HK1120342A

HK1120342A - Rf receiver and operation method thereof

Info

Publication number: HK1120342A
Application number: HK08108377.8A
Authority: HK
Inventors: 李俊强; 纳尔逊．R．索伦伯格; 约瑟夫．博库兹; 苌立枫
Original assignee: 美国博通公司
Priority date: 2006-09-21
Filing date: 2008-07-29
Publication date: 2009-03-27

Description

Radio frequency receiver and operation method thereof

Technical Field

The present invention relates to wireless communication systems, and more particularly, to equalization of data communications by a radio transceiver in a wireless communication system.

Background

Cellular wireless communication systems support wireless communication services in many regions of the world. Cellular wireless communication systems include a "network infrastructure" that communicates with wireless terminals within respective service areas. A typical network infrastructure includes a multitude of base stations distributed throughout a service area, each supporting wireless communications within a respective cell (or sector group). The base stations are coupled to Base Station Controllers (BSCs), each of which serves a number of base stations. Each BSC is connected to a Mobile Switching Center (MSC). Each BSC is also connected to the internet, either directly or indirectly.

In operation, each base station communicates with a number of wireless terminals within its serving cell/sector. A BSC coupled to the base station conveys voice information between the MSC and the serving base station. The MSC routes the voice communication to another MSC or to the PSTN. The BSC routes data communications between the serving base station and a packet data network that may include or be connected to the internet. Transmissions from a base station to a wireless terminal are referred to as "forward link" transmissions, while transmissions from a wireless terminal to a base station are referred to as "reverse link" transmissions. The amount of data transmitted on the forward link is often greater than the amount of data on the reverse link. This is because data users often request data from a data source, such as a network server, which provides the data to the wireless terminal.

The radio link between a base station and its serving wireless terminals often operates in accordance with one or more operating standards. These operating standards define the manner in which wireless links are allocated, installed, serviced, and removed. Cellular standards currently in widespread use include the global system for mobile communications (GSM) standard, the north american Code Division Multiple Access (CDMA) standard, and the north american Time Division Multiple Access (TDMA) standard, among others. These operating standards support both voice and data communications. The recently introduced standard is the Universal Mobile Telecommunication Service (UMTS)/Wideband CDMA (WCDMA) standard. The UMTS and WCDMA standards adopt the principles of CDMA and support large throughputs of voice and data.

The radio link between a base station and a served wireless terminal is called a "channel". The channel may distort or add noise to the wireless transmissions it serves. "channel equalization" is a processing step employed by a wireless receiver (e.g., a wireless terminal) to remove the effects of a channel. Although the channel equalization operation is certainly helpful in removing the influence of the channel, the characteristics of the channel are always changing. Therefore, the coefficients of the channel equalizer must be constantly updated. However, the generation of channel equalizer coefficients is a difficult and time consuming process. Accordingly, there is a need for improved methods of determining channel equalizer coefficients.

Disclosure of Invention

According to an aspect of the present invention, the present invention provides a method for operating a radio frequency receiver, including:

for a first diversity path processing a first time domain signal:

generating a first expected signal time domain channel estimation value, and converting the first expected signal time domain channel estimation value into a frequency domain to generate a first expected signal frequency domain channel estimation value;

generating a first interference signal time domain channel estimation value, and converting the first interference signal time domain channel estimation value into a frequency domain to generate a first interference signal frequency domain channel estimation value;

for a second diversity path processing a second time domain signal:

generating a second expected signal time domain channel estimation value, and converting the second expected signal time domain channel estimation value into a frequency domain to generate a second expected signal frequency domain channel estimation value;

generating a second interference signal time domain channel estimation value, and converting the second interference signal time domain channel estimation value into a frequency domain to generate a second interference signal frequency domain channel estimation value; and

for a second diversity path of the radio frequency receiver:

generating a first frequency domain equalizer coefficient and a second frequency domain equalizer coefficient according to the first expected signal frequency domain channel estimation value, the first interference signal frequency domain channel estimation value, the second expected signal frequency domain channel estimation value and the second interference signal frequency domain channel estimation value;

converting the first frequency domain equalizer coefficients to the time domain, producing first time domain equalizer coefficients;

converting the second frequency domain equalizer coefficients to the time domain, producing second time domain equalizer coefficients;

equalizing a first desired signal time-domain data symbol of the first time-domain signal using the first time-domain equalizer coefficients;

equalizing a second desired signal time domain data symbol of the second time domain signal using the second time domain equalizer coefficients.

Preferably, the generating the first desired signal frequency domain channel estimate further comprises:

performing a clustering path processing on a first expected signal time domain training symbol of the first time domain signal;

generating the first expected signal time domain channel estimate based on the first expected signal time domain training symbol after the clustering path processing;

performing a fast fourier transform on the first desired signal time domain channel estimate to generate the first desired signal frequency domain channel estimate; and is

Said generating a second desired signal frequency domain channel estimate further comprises:

performing clustering path processing on a second expected signal time domain training symbol of the second time domain signal;

generating a second expected signal time domain channel estimate based on the second expected signal time domain training symbols after the clustering path processing;

performing a fast Fourier transform on the second desired signal time domain channel estimate to produce the second desired signal frequency domain channel estimate.

Preferably, the generating the first interference signal frequency domain channel estimation value further comprises:

performing a clustering path processing on a first interfering signal time domain training symbol of the first time domain signal;

generating the first interference signal time domain channel estimation value based on the first interference signal time domain training symbol after the clustering path processing;

performing fast fourier transform on the first interfering signal time domain channel estimate to generate the first interfering signal frequency domain channel estimate; and is

The generating a second interfering signal frequency domain channel estimate further comprises:

performing clustering path processing on a second interference signal time domain training symbol of the second time domain signal;

generating the second interference signal time domain channel estimation value based on the second interference signal time domain training symbol after the clustering path processing;

and performing fast Fourier transform on the second interference signal time domain channel estimation value to generate the second interference signal frequency domain channel estimation value.

Preferably, said converting the first frequency domain equalizer coefficients to first time domain equalizer coefficients further comprises:

inverse fast Fourier transform the first frequency domain equalizer coefficients to generate the first time domain equalizer coefficients;

selectively ordering the first time domain equalizer coefficients; and is provided with

Said converting the second frequency domain equalizer coefficients to second time domain equalizer coefficients further comprises:

inverse fast Fourier transform the second frequency-domain equalizer coefficients to generate the second time-domain equalizer coefficients;

and selecting and ordering the second time domain equalizer coefficients.

Preferably, the method further comprises: combining the equalized first desired signal time domain data symbol and the second desired signal time domain data symbol to produce a composite equalized time domain data symbol.

Preferably, the generating the first frequency-domain equalizer coefficients and the second frequency-domain equalizer coefficients based on the first frequency-domain channel estimate and the second frequency-domain channel estimate further comprises: performing a minimum mean square error algorithm to generate the first and second frequency-domain equalizer coefficients.

Preferably, the first diversity path and the second diversity path receive different multipath versions of a single radio frequency transmitted time domain signal.

Preferably, the radio frequency receiver supports the following wireless operations: cellular wireless communication, wireless metropolitan area network communication, wireless local area network communication, wireless personal network communication.

According to another aspect of the present invention, the present invention provides a method for operating a radio frequency receiver, including:

receiving a time domain signal, wherein the time domain signal comprises a desired signal time domain training symbol and a data symbol and an interference signal time domain training symbol and a data symbol;

processing the desired signal time domain training symbols to produce desired signal time domain channel estimates;

processing the interference signal time domain training symbols to generate an interference signal time domain channel estimate;

converting the desired signal time domain channel estimate to the frequency domain to produce a desired signal frequency domain channel estimate;

converting the interference signal time domain channel estimation value into a frequency domain to generate an interference signal frequency domain channel estimation value;

generating a frequency domain equalizer coefficient based on the desired signal time domain channel estimate and an interfering signal time domain channel estimate;

converting the frequency domain equalizer coefficients to the time domain to produce time domain equalizer coefficients;

and carrying out equalization processing on the time domain data symbols of the expected signals by using the time domain equalizer coefficients.

Preferably, the processing the desired signal time domain training symbols to generate the desired signal time domain channel estimate further comprises:

performing clustering path processing on the expected signal time domain training symbol;

and generating the expected signal time domain channel estimation value based on the expected signal time domain training symbols after the clustering path processing.

Preferably, the processing the interfering signal time domain training symbols to generate the interfering signal time domain channel estimate further comprises:

performing clustering path processing on the interference signal time domain training symbol;

and generating the interference signal time domain channel estimation value based on the interference signal time domain training symbol processed by the clustering path.

