MXPA00005715A - Receiver with parallel correlator for spread spectrum digital transmission - Google Patents

Abstract

A receiver of a handset transceiver, in a wireless telephone system having a plurality of wireless handset and a base unit having a base transceiver. Each handset has a handset transceiver for establishing a time-division multiple access (TDMA) link over a shared channel with the base unit via the base transceiver, in which each handset communicates during an exclusive time slot of a TDMA scheme that allocates time slots to active handsets for receiving a spread spectrum signal comprising successive chips representing successive symbols. Each receiver has one or more demodulation loops for demodulating the received spread spectrum signal, wherein each demodulation loop is characterized by one or more demodulation loop parameters. A parallel correlator of the receiver detects a peak bin and provides the peak bin and two adjacent bins. An error estimator of the receiver adjusts the demodulation parameters in accordance with the peak bin and the two adjacent bins to optimize demodulation of the spread spectrum signal.

Description

RECEIVER WITH PARALLEL CORRELATION FOR EXTENDED SPECTRUM DIGITAL TRANSMISSION BACKGROUND OF THE INVENTION Field of the Invention The present invention relates to the acquisition or reception of data, and in particular, to the acquisition of data in a multi-line wireless telephone system multiplexed in time division, of spread spectrum.

Description of the Related Art The transmission of digital data from a transmitter to a receiver requires a variety of digital signal processing techniques, to allow data to be transmitted by the transmitter and recovered or successfully acquired by the receiver. In digital wireless telephone systems, for example, a cordless telephone headset unit (wireless) communicates by means of digital radio signals with a base unit, which is normally connected by a conventional telephone line to an external telephone network . Each handset and base comprise a transceiver, which has a transmitter and a receiver. In this system, a user can use the wireless headset to engage in a telephone call with another user through the base unit and the telephone network.

Multi-line wireless telephone systems are being used in different situations, such as business with many phone users. These systems employ a base unit that communicates with up to N real-time headphones, typically with digital communication schemes, such as extended-spectrum time division multiplexing (TDM) schemes, such as time division multiple access (TDMA). In an extended spectrum system, bandwidth resources are traded for operating gains, according to the so-called Shannon theory. The advantages of an extended spectrum system include low power spectral density, better rejection of narrowband interference, integrated selective addressing capability (with code selection), and inherent channel multiple access capability. The extended spectrum systems employ a variety of techniques, including direct sequencing (DS), frequency hopping (FH), chirp systems, and hybrid DS / FH systems. In a TDMA system, a single radiofrequency channel is used, and each headset transmits and receives audio data packets, as well as non-audio data packets, during dedicated slices of time, or time slots, within of a global TDMA cycle or era. Other communication schemes include frequency division multiple access (FDMA), code division multiple access (CDMA), and combinations of these schemes. Different modulation schemes are used, such as amplitude / phase modulation without carrier (CAP), and quadrature amplitude modulation (QAM). The digital data is normally transmitted as modulated signals on a transmission medium, such as the radio frequency channel, in the form of binary data bits. (Other transmission media used frequently for digital communications include asymmetric digital subscriber cycle (ADSL) systems, or cable modem systems). The digital data is often modulated and transmitted in a complex digital data form, wherein the transmitted data comprises symbols from which the original data can be reconstructed by the receiver. The complex digital symbol data usually comprises real data (in phase, or "I"), and imaginary data (in quadrature, or "Q"), (pairs, I, Q). Each symbol of a pair I, Q can be a number of multiple bits, and represents a location of a constellation, mapped against a quadrant. Each symbol is mapped or mapped to a prescribed coordinate in a constellation of four quadrant grid type, using a look-up table (eg, a ROM). A prescribed number of symbols occupy the assigned areas in each quadrant, depending on the coding scheme. Depending on the number of bits / symbols of a given coding scheme, each quadrant of the constellation contains a number of symbols at prescribed coordinates with respect to the quadrature axes I and Q. For example, in the QPSK coding scheme, each sample It has 1 to 4 phase positions, one for each quadrant, such that each pair of symbols represents two data bits. To transmit a given input data value in a complex data system, the input data value to be transmitted is mapped to a pair of symbols or to a pair of I, Q coordinates of a corresponding constellation point on a complex signal constellation, having the real and imaginary axes I and Q. These symbols I, Q which represent the original data values, are then transmitted as part of the data packets by means of a modulated channel. A receiver can retrieve the I, Q pairs and determine the location in the constellation from them, and perform a reverse mapping to provide the original input data value or a close approximation thereof. In an extended spectrum system, each symbol is transmitted by a string of "sub-symbols" or "chips", which are usually derived by multiplying the symbol (which can be a 1 or -1, in some schemes) by a binary string of a pseudo-random number (PN) of a certain length (number of C chips). Accordingly, these systems are characterized by a chip speed, which is related to the symbol rate. Extended spectrum systems, in general, can also be used to transmit any digital data, whether in a complex format or not, and whether in a TDMA system or not. In an extended spectrum system, a signal represents successive symbols, by means of successive symbol chips. A received signal is displayed to provide samples. In this way, the samples represent a signal, which itself represents chips, which represent symbols. The receiver side of a transceiver samples a signal received with an analog-to-digital converter (ADC), which provides representative samples of the signal, which in turn represents symbols. The transmitter side of a transceiver converts the symbols into an analog samples that constitute a signal, with a digital-to-analog converter (DAC). As noted above, the transmission of digital data requires a variety of digital signal processing techniques to allow data to be transmitted by the transmitter (eg, the transceiver of the base unit), and successfully retrieved by the receiver (for example, the receiver of a given handset transceiver). For example, the receiver side of a data transmission in an extended spectrum digital wireless telephone system employs a variety of functions to recover data from a transmitted radio frequency signal. These functions may include: time recovery for symbol synchronization, carrier recovery (frequency demodulation), and gain. The receiver thus includes, among other things, an automatic gain control (AGC) cycle, a carrier tracking cycle (CTL), and a time cycle for each link. Time recovery is the process by which the receiver's clock (time base) is synchronized with the transmitter's clock. This allows the received signal to be sampled at the optimum point in time, to reduce the opportunity for a splice error associated with the processing directed to the decision of received symbol values. In some receivers, the received signal is sampled at a multitude of transmitter symbol rates. For example, some receivers sample the received signal at twice the symbol speed of the transmitter. In any case, the receiver's sampling clock must be synchronized with the symbol clock of the transmitter. Carrier recovery is the process by which a received radiofrequency signal, after changing in frequency to a lower intermediate pass band, is changed in frequency to the baseband, to allow the retrieval of information from the modulation baseband.

