low relative to the RF energy of the desired RF signal so that the desired RF signal can be received at the receiver. Furthermore, although a predetermined desired threshold of interfering RF energy relative to the desired RF energy is described, the underlying principles of the present invention are not limited to a predetermined threshold.

[0077] There are different types of interference zones 8010 as shown in Figure 4. For example, a "Type A" region (denoted by the letter "A" in Figure 4) is affected by interference from only one nearby cluster, while a "Type B" region (denoted by the letter "B") corresponds to interference from two or more nearby clusters.

[0078] Figure 5 illustrates a framework used in one embodiment of the present invention. Dots indicate DIDO distributed antennas, crosses point to DIDO clients, and arrows indicate the direction of RF energy propagation. DIDO antennas in the main cluster transmit pre-coded data signals to clients MC 501 in that cluster. Similarly, DIDO antennas in the interference cluster serve clients IC 502 in that cluster through conventional DIDO pre-coding. A green cross 503 indicates a target client TC 503 in the interference zone. DIDO antennas in the main cluster 511 transmit pre-coded data signals to the target client (black arrow) through conventional DIDO pre-coding. DIDO antennas in the interference cluster 512 use pre-coding to create zero RF energy in the direction of the target client 503 (green arrow).

[0079] The received signal at target client k in any interference zone 410A (B in FIG. 4) is TIFF0007621401000009.tif7110(1), where k=1, ..., K, K is the number of clients in the interference zones 8010A, B, U is the number of clients in the main DIDO cluster, and C is the number of interference DIDO clusters 412-413. TIFF0007621401000010.tif43 is the number of clients in interference cluster c. Furthermore, assuming M transmit DIDO antennas and N receive antennas at the client device, TIFF0007621401000011.tif416 is a vector containing the incoming data stream at client k, TIFF0007621401000012.tif415 is a vector of outgoing data streams to client k in the main DIDO cluster, TIFF0007621401000013.tif415 is a vector of the outgoing data stream to client u in the main DIDO cluster, TIFF0007621401000014.tif418 is the vector of outgoing data streams to client i in the c-th interfering DIDO cluster, TIFF0007621401000015.tif416 is a vector of additive white Gaussian noise (AWGN) at the N receive antennas of client k, TIFF0007621401000016.tif420 is the DIDO channel matrix from M transmit DIDO antennas to N receive antennas at client k in the primary DIDO cluster, TIFF0007621401000017.tif421 is the DIDO channel matrix from M transmit DIDO antennas to N receive antennas of client k in the c-th interfering DIDO cluster, TIFF0007621401000018.tif420 is the matrix of DIDO precoding weights for client k in the main DIDO cluster, TIFF0007621401000019.tif420 is the matrix of DIDO precoding weights for client u in the main DIDO cluster, TIFF0007621401000020.tif421 is the matrix of DIDO precoding weights for client i in the cth interfering DIDO cluster.

[0080] For simplicity of notation and without loss of generality, we assume that every client is equipped with N receive antennas, and there are M DIDO distributed antennas in every DIDO cluster. TIFF0007621401000021.tif521 and Assume that M is greater than the total number of receive antennas in the cluster. If M is greater than the total number of receive antennas in the cluster, the extra transmit antennas are used to proactively cancel interference for target clients in the interference zone or to improve link robustness for clients in the same cluster through diversity schemes described in related patents and applications, including U.S. Pat. Nos. 7,599,420, 7,633,994, 7,636,381, and U.S. patent application Ser. No. 12 / 143,503.

[0081] The DIDO precoding weights are calculated to pre-cancel the inter-client interference within the same DIDO cluster. For example, block diagonalization (BD) precoding, as described in U.S. Pat. Nos. 7,599,420, 7,633,994, 7,636,381, and U.S. patent application Ser. No. 12 / 143,503, and related patents and applications including [7], can be used to remove the inter-client interference such that the following condition is satisfied in the main cluster (2): TIFF0007621401000023.tif479The precoding weight matrix in adjacent DIDO clusters is designed so that the following condition is satisfied (3). TIFF0007621401000024.tif493 precoding matrix TIFF0007621401000025.tif46 in the interference cluster from M transmit antennas. The downlink channels to the i-th client in the interference zone and to the k-th client in the interference zone are estimated, and the precoding matrix is ​​calculated by the DIDO BTS in the interference cluster. When the BD method is used to calculate the precoding matrix in the interference cluster, the following effective channel matrix is ​​constructed to calculate the weight for the i-th client in the neighboring cluster (4): TIFF0007621401000027.tif1221 where, TIFF0007621401000028.tif56 is the channel matrix for interference cluster c. TIFF0007621401000029.tif425, where the column corresponding to the i-th client is removed. Substituting conditions (2) and (3) into (1), we obtain the received data stream for target client k, and the intra-cluster and inter-cluster interference is removed (5). TIFF0007621401000030.tif434 Precoding weights in (1) calculated for neighboring clusters TIFF0007621401000031.tif46 is designed to transmit pre-encoded data streams to all clients in their clusters while proactively canceling interference to target clients in their interference zones. The target clients receive pre-encoded data only from their primary cluster. In a different embodiment, the same data stream is sent to the target client from the primary cluster and neighboring clusters to obtain diversity gain. In this case, the signal model in (5) is It is represented as TIFF0007621401000032.tif679(6), where: TIFF0007621401000033.tif46 is the DIDO precoding matrix from the DIDO transmitter in the cth cluster to target client k in the interference zone. Note that the method in (6) requires time synchronization across neighboring clusters, which may be complex to achieve in large systems, but is still quite achievable if the diversity gain benefits justify the cost of implementation.

[0082] We begin by evaluating the performance of the proposed method in terms of the bit error rate (SER) as a function of the signal-to-noise ratio (SNR). Without loss of generality, we define the following signal model (7) assuming a single antenna per client and reformulating (1): TIFF0007621401000034.tif6103Here, INR is the interference-to-noise ratio defined as INR=SNR / SIR, and SIR is the signal-to-interference ratio.

[0083] Figure 6 shows the SER as a function of SNR, assuming an SIR of 10 dB for a target client in the interference zone. Without loss of generality, the SER was measured for 4-QAM and 16-QAM without forward error correction (FEC) coding. We fix the target SER at 1% for the uncoded system. This target corresponds to different values ​​of SNR based on the modulation order (i.e., SNR = 20 dB for 4-QAM and SNR = 28 dB for 16-QAM). When using FEC coding for coding gain, a lower SER target can be met for the same value of SNR. We consider a scenario with two clusters (one main cluster and one interference cluster) with two DIDO antennas per cluster and two clients (each equipped with a single antenna). One of the clients in the main cluster is in the interference zone. While assuming a flat-fading narrowband channel, the following results can be extended to frequency-selective multicarrier (OFDM) systems, where each subcarrier experiences flat fading. Two scenarios are considered: (i) precoding weights TIFF0007621401000035.tif66 shows one scenario with DIDO inter-cluster interference (IDCI) calculated without corresponding target clients in the interference zone, and (ii) weights to remove IDCI to the target clients. Consider the other scenario where IDCI is removed by calculating TIFF0007621401000036.tif66. It is observed that when IDCI is present, the SER is high and greater than the predetermined target. With IDCI precoding in the neighboring cluster, the interference to the target client is removed and the SER target is reached such that an SNR > 20 dB is obtained.

[0084] The results in Figure 6 assume IDCI precoding as in (5). Further diversity gain is obtained when IDCI precoding in the neighboring cluster is also used to precode the data stream to the target client in the interference zone as in (6). Figure 7 compares the SER derived from two techniques: (i) "Method 1" using IDCI precoding in (5), and (ii) "Method 2" using IDCI precoding in (6), where the neighboring cluster also transmits precoded data streams to the target client. Method 2 achieves a ~3 dB gain compared to conventional IDCI precoding due to the additional array gain provided by the DIDO antennas in the neighboring cluster used to transmit the precoded data stream to the target client. More generally, the array gain of Method 2 over Method 1 is 10 dB. * It is proportional to log10(C+1), where C is the number of neighboring clusters and the coefficient "1" refers to the main cluster.

[0085] Next, we evaluate the performance of previous methods as a function of the target client's location relative to the interference zone. Consider a simple scenario in which a target client 8401 moves from a primary DIDO cluster 802 to an interference cluster 803, as shown in Figure 8. We assume that all DIDO antennas 812 in the primary cluster 802 use BD precoding to cancel intra-cluster interference, so that condition (2) is satisfied. We assume a single interfering DIDO cluster, a single receive antenna at the client device 801, and equal path loss from all DIDO antennas in the primary or interference cluster to the client (i.e., DIDO antennas arranged around the client, representing a circle). We use a simplified path loss model

[11] with a path loss exponent of 4 (as in a typical urban environment). The following analysis is based on the following simplified signal model (8), which extends (7) to accommodate path loss: TIFF0007621401000037.tif21124Here, the signal-to-interference ratio (SIR) is derived as SIR = ((1-D) / D)4. In modeling IDCI, three scenarios are considered: i) the ideal case without IDCI, ii) IDCI pre-cancelled through BD pre-coding in the interfering cluster to satisfy condition (3), and iii) with IDCI and no pre-cancellation by neighboring clusters.