Preferably, the generating the frequency domain equalizer coefficients based on the desired signal frequency domain channel estimate and the interfering signal frequency domain channel estimate further comprises: a minimum mean square error algorithm is performed to generate the frequency domain equalizer coefficients.

According to an aspect of the present invention, the present invention also provides a radio frequency receiver, including:

a radio frequency front end;

a baseband processing module coupled with the radio frequency front end, the baseband processing module further comprising:

a first diversity path for generating a first desired signal time domain channel estimate and transforming the first desired signal time domain channel estimate to the frequency domain to generate a first desired signal frequency domain channel estimate; generating a first interference signal time domain channel estimation value, and converting the first interference signal time domain channel estimation value into a frequency domain to generate a first interference signal frequency domain channel estimation value;

a second diversity path for generating a second desired signal time domain channel estimate and transforming the second desired signal time domain channel estimate to the frequency domain to generate a second desired signal frequency domain channel estimate; generating a second interference signal time domain channel estimation value, and converting the second interference signal time domain channel estimation value into a frequency domain to generate a second interference signal frequency domain channel estimation value;

the equalizer weight calculation module is used for generating a first frequency domain equalizer coefficient and a second frequency domain equalizer coefficient according to the first expected signal frequency domain channel estimation value, the first interference signal frequency domain channel estimation value, the second expected signal frequency domain channel estimation value and the second interference signal frequency domain channel estimation value;

the first diversity path further converts the first frequency domain equalizer coefficients to the time domain to produce first time domain equalizer coefficients and equalizes first desired signal time domain data symbols of the first time domain signal using the first time domain equalizer coefficients;

the second diversity path further converts the second frequency-domain equalizer coefficients to the time domain to produce second time-domain equalizer coefficients and equalizes a second desired signal time-domain data symbol of the second time-domain signal using the second time-domain equalizer coefficients.

Preferably, in generating the first desired signal time domain channel estimate and converting the first desired signal time domain channel estimate to the frequency domain to generate the first desired signal frequency domain channel estimate, the first diversity path is further configured to: performing a clustering path processing on a first expected signal time domain training symbol of the first time domain signal; generating the first expected signal time domain channel estimation value based on the first expected signal time domain training symbol after the clustering path processing; and performing fast fourier transform on the first desired signal time domain channel estimate to generate the first desired signal frequency domain channel estimate: and is

In generating the second desired signal time domain channel estimate and converting the second desired signal time domain channel estimate to the frequency domain to generate the second desired signal frequency domain channel estimate, the second diversity path is further for: performing a clustering path processing on a second desired signal time domain training symbol of the second time domain signal; generating a second expected signal time domain channel estimate based on the second expected signal time domain training symbols after the clustering path processing; and performing fast fourier transform on the second desired signal time domain channel estimate to generate the second desired signal frequency domain channel estimate.

Preferably, when generating the first interfering signal time domain channel estimate and converting the first interfering signal time domain channel estimate to the frequency domain to generate the first interfering signal frequency domain channel estimate, the first diversity path is further configured to: performing a clustering path processing on a first interfering signal time domain training symbol of the first time domain signal; generating the first interference signal time domain channel estimation value based on the first interference signal time domain training symbol after the clustering path processing; and performing fast fourier transform on the first interfering signal time domain channel estimate to generate the first interfering signal frequency domain channel estimate; and is

In generating the second interferer time-domain channel estimate and converting the second interferer time-domain channel estimate to the frequency domain to generate the second interferer frequency-domain channel estimate, the second diversity path is further to: performing a clustering path processing on a second interference signal time domain training symbol of the second time domain signal; generating the second interference signal time domain channel estimation value based on the second interference signal time domain training symbol after the clustering path processing; and performing fast Fourier transform on the second interference signal time domain channel estimation value to generate the second interference signal frequency domain channel estimation value.

Preferably, the radio frequency receiver further comprises:

a combiner that combines the equalized first desired signal time domain data symbol and second desired signal time domain data symbol to produce a composite equalized time domain data symbol.

Preferably, the equalizer weight calculation module further performs a minimum mean square error algorithm to generate the first and second frequency domain equalizer coefficients when generating the first and second frequency domain equalizer coefficients based on the first desired signal frequency domain channel estimate, the first interfering signal frequency domain channel estimate, the second desired signal frequency domain channel estimate, and the second interfering signal frequency domain channel estimate.

The first diversity path and the second diversity path receive different multipath versions of a single radio frequency transmitted time domain signal.

According to another aspect of the present invention, the present invention also provides a radio frequency receiver, including:

a radio frequency front end;

a baseband processing module coupled to the rf front end, the baseband processing module receiving a time domain signal including a desired signal time domain training symbol and a data symbol and an interfering signal time domain training symbol and a data symbol, and the baseband processing module further comprising:

at least one channel estimator for processing the desired signal time domain training symbols to produce desired signal time domain channel estimates and for processing the interfering signal time domain training symbols to produce interfering signal time domain channel estimates;

at least one fast fourier transformer for transforming the desired signal time domain channel estimate to the frequency domain to produce a desired signal frequency domain channel estimate and for transforming the interfering signal time domain channel estimate to the frequency domain to produce an interfering signal frequency domain channel estimate; (ii) a

A weight calculator for generating a frequency domain equalizer coefficient based on the desired signal time domain channel estimate and an interfering signal time domain channel estimate;

an inverse fast fourier transformer for transforming the frequency domain equalizer coefficients to the time domain to produce time domain equalizer coefficients;

an equalizer for equalizing the time domain data symbols using the time domain equalizer coefficients.

Preferably, the baseband processing module is further configured to perform a clustering path process on the desired signal time domain training symbols when processing the desired signal time domain training symbols to generate the desired signal time domain channel estimate; and generating the expected signal time domain channel estimation value based on the expected signal time domain training symbol after the clustering path processing; and is

The baseband processing module is further configured to perform a cluster path processing on the interfering signal time domain training symbols when processing the interfering signal time domain training symbols to generate the interfering signal time domain channel estimate; and generating the interference signal time domain channel estimation value based on the interference signal time domain training symbol processed by the clustering path.

Preferably, when converting the frequency domain equalizer coefficients into time domain equalizer coefficients, the baseband processing module is further configured to perform an inverse fast fourier transform on the frequency domain equalizer coefficients to generate the time domain and time domain equalizer coefficients, and to select and order the time domain equalizer coefficients.

Preferably, in generating the frequency domain equalizer coefficients based on the desired signal frequency domain channel estimate and the interfering signal frequency domain channel estimate, the baseband processing module is further configured to perform a minimum mean square error algorithm to generate the frequency domain equalizer coefficients.

Preferably, the radio frequency front end and the baseband processing module support the following wireless operations: cellular wireless communications, wireless metropolitan area network communications, wireless local area network communications, wireless personal network communications.

Other features and advantages of the present invention will be described in further detail in the following specification, taken in conjunction with the accompanying drawings and specific embodiments.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

fig. 1 is a partial schematic diagram of a cellular wireless communication system supporting wireless terminals operating in accordance with the present invention;

FIG. 2 is a functional block diagram of a wireless terminal constructed in accordance with the present invention;

FIG. 3 is a block diagram of a multi-radio frequency front end (receiver/transmitter) radio in accordance with one embodiment of the present invention;

FIG. 4 is a schematic diagram of the components of a baseband processing module according to an embodiment of the invention;

FIG. 5 is a schematic diagram of an equalization portion of a baseband processing module according to a first embodiment of the present invention;

FIG. 6 is a schematic diagram of an equalization portion of a baseband processing module according to a second embodiment of the present invention;

FIG. 7 is a flow chart of an equalization operation according to an embodiment of the present invention;

FIG. 8 is a flow chart of an equalization operation according to an embodiment of the present invention;

FIG. 9 is a schematic diagram of a composite channel model for performing equalization operations according to an embodiment of the present invention;

FIG. 10 is a schematic diagram of an equalization portion of a baseband processing module according to a third embodiment of the present invention;

fig. 11 is a flowchart of an equalizing operation according to the third embodiment of the present invention;

fig. 12 is a schematic diagram of an equalizing part of a baseband processing module according to a fourth embodiment of the present invention;

fig. 13 is a flowchart of an equalizing operation according to the fourth embodiment of the present invention.