An AGC tracks the strength of the signal, and adjusts the gain, for example, to help compensate for the effects of the alterations of the transmission channel on the received signal. The AGC, along with the other equalization techniques, can help remove intersymbol interference (ISI) caused by alterations in the transmission channel. The ISI causes the value of a symbol given by the values of the previous and next symbols to be distorted. Therefore, the AGC is important, because multiple headsets and / or base stations in close proximity can interfere with each other, and therefore, system transceivers must use a minimum gain necessary to avoid saturation of the driving system. interference, and also to use battery power more efficiently. These and related functions, and the related modulation schemes and systems are discussed in more detail in Edward A. Lee and David G. Messerschmitt, Digi tal Copimunication, 2nd edition (Boston: Kluwer Academic Publishers, 1994). The receivers require a relatively stable source of a sampling clock signal, which can also be controlled, in such a way that it can be secured to the symbol clock of the transmitter. For this function, voltage-controlled crystal oscillators (VCXO) have been used, because the clock signal produced by a VCXO is stable but controllable over a relatively narrow range, to allow the symbol clock of the transmitter to be secured. Other types of time recovery systems, such as the "Timing Recovery System for a Digital Signal Processor" (Time Recovery System for a Digital Signal Processor), described in the European Patent Application Number EP- can also be used. 0,793,363, European filing date on February 20, 1997, Thomson Consumer Electronics, Inc., inventors Knutson, Ramaswamy, and McNeely (Knutson et al.). In a multi-line extended spectrum wireless telephone system, as in all extended spectrum systems, it is important that each transceiver in the system can accurately receive the transmitted signals, and in particular, sample at the appropriate frequency and phase, to improve the reception and recovery of the signal. In an extended-spectrum TDMA system, it is also important that each transceiver be able to detect a valid data signal, and also detect the guardband indicating the end of the transmission of the data packet. International Application Number WO 96/14697 (Aeronet), published on May 17, 1996, discloses a correlation system for use in wireless direct sequence extended spectrum systems. U.S. Patent No. 4,587,662 (Langewellpott), issued May 6, 1986, discloses a TDMA spread spectrum receiver with coherent detection, to use indirect path signals in a receiver for transceiver stations fixed and mobile of a TDMA extended-spectrum digital radio system. At the same time, the signal forming the image of the multipath profile is continuously regenerated by a regeneration circuit. The regeneration circuit provides a coherent phase signal used to demodulate the received signal. The European Patent Application Number 0,505,771 Al (Hughes Aircraft) published September 30, 1992, discloses a communication satellite system having an increased power output density per unit of bandwidth in an extended-spectrum signal transmission.

COMPENDIUM A receiver of a headset transceiver, in a wireless telephone system having a plurality of wireless headsets and a base unit having a base transceiver. Each headset has a headset transceiver to establish a time division multiple access (TDMA) link over a channel shared with the base unit by means of the base transceiver, where each headset communicates during a time slot exclusive to a TDMA scheme that allocates time slots to the headset during a time slot exclusive to a TDMA scheme that allocates time slots to activate the headset to receive a spectrum signal extended comprising successive chips representing successive symbols. Each receiver has one or more demodulation cycles to demodulate the received spread spectrum signal, wherein each demodulation cycle is characterized by one or more demodulation cycle parameters. A parallel correlator of the receiver detects a peak bin, and provides the bin peak and two adjacent bins. A receiver error estimator adjusts the demodulation parameters according to the bin peak and the two adjacent bins, to optimize the demodulation of the spread spectrum signal.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram of a multi-line TDMA multi-line cordless telephone system, in accordance with one embodiment of the present invention. Figure 2 is a schematic representation of the TDMA slot structure used in the TDMA scheme of the system of Figure 1, in accordance with one embodiment of the present invention. Figure 3 is a block diagram illustrating a parallel correlator of a transceiver receiver of the system of Figure 1 in greater detail, in accordance with one embodiment of the present invention. Figure 4 is a flow chart illustrating the method of operation of the receiver of Figure 3, in accordance with one embodiment of the present invention. Figure 5 is a block diagram illustrating the receiver of Figure 3 in greater detail, in accordance with one embodiment of the present invention. Figure 6 illustrates exemplary time phase diagrams showing the correlation peak sampling against the time error.