[0086] Figure 9 shows the signal-to-interference-and-noise ratio (SINR) as a function of distance (D) (i.e., as the target client moves from the main cluster 802 toward the DIDO antenna 813 in the interference cluster 8403). SINR is derived as the ratio of signal power and interference plus noise power using the signal model in (8). D = D o Towards D o Assume σ = 0.1 and SNR = 50 dB. Without IDCI, the wireless link performance is affected by noise alone, and the SINR decreases due to propagation loss. When IDCI is present (i.e., without IDCI precoding), interference from DIDO antennas in neighboring clusters contributes to reducing the SINR.

[0087] Figure 10 shows the symbol error rate (SER) performance of three scenarios of 4-QAM modulation in a flat-fading narrowband channel. These SER results correspond to the SINR thresholds in Figure 9. T Assume a 1% SER threshold for an uncoded system (i.e., without FEC), corresponding to =20 dB. The SINR threshold depends on the modulation order used for data transmission. Typically, a higher modulation order will result in a higher SINR to yield the same target error rate. TFEC can result in a lower target SER for the same SINR value due to coding gain. In the case of IDCI without precoding, the target SER is only achieved within the range D<0.25. With IDCI precoding in neighboring clusters, the range for meeting the target SER extends to D<0.6. Above that range, the SINR increases due to propagation loss and the target SER is not met.

[0088] An embodiment of the IDCI precoding method is shown in FIG. 11 and consists of the following steps: SIR Estimation 1101: The client estimates the signal power from the primary DIDO cluster (i.e., based on the received pre-coded data) and the noise-plus-interference signal power from neighboring DIDO clusters. In a single-carrier DIDO system, the frame structure can be designed with short silence periods. For example, a silence period can be defined between channel estimation training and pre-coded data transmission during channel state information (CSI) feedback. In one embodiment, the noise-plus-interference signal power from neighboring clusters is measured during the silence period from a DIDO antenna in the primary cluster. In practical DIDO multi-carrier (OFDM) systems, null tones are typically used to prevent direct current (DC) and band edge attenuation due to filtering at the transmit and receive sides. In another embodiment using a multi-carrier system, the noise-plus-interference signal power is estimated from the null tones. A correction factor can be used to compensate for transmit / receive filter attenuation at the band edges. The signal-to-noise-plus-interference power (P) from the primary cluster is calculated as follows: S ) and neighboring clusters (P IN ) is estimated, the client SINR is calculated as (9). Alternatively, the SINR estimate can be derived from the received signal strength indication (RSSI), which is used in common wireless communication systems to measure wireless signal power. It is recognized that the metric in (9) cannot distinguish between noise and interference power levels. For example, a client affected by shadowing in an interference-free environment (i.e., behind an obstacle that attenuates the signal power from all DIDO distributed antennas in the main cluster) can estimate a low SINR even if it is not affected by inter-cluster interference. A more reliable metric for the proposed method is TIFF0007621401000039.tif827(10) where P N is the noise power. In a practical multi-carrier OFDM system, it is assumed that all DIDO antennas in the main cluster and neighboring clusters use the same set of null tones, and the noise power P N is estimated from the null tone. As mentioned above, the noise plus interference power (P IN ) is estimated from the silence period. Finally, the signal-to-noise-plus-interference power (P S ) is derived from the data tones. From these estimates, the client calculates the SIR in (10). Channel estimation in neighboring clusters 1102-1103: The estimated SIR in (10) is compared with a predetermined threshold (SIR T ), the client starts listening to training signals from neighboring clusters. Tis dependent on the modulation and FEC coding scheme (MCS) used for data transmission. Different SIR targets are defined by the client's MCS. When DIDO distributed antennas in different clusters are time-synchronized (i.e., locked to the same pulses per second PPS time reference), the client utilizes training sequences to distribute its channel estimates to DIDO antennas in neighboring clusters at 8703. The training sequences for channel estimation in neighboring clusters are designed to be orthogonal to the training from the main cluster. Alternatively, when DIDO antennas in different clusters are not time-synchronized, orthogonal sequences (with good cross-correlation properties) are used for time synchronization in different DIDO clusters. With the client locked to the time / frequency reference of the neighboring cluster, channel estimation is performed at 1103. IDC precoding I1104: Once channel estimates are available to the DIDO BTSs in the neighboring cluster, IDCI precoding is calculated to satisfy condition (3). The DIDO antennas in the neighboring cluster transmit precoded data streams only to clients in their cluster while proactively canceling interference to clients in the interference zone 410 of Figure 4. It is recognized that when a client is in the Type B interference zone 410 of Figure 4, interference to the client is generated by multiple clusters, and IDCI precoding is performed by all neighboring clusters simultaneously.

[0089] Handoff Method Hereafter, different handoff methods are described for a client moving across a DIDO cluster populated by distributed antennas located in different areas providing different types of service (i.e., low mobility or high mobility service).

[0090] a. Handoff between adjacent DIDO clusters In one embodiment, the above-described IDCI-precoder for eliminating inter-cluster interference is used as a baseline for handoff methods in DIDO systems. Traditional handoffs in cellular systems are designed for clients to seamlessly switch between cells served by different base stations. DIDO systems allow clients to move between clusters without losing connectivity.

[0091] To illustrate one embodiment of a handoff strategy for a DIDO system, consider again the example of Figure 8 with only two clusters 802 and 803. When a client 801 moves from a primary cluster (C1) 802 to a neighboring cluster (C2) 803, an embodiment of the handoff method dynamically calculates the signal quality of the different clusters and selects the cluster that provides the lowest error rate performance for the client.

[0092] Figure 12 shows the SINR variation as a function of a client's distance from the center of cluster C1. Consider a target SINR of 20 dB for 4-QAM modulation without FEC coding. The line identified by the circles represents the SINR of a target client served by the DIDO antenna in C1 when C1 and C2 use DIDO precoding without interference cancellation. The SINR decreases as a function of D due to propagation loss and interference from neighboring clusters. When DIDO precoding is performed on neighboring clusters, the SINR reduction is only due to propagation loss (as shown by the line with triangles) because the interference is completely eliminated. Symmetric behavior is experienced when a client is served from a neighboring cluster. One embodiment of a handoff strategy is defined such that when a client moves from C1 to C2, the algorithm switches between different DIDO schemes that maintain the SINR above a predetermined target.

[0093] From the plots in Figure 12, we derive the SER for 4-QAM modulation in Figure 13. It can be seen that by switching between different precoding strategies, the SER was maintained within the predetermined target.

[0094] One embodiment of the handoff strategy is as follows. · C1-DIDO and C2-DIDO precoding: When a client is within C1 in the direction away from the interference zone, clusters C1 and C2 operate independently with conventional DIDO precoding. C1-DIDO and C2-IDCI precoding: When a client moves towards the interference zone, its SIR or SINR will degrade. When it reaches the target SINRT1, the target client starts to estimate the channel from all DIDO antennas in C2 and provides CSI to the BTS in C2. The BTS in C2 calculates IDCI precoding and transmits it to all clients in C2 while preventing interference to the target client. As long as the target client is in the interference zone, it continues to provide CSI to C1 and C2. C1-IDCI and C2-DIDO precoding: As the client moves toward C2, the SIR or SINR continues to decrease until it reaches the target again. At this point, the client decides to switch to a neighboring cluster. In this case, C1 begins using CSI from the target client to create zero interference in that direction with IDCI precoding, while the neighboring cluster uses the CSI for conventional DIDO precoding. In one embodiment, when the SIR estimate approaches the target, clusters C1 and C2 alternately try DIDO and IDCI precoding schemes to allow the client to estimate the SIR in both cases. The client then selects the best scheme that maximizes a predetermined error rate performance metric. When this method is applied, the intersection of handoff strategies occurs at the intersection of the triangle and diamond curves in FIG. 12. One embodiment uses the modified IDCI precoding method described in (6), and the neighboring cluster also transmits precoded data streams to the target client to achieve array gain. In this way, the handoff strategy is simplified because the client does not need to estimate the SINR for both strategies at the intersection. C1-DIDO and C2-DIDO precoding: When a client moves out of the interference zone in the direction of C2, the main cluster C1 stops pre-cancelling interference in the direction of that target client through IDCI precoding and switches back to conventional DIDO precoding for all clients remaining in C1. This final crossover point in the handoff strategy is useful to avoid unnecessary CSI feedback from the target client to C1, thus reducing the overhead over the feedback channel. In one embodiment, the second target SINR T2is defined. When the SINR (or SIR) exceeds this target, the strategy switches to C1-DIDO and C2-DIDO. In one embodiment, cluster C1 keeps alternating between DIDO precoding and IDCI precoding to allow the client to estimate the SINR. The client then determines the target SINR for C1 from above. T Choose the method that most closely approaches 1.

[0095] The above method is used to calculate SINR or SIR estimates for different schemes in real time and select the optimal scheme. In one embodiment, the handoff algorithm is designed based on the finite state machine shown in Figure 14. The client tracks the current state and switches to the next state when the SINR or SIR is below or above a predetermined threshold shown in Figure 12. As described above, in state 1201, clusters C1 and C2 operate independently with conventional DIDO precoding, and the client is served by cluster C1; in state 1202, the client is served by cluster C1, the BTSs in C2 calculate IDCI precoding, and cluster C1 operates using conventional DIDO precoding; in state 1203, the client is served by cluster C2, the BTSs in C1 calculate IDCI precoding, and cluster C2 operates using conventional DIDO precoding; and in state 1204, the client is served by cluster C2, and clusters C1 and C2 operate independently with conventional DIDO precoding.