Detailed Description

Fig. 1 is a system diagram of a portion of a cellular wireless communication system 100 that supports wireless terminals operating in accordance with the present invention. The cellular wireless communication system 100 includes a Public Switched Telephone Network (PSTN) interface 101 (e.g., a mobile switching center), a wireless network packet data network (N/W PDN) 102 including GPRS support nodes, EDGE support nodes, WCDMA support nodes, and other components, radio network controllers/base station controllers (RNC/BSC) 152 and 154, and base stations/nodes 103, 104, 105, and 106. The wireless network packet data network 102 is connected to additional private and public data networks 114 (e.g., the internet, wide area networks, local area networks, etc.). A conventional voice terminal 121 is connected to the PSTN 110. A voice over internet protocol (VoIP) terminal 123 and a personal computer 125 are connected to the internet/local area network 114. PSTN interface 101 is connected to PSTN 110. Of course, this particular configuration may vary from system to system.

Each base station/node 103-106 serves a cell/group of sectors within which wireless communications are supported. The wireless links, which include forward and reverse links, support wireless communications between the base stations and the wireless terminals that they serve. These wireless links support digital data communications, voIP communications, digital multimedia communications. The cellular radio communication system 100 is also backward compatible, supporting analog operation. The cellular wireless communication system 100 supports one or more of the UMTS/WCDMA standards, the GSM General Packet Radio Service (GPRS) extensions to the GSM, EDGE standards, one or more of the WCDMA standards, and/or various other CDMA standards, TDMA standards, and/or FDMA standards, among others. Wireless terminals 116, 118, 120, 122, 124, 126, 128 and 130 are connected to the cellular radio communication system 100 via radio links with base stations/node bs 103-106. As shown, the wireless terminals include cellular telephones 116 and 118, notebook computers 120 and 122, desktop computers 124 and 126, and data terminals 128 and 130. However, the cellular radio communication system 100 also supports other types of radio terminals. Generally, like notebook computers 120 and 122, desktop computers 124 and 126, data terminals 128 and 130, and cellular telephones 116 and 118, are capable of "surfing" the Internet 114 (a packet data network). Sending and receiving data communications, such as e-mail, sending and receiving files, and performing other data operations. Most of these data operations have significant download data rate requirements and much smaller upload data rate requirements. Some or all of the wireless terminals 116-130 are thus capable of supporting the EDGE operating standard, the GPRS standard, the UMTS/WCDMA standard, the HSDPA standard, the WCDMA standard, and/or the GSM standard. Further, some or all of wireless terminals 116-130 may be capable of performing the equalization operations of the present invention to support high-speed operation standards.

Fig. 2 is a functional block diagram of a wireless terminal constructed in accordance with the present invention. The wireless terminal includes a host processing component 202 and an associated radio frequency module 204. For a cellular telephone, the host processing components and the radio frequency module 204 are in one housing. In some cellular telephones, the host processing component 202 and some or all of the components of the radio frequency module 204 are formed on one Integrated Circuit (IC). For a pda host, a notebook computer host, and/or a pc host, the rf module 204 may be disposed on an expansion card or mother board so that they are in a different housing than the host processing component 202. The host processing component 202 includes at least one processing module 206, memory 208, radio frequency interface 210, input interface 212, output interface 214. The processing module 206 and memory 208 execute instructions to support the functions of the host terminal. For example, in a cellular telephone host device, the processing module 206 performs user interface operations and executes host software programs and the like.

The radio frequency interface 210 allows data to be received and transmitted from the radio frequency module 204. For data received from the rf module 204 (e.g., inbound data), the rf interface 210 provides the data to the processing module 206 for further processing and/or transmission to the output interface 214. Output interface 214 provides a connection to a display device, such as a display, monitor, speaker, etc., that displays the received data. Rf interface 210 also provides data from processing module 206 to rf module 204. The processing module 206 may receive outbound data from input devices such as a keyboard, keypad, microphone, etc., or generate data itself via the input interface 212. For data received via the input interface 212, the processing module 206 may perform a corresponding host function on the data and/or transmit it to the radio frequency module 204 via the radio frequency interface 210.

The rf module 204 includes a host interface 220, a baseband processing module (baseband processor) 222, an analog-to-digital converter 224, a filtering/gain module 226, a down conversion module 228, a low noise amplifier 230, a local oscillation module 232, a memory 234, a digital-to-analog converter 236, a filtering/gain module 238, an up conversion module 240, a power amplifier 242, an RX filtering module 264, a TX filtering module 258, a TX/RX switching module 260, and an antenna 248. Antenna 248 may be a single antenna shared by the transmit and receive paths (half-duplex) or include antennas in which the transmit path and receive path are separate (duplex). The implementation of the antenna will be determined by the particular standard in accordance with which the wireless communication device is to operate.

The baseband processing module 222, in combination with executable instructions stored in the memory 234, performs digital receiver functions and digital transmitter functions. Digital receiver functions include, but are not limited to, digital intermediate frequency to baseband conversion, demodulation, constellation inverse mapping, descrambling, and/or decoding. Digital transmitter functions include, but are not limited to, decoding, scrambling, constellation mapping, modulation, and/or conversion of a digital baseband to an intermediate frequency. The transmit and receive functions provided by the baseband processing module 222 may be implemented using a common device or separate processing devices. A processing device may include a microprocessor, microcontroller, digital signal processor, microcomputer, central processing unit, field programmable gate array, programmable logic device, state machine, logic circuitry, analog circuitry, digital circuitry, and/or any device that controls signals (analog and/or digital) based on operational instructions. The memory 234 may be a storage device or an aggregate of several storage devices. The storage device may be read-only memory, random access memory, volatile memory, non-volatile memory, solid-state memory, dynamic memory, flash memory, and/or any device that can store digital information. Note that when the baseband processing module 222 implements one or more of its functions via a state machine, analog circuitry, digital circuitry, and/or logic circuitry, the memory storing the corresponding operational instructions is embedded within the circuitry comprising the state machine, analog circuitry, digital circuitry, and/or logic circuitry.

In operation, the rf module 204 receives outbound data 250 from the host processing component through the host interface 220. The host interface 220 sends the outbound data 250 to the baseband processing module 222, which processes the outbound data 250 according to a particular wireless standard (e.g., UMTS/WCDMA, GSM, GPRS, EDGE, etc.) to generate digital transmission format data 252. The digital transmission format data 252 is a digital baseband signal or a digital low intermediate frequency signal having a frequency of a low intermediate frequency ranging from 0 to several kilohertz/megahertz.

Digital to analog converter 236 converts digital transmission format data 252 from the digital domain to the analog domain. The filter gain module 238 filters and/or adjusts the gain of the analog signal before providing it to the up-conversion module 240. The up-conversion module 240 directly converts the analog baseband or low if signal to a rf signal based on the local oscillation 254 provided by the local oscillation module 232. The power amplifier 242 amplifies the rf signal, generates an outbound rf signal 256 and filters this signal with a TX filtering module 258. The TX/RX switching module 260 receives the amplified and filtered signals from the TX filtering module 258 and provides an output radio frequency signal 256 to the antenna 248, and the antenna 248 transmits the outbound radio frequency signal 256 to a target device, such as the base stations 103-106.

The rf module 204 also receives an inbound rf signal 262 transmitted by the base station via the antenna 248, the TX/RX switching module 260, and the RX filtering module 264. The low noise amplifier 230 receives and amplifies the inbound radio frequency signal 262. The low noise amplifier 230 provides the amplified inbound rf signal to the down conversion module 228, which converts the amplified inbound rf signal to an inbound low frequency signal or baseband signal based on a receiver local oscillation 266 provided by the local oscillation module 232. The down conversion module 228 provides the inbound low frequency signal (or baseband signal) to the filtering/gain module 226, which filters and/or adjusts the gain of the signal before providing it to the analog-to-digital converter 224. Analog-to-digital converter 224 converts the filtered low frequency signal (or baseband signal) from the analog domain to the digital domain to produce data 268 in a digital receive format. The baseband processing module 222 demodulates, demaps, descrambles, and/or decodes the received formatted data 268 to recover the inbound data 270 according to a particular wireless communication standard performed by the radio frequency module 204. The host interface 220 provides the retrieved inbound data 270 to the host processing component 202 via the radio interface 210.

It is understood that all components of the rf module 204, including the baseband processing module 222 and the rf front-end components, may be formed on one integrated circuit. In other constructions, the baseband processing module 222 and the rf front-end components of the rf module 204 may be formed on different integrated circuits. Rf module 204 may be formed on the same integrated circuit as host processing component 202. In further embodiments, baseband processing module 222 and host processing component 202 may be formed on different integrated circuits. Therefore, all the components in fig. 2, except for the antenna, display, speaker, etc. and the keyboard, keypad, microphone, etc., can be formed on the same integrated circuit. Many different integrated circuit configurations may be suitable without departing from the teachings of the present invention. The baseband processing module 222 performs equalization processing on the digital transport format data (baseband TX signal) 252 in a new manner in accordance with the present invention. Various equalization operation techniques are further described in fig. 3-13.