DESCRIPTION OF THE PREFERRED MODE As described in more detail below, the present invention employs a demodulator architecture in the receiver of the base unit and the handset of the wireless telephone system, which employs a parallel correlator to generate symbols and indexes for the error estimators of the demodulator architecture. The error estimators use the general data from the parallel correlator, to track / adjust the time, the carrier phase shift, and the AGC cycles, in order to optimize or improve the reception and recovery (acquisition) or demodulation of the signal. The present invention also uses the data generated from the parallel correlator, in order to detect a valid data signal, and also to detect the guardband indicating the end of the transmission of the data packet. Referring now to Figure 1, a block diagram of the multi-line wireless telephone system TDMA 100 is shown, in accordance with one embodiment of the present invention. The TDMA system 100 comprises a base unit 110, which has the receiver and transmitter units 112 and 111, respectively, and is coupled to the external telephone network 116 via the telephone lines 115. The base unit 110 has a memory device or storage (not shown), such as a RAM or a hard disk drive, for storing data. The system 100 also comprises N wireless headphones 120 ^ 1202, • ..120N. Each does not have a transmitter and receiver unit (transceiver), such as transmitter 121 and receiver 122 of headset 120 !. At any given time, some number (or none) of the headphones are operating or off-hook (ie, in the process of conducting a telephone call). Accordingly, the system 100 provides a wireless network between the base station 110 and each earphone 12 (l <i> N). In one embodiment, system 100 comprises 4 headphones 120. ^ 1204, all of which may be active in a simultaneous manner. In another embodiment, the system 100 comprises a different number of headphones, for example N = 12, of which up to 8 may be active or operational at the same time. In one embodiment, the present invention comprises an extended spectrum TDMA system for connecting multiple transceivers to a base station on a single radiofrequency channel. In particular, the system 100 employs a TDMA extended-spectrum digital scheme, which allows energy to be used efficiently, because each operating headset is "off" (ie, it is not transmitting or receiving data, and therefore, it is not using as much battery power) during most portions of the TDMA era, and is only "on" during its own slices or time slots. In a modality, a headset is turned off by turning off the power to at least its CPU and transceiver units (receiver and transmitter), while remaining energized only on one clock and the associated timer or sequence controller circuit, enough to arouse the CPU in a previously determined time slot. In the present invention, the receiver of each transceiver of the base unit 110 and each handset 12Ü employs the demodulator architecture of the present invention to improve signal acquisition. In general, the adjustment of the demodulation parameters, such as time, carrier phase shift, and AGC cycles, according to the data generated from the parallel correlator (such as the bin peak and the adjacent bins) to optimize or improve the reception and recovery (acquisition) of signals, it can be referred to herein as to optimize signal demodulation. Referring now to Figure 2, a schematic representation of the TDMA slot structure 200 used in the TDMA scheme 200 of the system of Figure 1 is shown, in accordance with one embodiment of the present invention. System 100 employs a TDMA era that has structure 200, which is illustrated by assuming 12 total headphones 1201-12012 of which 8 can be active or operational at the same time, for example headphones 1201-1208. The TDMA 200 time structure comprises a number of rows and columns. Each row of the structure TDMA 200 represents a field of 2 milliseconds of digital data, and is par or non, and is grouped into a pair with a row or field non or pair, respectively. The TDMA 200 epoch structure is a time of 48 milliseconds. In a normal mode of operation, each field comprises 9 total packets: a data packet (not audio) in the first column (either transmitted from the base or from an earphone), and 8 audio packets, grouped into four pairs of two. Each pair of audio packets in a row includes a base audio transmission packet (time slot) (to a given handset from the base unit 110), and an audio transmission packet from the handset (from the given handset) to the base). Each type of package contains different sub-fields or sections. For example, a data packet comprises a 32-bit synchronization field, a data field, a FEC field (forward error correction), and guard time or band / silence barrier of approximately 5 milliseconds. The data in a data packet is used to communicate between the base unit and a particular handset, and contains different types of information, such as caller identification type information, rank and power information, and the like. An audio package comprises an audio packet header, FEC data section, and guard time. An audio packet header, for example, contains information that identifies the audio packet (such as the handset), the current location at the time, and the like. Accordingly, within each era, a pair of data packet slots, and several pairs of audio packet slots, are assigned to each handset. The data packet slots are used to establish a "data link" with each respective handset, and the audio packet slots are used to establish an "audio link" with each respective handset. The data links together constitute the system's data channel, while the audio links constitute the audio channel of the system. In other words, the first column of the TDMA 200 vintage structure corresponds to the data channel (data links), and the remaining columns correspond to the audio channel (audio links). The data link for a given handset is used to transmit, by means of data packets, non-audio data, generally referred to herein as signaling information. Each data packet is a set of data transmitted either to a given handset from the base unit or vice versa, during a separate time slot during which time no other handset receives or transmits data on the system data channel. These data packets may contain different types of data, such as data or synchronization words with time stamp information transmitted to a headset in sleep mode, caller identification information, incoming call information, being the telephone number that is dialed by the handset, and the like. The signaling information transmitted by the data link is used to establish calls, inform the receiver of incoming calls, maintain communication links between the headphones and the base, and the like. Voice data, i.e., audio packets containing audio data for a real-time telephone conversation, are transmitted over the audio link for a given handset. The bandwidth for the audio link for a handset is much higher than the bandwidth for the data link. This is because, within each era, a pair of data packet slots and several pairs of audio packet slots are assigned to each handset. For example, for N = 12 total headphones 1201-12012 with up to 8 off-hooks at a time, there are 12 pairs of audio packets per headset per time, comparing with a pair of data packets per headphone per time, for a bandwidth of audio channel (or link) twelve times greater than the bandwidth of the data channel (or link). The audio packages contain digitized (and possibly compressed) voice information. Therefore, for example, the pair in row 0 comprises an even row and a non row. In the even row, the base transmits data in the first time slot (slot 251), to one of the twelve headphones, for example the headset 120 !. There are a couple of rows at time 200 for each handset, so that each handset can receive and transmit data to the base unit 110, once per time. After the first data slot 251, assuming the handset 1201 # is operational (off hook) an audio