[0096] In the presence of shadowing effects, the signal quality or SIR may fluctuate around a threshold as shown in Figure 15, resulting in repetitive switching between the successive states of Figure 14. Repetitive state changes are an undesirable effect because they result in significant overhead for the control channel between the client and the BTS that enables switching between transmission modes. Figure 15 illustrates one embodiment of a handoff strategy in the presence of shadowing. In one embodiment, the shadowing coefficient is simulated to follow a log-normal distribution with variance 3 [3]. Hereafter, we define several methods to prevent repetitive switching effects during DIDO handoff.

[0097] One embodiment of the present invention uses a hysteresis loop to address state switching effects. For example, when switching between states "C1-DIDO, C2-IDCI" 9302 and "C1-IDCI, C2-DIDO" 9303 (or alternatively, vice versa) in FIG. 14, the threshold SINR T 1 can be adjusted within the range A1. This method is suitable for when the signal quality is SINR T To avoid repetitive switching between states when fluctuating around 1, for example, Figure 16 shows the hysteresis loop mechanism when switching between any two states in Figure 14. To switch from state B to A, the SIR must be greater than (SIRT1 + A1 / 2), but to switch from A to B again, the SIR must be greater than (SIRT1 + A1 / 2). T 1-A1 / 2).

[0098] In a different embodiment, the threshold SINR T 2 is adjusted to avoid repeated switching between the first and second (or third and fourth) states of the finite state machine in FIG. 14. For example, the range of values ​​A2 is T2 can be defined to be selected in that range based on the channel condition and shadowing effect.

[0099] In one embodiment, based on the variance of shadowing estimated over the wireless link, the SINR threshold is set to a value in the range [SINR T 2. SINR T 2+A2]. The variance of the log-normal distribution can be estimated from the variance of the received signal strength (or RSSI) as the client moves from its current cluster to a neighboring cluster.

[0100] Previous methods assume that the client triggers the handoff strategy. In one embodiment, the handoff decision is left to the DIDO BTS, assuming communication across multiple BTSs is enabled.

[0101] For simplicity, the previous methods are derived assuming no FEC coding and 4-QAM. More generally, SINR or SIR thresholds are derived for different modulation and coding schemes (MCS), and handoff strategies are designed in combination with link adaptation to optimize the downlink data rate for each client in the interference zone (see, e.g., U.S. Pat. No. 7,636,381).

[0102] b. Handoff between low and high Doppler DIDO networks DIDO systems use a closed-loop transmission scheme to pre-code the data stream over the downlink channel. Closed-loop schemes are inherently constrained by latency over the feedback channel. A practical DIDO system can reduce computational time with transceivers that have high processing power; it is estimated that most of the latency is introduced by the DIDO BSN when delivering the CSI and baseband pre-coded data from the BTS to the distributed antennas. A BSN can be formed with various network technologies, including, but not limited to, digital subscriber line (DSL), cable modem, fiber ring, T1 line, hybrid fiber-coaxial (HFC) network, and / or fixed wireless (e.g., Wi-Fi). Dedicated fiber typically offers very high bandwidth and low latency (potentially less than 1 millisecond in a local area), but is not as widespread as DSL and cable modem. Today, DSL and cable modem connections are very widespread, with last-mile latency typically between 10 and 25 ms in the United States.

[0103] The maximum latency across the BSN determines the maximum Doppler shift that can be satisfied across a DIDO radio link without performance degradation due to DIDO precoding. For example, in [1] it was found that at a carrier frequency of 400 MHz, a network with a latency of approximately 10 ms (i.e., DSL) can accommodate client speeds of up to 8 mph (operating speed), while a network with a latency of 1 ms (i.e., fiber ring) can support speeds of up to 70 mph (i.e., freeway traffic).

[0104] Two or more DIDO subnetworks are defined based on the maximum Doppler shift that can be satisfied across the BSN. For example, a BSN with high-latency DSL connections between the DIDO BTSs and distributed antennas can only deliver low-mobility or fixed-wireless services (i.e., a low-Doppler network), while a low-latency BSN over low-latency fiber rings can satisfy high mobility (i.e., a high-Doppler network). It is recognized that most broadband users are not mobile when using broadband, and furthermore, most users are unlikely to be located near areas where many high-speed objects pass through (e.g., next to a highway) because such locations are typically less desirable for living or operating an office. However, there are broadband users who may be using broadband at high speeds (e.g., in a car traveling on a highway) or near high-speed objects (e.g., in a store located near a highway). To address these two different user Doppler scenarios, in one embodiment, a low-Doppler DIDO network typically consists of many DIDO antennas with relatively low power (i.e., 1 W to 100 W, for indoor or rooftop deployments) spread over a wide area, while a high-Doppler network typically consists of a few DIDO antennas with high power transmissions (i.e., 100 W, for rooftop or tower deployments). A low-Doppler DIDO network typically serves many low-Doppler users and can do so at typically low connection costs using inexpensive, high-latency broadband connections such as DSL and cable modems. A high-Doppler DIDO network typically serves a small number of high-Doppler users and can do so at typically high connection costs using expensive, low-latency broadband connections such as fiber.

[0105] To avoid interference across different types of DIDO networks (e.g., low Doppler and high Doppler), different multiple access techniques such as time division multiple access (TDMA), frequency division multiple access (FDMA) or code division multiple access (CDMA) can be used.

[0106] We propose a method to assign clients to different types of DIDO networks and enable handoff between them. Network selection is based on each client's mobility type. The client's velocity (v) is proportional to the maximum Doppler shift according to the following equation [6] (11): TIFF0007621401000040.tif623where, f d is the maximum Doppler shift, λ is the wavelength corresponding to the carrier frequency, and θ is the angle between the vector indicating the direction transmitter-client and the velocity vector.

[0107] In one embodiment, the Doppler shift of every client is calculated through a blind estimation technique. For example, the Doppler shift can be estimated by transmitting RF energy to the client and analyzing the reflected signal similar to a Doppler radar system.

[0108] In another embodiment, one or more DIDO antennas send training signals to the client. Based on these training signals, the client estimates the Doppler shift using techniques such as counting the number of zero-crossings of the channel gain or performing spectral analysis. It is noted that for a constant velocity v and client trajectory, the angular velocity vs sin θ in (11) may depend on the client's relative distance from any DIDO antenna. For example, a DIDO antenna closer to a moving client will experience a larger angular velocity and Doppler shift than a more distant antenna. In one embodiment, Doppler velocity is estimated from multiple DIDO antennas at different distances from the client, and the average, weighted average, or standard deviation is used as an indicator of the client's mobility. Based on the estimated Doppler indicator, the DIDO BTS determines whether to assign the client to a low-Doppler or high-Doppler network.

[0109] Doppler signatures are periodically monitored for all clients and sent back to the BTS, and when one or more clients change their Doppler velocity (i.e., a client on a bus versus a client walking or sitting), they are dynamically reassigned to a different DIDO network that can accommodate their level of mobility.

[0110] Although the Doppler of a slow-moving client may be affected by being near a high-speed object (e.g., near a highway), the Doppler is typically much less than that of a client that is itself in motion. Thus, in one embodiment, the client's velocity is estimated (e.g., by using means such as monitoring the client's location using GPS), and if the velocity is low, the client is assigned to a low-Doppler network, and if the velocity is high, the client is assigned to a high-Doppler network.

[0111] Power control and antenna grouping methods A block diagram of a DIDO system with power control is shown in Figure 17. One or more data streams (s k ) are multiplied by weights generated by the DIDO precoding unit. The precoded data streams are multiplied by power scaling factors calculated by the power controller based on the input channel quality information (CQI). The CQI is fed back from the client in the DIDOBTS or derived from the uplink channel assuming uplink-downlink channel reciprocity. The U precoded streams of different clients are then multiplied by the M data streams (t m ), which are then combined and multiplexed, one for each of the M transmit antennas. Finally, stream t m is sent to a digital-to-analog converter (DAC) unit, a radio frequency (RF) unit, a power amplifier (PA) unit and finally an antenna.

[0112] The power control unit measures the CQI for all clients. In one embodiment, the CQI is an average SNR or RSSI. The CQI varies for different clients based on propagation loss or shadowing. The power control method uses a transmit power scaling factor P k and multiplying it by the pre-coded data streams generated for the different clients, where one or more data streams can be generated for every client based on the number of receive antennas of the client.

[0113] To evaluate the performance of the proposed method, we define the following signal model based on (5) including propagation loss and power control parameters (12): TIFF0007621401000041.tif655where k=1, …, U, U is the number of clients, and SNR=P o / N o , P o is the average transmit power, and N o is the noise power, TIFF0007621401000042.tif53 is the propagation loss / shadowing coefficient. To model the propagation loss / shadowing, we use the following simplified model (13): TIFF0007621401000043.tif825Here, a=4 is the path loss exponent, and we assume that the path loss increases with the client's exponent (i.e., clients are located at increasing distances from the DIDO antenna).