Fig. 3 is a schematic diagram of a multi-radio front-end (receiver/transmitter) radio module 300 constructed in accordance with an embodiment of the invention. The rf module 300 includes a baseband processing module 222 and a plurality of rf front ends, including an rf front end 1302, an rf front end 2 304, an rf front end 3 306, and an rf front end N308. The rf front ends 302, 304, 306, and 308 are served by antennas 310, 312, 318, 316, respectively. The rf module 300 may serve several diversity paths of a single signal being propagated. Therefore, in a simple embodiment of diversity path, the rf module 300 includes a first rf front end 302, a second rf front end 304, and a baseband processing module 222. This embodiment will be further described with reference to fig. 5. Alternatively, multiple radio frequency front ends 302-308 may serve Multiple Input Multiple Output (MIMO) communications, with each front end 302-308 being assigned to a respective MIMO data path. MIMO communication is currently implemented in the form of WLAN, such as IEEE 802.11n. In each case, the principles of the present invention are also applicable to rf modules 300 having two or more rf front ends.

Fig. 4 is a schematic diagram of the components of a baseband processing module according to one embodiment of the present invention. The baseband processing module (baseband processor) 222 includes a processor 402, a memory interface 404, an on-board memory 406, an uplink/downlink interface 408, TX processing components 410, a TX interface 412. The baseband processing module 222 further includes an RX interface 414, a cell search module 416, a multipath scanning module 418, a rake receiver combiner 420, and a Turbo (Turbo) decoding module 422. The baseband processing module 222 is in some embodiments connected to an external memory 234. However, in other embodiments, the memory 406 meets the memory requirements of the baseband processing module 402.

As already described in fig. 2, the baseband processing module receives outbound data 250 from the connected host processing component 202 and provides inbound data 270 to the connected host processing component 202. In addition, the baseband processing module 222 provides transmit data (baseband TX signal) 252 in digital format to the associated rf front end. The baseband processing module 222 receives digital receive format data (baseband RX signals) 268 from an associated rf front end. As already described in fig. 2, an ADC 224 generates receive format data (baseband RX data) 268, while a DAC 236 in the rf front end receives digital transmit format data (baseband TX signal) 252 from the baseband processing module 222.

In fig. 4, the downlink/uplink interface 408 is capable of receiving outbound data 250 from a connected host processing component (e.g., host component 202 through host interface 220), according to a specific embodiment of the present invention. In addition, the downlink/uplink interface 408 is capable of providing inbound data 270 to the connected host processing component 202 through the host interface 220. TX processing component 410 and TX interface 412 are connected to the rf front end and downlink/uplink interface 408 as in fig. 2. The TX processing component 410 and the TX interface 412 can receive outbound data from the downlink/uplink interface 404, process the outbound data to generate a baseband TX signal 252 as shown in fig. 2, and output the baseband TX signal 252 to the radio frequency front end. The RX processing components include an RX interface 414, a finger receiver combiner 420, and in some embodiments the processor 402 is capable of receiving an RX baseband signal 268 from a radio frequency front end.

Equalization processing operations in a radio frequency receiver, in accordance with the present invention, may be implemented with one or more components of the baseband processing module 222. In a first configuration, the equalization operation is implemented with equalization operation 415a of processor 402. The equalization operation 415a may be implemented in software, hardware, or a combination of software and hardware. When equalization operation 415a is implemented in software instructions, processor 402 retrieves the instructions through memory interface 404 and executes the instructions to implement equalization operation 415a.

In a further configuration, a dedicated equalization component 415b is provided between RX interface 414 and modules 416, 418 and 420 and performs the equalization operations of the present invention. In this configuration, the equalization operation 415b may be implemented by hardware, software, or a combination of hardware and software. In another equalization operation configuration according to the present invention, the equalization operation 415c is implemented in the finger receiver combiner module 420 by an equalization operation 415 c. The equalization operation 415c may be implemented in hardware, software, or a combination of hardware and software to perform the equalization operations of the present invention.

As further shown in fig. 4, data 268 in a data reception format includes a plurality of signal paths. Each signal path may be received from a corresponding radio frequency front end as shown in fig. 3. Thus, the different forms 268 of each digital reception format data may be different multipath versions of a single received signal, or different multipath forms like different radio frequency signals in a MIMO system.

Fig. 5 is a schematic diagram of an equalizing component in a baseband processing module according to a first embodiment of the present invention. These components of the baseband processing module 222 perform the equalization operations of the present invention. Of course, the baseband processing module 222 may include other components in addition to those shown in fig. 5. The functional elements of fig. 5 may be implemented in dedicated hardware, general purpose hardware, software, or combinations thereof.

The components of the baseband processing module 222 in fig. 5 include a first diversity path component, a second diversity path component, and common components. As depicted in fig. 3, a radio frequency transceiver (transmitter/receiver) may include multiple receive signal paths. The multiple receive signal paths may include components that operate on different multipath versions of a transmitted signal or components that operate on signals containing different data. In the embodiment according to fig. 5, the functional components operate on different multipath versions of one radio frequency transmit time domain signal.

The first diversity path components include a Cluster Path Processor (CPP)/channel estimation module 504, a fourier transform (FFT) module 506, a multiplier 512, an inverse fourier transform (IFFT) module 514, a select ordering (tap ordering) module 516, and a time domain equalizer 518. The second diversity path components include a cluster path processor/channel estimation module 524, a fourier transform (FFT) module 526, a multiplier 530, an inverse fourier transform module 532, a selection ordering module 534, and a time domain equalizer 536. The common processing blocks of the radio receiver of fig. 5 include a Minimum Mean Square Error (MMSE) weight calculation block 510, a noise variance estimation block 502, and a combiner 538.

During its operation, the first diversity path processes the first time domain signal 502. The first time domain signal 502 includes first time domain training symbols and first time domain data symbols. In general, a frame of transmitted symbols in a radio frequency system typically includes a preamble portion having training symbols and a payload portion carrying data symbols. The training symbols are used by a channel estimation operation to generate equalizer coefficients, which are then used for equalization of the data symbols. CPP/channel estimation module 504 is capable of processing first time domain training symbols of first time domain signal 502 to thereby generate a first time domain channel estimate 508. The FFT section 506 can convert the first time domain channel estimate to the frequency domain to generate a first frequency domain channel estimate 508.

Similarly, the second diversity path receives a second time domain signal 522 that includes a second time domain training symbol and a second time domain data symbol. CPP/channel estimation module 524 processes second time domain training symbols of second time domain signal 502 to generate a second time domain channel estimate. FFT module 526 converts the second time domain channel estimate to the frequency domain to produce a second frequency domain channel estimate 528.

MMSE/weight calculation module 510 receives the noise variance estimation parameters from noise variance estimation module 502 and generates first frequency domain equalizer coefficients 511 and second frequency domain equalizer coefficients 513 based on first frequency domain channel estimate 508 and second frequency domain channel estimate 528.

Looking again at the first diversity path, multiplier 512 multiplies the output of FFT module 506 with first frequency domain equalizer coefficients 511. However, in other embodiments, multiplier 512 simply passes first frequency domain equalizer coefficients 511. The IFFT module then converts the first frequency domain equalizer coefficients 511 processed by the multiplier 512 to the time domain to produce first time domain equalizer coefficients. The select ordering module 516 then orders the first time domain equalizer coefficients, resulting in selectively ordered time domain equalizer coefficients to the time domain equalizer 518. The time domain equalizer 518 equalizes the first time domain data symbols using the first time domain equalizer coefficients obtained from the selective ordering module 516.

Looking again at the second diversity path, multiplier 530 multiplies the second frequency domain equalizer coefficients 513 with the output from FFT module 526. In other embodiments, multiplier 530 simply passes second frequency domain equalizer coefficients 513. The IFFT module 532 then converts its input from the frequency domain to the time domain to produce second time domain equalizer coefficients. The select ordering module 534 orders the second time-domain equalizer coefficients, producing an output to the time-domain equalizer. The time domain equalizer 536 equalizes the second time domain data symbol using the second time domain equalizer coefficients. Finally, a combiner 538 combines the equalized first time domain data symbol obtained from the first time domain equalizer 518 and the equalized second time domain data symbol obtained from the second time domain equalizer 536 to produce a composite time domain data symbol 540.