packet is transmitted to the 120-L headset in the audio pack slot 253, then an audio packet is transmitted through the headset 120-_ to the base unit 110 in the audio pack slot 154, and so on for three of the other headphones to the end of the field or the row. In the non row for the pair of rows 0, the data slot 252 is used to receive the data transmitted from the earphone 120 to the base unit 110, and the audio packets for the remaining active headphones are transmitted. In the pairs of rows 1-11, the same sequence occurs, except that the data packets are to and from different headphones than to the pair of rows 0. In one mode of operation, each handset receives 16 samples of ADPCM (modulation in adaptive differential pulse code) of 4 bits during each time slice of the time allocated for the handset to receive audio data; and transmits to the base unit 16 samples of ADPCM during each time slice of the time allocated for the handset, to transmit audio data. In another mode of operation, the number of samples can be doubled up to 32 per slice of time, lowering the quality of each sample to two-bit samples. ADPCM and related technical issues are described in detail in International Telecommunication Union (ITU), Recommendation G.727, 12/1990), "5-, 4-, 3- and 2-Bits Sample Embedded Adaptive Differential Pulse Code Modulation (ADPCM ), "http: // ww. and you . ch. The present invention uses independent automatic gain control (AGC), carrier tracking cycles (CTL), and time cycles for the receiver side of each link, in order to independently track the demodulation parameters and the states associated with these three cycles or blocks. The states and parameters associated with these three cycles, and which are adjusted by error estimators, using indices generated by the parallel correlator, to improve or allow synchronization or acquisition of the received signal, are referred to herein, in general. , as demodulation parameters. In a preferred embodiment, the present invention is implemented in a digital system where the state of the three cycles or blocks of the demodulator (AGC, CTL, and time) exists in the digital domain. In an alternative mode, some analog domains can be used. Referring now to Figure 3, a block diagram illustrating a parallel correlator 300 of a transceiver receiver of system 100 of Figure 1 is shown in greater detail, in accordance with one embodiment of the present invention. Referring now to Figure 5, a block diagram illustrating a receiver 500 comprising the parallel correlator 300 of Figure 3 is shown, in accordance with one embodiment of the present invention. The receiver 500 may be a receiver 122 of a receiver or receiver 112 of the base unit 110, and further comprises the radio frequency circuit 510 for receiving the radio frequency signal, in accordance with the feedback of the AGC from the DAC 541 and the cycle of AGC 533, and to provide a near baseband signal to the ADC 521. The ADC 521 samples the near baseband signal using a fixed clock that is applied to it, for example at a 2x chip rate. As will be appreciated, time error estimation is usually employed for clock recovery in communication systems. Conventional time recovery methods include a feedback control system for estimating the time error, based on the input signal, and filter the error and drive a VCXO to adjust the phase of the locally generated clock. For example, decision time error estimation is sometimes used, using non-decision-driven techniques, such as Gardner's algorithm, "A BPSK / QPSK Timing-Error Detector for Sampled Receivers," F.M. Gardner, IEEE Trans. on Comm. , May 1986, pages 423-429. Decision-oriented techniques, such as the Müller and Mueller algorithm are also sometimes used, "Timing Recovery in Digital Synchronous Data Receivers," K.H. Mueller and M. Müller, IEEE Trans. on Comm. May 1976, pages 516-530. However, in the present invention, the error estimators use the data generated from the parallel correlator to track / adjust the time, the carrier phase shift, and the AGC cycles, in order to improve reception and recovery of the signal. In particular, the interpolation / time recovery 522 adjusts the sampling phase for symbol synchronization according to the time feedback from time cycle 532; and the 523 derailor, which shifts the signal and changes the frequency of the signal to baseband according to the carrier feedback, from the carrier or CTL cycle 531. Accordingly, the derailor 523 presents a corrected phase signal to the parallel correlator 300. The parallel correlator 300 provides peak symbol data and correlation to the error estimators 524, as described in more detail below, based on which, the error estimators 524 adjust the demodulation parameters so that the control cycles 531, 532, 533 improve the acquisition of the signal. Time cycle 532 is used to establish sampling synchronization at the receiver, such that sampling is presented at the appropriate time. Referring once again to Figure 3, the parallel correlator 300 in one embodiment comprises an array of 32 I-correlators, 3100-31031, and an array of 32 Q-correlators 3200-32031, each of which produces a pair I , Q, such as I0, Q0. 32 correlators are used for each of I and Q, because a sampling rate of 2x the chip rate is used, and there are 16 chips / symbol (ie, the length of the "extension code" or the PN sequence). is 16). Therefore, each symbol corresponds to 16 chips or 32 samples. In general, the number of correlators is equal to the length of the extension code per number of samples per chip. Therefore, 64 total correlators are shown in Figure 3. If the complex number extension code is used, only 32 complex correlators are required in total, because each chip repeats once to generate 32 sample chips for 16 PN chips, since there are two samples per chip. The present invention updates the symbol time, the offset of the carrier, and the AGC at the sampling rate with the error signals generated at the symbol rate. As will be appreciated, in the alternative embodiments, different numbers of PN chips and samples other than 16, such as 15 or 17, may be employed. Each I 310 correlator! comprises the multiplier node 311i (the integration and emptying block 312i # and the accumulation block (average) 313 ^ Each correlator Q 32ÜÍ comprises similar components Each product or multiplier node 31 ^ or 321¿ receives, as an input, the Relevant sample (I ± or Qi respectively), and the first binary digit of the sequence PN, PNi The pairs I, Q, are given to the adders of absolute magnitude 3400-34031, each of which produces an output from "bin", bin0-bin 31. The bins are provided to reliably search and decide block 345, which provides an index to the multiplexer (MUX) 346, and indicates valid, binL, and binR to the error estimators 524. As will be appreciated, the parallel correlator 300 performs the "correlation" function, by which, a "correlation peak" is detected, which allows the receiver 500 to know the beginning and the end of each chip sequence, in such a way that you can derive symbols ap Artir of the same. For example, if the receiving system 300 is properly synchronized, and is receiving a valid signal, I2, Q2 / may be the valid symbol pair representing the current symbol being received. In this case, bin0, bin-L, bin3, ... bin31 will all have a very low value, and bin2 will have a relatively high value or "peak". This is because I.¡_ and Qi t in this example, ideally have a value of +1, for i =. { ?, l, 3, ... 3l} , while I2 and Q2 have a value of + 32, due to the effect of the multiplication with an appropriately aligned PN sequence. As will be appreciated, the absolute magnitude is employed by the adders 340 ^ because the received signals may be rotating, and may have different phases, because complex I, Q pairs are used. The adders 340 ^ therefore measure the power of the complex signal. Therefore, a correlation peak can be detected by the search block 345, comparing bin0-bin31, even when the received signal has some phase rotation.