[0114] Figure 18 shows SER vs. SNR assuming four DIDO transmit antennas and four clients in different scenarios. In the ideal case, all clients have the same path loss (i.e., a = 0) and P k = 1. The plots with squares refer to the cases where the clients have different path loss coefficients and no power control. The curves with dots are derived from the same scenario (with path loss) and the power control coefficient is TIFF0007621401000044.tif516. In the power control method, more power is allocated to the data streams that are for clients that experience higher propagation loss / shadowing, thus resulting in a 9 dB SNR gain compared to no power control (for this given scenario).

[0115] The US Federal Communications Commission (FCC) (and other international regulatory bodies) define constraints on the maximum power that can be transmitted from wireless devices to limit human exposure to electromagnetic (EM) radiation. There are two types of limits: i) occupational / controlled environment limits, where people are made fully aware of radio frequency (RF) sources through barriers, warnings, or labels, and ii) "general public / uncontrolled" limits, where there are no controls on exposure [2].

[0116] Different emission levels are defined for different types of wireless devices. Generally, DIDO distributed antennas used for indoor / outdoor applications fall under the FCC category of "mobile" devices, which is defined as [2]: "Transmitting devices designed for use at other than fixed locations, typically expected to be used with a radiating structure maintained 20 cm or more from the body of the user or nearby persons."

[0117] The EM emissions of "mobile" devices are in mW / cm 2 This is measured in terms of maximum permissible exposure (MPE), which is expressed as radiated power (W). Figure 19 shows the MPE power density as a function of distance from the RF radiation source for different values ​​of transmit power at a carrier frequency of 700 MHz. The maximum permissible transmit power to meet the FCC "uncontrolled" limits for devices operating greater than 20 cm from the human body is typically 1 watt.

[0118] Less restrictive power emission limits are defined for transmitters located on rooftops or buildings away from the "general public." For these "rooftop transmitters," the FCC defines a more relaxed emission limit of 1000 W, measured in terms of effective radiated power (ERP).

[0119] Based on previous FCC constraints, in one embodiment, we define two types of DIDO distributed antennas for practical systems. Low Power (LP) Transmitter: 1W and 5Mbps maximum transmit power for consumer-grade broadband anywhere (e.g., DSL, cable modem, fiber-to-the-home (FTTH) backhaul connectivity located at any elevation (i.e., indoor or outdoor). High Power (HP) Transmitter: Rooftop or building-mounted antenna at a height of approximately 10 meters (effectively "unlimited" data rates compared to the yield available over DIDO radio links) with a transmit power of 100W and commercially available wideband (e.g., fiber optic ring) backhaul.

[0120] Note that LP transmitters with DSL or cable modem connectivity are good candidates for low-Doppler DIDO networks (as discussed in the previous section) because the clients are mostly stationary or have low mobility. HP transmitters with commercially available fiber connectivity can tolerate higher client mobility and can be used for high-Doppler DIDO networks.

[0121] To gain practical insight into the performance of DIDO systems with different types of LP / HP transmitters, consider the practical example of a DIDO antenna placement in downtown Palo Alto, California. Figure 20a shows the N LP In Fig. 20b, the 50 LP antennas are randomly distributed over N = 100 low-power DIDO distributed antennas. HP = 50 high power transmitters.

[0122] Based on the DIDO antenna distributions in Figures 20a-b, we derive coverage maps for a system using DIDO technology in Palo Alto. Figures 21a and 21b show two power distributions corresponding to the configurations in Figures 20a and 20b, respectively. The received power distributions (expressed in dBm) are derived for an urban environment defined by the 3GPP standard [3] at a carrier frequency of 700 MHz, assuming a propagation loss / shadowing model. It can be seen that using 50% of the HP transmitters improves coverage in the selected area.

[0123] Figures 22a-b show the speed distributions for the two previous scenarios. The yields (expressed in Mbps) are derived based on power thresholds for different modulation and coding schemes defined in the 3GPP Long Term Evolution (LTE) standard in [4, 5]. The bandwidth for all transmissions is fixed at 10 MHz with a carrier frequency of 700 MHz. Two different frequency allocation plans are considered: i) 5 MHz of spectrum allocated only to LP base stations; ii) 9 MHz for HP transmitters and 1 MHz for LP transmitters. Note that LP fields are typically allocated lower bandwidth due to DSL backhaul connectivity, which limits the yield. Figures 22a-b show that the speed distribution can be significantly increased by increasing the average data rate per client from 2.4 Mbps in Figure 22a to 38 Mbps in Figure 22b when using 50% of the HP transmitters.

[0124] We then defined an algorithm to control the power transmission of the LP base station so that higher power is allowed at any given time, thus increasing the throughput across the downlink channel of the DIDO system of Figure 22b. It is recognized that the FCC limits on power density are defined based on averages over time as [2] (14). TIFF0007621401000045.tif1230 where, TIFF0007621401000046.tif632 exists in the MPE averaging time, t n is the power density S n The MPE limit is the duration of exposure to radiation at 100 Hz. For "controlled" exposure, the average time is 6 minutes, while for "uncontrolled" exposure, it is increased to 30 minutes. Any power source is then allowed to transmit at a power level greater than the MPE limit as long as the average power density of (14) meets the FCC limit for a 30-minute average of "uncontrolled" exposure.

[0125] Based on this analysis, we define an adaptive power control method that increases the instantaneous transmit power per antenna while maintaining the average power per DIDO antenna below the MPE limit. We consider a DIDO system with more transmit antennas than active clients. This is a reasonable assumption considering that DIDO antennas can be thought of as inexpensive wireless devices (similar to WiFi access points) and can be located anywhere there is DSL, cable modem, fiber optic, or other Internet connectivity.

[0126] The framework of a DIDO system with adaptive per-antenna power control is shown in Figure 23. The amplitude of the digital signal emerging from multiplexer 234 is scaled by power scaling factors S1...S before being sent to DAC unit 235. M The power scaling factor is calculated by the power control unit 232 based on the CQI 233.

[0127] In one embodiment, N g DIDO antenna groups are defined. Every group contains at least as many DIDO antennas as there are active clients (K). At any given time, only one group is configured to achieve the MPE limit ( TIFF0007621401000047.tif79) o ) to send to the client a >K active DIDO antennas. The method is repeated across all antenna groups according to a round-robin scheduling policy as shown in Figure 24. In another embodiment, a different scheduling method (i.e., proportional fair scheduling [8]) is used for cluster selection to optimize error rate or yield performance.

[0128] Assuming round-robin power allocation, from (14) TIFF0007621401000048.tif844(15) derive the average transmit power for every DIDO antenna, where t oexists during the period when the antenna group is active, and T MPE = 30 min is in the averaging time defined by FCC guidelines [2]. The ratio of (15) is the ratio of the average transmit power from any DIDO antenna to the MPE limit ( The duty factor (DF) of the group is defined to satisfy ∑ ... TIFF0007621401000050.tif1244 The SNR gain (in dB) obtained in a DIDO system with power control and antenna grouping is: TIFF0007621401000051.tif1047(17) as a function of duty cycle. The gain of (17) is G across all DIDO antennas. dB It will be appreciated that this is achieved at the cost of additional transmit power of . In general, for all N g All N in the group a The total transmitted power from TIFF0007621401000052.tif1043(18), where P ij teeth, TIFF0007621401000053.tif1171(19) is the average transmit power per antenna, and S ij (t) is the power spectral density of the ith transmit antenna in the jth group. In one embodiment, the power spectral density in (19) is designed to optimize the error rate or yield performance of every antenna.

[0129] To get some intuition about the performance of the proposed method, consider 400 DIDO distributed antennas within a given coverage area and 400 clients that subscribe to wireless Internet service provided by the DIDO system. It is unlikely that every Internet connection will be fully utilized at all times. Assume that 10% of the clients actively use their wireless Internet connection at any given time. Then, we assign each of the 400 DIDO antennas to N a N = 40 antennas g The DIDO antennas can be partitioned into G = 10 groups, with every group serving K = 40 active clients at any given time with a duty cycle DF = 0.1. The SNR gain resulting from this transmission scheme is G = 10 dB, provided by the additional transmit power of 10 dB from all DIDO antennas. dB =10log 10 (1 / DF)=10 dB. However, it is observed that the average transmit power per antenna is constant and within the MPE limit.

[0130] Figure 25 compares the (uncoded) SER performance of previous power control with antenna grouping with the conventional eigenmode selection of U.S. Patent No. 7,636,381. All schemes use BD precoding with four clients, each equipped with a single antenna. SNR refers to the ratio of power per transmit antenna to noise power (i.e., transmit SNR per antenna). The curves shown for DIDO4x4 assume four transmit antennas and BD precoding. The curve with squares shows the SER performance with two extra transmit antennas and BD with eigenmode selection, which provides a 10 dB SNR gain over conventional BD precoding (with a 1% SER target). Power control with antenna grouping and DF=1 / 10 also provides a 10 dB gain with the same SER. Note that eigenmode selection changes the slope of the SER curve due to diversity gain, while this power control method shifts the SER curve to the left (while maintaining the same slope) due to increased average transmit power. For comparison, it is found that the SER at the larger duty cycle DF=1 / 50 provides another 7 dB gain compared to DF=1 / 10.

[0131] It should be noted that our power control may be less complex than conventional eigenmode selection methods. In fact, the antenna IDs for every group can be pre-calculated and shared among DIDO antennas and clients via a lookup table, so that only K channel estimates are needed at any given time. For eigenmode selection, (K+2) channel estimates are calculated, and an additional computational step is required to select the eigenmode that minimizes the SER for all clients at any given time.