In accordance with another feature of baseband processing module 222 of fig. 5, CPP/channel estimation module 504 performs a clustered path processing on a first time domain training signal of first time domain signal 502. Clustered Path Processing (CPP) is an operation that processes multipath signal components that are relatively close in time. A complete description of how to perform the clustering path processing is given in U.S. patent application Ser. No. 11/173,854 (30/6/2005) with "METHOD AND SYSTEM FOR MANAGING, CONTROLLING, AND COMMINING SIGNALS IN A FREQUENCEY SELECTIVE MULTIPATH FADING CHANNEL", which is incorporated herein by reference in its entirety.

After completion of the cluster path processing operations, CPP/channel estimation module 504 generates a first time domain channel estimate based on the first time domain training symbols processed by the cluster path. Further, in the second diversity path, CPP/channel estimation module 522 performs a clustering path process on second time domain training symbols of second time domain signal 522. CPP/channel estimation module 524 then generates a second time-domain channel estimate based on the second time-domain training symbols after processing by the clustered paths.

In its operation, MMSE weight calculation module 510 performs an MMSE algorithm on first and second frequency-domain channel estimates 508, 528, resulting in first and second frequency-domain equalizer coefficients 511, 513. One implementation of these operations will be described below. Other operations than those described below may also be used to generate equalizer coefficients in accordance with the present invention.

In the particular implementation described herein, in the time domain, the matrix signal model for each antenna serving the dual diversity path structure of fig. 5 is represented by:

y _i ＝H _i x+n _i i＝1，2 (1)

channel matrix H _i It can be modeled with a circulant matrix that satisfies the following equation:

H ₁ ＝F ^-1 Λ ₁ F；H ₂ ＝F ^-1 Λ ₂ F (2)

where F is an orthogonal discrete fourier transform matrix.

The matrix F is multiplied on both sides of equation (1) to obtain a frequency domain channel model as shown in the following equation:

Y _i ＝Fy _i ＝Λ ⁱ X+N _i (3)

wherein X = Fx; n is a radical of hydrogen _i ＝Fn _i ，i＝1，2

The channel model may be represented on the kth subcarrier in the frequency domain as:

Y[k]＝Λ _k X[k]+N[k] (4)

wherein:

and is

Is a 2 x 1 vector.

Therefore, the MMSE optimal weight on the kth subcarrier is:

C[k]＝E(Y[k] ^* Y[k] ^T ) ^-1 E(Y[k] ^* X)＝(Λ _k ^* Λ _k ^T +C _nn ) ^-1 Λ _k (6)

therefore, the estimated transmitted signal is:

after simplified expression (7), the MMSE-FDE weight for the dual diversity path arrangement of fig. 5 is:

the equalized time domain signal is:

fig. 6 is a schematic diagram of an equalizing portion of a baseband processing module according to a second embodiment of the present invention. The components of the baseband processing module 222 receive the time domain signal 602 from the radio frequency front end shown in fig. 2. Time domain signal 602 includes time domain training symbols and time domain data symbols. The components of fig. 6 include a channel estimation module 604, an FFT module 606, a weight calculation module 610, an IFFT module 614, a selection ordering module 616, and a time domain equalizer 618. Channel estimation module 604 processes the time domain training symbols of time domain signal 602 to produce time domain channel estimate 603.FFT module 606 converts time domain channel estimate 603 to the frequency domain to produce frequency domain channel estimate 608. The weight calculation module 610 generates frequency domain equalizer coefficients based on the frequency domain channel estimate 608 and the noise mean square estimate from the noise variance estimation module 602. Multiplier 612 receives frequency domain equalizer coefficients 611 and receives input from FFT module 606. Multiplier 612 produces an output to IFFT module 614, IFFT module 614 converts frequency domain equalizer coefficients 611, which may have been modified by multiplier 612, to time domain equalizer coefficients. The selective ordering module 616 selectively orders the time-domain equalizer coefficients to generate selectively ordered time-domain equalizer coefficients for the time-domain equalizer 616. Time-domain equalizer 616 equalizes the time-domain data symbols of time-domain signal 602 with the time-domain equalizer coefficients to produce equalized time-domain symbols 640.

The channel estimation module 604 may perform the trunking path processing operations as described in fig. 5. When performing a clustered path processing operation to generate time domain training symbols, CPP/channel estimation module 604 may generate time domain channel estimates from the clustered path processed time domain training symbols. MMSE weight calculation module 610 may perform an MMSE algorithm on the frequency-domain channel estimates to generate frequency-domain equalizer coefficients.

Fig. 7 is a flow diagram of an equalization operation according to one embodiment of the present invention. The operational flow 700 begins with operation on at least two diversity paths (step 702). As already described in fig. 3, the rf module may include a plurality of rf front ends 302-308, each serving a corresponding one of the diversity paths. Therefore, in fig. 7, operation steps 704 through 708 are performed for each diversity path. In particular, for each diversity path, the baseband processing module receives a respective time domain signal comprising a time domain training symbol and a time domain data symbol.

For the first diversity path, the operations include receiving a first time domain signal including first time domain training symbols and first time domain data symbols. Next operations include processing the first time domain training symbols to generate a first time domain channel estimate (step 706). Further, operations include converting the time-domain channel estimate to the frequency domain to generate a first frequency-domain channel estimate (step 708).

For the second diversity path, the operations include receiving a second time domain signal including a second time domain training symbol and a second time domain data symbol (step 704). Next operations include processing the second time domain training symbol to produce a second time domain channel estimate (step 706). Further, the operations include converting the second time-domain channel estimate to the frequency domain to produce a second frequency-domain channel estimate (step 708).

After operations 702-708 have been completed for each diversity path, the process proceeds to step 710 where frequency domain equalizer coefficients are generated for each diversity path. For the particular example of fig. 5 where there are two diversity paths, the operation of step 710 includes generating first frequency-domain equalizer coefficients and second frequency-domain equalizer coefficients based on the first frequency-domain channel estimate and the second frequency-domain channel estimate. The frequency domain equalizer coefficients are then converted to time domain equalizer coefficients in step 712. For the example of a first diversity path and a second diversity path, the operation of step 712 may include converting the first frequency domain equalizer coefficients to the time domain to produce first time domain equalizer coefficients and converting the second frequency domain equalizer coefficients to the time domain to produce second time domain equalizer coefficients.

The operation then includes, for each diversity path, performing time-domain equalization of the respective time-domain data symbols (step 714). For the example with first and second diversity paths, the operation of step 714 includes equalizing the first time domain data symbol with first time domain equalizer coefficients and equalizing the second time domain data symbol with second time domain equalizer coefficients. Finally, in step 716, the operation includes combining the equalized time domain data symbols obtained from the multiple diversity paths. For the particular case with first and second diversity paths, the operation of step 716 includes combining the equalized first time domain data symbol and the equalized second time domain data symbol to produce a composite time domain data symbol.

The operations of steps 702-716 are repeated each time a new equalizer coefficient is generated based on the received physical layer frame including the training symbols. In most rf receivers, the operational flow 700 shown in fig. 7 is repeated for each received physical layer frame. However, in other embodiments, channel estimation is performed periodically based on detecting a change in channel conditions or when time constraints are met.

The operation of step 706 includes the cluster path processing already described previously. After the clustering process is performed, the time domain channel estimate comprises the time domain training symbols processed by the clustering paths. The transformation from the time domain to the frequency domain employs a fast fourier transform, and the transformation from the frequency domain to the time domain employs an inverse fast fourier transform. The operations of step 710 include generating frequency domain equalizer coefficients using an MMSE algorithm based on the received channel estimates. The operation of fig. 7 may support various types of systems including cellular wireless communication systems, wireless metropolitan area network system (e.g., wiMAX) standards, WLAN communication operations, WPAN communication operations.

Fig. 8 is a flow diagram of an equalization operation according to one embodiment of the invention. Operational flow 800 includes a first received time domain signal including time domain training symbols and time domain data symbols (step 802). Next, the time domain training symbols are processed to generate time domain channel estimates (step 804). The time domain channel estimate is then converted to the frequency domain to generate a time domain channel estimate (step 806).