As illustrated, the MUX 346 receives all 31 I, Q pairs as one input, and selects one of these Isim QSim pairs to be the acquired symbol, according to the index signal provided by the search block 345. In one mode, the search block 345 determines the index, determining which bin signal is the peak, that is, relatively higher by comparing with its adjacent neighbors. For example, in the previous example, bin2 is a peak, and therefore, index 2, such that ISim'Qsim = I2 / Q2- The symbol stream at the output of parallel correlator 300 is applied to a system FEC (not shown). As will be understood, the radiofrequency signal received from a remote transmitter is asynchronous to the receiver, and therefore, may not be sampled by the radio frequency circuit 510, the ADC 521, and the interpolation / time recovery 522 of the receiver 500 exactly at the appropriate point of the radio frequency signal. However, because the sampling rate is 2x the chip rate, in general the search block 345 will detect some peak, even when the relative contrast between the peak and its neighboring bins would be higher if a better sampling synchronization was achieved . In the worst case, two neighboring bins will have equal peaks. In this case, one can be selected arbitrarily, and the demodulation parameters can be adjusted in accordance with the same, as described below.

The search block 345 also provides the error estimators 524 with the valid signals, binL, and binR. These signals are the same as binp, binp.lf and binp + 1, respectively, where "p" represents the peak bin. Accordingly, the bin signal for the peak bin is transmitted, along with its immediate adjacent neighboring bin signals "left" and "right". These signals provided by the parallel correlator 300 are used by the error estimators 524 to develop the error signals used to adjust the respective demodulation parameters that control the cycles 531, 532, 533, in order to optimize the demodulation of the signal . The error estimators 524 may be a functionality implemented by a processor or specialized hardware, as will be appreciated. Accordingly, the error estimators 524 analyze the adjacent bins (binL, binR) in relation to the sampled correlation peak, to adjust the time cycle 532, in order to optimize the time of the sample phase. Additionally, error estimators 524 analyze the peak bin to adjust the CTL 531 cycle, in order to optimize derotation. The error estimators 524 also analyze the power of the peak bin to adjust the AGC 533 cycle, in order to optimize the gain of the radio frequency circuit 510. These changes optimize the correlation to increase the robustness of the spread spectrum communications.