[0132] Next, we describe another method involving DIDO antenna grouping to reduce CSI feedback overhead in some special scenarios. Figure 26a shows one scenario where clients (dots) are randomly spread in an area covered by multiple DIDO distributed antennas (cross symbols). The average power across all transmit-receive radio links is TIFF0007621401000054.tif521(20), where H is the channel estimation matrix available in the DIDO BTS.

[0133] Matrix A in Figures 26a-26c is obtained numerically by averaging the channel matrix over 1000 cases. Two alternative scenarios are shown in Figures 26b and 26c, respectively, where clients are grouped together around a subset of DIDO antennas and receive negligible power from DIDO antennas located farther away. For example, Figure 26b shows two groups of antennas from which a block-diagonal matrix A is obtained. One extreme scenario is when every client is very close to only one transmitter, and the transmitters are oriented far away from each other so that the power from all other DIDO antennas is negligible. In this case, the DIDO link deteriorates over multiple SISO links, and A is a diagonal matrix as in Figure 26c.

[0134] In all three previous scenarios, BD precoding dynamically adjusts precoding weights to accommodate different power levels across the wireless link between the DIDO antenna and the client. However, it is advantageous to identify multiple groups within a DIDO cluster and operate DIDO precoding only within each group. This proposed grouping method offers the following advantages: Computational gain: DIDO precoding is computed only within every group in a cluster. For example, when BD precoding is used, singular value decomposition (SVD) has a complexity of O(n 3), where n is the minimum dimension of the channel matrix H. If H can be made a block diagonal matrix, then the SVD can be computed for every block with reduced complexity. In practice, if the channel matrix is ​​split into two block matrices with dimensions n1 and n2 such that n=n1+n2, the complexity of SVD can be reduced to O(n1 3 )+O(n2 3 ) <O(n 3 ) In the extreme case, when H is a diagonal matrix, the DIDO links become multiple SISO links and no SVD calculation is required. Reduced CSI feedback overhead: When DIDO antennas and clients are divided into groups in one embodiment, CSI is calculated from clients to antennas only within the same group. For TDD systems, by assuming channel reciprocity, antenna grouping reduces the number of channel estimates to calculate the channel matrix H. For FDD systems, where CSI is fed back over the air link, antenna grouping further reduces CSI feedback overhead across the air link between DIDO antennas and clients.

[0135] Multiple access techniques for DIDO uplink channels. In one embodiment of the present invention, different multiple access techniques are defined for the DIDO uplink channel. These techniques can be used to feed back CSI or transmit data streams from clients to DIDO antennas in the uplink. Hereinafter, the feedback CSI and data streams are referred to as uplink streams. Multiple Input Multiple Output (MIMO): Uplink streams are transmitted from clients to DIDO antennas via open-loop MIMO multiplexing. This method assumes that all clients are time / frequency synchronized. In one embodiment, synchronization between clients is achieved through training from the downlink, and all DIDO antennas are assumed to be locked to the same time / frequency reference clock. Note that variations in delay spread at different clients can cause jitter between different client clocks, which can affect the performance of the MIMO uplink scheme. After clients send their uplink streams via MIMO multiplexing, the receiving DIDO antennas can use nonlinear (i.e., maximum likelihood, ML) or linear (i.e., zero-forcing, minimum mean square error) receivers to remove co-channel interference and individually demodulate the uplink streams. Time Division Multiple Access (TDMA): Different clients are assigned different time slots. Every client sends its uplink stream when a time slot is available. Frequency Division Multiple Access (FDMA): Different clients are assigned different carrier frequencies. In multi-carrier (OFDM) systems, subsets of tones are assigned to different clients that simultaneously transmit uplink streams, thus reducing latency. Code Division Multiple Access (CDMA): Every client is assigned a different pseudo-random number and orthogonality across clients is achieved in the code domain.

[0136] In one embodiment of the present invention, clients are wireless devices that transmit at much lower power than the DIDO antennas. In this case, the DIDO BTS defines client subgroups based on uplink SNR information so that interference across the subgroups is minimized. Within any subgroup, previous multiple access techniques are used to create orthogonal channels in the time, frequency, space, or code domains, thus avoiding uplink interference across different clients.

[0137] In another embodiment, the uplink multiple access techniques described above are used in combination with the antenna grouping method presented in the previous section to define different client groups within a DIDO cluster. System and method for link adaptation in a DIDO multi-carrier system

[0138] System and method for link adaptation in a DIDO multi-carrier system A link adaptation method for a DIDO system that exploits the time, frequency, and spatial selectivity of the wireless channel is defined in U.S. Patent No. 7,636,381. An embodiment of the present invention for link adaptation in a multi-carrier (OFDM) DIDO system that exploits the time / frequency selectivity of the wireless channel is described below.

[0139] In [9], we provide a simulation of a Rayleigh fading channel according to the exponentially decaying power delay profile (PDP) or Saleh-Valenzuela model. For simplicity, Assume a single cluster channel with a multipath PDP defined as TIFF0007621401000055.tif625, where n=0...,L-1) is the index of the channel taps, L is the number of channel taps, TIFF0007621401000056.tif518 shows the channel delay spread (σ DSβ is the PDP index, which is a measure of the channel interference zone width that is inversely proportional to β. A low value of β results in a frequency-nonselective channel, while a high value of β results in a frequency-selective channel. The PDP in (21) is normalized so that the total average power for all L channel taps is unity (22). Figure 27 shows the amplitude of the low frequency selective channel (assuming β=1) across the delay domain or instantaneous PDP (upper plot) and frequency domain (lower plot) for a DIDO 2x2 system. The first subscript indicates the client and the second subscript indicates the transmit antenna. The high frequency selective channel (with β=0.1) is shown in Figure 28.

[0140] Next, we study the performance of DIDO precoding in a frequency-selective channel. We calculate the DIDO precoding weights through BD assuming that the signal model in (1) satisfies the condition in (2). Under the condition in (2), we reformulate the DIDO received signal model in (5) as follows (23): TIFF0007621401000058.tif535

[0141] where: TIFF0007621401000059.tif521 is the effective channel matrix for user k. For DIDO2x2, with one antenna per client, the effective channel matrix reduces to a single value with the frequency response shown in Figure 29 and Figure 28 for a channel characterized by high frequency selectivity (e.g., with β = 0, 1). The solid line in Figure 29 refers to client 1, while the dotted line refers to client 2. Based on the channel quality metrics in Figure 29, we define a time / frequency domain link adaptation (LA) method that dynamically adjusts the MCS based on changing channel conditions.

[0142] We begin by evaluating the performance of different MCSs in AWGN and Rayleigh fading SISO channels. For simplicity, we assume no FEC coding, but the following LA method can be extended to systems that include FEC.

[0143] Figure 30 shows the SER for different QAM schemes (i.e., 4-QAM, 16-QAM, and 64-QAM). Without loss of generality, we assume a target SER of 1% for the uncoded system. The SNR thresholds to meet the target SER in an AWGN channel are 8 dB, 15.5 dB, and 22 dB for the three modulation schemes, respectively. In a Rayleigh fading channel, the SER performance of previous modulation schemes is known to be worse than AWGN

[13] , and the SNR thresholds are 18.6 dB, 27.3 dB, and 34.1 dB, respectively. We recognize that DIDO precoding transforms a multi-user downlink channel into a set of parallel SISO links. Therefore, the same SNR thresholds as in Figure 30 for the SISO system apply to the DIDO system on a per-client basis. Furthermore, when instantaneous LA is performed, the threshold in an AWGN channel is used.

[0144] The key idea of ​​the proposed LA method for DIDO systems is to use a low MCS order when the channel experiences deep fades in the time or frequency domain (as shown in Figure 28) to ensure link robustness. On the other hand, when the channel is characterized by large gains, the LA method switches to a higher MCS order to increase spectral efficiency. One contribution of this application compared to U.S. Patent No. 7,636,381 is the use of the effective channel matrix in (23) and Figure 29 as a metric to enable adaptation.