The operational flow further includes generating time-domain equalizer coefficients based on the frequency-domain channel estimates generated in step 806 (step 808). The frequency domain equalizer coefficients are then converted to the time domain to produce time domain equalizer coefficients (step 810). In step 812, the time domain data symbols are equalized by using the time domain equalizer coefficients generated in step 810, and then the operation is ended. Of course, the operational flow 800 of fig. 8 may be repeated for each received physical layer frame containing training symbols and data symbols. Various particular embodiments previously described in connection with fig. 1-7 may use the operational flow 800 of fig. 8 and are not described further below.

Fig. 9 is a schematic diagram of a composite channel model on which an equalization operation is performed according to one embodiment of the present invention. The desired signal 902 and the main interfering signal 906 are operated on by the radio frequency signal according to the channel model shown in fig. 9. In the model of fig. 9, RX signal 914 received by the rf receiver is the sum of the desired signal processed by desired channel 904, the interfering signal 906 processed by interfering signal channel 908, and noise 910 (shown as adder 912). In accordance with the present invention, the resulting equalizer coefficients significantly/completely cancel the components of RX signal 914 introduced by interfering signal 906.

In one dominant interference scenario, the signal model at the k-subcarrier of RX signal 914 in the frequency domain is represented as:

Y[k]＝H _d [k]S+H _I [k]I+N (11)

the structure and method of canceling some/all of the interfering signals using an equalizer constructed in accordance with the present invention will be further described in fig. 10-13.

Fig. 10 is a schematic diagram of an equalizing part of a baseband processing module according to a third embodiment of the present invention. These components of the baseband processing module 222 perform equalization operations in accordance with the present invention. Of course, the baseband processing module 222 may include other components in addition to those shown in fig. 10. The functional blocks of fig. 10 may be implemented in dedicated hardware, general purpose hardware, software, or combinations thereof.

The components of the baseband processing module 222 of fig. 10 include a first diversity path component, a second diversity path component, a shared component. As depicted in fig. 3, a radio frequency transceiver (transmitter/receiver) may include multiple receive signal paths. The multiple receive signal paths include various components that process different multipath versions of a single transmit signal or process signals that include different data. In accordance with the embodiment shown in fig. 10, these functional components operate on different multipath versions of a single radio frequency transmit time domain signal.

The first diversity path components include a desired signal constellation path processor/channel estimation module 1004, an interfering signal constellation path processor/channel estimation module 1042, an FFT module 1006, an FFT module 1044, a multiplier 1012, an IFFT module 1014, a selection ordering module 1016, a time domain equalizer 1018. The second diversity path components include a desired signal constellation path processing/channel estimation module 1024, an interfering signal constellation path processing/channel estimation module 1048, an FFT module 1026, an FFT module 1050, a multiplier 1030, an IFFT module 1032, a selection ordering module 1034, and a time domain equalizer 1036.

The common processing module of the RF receiver of fig. 10 includes a joint Delay Locked Loop (DLL) 1056, a joint scrambling and code tracking interference module 1054, a Minimum Mean Square Error (MMSE) weight calculation module 1010, a noise variance estimation module 1002, and a combiner 1038. As will be further described below with respect to the joint delay locked loop 1056. In general, the joint delay locked loop 1056 controls and sets the sampling positions of the CPP/channel estimation modules 1004, 1042, 1024 with CPP operation. The joint scrambling and code tracking interference module 1054 provides scrambling and code tracking information to the interfering signal CPP/channel estimation modules 1042 and 1048.

In operation, the first diversity path operates on the first time domain signal 1002. The first time domain signal 1002 includes a desired signal time domain training symbol and data symbol and an interfering signal time domain training symbol and data symbol. As is known to those skilled in the art, a frame of transmitted symbols in a radio frequency system includes a preamble portion having training symbols and a payload portion carrying data symbols. The training symbols are used by the channel estimation operation to generate equalizer coefficients, which are then used in the equalization operation of the data symbols. Desired signal CPP/channel estimation module 1004 is capable of processing desired signal time domain training symbols of first time domain signal 1002 to produce a first desired signal time domain channel estimate.

The interfering signal CPP/channel estimation module 1042 is capable of processing the interfering signal time domain training symbols of the first time domain signal 1002 to generate a first interfering signal time domain channel estimate. In generating the respective channel estimate values, CPP/channel estimation modules 1004 and 1042 may receive energy estimate values for the desired signal and the interfering signal from desired signal energy estimation module 1056 and interfering signal energy estimation module 1054, respectively. The CPP/channel estimation modules 1004 and/or 1042 are capable of performing cluster path processing. The FFT module 1006 converts the first desired signal time domain channel estimate to the frequency domain to produce a first desired signal frequency domain channel estimate 1008. The FFT module 1044 converts the first interfering signal time domain channel estimate to the frequency domain to generate a first interfering signal frequency domain channel estimate 1046.

Similarly, the second diversity path operates on the second time domain signal 1022. Second time domain signal 1022 includes desired signal time domain training symbols and data symbols and interfering signal time domain training symbols and data symbols. Desired signal CPP/channel estimation module 1024 is capable of processing desired signal time domain training symbols of second time domain signal 1022 to produce a second desired signal time domain channel estimate. Interfering signal CPP/channel estimation module 1048 is capable of processing interfering signal time domain training symbols of second time domain signal 1022 to produce a second interfering signal time domain channel estimate. In generating the respective channel estimates, CPP/channel estimation modules 1024 and 1048 may receive energy estimates for the desired signal and the interfering signal from desired signal energy estimation module 1056 and interfering signal energy estimation module 1054, respectively. CPP/channel estimation modules 1024 and/or 1048 are capable of performing cluster path processing. The FFT module 1026 converts the second desired signal time domain channel estimate to the frequency domain to produce a second desired signal frequency domain channel estimate 1028. The FFT module 1050 converts the second interfering signal time domain channel estimate to the frequency domain to produce a second interfering signal frequency domain channel estimate 1052.

MMSE/weight calculation module 1010 receives the noise variance estimation parameters from noise variance estimation module 1002 and generates first frequency domain equalizer coefficients 1011 and second frequency domain equalizer coefficients 1013 based on first desired signal time domain channel estimate 1008, first interfering signal frequency domain channel estimate 1046, second desired signal frequency domain channel estimate 1028, and second interfering signal frequency domain channel estimate 1052.

Referring to the first diversity path, multiplier 1012 multiplies the output of FFT module 1006 by a first frequency domain equalizer coefficient 1011. However, in other embodiments, multiplier 1018 simply transmits first frequency domain equalizer coefficients 1011. The IFFT module 1014 then converts the first frequency domain equalizer coefficients 1011, which have been operated on by the multiplier 1012, to the time domain, producing first frequency domain equalizer coefficients. Next, the select ordering module 1016 orders the first frequency domain equalizer coefficients to generate time domain equalizer coefficients after selecting and ordering for the time domain equalizer 1018. The time domain equalizer 1018 performs equalization processing on the first time domain data symbols using the first time domain equalizer coefficients obtained from the selection ordering module 1016.

Referring again to the second diversity path, multiplier 1030 multiplies the output from FFT module 1026 by second frequency-domain equalizer coefficients 1013. In other embodiments, multiplier module 1030 transmits only the second frequency-domain equalizer coefficients 1013.IFFT module 1032 converts its input from the frequency domain to the time domain, generating second time domain equalizer coefficients. The selection ordering module 1034 performs selection ordering on the second time domain equalizer coefficients to generate and output the second time domain equalizer coefficients to the time domain equalizer. The time domain equalizer 1036 equalizes the second time domain data symbol with the second time domain equalizing equalizer coefficient. Finally, a combiner 1038 combines the equalized first time-domain data symbol obtained from the first time-domain equalizer 1018 with the second equalized time-domain data symbol obtained from the time-domain equalizer 1036 to produce a composite time-domain data symbol 1040.

Fig. 11 is a flowchart of an equalizing operation according to the third embodiment of the present invention. The operation 1100 begins with operation for at least two diversity paths (step 1102). As already described in fig. 3, the rf module may include a plurality of rf front ends 302-308, each of which serves a diversity path. Thus, referring back to FIG. 11, operations 1104-1114 are performed for each diversity path. Specifically, for each diversity path, the baseband processing module receives a respective time domain signal that includes a desired signal time domain training symbol, a desired signal time domain data symbol, an interfering signal time domain training symbol, and an interfering signal time domain data symbol.

For the first diversity path, the performing operation includes receiving a first time domain signal. The first diversity path then estimates the energy of the desired signal and at least one interfering signal present in the first time domain signal (step 1106.) next operations include processing the first interfering signal (dominant interfering) time domain training symbols to produce a first interfering signal time domain channel estimate (step 1108). Further, the operations include converting the first interfering signal time domain channel estimate to the frequency domain to produce a first interfering signal frequency domain channel estimate (step 1110). Operations then include processing the first desired signal time domain training symbols to produce first desired signal time domain channel estimates (step 1112). Next, the operations include converting the first desired signal time domain channel estimate to the frequency domain to produce a first desired signal frequency domain channel estimate (step 1114).