With respect to the optimized sample phase time, if the demodulation parameters are optimally adjusted, the difference between the binp peak and its adjacent bins (binL and binR) will be maximized, and binL and binR will each differ from the binp by approximately the same quantity. In other words, the peak binp is presented in the correlation peak sampled when the time is optimally adjusted. Referring now to Figure 6, exemplary time phase diagrams 600 are shown which show the correlation of the peak sampling against the time error. In diagram 610 the correct or optimal time phase is illustrated, in diagram 620 the advanced time phase is illustrated, and in diagram 630 the delayed time phase is illustrated. Each diagram illustrates the de-strain correlation of the received signal, together with its peak value and its immediate left and right points, which correspond to binp, binL, and binR. For example, in diagram 610, bins 611, 612, and 613 correspond to binL, binR, and binp, respectively, as shown, bins 611 and 612 are symmetric, and both lower than the peak bin 613. When the time phase is advanced, and therefore, it is suboptimal, however, binL (bin 621 of diagram 620) has a greater magnitude than binR (bin 622). In a similar manner, when the time phase is delayed, it is also suboptimal, as illustrated in diagram 630. Accordingly, in the present invention, binp, binL, and binR are provided by the correlator parallel to the error estimators. 524. Whenever the error estimators 524 detect the advanced or delayed time phase, error signals are generated to adjust the time cycle 532, in order to achieve a correct sample time phase. Accordingly, in the present invention, if the sample time phase is advanced or delayed (not optimized), the error estimators 524 provide error signals to adjust the demodulation parameters associated with the time cycle 532, with the object to maximize binp, and therefore minimize binL and binR, as well as to equal binL and binR. With respect to optimizing derotation, the error estimators 524 compare the received peak bin with an optimal or ideal peak bin (non-rotating), and adjust the CTL 531 cycle as necessary to achieve a more optimal or ideal desrotation. With respect to the optimal gain, the error estimators 524 compare the power of the peak bin with an optimum or ideal power or bin peak gain, and adjust the AGC 533 cycle as necessary to achieve a more optimal or ideal demodulation gain. Referring once again to Figure 3, in the present invention, therefore, the parallel correlator 300 detects the time / carrier insurance, and generates symbols / indices for the error estimators 524. The error estimators 524 use the data generated from the parallel correlator 300 to track / adjust the carrier phase shift, time, and AGC cycles 531, 532, 533. With the implementation of a number of correlators, each operating in parallel, and correlating with a delayed version If the sample is different from the known PN sequence, the profile (correlation bins) of the received symbol can be obtained at the symbol rate. By using parallel correlators, the time error signals necessary for the sample time of the ADC 521, the carrier frequency shifts for the derailor 523, and the AGC feedback for the radio frequency circuit 510 can be obtained, all from the correlation bin (profile) and the indexes of the parallel correlators at the same time. The present invention, therefore, provides a more efficient and accurate method for estimating error signals, while maintaining the de-exhaustion processing gain. In one embodiment, a TDMA system is used, where several transceivers are coupled by means of links to the base station 110, on a single radio frequency channel. The present invention preferably employs a technique for maintaining the existing radiofrequency links, and establishing initial cold-start radiofrequency links on the same radio frequency channel, and for determining whether the communication link is secured at symbol time, tracking of carrier frequency, and the AGC. Referring now to Figure 4, there is shown a flow chart illustrating the method of operation 400 of the receiver of Figure 3, in accordance with one embodiment of the present invention. The method 400 provides a technique for determining whether the communication link has been established in symbol time, carrier tracking, and AGC (whether there is a valid data signal or not), and also to detect the band of guard that indicates the end of the transmission of the data package. The detection of the guard band is important for cold starts and efficient switching between signals, and allows handset synchronization to the base station in the correct TDMA epoch position after cold start. In a TDMA system, where only some headphones are supposed to be "activated" during their respective TDMA time slots, the method 400 of the present invention also allows the appropriate transceiver of the system 100 to be secured. The latter is achieved due to that the transceiver of each headset can detect when there is a valid data signal that is received at the beginning of its time slot, and can also detect the guard band that indicates the end of the transmission of the data packet. This allows the headset to turn off for a specified clock time (usually using a sequence controller timer), and it "wakes up" again at the beginning of its next slot in the TDMA era. Some time before the reception of data, or simultaneously with the initialization, a peak detection threshold has been calculated by restoring the threshold, and calculating an average execution of M (for example 5) peak bins, without counting the band of save (steps 401, 402). Then the threshold is calculated as the running average of the previous M bins, and then it is scaled to be a smaller order of magnitude (step 403). At the beginning of a data slot of a given transceiver, its sequence controller timer wakes up to the transceiver, and starts to initialize; in an alternative way, the initialization may begin when a valid signal is received (steps 401, 402). After initialization and that the peak detection threshold has been determined, method 400 determines whether valid data is being received, or whether the guardband has been found. Whenever valid data is being received, the operation is proceeding in a satisfactory manner. If too many errors are detected, steps are taken to try to secure. If a guardband is detected, the transceiver can activate a countdown timer, and deactivate it to its next data or audio slot. Therefore, after initialization, if the search indicator is not set (looking for the guard band), then the method looks for a period of time for M peaks (steps 441, 442, 444), to determine if they are receiving still valid data. If no valid data is detected before the timeout, then a different carrier, AGC, or channel is tested, and partial initialization is done (steps 444-447), or a cold restart if necessary (step 448). Conveniently, the search for M peaks (or K non-peaks) is made at the symbol rate, because the parallel correlator 300 provides all the correlation bins for each symbol. Therefore, in one embodiment, the method 400 searches for the consistency of the bins peak correlation indexes. In particular, over the span of N symbols, if M bins are found consecutive correlation peak in the same index (if there are M bin consecutive correlation peaks in the same index, for example bin2 is a peak for more than M consecutive times) , you can declare an initial insurance. If M peaks are detected, indicating valid data, then the search indicator is set, and method 400 begins searching for the guardband, searching for each consistent peak bins below the peak detection threshold (steps 442, 443, 441, 421 ). The search for K peak bins below the threshold continues until a timeout is reached, at which point, different positions of carrier, AGC, or channel are tested (steps 422, 445-447). If K peaks are detected, then the guardband is found, and different indicators and counters are reset, because the data packet transmission is ending (steps 421, 423). Then the transceiver enters the standby mode, until its next data packet transmission slot is presented. At the beginning of the next data slot, the transceiver initiates the data reception again, and the parallel correlator 300 provides the binp, binL, and binR data to the error estimators 524, to adjust the demodulation parameters for the 531 cycles, 532, and 533, if necessary. In this step, the CTL cycle 531 corrects the frequency offset, and the time recovery cycle 532 corrects the time errors. Then the AGC 533 cycle tracks any fluctuations in signal strength. In order to obtain the most reliable link insurance, in one embodiment, a coupling of the bit pattern is executed, to couple the decoded bits with some known bit patterns, for example, a data packet header (steps 421). , 423, 424, 427). A false insurance will be asserted if the bit pattern match is not achieved within a previously determined time frame (step 425). At this point, the process begins again by going through the steps of the signal gains (AGC) and the frequencies of local carriers (steps 445-447). If the decoded bit pattern matches one of the stored / known bit patterns, the radio frequency link is established (step 427), and then it is determined where the system is in the TDMA epoch (step 428), and the appropriate counter values, and another TDMA time greeting takes place (step 429). Step 429 leads to processor implementation steps 431 et seq., At which point, CTL cycle 531 corrects frequency offset, and time recovery cycle 532 corrects time errors. Then the AGC 533 cycle tracks any fluctuations in signal strength. Once the error signals are declared valid, the symbol and carrier tracking time cycles 531, 532, which are operating as closed cycles, will correct for any time and frequency offset of the carrier. Of course, correlation peak bins can still be obtained, even when there are some phase shifts in the symbol time and the carrier frequency. In one embodiment, the carrier frequency may be out of phase by as much as 100 KHz, and still produce a peak correlation bin. If peak Mbins consisting of a given index are detected (step 442), a search indicator is set to indicate that a relevant data packet possibly exists (step 443). The search indicator causes a guardband search block to point towards the packet boundary. The threshold value is used from step 403 to decide if a guardband has been found after the search / freeze indicators are set. This value is calculated from the average of the previous peak M bins, and is scaled to be an order of magnitude smaller. Because the system 100 is a TDMA system, a counter can be established, so that it is known when the next possible relevant data packet will arrive. Then the next data packet will be used to generate error signals, and it will be decoded for identification. Once the correct identification (or coupling of the bit pattern) is achieved, the different counters used to point to the correct time slots are established by the microprocessor. If the correct identification can not be obtained, it is asserted a time out, and causes the local carrier frequency (LO) or the AGC cycles, to be scaled up to another value, for example, a step of 50 KHz for LO, and 10 dB for the AGC. The search for M consistent indices and the guard band also restarts. Confidence meters can be integrated into each of these blocks to ensure a high probability of establishing radio frequency links. For example, the search and freeze indicators are established only after we have obtained 3 indexes consistent with consecutive M. It should be noted that the duration for the search of the consistent index should last at least two fields. This is due to the fact that the base station 110 will transmit data every third field, as shown in Figure 2, even when there is no active headset. This operation is orchestrated by a microprocessor, that is, after the initial insurance is obtained, the microprocessor can provide some parameters or values for the coupling operation of the bit pattern. The initial insurance is achieved through the hardware, while the final insurance statement is determined one layer up by the micro. Once the link is established, the different counters used to point to the correct slots and frames in a TDMA system are established, in the correct numbers determined by the micro. In addition to the digital communications of a wireless telephone system as described hereinabove, the present invention is also applicable to BPSK, QPSK, CAP, and QAM, for example, as well as to VSB modulation systems, such as are employed by the Grand Alliance High Definition Television (HDTV) system proposed for use in the United States. . One skilled in the art will recognize that design changes are required to adapt the modulation system of the transmitter disclosed to the desired modulation scheme, and will understand how to design the illustrated components to operate with the desired modulation scheme. One skilled in the art will recognize that the wireless system described above according to the principles of the invention can be a cellular system, wherein the base unit 110 represents a base station serving one of the cells in a cellular telephone network. . It will be understood that those skilled in the art can make different changes in the details, materials, and configurations of the parts that have been described and illustrated above for the purpose of explaining the nature of this invention, without departing from the principle and scope of the invention. , as described in the following claims.