[0145] The general framework of the LA method is shown in Figure 31 and is defined as follows: CSI estimation: In 3171, the DIDO BTS calculates the CSI from all users. Users can be equipped with single or multiple receive antennas. DIDO Precoding: At 3172, the BTS calculates DIDO precoding weights for all users. In one embodiment, the BD is used to calculate these weights. The precoding weights are calculated on a per-tone basis. Link quality metric calculation: At 3173, the BTS calculates a frequency-domain link quality metric. In an OFDM system, the metric is calculated from the CSI, and DIDO precoding weights are calculated for every tone. In one embodiment of the present invention, the link quality metric is the average SNR across all OFDM tones. We define this method as LA1 (based on average SNR performance). In another embodiment, the link quality metric is the frequency response of the effective channel in (23). We define this method as LA2 (based on per-tone performance) to exploit frequency diversity. The frequency-domain effective channel is shown in Figure 29 when every client has a single antenna. When a client has multiple receive antennas, the link quality metric is defined as the Frobenius norm of the effective channel matrix for every tone. Alternatively, multiple link quality metrics are defined for every client as the singular values ​​of the effective channel matrix in (23). Bitloading algorithm: In 3174, based on the link quality metric, the BTS determines the MCS for different clients and different OFDM tones. For the LA1 method, the same MCS is used for all clients and all OFDM tones based on the SNR threshold of the Rayleigh fading channel in Figure 30. For the Rayleigh fading channel LA2, different MCSs are assigned to different OFDM tones to take advantage of channel frequency diversity. Precoded Data Transmission: At 3175, the BTS transmits precoded data streams from the DIDO distributed antennas to the clients using the MCS derived from the bitloading algorithm. A header is attached to the precoded data to inform the clients of the MCS for different tones. For example, if 8 MCSs are available and an OFDM code is defined with N=64 tones, then log2(8) * N = 192 bits are required to convey the current MCS to any client. Assuming 4-QAM (2 bit / symbol spectral efficiency) is used to map these bits to symbols, only 192 / 2 / N = 1.5 OFDM symbols are required to map the MCS information. In another embodiment, multiple subcarriers (or OFDM tones) are grouped into subbands, and the same MCS is assigned to all tones within the same subband to reduce the overhead due to control information. Furthermore, the MC is adjusted based on the time variation of the channel gain (proportional to the interference time). In fixed wireless channels (characterized by low Doppler effect), the MC is recalculated every small fraction of the channel interference time, thus reducing the overhead required for control information.

[0146] Figure 32 shows the SER performance of the LA method described above. For comparison, the SER performance in a Rayleigh fading channel is plotted for each of the three QAM schemes used. The LA2 method adapts the MCS to the effective channel variations in the frequency domain, thus achieving a 1.8 bps / Hz gain in spectral efficiency for low SNRs (i.e., SNR = 20 dB) and a 15 dB gain in SNR (towards SNR > 35 dB) compared to LA1.

[0147] System and method for DIDO precoding interpolation in multi-carrier systems The computational complexity of a DIDO system is mostly localized to a centralized processor or BTS. The most computationally expensive task is the calculation of precoding weights for all clients from the CSI. When BD precoding is used, the BTS needs to perform a singular value decomposition (SVD) task for as many clients as there are in the system. A way to reduce complexity is through parallelized processing, where the SVD is calculated for every client on a separate processor.

[0148] In a multi-carrier DIDO system, each subcarrier is subject to flat fading, and SVD is performed for every client across every subcarrier. Clearly, the system complexity increases linearly with the number of subcarriers. For example, in an OFDM system with a signal bandwidth of 1 MHz, the cyclic prefix (L0) must have at least 8 channel taps (i.e., 8 microseconds duration) to avoid inter-symbol interference in outdoor urban macrocell environments with large delay spread [3]. The size (N) of the fast Fourier transform (FFT) used to generate the OFDM symbols FFT ) is typically set to a multiple of L0 to reduce data rate degradation. FFT = 64, the effective spectral efficiency of the system is increased by a factor N FFT / (N FFT +L0) = 89%. FFT As the value of σ increases, higher spectral efficiency is obtained at the cost of higher computational complexity in the DIDO precoder.

[0149] A method to reduce computational complexity in a DIDO precoder is to perform an SVD operation over a subset of tones (called pilot tones) and derive precoding weights for the remaining tones through interpolation. Weight interpolation is one error factor that results in inter-client interference. In one embodiment, an optimal weight interpolation method is used to reduce inter-client interference, improving error rate performance and reducing computational complexity in multi-carrier systems. In a DIDO system with M transmit antennas, U clients, and N receive antennas per client, the kth client (W) can be assigned a weight that ensures zero interference to other clients u. k ) precoding weight condition is TIFF0007621401000060.tif691(24) is derived from (2), where TIFF0007621401000061.tif54 is the channel matrix corresponding to other DIDO clients in the system.

[0150] In one embodiment of the present invention, the objective function of the weighted interpolation method is: TIFF0007621401000062.tif1157(25), where: TIFF0007621401000063.tif53 is a set of parameters to be optimized for user k, TIFF0007621401000064.tif614 is the weighted interpolation matrix, TIFF0007621401000065.tif56 shows the Frobenius norm of a matrix. The optimization problem is It is formulated as TIFF0007621401000066.tif657(26), where: TIFF0007621401000067.tif54 is the feasible set of the optimization problem, TIFF0007621401000068.tif611 is the optimal solution.

[0151] The objective function in (25) is defined for one OFDM tone. In another embodiment of the invention, the objective function is defined as a linear combination of the Frobenius norms of the matrix (25) for all OFDM tones to be interpolated. In another embodiment, the OFDM spectrum is partitioned into a subset of tones and the optimal solution is TIFF0007621401000069.tif677(27), where n is the OFDM tone index and A is the subset of tones.

[0152] Weighted interpolation matrix of (25) TIFF0007621401000070.tif514 is a set of parameters TIFF0007621401000071.tif53. Once the optimal set is determined according to (26) or (27), the optimal weight matrix is ​​calculated. In one embodiment of the present invention, the weight interpolation matrix for a given OFDM tone n is defined as a linear combination of the weight matrices of the pilot tones. An example of a weight interpolation function for a beam pulse forming system with a single client is defined in

[11] . In a DIDO multi-client system, the inventors define the weight interpolation matrix as TIFF0007621401000072.tif6116(28), where 0≦l≦(L0-1), and L0 is the number of pilot tones. as TIFF0007621401000073.tif525 TIFF0007621401000074.tif530. The weight matrix in (28) is then TIFF0007621401000075.tif626. When N=1 (single receive antenna per client), the matrix in (28) becomes a vector that is normalized with respect to its norm. In one embodiment of the present invention, the pilot tones are selected uniformly across the range of OFDM tones. In another embodiment, the pilot tones are adaptively selected based on the CSI to minimize the interpolation error.

[0153] The inventors recognize that the key difference between the system and method of

[11] and the proposed system and method of this patent application is the objective function. In particular, the system of

[11] assumes multiple transmit antennas and a single client, and therefore the related method is designed to maximize the product of the precoding weights by the channel to maximize the received SNR for the client. However, this method does not work in a multi-client scenario because inter-client interference occurs due to interpolation errors. In contrast, the method of this application is designed to minimize inter-client interference, thus improving the error rate performance for all clients.

[0154] Figure 33 shows N FFT Figure 33 shows the matrix entries in (28) as a function of OFDM tone index for a DIDO2x2 system with β = 64 and L0 = 8. The channel PDP is generated according to the model in (21) with β = 1, and the channel consists of only 8 channel taps. It is noted that L0 should be chosen to be larger than the number of channel taps. The solid line in Figure 33 represents the ideal function, while the dotted line is the interpolated function. The interpolated weights fit the ideal weights of the pilot tones according to the definition in (28). The weights calculated over the remaining tones only approximate the ideal case due to estimation errors.

[0155] The method for performing weighted interpolation is to find the feasible set in (26). TIFF0007621401000076.tif54. To reduce the complexity of the search, we quantize the feasible set to P values ​​uniformly in the range [0, 2π]. Figure 34 shows the case where L0=8, M=N tFigure 1 shows SNR vs. SER for 2 transmit antennas and a variable P. As the number of quantization levels increases, the SER performance improves. It can be seen that the case P=10 approaches the performance of P=100, with much lower computational complexity due to the reduced number of searches.

[0156] Figure 35 shows the SER performance for different DIDO orders and L0 = 16 interpolation methods. The number of clients is assumed to be the same as the number of transmit antennas, and every client is equipped with a single antenna. As the number of clients increases, the SER performance degrades due to the increase in inter-client interference generated by weight interpolation errors.

[0157] In other embodiments of the present invention, weight interpolation functions other than that of (28) are used. For example, a linear predictive autoregressive model

[12] can be used to interpolate weights across different OFDM tones based on an estimate of the channel frequency correlation.

[0158] References

[0159] [1] U.S. Patent Application Serial No. 12 / 630,627, filed December 2, 2009, by A. Forenza and S.G. Perlman, entitled "System and Method for Distributed Antenna Wireless Communications."

[0160] [2] FCC, "Evaluating Compliance with FCC Guidelines for Human Exposure to Radio Frequency Electromagnetic Fields," OET Bulletin 65, 1997-01, August 1997.

[0161] [3] 3GPP, "Spatial Channel Model AHG (Ad Hoc Combination from 3GPP and 3GPP2)", SCM, Text V6.0, April 22, 2003.

[0162] [4] 3GPP TR 25.912: “Feasibility Study for Evolved UTRA and UTRAN”, V9.0.0 (2009-10).

[0163] [5] 3GPP TR 25.913: "Evolved UTRA (E-UTRA) and Evolved UTRAN (E-UTRAN) Requirements", V8.0.0 (2009-01).

[0164] [6] W.C. Jakes, "Microwave Mobile Communications," IEEE Press, 1974.

[0165] [7] K.K. Wong et al., "Joint Channel Orthogonalization for Multiuser MIMO Antenna Systems," IEEE Transactions on Wireless Communications, Vol. 2, pp. 773-786, July 2003.

[0166] [8] P. Viswanath et al., "Beam pulse forming with secondary user awareness of primary users using damped antennas," IEEE Transactions on Information Theory, Vol. 48, pp. 1277-1294, June 2002.

[0167] [9] AAM Saleh et al., "Statistical Model of Indoor Multipath Propagation," Transactions of the IEEE Society for Selected Areas in Communications, Vol. 195, SAC-5, No. 2, pp. 128-137, February 1987.