For the second diversity path, the operations performed include receiving a second time domain signal. The second diversity path then estimates the energy of the desired signal and at least one interfering signal present in the second time domain signal (step 1106.) the next operation includes processing the second interfering signal (dominant interfering) time domain training symbols to produce a second interfering signal time domain channel estimate (step 1108). Further, the operations include converting the second interfering signal time domain channel estimate to the frequency domain to produce a second interfering signal frequency domain channel estimate (step 1110). Operations then include processing the second desired signal time domain training symbols to produce second desired signal time domain channel estimates (step 1112). Next, the operations include converting the second desired signal time domain channel estimate to the frequency domain to produce a second desired signal frequency domain channel estimate (step 1114).

After performing operations 1104-1114 for each diversity path, the process proceeds to step 1116, where frequency domain equalizer coefficients are generated for each diversity path. For the particular embodiment shown in fig. 10 that includes two diversity paths, the operation of step 1116 includes generating first and second frequency-domain equalizer coefficients based on the first and second frequency-domain channel estimates. The next operations include converting the frequency domain equalizer coefficients to time domain equalizer coefficients (step 1118). For the special case of the first and second diversity paths, the operations of step 1118 include converting the first frequency domain equalizer coefficients to the time domain to produce first time domain equalizer coefficients and converting the second frequency domain equalizer coefficients to the time domain to produce second time domain equalizer coefficients.

The next operation includes, for each diversity path, time-domain averaging the respective time-domain data symbols (step 1120). For a particular example of the first and second diversity paths, the operation of step 1120 includes equalizing the first time domain data symbol with first time domain equalizer coefficients and equalizing the second time domain data symbol with second time domain equalizer coefficients. The final operation includes combining equalized time domain data symbols resulting from the multiple diversity paths (step 1122). For the particular case of the first and second diversity paths, the operation of step 1122 includes combining the equalized first time domain data symbol and the equalized second time domain data symbol to produce a composite time domain data symbol.

The operation steps 1102-1122 are repeated after each generation of a new equalizer coefficient based on the received physical layer frame including the training symbols. In many RF receivers, the operations 1100 of fig. 11 are repeated for each received physical layer frame. However, in further embodiments, channel estimation may be performed periodically based on detected changes in channel conditions or when time constraints are met.

The operations of steps 1108 and 1112 include the cluster path processing already described previously. After the cluster path processing is performed, the time domain channel estimation value comprises the time domain training symbol processed by the cluster path. The conversion from the time domain to the frequency domain is by a fast fourier transform and the conversion from the frequency domain to the time domain is by an inverse fast fourier transform. The operations of step 1116 may use an MMSE algorithm to generate frequency domain equalizer coefficients based on the received channel estimates. The operations in fig. 11 may support various types of systems including cellular wireless communication systems, wireless metropolitan area communication system standards (e.g., wiMAX), WLAN communication operations, WPAN communication operations.

Fig. 12 is a schematic diagram of an equalizing part of a baseband processing module according to a fourth embodiment of the present invention. The components of the baseband processing module 222 receive the time domain signal 1202 from the radio frequency front end shown in fig. 2. Time domain signal 1202 includes desired signal time domain training symbols and data symbols and interfering signal time domain training symbols and data symbols. The components shown in fig. 12 include a desired signal channel estimation module 1204, an interfering signal channel estimation module 1242, an FFT module 1206, an FFT module 1244, a weight calculation module 1210, an IFFT module 1214, a selection ordering module 1216, and a time domain equalizer 1218. Channel estimation module 1204 is configured to process the desired signal time domain training symbols of time domain signal 1202 to generate time domain channel estimate 1203. Interfering signal channel estimation module 1204 is configured to process interfering signal time domain training symbols of time domain signal 1202 to generate a time domain channel estimate. An FFT module 1206 converts the desired signal time domain channel estimate 1203 to the frequency domain to produce a desired signal frequency domain channel estimate 1208.FFT module 1244 converts the interfering signal time domain channel estimate to the frequency domain to produce an interfering signal frequency domain channel estimate 1246. The weight calculation module 1210 generates frequency domain equalizer coefficients 1212 from the desired signal frequency domain channel estimate 1208, the interfering signal frequency domain channel estimate 1246, and the noise variance estimate from the noise variance estimation module 1202. Multiplier 1012 receives frequency domain equalizer coefficients 1212 and the resulting input from FFT module 1206. The multipliers 1012 produce outputs to IFFT module 1214 that convert the frequency domain equalizer coefficients 1212 (which may have been modified by multipliers 1012) to the time domain, producing time domain equalizer coefficients. The selection ordering module 1216 selectively orders the time-domain equalizer coefficients to generate selectively ordered time-domain equalizer coefficients for the time-domain equalizer 1218. Time domain equalizer 1218 equalizes the time domain data symbols of time domain signal 1202 using the time domain equalizer coefficients, producing equalized time domain symbols 1240.

Channel estimation module 1204 and/or 1242 may perform the cluster path processing operations as described in fig. 5. When performing clustered path processing to generate time domain training symbols, CPP/channel estimation modules 1204 and 1242 may generate time domain channel estimates based on the clustered path processed time domain training symbols. MMSE weight calculation module 1210 may perform an MMSE algorithm on the channel estimates to generate frequency domain equalizer coefficients.

Fig. 13 is a flowchart of an equalizing operation according to the fourth embodiment of the present invention. Operation 1300 comprises first receiving a time domain signal comprising desired signal time domain training symbols and data symbols and interfering signal time domain training symbols and data symbols (step 1302). Next, the energy of the desired signal and the at least one interfering signal present in the time domain signal is estimated (step 1304). Next, the time domain training symbols are processed to generate an interference signal time domain channel estimate (step 1306). Next, the interfering signal time domain channel estimate is converted to the frequency domain to generate an interfering signal frequency domain channel estimate (step 1308). Next, the time domain training symbols are processed to produce a desired signal time domain channel estimate (step 1312). Next, the desired signal time domain channel estimate is converted to the frequency domain to produce a desired signal frequency domain channel estimate (step 1314).

Then, the operations further include generating frequency domain equalizer coefficients based on the frequency domain channel estimates generated at steps 1308 and 1314 (step 1316). The frequency domain equalizer coefficients are then converted to the time domain to produce time domain equalizer coefficients (step 1318). In the final end step, the time domain data symbols are equalized using the time domain equalizer coefficients generated in step 1318 (step 1320). Of course, the operational flow 1300 of fig. 13 may be repeatedly performed for each received physical layer frame including training symbols and data symbols.

The operational flow shown in fig. 11 and 13 and the corresponding structure of fig. 10 and 12 can be implemented by executing the following formulas and methods. In particular, the modules 1010 and 1210 and the performed steps 1116 and 1316 may be implemented according to the following. The MMSE optimal weight at each subcarrier is calculated as:

W[k]＝E(Y[k] ^* Y[k] ^T ) ^-1 E(Y[k] ^* S)＝(H _d [k] ^* H _d [k] ^T +H _I [k] ^* H _I [k] ^T +C _nn ) ^-1 H _d [k] (12)

then, time domain equalizer coefficients are obtained using an IFFT operation. The interference suppression capability is represented by the following formula:

more specifically, there are:

ignoring the index k, equation (12) can be written as:

by definition:

the equation can be simplified as:

to simplify the direct inversion of the 2 × 2 matrix, a simplified weight calculation method is as follows:

definition ofThe FDE-IS weight at each subcarrier IS:

wherein:

in equation (12), it is assumed that the transmission power of the dominant interference is equal to the transmission power of the desired signal, and the CPICH power allocation of the dominant interference is the same as the power allocation of the desired signal. These estimates all affect the operation of the system of the present invention. The present invention therefore solves the problem of robustness of the impact due to drift in the transmission power and CPICH power allocation estimates for the dominant interference. By defining the offset factors γ and ζ as the estimated offsets of the transmission power and CPICH power allocations, respectively, equation (12) can be written as:

W[k]＝E(Y[k] ^* Y[k] ^T ) ^-1 E(Y[k] ^* S)