Claims (18)

1. A receiver (500) for receiving a sampled spread spectrum signal comprising successive chips representing successive symbol I, Q pairs generated by multiplying each symbol by a sequence of pseudo-random number (PN) of C-bits, in where each symbol corresponds to C chips, and each chip corresponds to S samples, where C and S are integers, the receiver comprising: (a) at least one of a cycle of automatic gain control (AGC) (533), a carrier tracking cycle (CTL) (531), and a time cycle (532), to demodulate the received extended spectrum signal, characterized in that it further comprises: (b) a parallel correlator (300) comprising: (1) (SxC) I correlators to provide (SxC) I symbols, one for each of the (SxC) I samples, and a corresponding PN bit, and (SxC) Q correlators to provide (SxC) Q symbols, one for each of the (SxC) samples Q, and a corresponding PN bit; (2) (SxC) absolute magnitude adders to receive the corresponding I, Q symbol pairs from the symbols I, Q provided by the I and Q correlators, and to provide a bin for each of these (SxC) symbol pairs I, Q, corresponding to the absolute magnitude of the sum of each respective symbol pair; (3) an element to detect a bin peak (binp) of the (SxC) bins, and to provide the bin peak and two adjacent bins (bin ^, bin x); and (c) an error estimator (524) comprising at least one of: (1) an element to determine if the adjacent bins (binp_lf binp + 1) are unequal, and to adjust the time cycle (532), to equalize the adjacent bins (binp_1, binp + 1); (2) an element to compare the pair of symbols I, Q corresponding to the bin peak (binp), with a pair of optimal non-rotating I, Q symbols, and to adjust the CTL (531), in order to reduce the difference between them, to optimize the desrotation of the signal; and (3) an element for comparing a peak bin power measurement (binp), with an optimal power measurement, and for adjusting the AGC cycle (533), in order to reduce the difference between the power measurement and the measurement of optimal power, in order to optimize the gain of the signal. The receiver of claim 1, wherein the receiver (500 is part of a headset transceiver (121, 122) of a wireless headset (12 OI) of a wireless telephone system (100), comprising the wireless headset , a plurality of other wireless headphones (120) and a base unit (110) having a base transceiver (111, 112) to communicate on a radiofrequency channel with each handset. The receiver of claim 2, wherein each handset comprising a headset transceiver for establishing a time division multiple access link (TDMA) on a channel shared with the base unit by means of the base transceiver, where each handset communicates during a time slot exclusive to a TDMA scheme (200) that allocates time slots (251-253) to the headphones during a time slot unique to a TDMA scheme that allocates time slots to active headphones. The receiver of claim 3, which further comprises a processor having an element for searching a first plurality of consecutive peak bins above a peak threshold, for determining whether the receiver is receiving valid data, and an element for searching a second plurality of consecutive peak bins below the peak threshold, to determine if the receiver is receiving guardband data, wherein the peak threshold is determined by calculating average execution of a third plurality of previous peak bins, and then scale down the average of execution by an order of magnitude. 5. The receiver of claim 4, wherein the receiver initiates a count down timer, and the end of a transmission of a data packet is disabled, and it wakes up in the next time slot for the handset according to the countdown timer down. The receiver of claim 1, wherein each cycle is characterized by one or more respective cycle control parameters, and the error estimator adjusts the cycles by adjusting the respective cycle control parameters for each cycle. 7. The receiver of claim 1, wherein C = 16 and S =

2. The receiver of claim 1, wherein each of the correlators I and Q comprises a multiplier node for multiplying the sample I or Q for that correlator, respectively, by the corresponding PN bit. The receiver of claim 1, wherein each cycle is characterized by one or more parameters of the demodulation cycle, and the error estimator adjusts the cycles by adjusting the demodulation parameters according to the error signals developed from according to the peak bin and the at least two adjacent bins. The receiver of claim 1, which further comprises a processor having an element for searching a first plurality of consecutive peak bins above a peak threshold, in order to determine if the receiver is receiving valid data, wherein the The peak threshold is determined by calculating the average execution of a second plurality of previous peak bins, and then the execution average is scaled down by an order of magnitude. The receiver of claim 10, the processor further comprising an element for searching a third plurality of consecutive peak bins below the peak threshold to determine if the receiver is receiving data from the guardband. 12. The receiver of claim 1, wherein the error estimator comprises elements (c) (1), (c) (2), and (c) (3). 1