[0168]

[10] A. Paulraj et al., "Introduction to Space-Time Radio Communications," Cambridge University Press, 40 West 20th Street, New York, NY, USA, 2003.

[0169]

[11] J. Choi et al., "Interpolation-based transmit beampulse forming for feedback-limited MIMO-OFDM," IEEE Transactions on Signal Processing, Vol. 53, No. 11, pp. 4125-4135, November 2005.

[0170]

[12] I. Wong et al., "Long-Range Channel Prediction for Adaptive OFDM Systems," Proceedings of the IEEE 2006 Asilomar Conference on Signals, Systems, and Computers, Vol. 1, pp. 723-736, Pacific Grove, California, USA, November 7-10, 2004.

[0171]

[13] J.G. Proakis, "Communications Systems Engineering," Prentice Hall, 1994.

[0172]

[14] B.D. Van Veen et al., "Beam Pulse Forming: A Flexible Approach for Spatial Filtering Applications," IEEE ASSP Magazine, April 1988.

[0173]

[15] R.G. Vaughan, "On optimal coupling in mobile," IEEE Transactions on Vehicular Technology, Vol. 37, No. 4, pp. 181-188, November 1988.

[0174]

[16] F. Qian, "Partially Adaptive Beam Pulse Forming for Correlated Interference Rejection," IEEE Transactions on Signal Processing, Vol. 43, No. 2, pp. 506-515, February 19, 1995.

[0175]

[17] H. Krim et al., "Two Decades of Array Signal Processing Research," IEEE Signal Processing Journal, pp. 67-94, July 1996.

[0176]

[19] W.R. Remley, "Digital Beam Pulse Forming System," U.S. Patent No. 4,003,016, January 1977.

[0177]

[18] RJ Masak, "Beam Pulse Forming / Null Steering Adaptive Array," U.S. Patent No. 4,771,289, September 1988.

[0178]

[20] K.-B. Yu et al., "Adaptive Digital Beam Pulse Forming Architecture and Algorithm for Nulling Mainlobe and Multiple Sidelobe Radar Jammers While Maintaining Monopulse Ratio Angle Estimation Accuracy," U.S. Patent No. 5,600,326, February 19, 1997.

[0179]

[21] H. Boche et al., "Analysis of Different Precoding / Decoding Strategies for Multiuser Beampulse Forming," IEEE Vehicle Technology Conference, Vol. 1, April 2003.

[0180]

[22] M. Schubert et al., "Joint" Dirty Paper Precoding and Downlink Beam Pulse Forming," Vol. 2, pp. 536-540, December 2002.

[0181]

[23] H. Boche et al., "General Duality Theory of Uplink and Downlink Beam Pulse Forming," Vol. 1, pp. 87-91, December 2002.

[0182]

[24] K.K. Wong, R.D. Murch, and K.B. Letaief, "Joint Channel Orthogonalization for Multiuser MIMO Antenna Systems," IEEE Transactions on Wireless Communications, Vol. 2, pp. 773-786, July 2003.

[0183]

[25] Q. H. Pencer, A. L. Windlehurst, and M. Haardt, "Zero-Forcing Methods for Downlink Spatial Multiplexing in Multiuser MIMO Channels," IEEE Transactions on Signal Processing, Vol. 52, pp. 461-471, February 2004.

[0184] II. Disclosure of the Present Application A wireless radio frequency (RF) communication system and method is described below that uses multiple distributed transmit antennas that work in concert to create a wireless link with a given user while suppressing interference to other users. Coordination across different transmit antennas is enabled through user clustering. A user cluster is a subset of transmit antennas whose signals can be reliably detected by a given user (i.e., received signal strength greater than noise or interference levels). Every user in the system defines their own user cluster. Waveforms sent by transmit antennas in the same user cluster are coherently combined to create a point of zero RF interference at the target user's location with RF energy at every other user's location reachable by those antennas. Consider a system with M transmit antennas in one user cluster and K users reachable by those M antennas in TIFF0007621401000077.tif412. The transmitter calculates the CSI ( TIFF0007621401000078.tif415). For simplicity, we assume that every user is equipped with a single antenna, but the same method can be extended to multiple receive antennas per user. TIFF0007621401000079.tif2115 is the channel vector ( Consider the channel matrix H obtained by combining the precoding weights ( TIFF0007621401000081.tif518) under the following conditions: Calculate to satisfy TIFF0007621401000082.tif423, where: TIFF0007621401000083.tif44 is the effective channel matrix for user k obtained by removing the k-th row of matrix H, TIFF0007621401000084.tif46 is a vector with all zero entries.

[0185] In one embodiment, the wireless system is a DIDO system, and user clustering is used to create a wireless communication link with a target user while proactively canceling interference to any other users reachable by antennas in the user cluster. In U.S. patent application Ser. No. 12 / 630,627, a DIDO system is described that includes: · DIDO Client: A user terminal equipped with one or more antennas. · DIDO Distributed Antenna: Transceiver base stations that work cooperatively to transmit pre-coded data streams to multiple users, thus suppressing inter-user interference. DIDO Base Transceiver Station (BTS): A centralized processor that generates pre-coded waveforms for the DIDO distributed antennas. DIDO Base Station Network (BSN): A wired backhaul that connects BTSs to DIDO distributed antennas or to other BTSs. DIDO distributed antennas are grouped into different subsets based on their spatial distribution relative to the BTS or DIDO client locations. We define three types of clusters as shown in Figure 36. · Supercluster 3640: A set of DIDO distributed antennas connected to one or more BTSs such that the round trip latency between every BTS and each user is within the constraints of the DIDO precoding loop. · DIDO Cluster 3641: A set of DIDO distributed antennas connected to the same BTS. When a supercluster contains only one BTS, the definition coincides with a DIDO cluster. User Cluster 3642: A set of DIDO distributed antennas that cooperatively transmit pre-coded data to a given user.

[0186] For example, a BTS may be a local hub connected to other BTSs and DIDO distributed antennas through a BSN. The BSN may consist of a variety of network technologies, including but not limited to digital subscriber line (DSL), ADSL, VDSL [6], cable modem, fiber ring, T1 line, hybrid fiber-coaxial (HFC) network, and / or fixed wireless (e.g., WiFi). All BTSs in the same supercluster share information about DIDO precoding through the BSN, such that round-trip latency is within the DIDO precoding loop.

[0187] In Figure 37, dots represent DIDO distributed antennas, crosses represent users, and dashed lines represent user clusters for users U1 and U8, respectively. The method described below is designed to create a communication link with target user U1 while creating a point of zero RF energy for any other users (U2-U8) inside or outside the user cluster.

[0188] The inventors proposed a similar method in [5] in which points of zero RF energy are created to eliminate interference in overlapping zones between DIDO clusters. Extra antennas are needed to transmit signals to clients within the DIDO cluster while suppressing inter-cluster interference. In embodiments of the method proposed in this application, we do not attempt to eliminate DIDO inter-cluster interference, but rather assume that a cluster associates with a client (i.e., a user cluster) to ensure that no interference (or the interference is negligible) is generated for any other client within its vicinity.

[0189] One idea related to the proposed method is that due to large propagation losses, users far enough away from the user cluster are not affected by radiation from the transmitting antenna. Users close to or within the user cluster receive signals without interference due to precoding. Furthermore, the condition Further transmit antennas can be added to the user cluster (as shown in FIG. 37) so that TIFF0007621401000085.tif512 is satisfied.

[0190] An embodiment of the method using user clustering comprises the following steps. a. Link quality measurement: Report the link quality between every DIDO distributed antenna and every user to the BTS. Link quality consists of the signal-to-noise ratio (SNR) or signal-to-interference-and-noise ratio (SINR). In one embodiment, the DIDO distributed antenna transmits a training signal, and users estimate their received signal quality based on the training. The training signals are designed to be orthogonal in the time, frequency, or code domain so that users can distinguish across different transmitters. Alternatively, the DIDO antenna transmits a narrowband signal (i.e., a single tone) at one predetermined frequency (i.e., a beacon channel), and users estimate link quality based on the beacon signal. A threshold is defined as the minimum signal amplitude (or power) above the noise level to successfully demodulate data, as shown in Figure 38a. Any link quality metric value below this threshold is assumed to be zero. The link quality metric is quantized over a finite number of bits and fed back to the transmitter. In a different embodiment, training signals or beacons are sent from the users and link quality is estimated at the DIDO transmit antennas (as in Figure 38b) assuming a correlation between uplink (UL) and downlink (DL) path loss, which is a realistic assumption in time division duplex (TDD) and frequency division duplex (FDD) systems when the UL and DL frequency bands are relatively close (with UL and DL channels at the same frequency). Information about link quality metrics is shared across different BTSs through the BSN so that all BTSs are aware of the link quality between every antenna / user pair across different DIDO clusters, as shown in Figure 37. b. User Cluster Definition: The link quality metrics of all radio links in a DIDO cluster are input into a link quality matrix shared across all BTSs throughout the BSN. An example link quality matrix for the scenario of Figure 37 is shown in Figure 39. The link quality matrix is ​​used to define user clusters. For example, Figure 39 shows the selection of a user cluster for user U8. The subset of transmitters with non-zero link quality metrics (i.e., active transmitters) for user U8 is first identified. These transmitters populate the user cluster for user U8. Next, the submatrix containing non-zero inputs from transmitters in the user cluster to other users is selected. Note that because the link quality metric is only used to select the user cluster, it can be quantized with only 2 bits (i.e., to identify above- or below-threshold states in Figure 38), thus reducing feedback overhead.