＝(H _d [k] ^* H _d [k] ^T +γζ ² H _I [k] ^* H _I [k] ^T +C _nn ) ^-1 H _d [k] (16)

generally, by setting at any time P _I ＝P _S If =4.0, then γ ≦ 1, where there is an estimation bias in either WCDMA compression mode or HSDPA discontinuous transmission mode; estimate E of assumed interference _C A CPICH/I equals-10 dB and an estimated deviation is within +/-4dB, then 0.4 < ζ ² Is less than 2.5. Ignoring the subscript index k, equation (16) can be written as:

by definition:

then there are:

definition ofThe weight of the FDE-IS at each subcarrier can be represented by:

Y ₁ ＝H _d1 S+H _I1 I+N ₁ ；Y ₂ ＝H _d2 S+H _I2 I+N ₂

wherein:

in the high signal-to-noise ratio range, there are:

the timing of the CPP processing operations of the present invention (e.g., modules 1004, 1042, 1024, and 1048) is also important, for example, the alignment of the CPP operations with the desired and interfering signals is important. Assume that the scan module 418 of the baseband processing module 222 provides a time reference to the CPP/channel estimation modules 1004, 1024, 1042, 1048, approximately NTc and MTc, respectively, for the desired signal and the interfering signal. With these assumptions, the signal model in the time domain can be represented by:

y(t)＝h _d (t-τ _d )s(t)+h _i (t-τ _i )i(t)+n(t)

＝h _d (t-p1)s(t-NTc)+h _i (t-p2)i(t-MTc)+n(t) (18)

suppose that:

t-p1＝T _s τ _d ＝NTc+p1；τ _i ＝MTc+p2

equation (18) can be written as:

y(t)＝h _d (t-τ _d )s(t)+h _i (t-τ _i )i(t)+n(t)

＝h _d (Ts)s(Ts+p1-NTc)+h _i (Ts+p1-p2)i((Ts+p1-MTc)+n((Ts+p1)

with these assumptions, since we only compensate for the total energy of the desired signal and its channel response corresponding to the sampling locations of both the desired signal and the jammer signal, the optimum time associated with the desired maximum energy output can be obtained by using a maximum energy delay phase locked loop for CPP operation. So that the sampling position of the interference signal depends on the desired signal DLL. By using the on-time sampling information from the desired DLL, the channel responses of the desired signal and the interfering signal at the same sampling location can be obtained.

Further, by adjusting the scrambling code N chips (N chip), a channel estimation value of the desired signal at the sampling phase P1 can be obtained. Similarly, when the interference signal CPP is processed, the scrambling code M chips (M chip) is adjusted, and the interference signal channel estimation value at the same sampling phase P1 can be obtained. Furthermore, considering that the DLL can track the primary path that can be shifted by 7 chips (chip) during the scan update period, and the desired sum interference can be shifted in different directions, an additional DLL must be added to lock on the interfering primary path. This additional DLL provides only the interfering SC code phase. For example, the interfering SC is shifted by one chip, i.e., m → m +1, since the sampling point depends on the desired DLL. The channel profile of the interference is shifted by a total of 1 chip. Assuming that an IIR filter is used to measure power, the previous and adjacent pointer (L + 1) must be used as the historical value for the current pointer L.

As one of ordinary skill in the art will appreciate, the phrase "communicatively coupled" as used herein includes wired and wireless, direct connections, and indirect connections via other components, elements, or modules. As will also be appreciated by one of ordinary skill in the art, a presumptive connection (i.e., presuming that one component is connected to another component) includes both wireless and wired, direct and indirect connections between the two components in the same manner as a "communicative connection".

The present invention demonstrates the specific functionality and relationship thereof through the use of method steps. The scope and order of the method steps have been arbitrarily defined for convenience of description. Other boundaries and sequences may be applicable so long as the specified functions and sequences are performed. Any such stated or selected limits or sequences therefore fall within the scope and spirit of the invention.

The invention has also been described with the aid of functional blocks illustrating some of the important functions. The boundaries of the functional blocks and the relationships of the various functional blocks have been arbitrarily defined for the convenience of the description. Other boundaries or relationships may be applicable so long as the specified functions are performed. Such other limits or relationships are therefore within the scope and spirit of the invention.

One of ordinary skill in the art will also recognize that the functional blocks and other illustrative blocks and components in the present application can be implemented as discrete components, application specific integrated circuits, processors executing appropriate software, and any combination of the foregoing.

Furthermore, although the present invention has been described in terms of several embodiments, it will be understood by those skilled in the art that the present invention is not limited to these embodiments, and various changes or equivalent substitutions may be made in these features and embodiments without departing from the spirit and scope of the invention. The scope of the invention is only limited by the claims of the present application.

Claims

1. A method of operating a radio frequency receiver, comprising:

for a first diversity path processing a first time domain signal:

for a second diversity path processing a second time domain signal:

generating a second expected signal time domain channel estimation value, converting the second expected signal time domain channel estimation value into a frequency domain, and generating a second expected signal frequency domain channel estimation value;

for a second diversity path of the radio frequency receiver:

2. The method of claim 1, wherein the generating a first desired signal frequency domain channel estimate further comprises:

performing a clustering path processing on a second desired signal time domain training symbol of the second time domain signal;

3. The method of claim 1, wherein the generating the first interfering signal frequency domain channel estimate further comprises:

performing a clustering path processing on a first interference signal time domain training symbol of the first time domain signal;

4. A method of operating a radio frequency receiver, comprising:

receiving a time domain signal, wherein the time domain signal comprises a desired signal time domain training symbol and a data symbol, and an interference signal time domain training symbol and a data symbol;

5. The method of claim 4, wherein the processing the desired signal time domain training symbols to generate the desired signal time domain channel estimate further comprises:

6. A radio frequency receiver, comprising:

a radio frequency front end;

a first diversity path for generating a first desired signal time-domain channel estimate and transforming the first desired signal time-domain channel estimate to the frequency domain to generate a first desired signal frequency-domain channel estimate; generating a first interference signal time domain channel estimation value, and converting the first interference signal time domain channel estimation value into a frequency domain to generate a first interference signal frequency domain channel estimation value;

a second diversity path for generating a second desired signal time-domain channel estimate and transforming the second desired signal time-domain channel estimate to the frequency domain to generate a second desired signal frequency-domain channel estimate; generating a second interference signal time domain channel estimation value, and converting the second interference signal time domain channel estimation value into a frequency domain to generate a second interference signal frequency domain channel estimation value;

the first diversity path further converts the first frequency domain equalizer coefficients to the time domain to produce first time domain equalizer coefficients and equalizes a first desired signal time domain data symbol of the first time domain signal using the first time domain equalizer coefficients;

7. The RF receiver of claim 6, wherein in generating the first desired signal time domain channel estimate and converting the first desired signal time domain channel estimate to the frequency domain to generate the first desired signal frequency domain channel estimate, the first diversity path is further configured to: performing a clustering path processing on a first expected signal time domain training symbol of the first time domain signal; generating the first expected signal time domain channel estimation value based on the first expected signal time domain training symbol after the clustering path processing; and performing fast fourier transform on the first desired signal time domain channel estimate to generate the first desired signal frequency domain channel estimate: and is

8. The RF receiver of claim 6, wherein in generating the first jammer time domain channel estimate and converting the first jammer time domain channel estimate to the frequency domain to generate the first jammer frequency domain channel estimate, the first diversity path is further configured to: performing a clustering path processing on a first interfering signal time domain training symbol of the first time domain signal; generating the first interference signal time domain channel estimation value based on the first interference signal time domain training symbol processed by the clustering path; and performing fast fourier transform on the first interfering signal time domain channel estimate to generate the first interfering signal frequency domain channel estimate; and is

In generating the second interfering signal time domain channel estimate and converting the second interfering signal time domain channel estimate to the frequency domain to generate the second interfering signal frequency domain channel estimate, the second diversity path is further to: performing clustering path processing on a second interference signal time domain training symbol of the second time domain signal; generating a second interference signal time domain channel estimation value based on the second interference signal time domain training symbol processed by the clustering path; and carrying out fast Fourier transform on the second interference signal time domain channel estimation value to generate the second interference signal frequency domain channel estimation value.

9. A radio frequency receiver, comprising:

a radio frequency front end;

A weight calculator for generating frequency domain equalizer coefficients based on the desired signal time domain channel estimate and an interfering signal time domain channel estimate;

10. The rf receiver of claim 9, wherein the baseband processing module is further configured to perform a constellation path processing on the desired signal time domain training symbols when processing the desired signal time domain training symbols to generate the desired signal time domain channel estimate; generating the expected signal time domain channel estimation value based on the expected signal time domain training symbol after the clustering path processing; and is