3. In a receiver (500), a method for receiving an extended-spectrum signal comprising successive chips representing successive symbol I, Q pairs generated by multiplying each symbol by a sequence of pseudo-random number (PN) of C-bits, where each symbol corresponds to C chips, and each chip corresponds to S samples, where C and S are estuaries, the method comprising the steps of: (a) demodulating the received extended-spectrum signal with at least one of an automatic gain control (AGC) cycle (533), a carrier tracking cycle (CTL) (531), and a time cycle 532, characterized by a receiver comprising a parallel correlator comprising: (1) (SxC) correlators I and (SxC) correlators Q, (2) (SxC) adders of absolute magnitude, and a search and decision unit with confidence; (b) provide, with the (SxC) correlators I, the (SxC) symbols I, one for each of the (SxC) samples I, and a corresponding PN bit, and provide, with the (SxC) Q correlators, the (SxC) Q symbols, one for each of the (SxC) Q samples, and a corresponding PN bit; (c) receiving, with the (SxC) adders of absolute magnitude, the corresponding pairs of symbols I, Q from the symbols I, Q provided by the I and Q correlators, and providing a bin for each of the (SxC) ) pairs of symbols I, Q corresponding to the absolute magnitude of the sum of each respective symbol pair. (d) detect, with the search and decision unit with confidence, a bin peak (binp) of the (SxC) bins, and provide the bin peak and two adjacent bin (bin 1 # binp + 1); And (e) perform, with an error estimator (524), at least one of the following steps: (1) determine if the adjacent bins (bin ^, binp + 1) are unequal, and adjust the time cycle (532) ) to equalize the adjacent bins (bin ^, binp + 1); (2) compare the pair of symbols I, Q corresponding to the peak bin (binp) with a pair of optimal non-rotating I, Q symbols, and adjust the CTL (531) to reduce the difference between them, in order to optimize the desrotation of the signal; (3) compare a peak bin power measurement (binp) with an optimal power measurement, and adjust the AGC cycle (533), in order to reduce the difference between the power measurement and the optimal power measurement, in order to optimize the gain of the signal. The method of claim 13, wherein step (e) comprises steps (e) (1), (e) (2), and (e) (3). 15. A wireless telephone system (100), which comprises: (a) a base unit (110) having a base transceiver (111, 112); and (b) a plurality of wireless headphones (120), each headset (12OI) comprising a headset transceiver (121, 122), to establish a wireless extended spectrum TDMA link on a channel shared with the base unit by means of the base transceiver, in accordance with a TDMA scheme (200), wherein each handset (12OÍ) communicates during a unique time slot (251-253) of a TDMA time (200), wherein the link comprises a sampled extended spectrum signal comprising successive chips representing pairs of successive I, Q symbols generated by multiplying each symbol by a sequence of pseudo-random number (PN) of C-bits, where each symbol corresponds to C chips, and each chip corresponds to S samples, where C and S are integers, each receiver transceiver comprising a receiver having: (1) at least one of an automatic gain control (AGC) cycle (533), a carrier tracking cycle (CTL) (531), and a Time cycle (532), pair to demodulate the received extended spectrum signal, characterized by: (2) a parallel correlator (300) comprising: (i) (SxC) I correlators to provide (SxC) symbols I, one for each of the (SxC) samples I, and a corresponding PN bit, and (SxC) Q correlators to provide (SxC) Q symbols, one for each of the (SxC) samples Q, and a corresponding PN bit; (ii) (SxC) absolute magnitude adders to receive the corresponding I, Q symbol pairs from the symbols I, Q provided by the I and Q correlators, and to provide a bin for each of the (SxC) pairs of symbols I, Q, corresponding to the absolute magnitude of the sum of each respective symbol pair; (iii) an element to detect a bin peak (binp) of the (SxC) bins, and to provide the bin peak and two adjacent bins (bin ^, binp + 1); and (3) an error estimator (524) comprising at least one of: (i) an element to determine whether the adjacent bins (bin ^, binp + 1) are unequal, and to adjust the time cycle (532) , in order to equalize the adjacent bins (bin ^, binp + 1); (ii) an element to compare the symbol pair I, Q corresponding to the peak bin (binp), with a pair of optimal non-rotating I, Q symbols, and to adjust the CTL (531), in order to reduce the difference between them, to optimize the desrotation of the signal; and (iii) an element for comparing a peak bin power measurement (binp), with an optimal power measurement, and for adjusting the AGC cycle (533), in order to reduce the difference between the power measurement and the measurement of optimal power, in order to optimize the gain of the signal. 16. The system of claim 15, wherein the error estimator comprises elements (3) (i), (3) (ii), and (3) (iii). 17. A receiver (500) for receiving a sampled spread spectrum signal comprising successive chips representing successive symbol I, Q pairs generated by multiplying each symbol by a sequence of pseudo-random (PN) numbers of C-bits , where each symbol corresponds to C chips, and each chip corresponds S samples, where C and S are integers, the receiver comprising: (a) at least one of a cycle of automatic gain control (AGC) (533), a carrier tracking cycle (CTL) (531), and a time cycle (532), for demodulating the received spread spectrum signal, characterized by: (b) a parallel correlator (300) comprising: (1) (SxC) I correlators to provide (SxC) symbols I, one for each of the (SxC) samples I, and a corresponding PN bit, and (SxC) Q correlators to provide (SxC) Q symbols, one for each of the (SxC) Q samples, and one PN bit correspondent; (2) (SxC) absolute magnitude adders to receive the corresponding I, Q symbol pairs from the symbols I, Q provided by the I and Q correlators, and to provide a bin for each of the (SxC) pairs of symbols I, Q, corresponding to the absolute magnitude of the sum of each respective symbol pair; (3) an element to detect a bin peak (binp) of the (SxC) bins, and to provide the bin peak and two adjacent bin (bin ^, binp + 1); and (c) an error estimator (524) comprising an element to determine if the adjacent bins (bin ^, binp + 1) are unequal, and to adjust the time cycle (532), in order to equalize the bins adjacent (bin ^, binp + 1). The receiver of claim 17, wherein the error estimator further comprises: an element for comparing the symbol pair I, Q corresponding to the peak bin (binp) with an optimal non-rotating symbol pair I, Q, and for adjust the CTL (531) in order to reduce the difference between them, to optimize the desrotation of the signal; and an element to compare a peak bin power measurement (binp) with an optimal power measurement, and to adjust the AGC cycle (533), to reduce the difference between the power measurement and the optimal power measurement, with In order to optimize the gain of the signal.