[0191] Another example is shown in Figure 40 for user U1. In this case, the number of active transmitters is lower than the number of users in the submatrix, thus violating the condition. Therefore, one or more columns are added to the submatrix to satisfy the condition. If the number of transmitters is greater than the number of users, the extra antennas can be used for diversity schemes (i.e., antenna or eigenmode selection).

[0192] Yet another example is shown for user U4 in Figure 41. It is recognized that a submatrix can be obtained as a combination of two submatrices. c. CSI reporting to BTS: Once a user cluster is selected, CSI from all transmitters in the user cluster to every user reached by those transmitters is made available to all BTSs. CSI information is shared across all BTSs through the BSN. TDD systems can exploit UL / DL channel reciprocity and derive CSI in the UL channel from training. To reduce the amount of feedback, FDD systems require feedback channels from all users to the BTS. Only CSI corresponding to non-zero entries in the link quality matrix is ​​fed back. d. DIDO Precoding: Finally, DIDO precoding is applied to any CSI where sub-matrices serving different users are clustered (e.g., as described in the related U.S. patent application). In one embodiment, the effective channel matrix The singular value decomposition (SVD) of TIFF0007621401000086.tif54 is calculated, and the precoding weights of user k are TIFF0007621401000087.tif54 is Alternatively, if M>K and SVD is When decomposing the effective channel matrix as TIFF0007621401000089.tif626, the DIDO precoding weights for user k are given by TIFF0007621401000090.tif639 where, TIFF0007621401000091.tif54 is is a matrix whose columns are singular vectors of the null subspace of TIFF0007621401000092.tif53. From basic linear algebra considerations, the matrix The right singular vectors in the null subspace of TIFF0007621401000093.tif53 are equal to the eigenvectors of C corresponding to the zero eigenvalue. TIFF0007621401000094.tif675 where the effective channel matrix is ​​given by It can be seen that the image is decomposed as TIFF0007621401000095.tif618. Next, One alternative to computing the SVD of TIFF0007621401000096.tif55 is to compute the eigenvalue decomposition of C. There are several ways to compute the eigenvalue decomposition, such as the power method. Because we are only interested in the eigenvectors corresponding to the null subspace of C, we will use the inverse power method described by iteration. TIFF0007621401000097.tif1239Here, the vector (u i ) is a random vector. Considering that the eigenvalues ​​(λ) of the null subspace are known (i.e., zero), the inverse power method requires only one iteration to converge, thus reducing the computational complexity. Then, Write the precoding weight vector as TIFF0007621401000098.tif521, where: TIFF0007621401000099.tif54 is a vector with true input equal to 1 (i.e., the pre-coding weight vector is TIFF0007621401000100.tif56). The DIDO precoding computation requires a single matrix inversion. There are several numerical methods to reduce the complexity of the matrix inversion, such as Strassen's algorithm [1] or the Coppersmith-Winograd algorithm [2,3]. Since C is essentially a Hermitian matrix, an alternative solution is to decompose C in its real and imaginary parts and compute the real matrix inversion according to the method of [4, Section 11.4].

[0193] Another feature of the proposed method and system is its reconfigurability: as a client moves across different DIDO clusters in Figure 42, the user cluster follows its movement. In other words, the subset of transmit antennas is constantly updated and the effective channel matrix (and corresponding precoding weights) are recalculated as the client changes position.

[0194] The method proposed here works within superclusters in Figure 36 because the links between BTSs through the BSN must have low latency. To suppress interference in overlapping zones of different superclusters, the method of [5] can be used, which uses redundant antennas to create points of zero RF energy in the interference zones between DIDO clusters.

[0195] It should be noted that the terms "user" and "client" are used interchangeably herein.

[0196] References [1] S. Robinson, "Towards an Optimal Algorithm for Matrix Multiplication," SIAM News, Vol. 38, No. 9, November 2005.

[0197] [2] D. Coppersmith and S. Winograd, "Matrix Multiplication via Arithmetic Progressions," J. Symb. Comp., vol. 9, pp. 251-280, 1990.

[0198] [3] H. Cohn, R. Kleinberg, B. Szegedy, and C. Umans, "Set-Theoretic Algorithms for Matrix Multiplication," pp. 379–388, November 2005.

[0199] [4] W.H. Press, sa. Teukolsky, W.T. Etcherling, and B.P. Flannery, "Numerical Computing Recipes in C: Scientific Computing Techniques," Cambridge University Press, 1992.

[0200] [5] U.S. patent application Ser. No. 12 / 802,988, filed June 16, 2010, entitled "Interference Management, Handoff, Power Control, and Link Adaptation in Distributed Input Distributed Output (DIDO) Communication Systems," by A. Forenza and S.G. Perlman.

[0201] [6] Per-Erik Eriksson and Bjorn Odenhammar, "VDSL2: The Next Key Broadband Technology," Ericsson Review, No. 1, 2006.

[0202] Embodiments of the present invention may include various steps as described above. These steps may be implemented with machine-executable instructions that cause a general-purpose or special-purpose processor to perform certain steps. For example, various components within the base station / AP and the client devices described above may be implemented as software running on a general-purpose or special-purpose processor. Various well-known personal computer components, such as computer memory, hard drives, and input devices, have been omitted from the figures to avoid obscuring relevant aspects of the present invention.

[0203] Alternatively, in one embodiment, the various functional modules and associated steps illustrated herein may be performed by predefined hardware components including hardwired logic that performs the steps, such as an application specific integrated circuit ("ASIC"), or by any combination of programmed computer components and custom hardware components.

[0204] In one embodiment, certain modules, such as the encoding, modulation, and signal processing logic 903 described above, may be implemented on a programmable digital signal processor ("DSP") (or a group of DSPs), such as a DSP using Texas Instruments' TMS320x architecture (e.g., TMS320C6000, TMS320C5000...). The DSP in this embodiment may be embedded within an add-on card to a personal computer, such as a PCI card. Of course, a variety of different DSP architectures may be used while still adhering to the underlying principles of the present invention.

[0205] Various embodiments of the present invention may be provided as a machine-readable medium storing machine-executable instructions. The machine-readable medium may include, but is not limited to, flash memory, optical disks, CD-ROMs, DVD-ROMs, RAM, EPROMs, EEPROMs, magnetic or optical cards, or any other type of machine-readable medium suitable for storing electronic instructions. For example, the present invention may be downloaded as a computer program product that may be transferred as a data signal embodied in a carrier or other propagation medium over a communications link (e.g., a modem or network connection) from a remote computer (e.g., a server) to a requesting computer (e.g., a client).

[0206] Throughout the foregoing description, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the systems and methods may be practiced without some of these specific details. Accordingly, the scope and spirit of the present invention should be determined in terms of the appended claims.

[0207] Furthermore, throughout the foregoing description, numerous references have been cited to provide a more complete understanding of the present invention, all of which are incorporated herein by reference. [Explanation of symbols]

[0208] DIDO Distributed Input Distributed Output

Claims

1. 1. A method implemented in a multi-user multi-antenna system (MU-MAS), comprising: measuring the quality of wireless links between one or more users and a plurality of distributed wireless transceiver stations within a coverage area and determining link quality measurements accordingly; using the link quality measurements to select one or more subsets of distributed wireless transceiver stations for one or more users; measuring channel state information (CSI) between one or more users and a respective subset of the distributed wireless transceiver stations, and using the CSI to pre-code data transmissions such that links for the one or more users are interference-free; adjusting the selection of the distributed wireless transceiver stations within one or more subsets by estimating the speed of the users as they move within the coverage area, adaptively recalculating CSI and precoding data transmissions; A method comprising:

2. The method of claim 1 , wherein the pre-encoding is performed by QR decomposition.

3. The method of claim 1 , wherein the pre-coding is performed by singular value decomposition (SVD).

4. The method of claim 1 , wherein the pre-encoding is performed by an inverse power method.

5. A multi-user multi-antenna system (MU-MAS), comprising: Multiple users and a plurality of distributed wireless transceiver stations communicatively coupled to said plurality of users via a plurality of wireless links; a plurality of processing units communicatively coupled to the distributed wireless transceiver stations via a network; Including, The processing unit includes: measuring the quality of wireless links between one or more users and a plurality of distributed wireless transceiver stations within a coverage area, thereby determining link quality measurements; using the link quality measurements to select one or more subsets of distributed wireless transceiver stations for one or more users; measuring channel state information (CSI) between one or more users and a respective subset of the distributed wireless transceiver stations, and using said CSI to precode data transmissions such that links for the one or more users are interference-free; adjusting the selection of the distributed wireless transceiver stations within one or more subsets by estimating the speed of the users as they move within the coverage area, adaptively recalculating CSI, and precoding data transmissions; system.

6. The system of claim 5 , wherein the pre-encoding is performed by QR decomposition.

7. The system of claim 5 , wherein the pre-encoding is performed by singular value decomposition (SVD).

8. The system of claim 5 , wherein the pre-encoding is performed by inverse power law.

Images